Cultural Models in Language and Thought

THE WELL-PROTECTED DOMAINS Ideology and the Legitimation of Power in the Ottoman Empire 1876-1909 Selim Deringil

LB.TAURIS Publishers LONDON-NEW YORK

This edition digitally reprinted in 2004 by Bookchase Printed and bound in the EU L.D.: SE-3846-2004 in Spain Paperback edition published in 1999 by I.B.Tauris & Co Ltd 6 Salem Road London W2 4BU www.ibtauris.com 175 Fifth Avenue New York NY 10010 In the United States of America and in Canada distributed by Palgrave MacmUlan, a divison of St. Martin's Press 175 Fifth Avenue, New York NY 10010 Copyright © 1998 Selim Deringil All rights reserved. Except for brief quotations in a review, this book or any part thereof, may not be reproduced in any form without permission in writing from the publisher. A full CIP record for this book is available from the British Library A full CIP record for this book is available from the Library of Congress ISBN 1 86064 472 4 Set in Adobe Jenson by Hepton Books, Oxford

Dedicated to my father Efdal Deringil

For God's sake, let us sit on the ground, And tell sad stories of the death of Kings: How some have been depos'd; some slain in war; Some haunted by the ghosts that have depos'd; Some poison'd by their wives; some sleeping kill'd King Richard II, Act III, Scene II

You may attack your enemy, you need not think whether you are right or wrong, sooner or later your judicial staff will find the best pretexts to prove you right. Nicolo Machiavelli, The Prince

The Sublime State finds itself stuck among Christian states ... Even if our material and diplomatic resources were unlimited, we would still be obliged to seek the aid of some of the Great Powers ... However, all aid among states is based upon mutual suspicion and calculated interest,.. Said Pasa, Memorandum prepared for the Sultan

Contents

Acknowledgements Map Introduction 1

x xii 1

'Long Live the Sultan!': Symbolism and Power in the Hamidian Regime

16

2

The Ottomani^ation of the Seriat

44

3

'To Enjoin the Good and to Forbid Evil': Conversion and Ideological Reinforcement

68

4

Education: the Answer to all Evil?

93

5

'They Confuse and Excite Minds': The Missionary Problem.

112

6

Ottoman Image Management and Damage Control

135

7

The Ottoman'Self Portrait'

150

8

Conclusion

166

Notes Bibliography Index

111 237 250

Acknowledgements

I owe thanks to a great many people for the help, advice, criticism, encouragement, and above all, the patience, that they were good enough to send my way. First and foremost thanks go to my colleague Selcuk Esenbel, who served as my 'inspiring muse', taskmaster, and first reader without whose help this book would simply not have been written. Similarly, my heartfelt thanks go to Daniel Goffman for his willingness to read drafts and provide incisive criticism, and to Edhern Eldem for making sure that I did not take myself too seriously. Ariel Salzmann also provided much appreciated criticism and made sure that I got my nose out of the archives and took at least an occasional look at the real world of writing. Caroline Finkel whose companionship in the Ottoman archives was invaluable, was also kind enough to provide logistical support on several visits to London, together with Andrew Finkel. Feroze Yasamee kindly read the manuscript and gave me an extremely thorough appraisal. Faruk Birtek was another friend and colleague who provided me with key insights, particularly in relation to the idea of'fine tuning' in imperial structures. I also owe agreat deal to Dru Gladney for providing me with an anthropological view of historical and other matters. Discussions with Professor $erif Mardin, and his constant encouragement of my work, have been invaluable and for them my sincerest gratitude is in order. Another colleague who has had a formative influence on me is Engin Akarh who encouraged me to learn Ottoman, and through whose work my interest in the Hamidian era begam I must also thank Professors Carl Brown, Norman Itzkowitz, and Avram Udovich for their support during my stay at Princeton University. In the same context owe thanks to Fikret Adanir for his efforts as coorganizer of a workshop on the Hamidian era, and for hosting me at Bochum. I have also profited greatly from discussions with the following: Mehmet Gene, liber Ortayli, $evket Pamuk, Niikhet Sirman, Nilufer Gole, Kemal Beydilli, Ya$ar Ocak, Ekmeleddin Ihsanoglu, Zafer Toprak, Butrus Abu Manneh, Hakan Erdem, Cem Behar, Philip Mansel, Giilru

x

Acknowledgements

xi

Necipoglu, Nevra Necipoglu, Cemal Kafadar, Alan Duben and Giiltekin Yikhz. Also, I am greatly in debt to my late father Efdal Deringil, to whom this book is dedicated, for his patient tutoring in my early days of learning the Ottoman script. Mehmet Gene;, Idris Bostan, Nejat Goyiinc. and Hayri Mutlucag, were more than generous with their time in helping me decipher documents in the archives. I thank them all for their patience. As someone who is almost computer-illiterate I must express my thanks to Giilen Aktas and Aydin Akkaya for the many hours of time they kindlv gave, as well as Nadir Ozbek and Emre Yalcjn. Giilen Aktas also drew the map. To Hatice Aynur I owe thanks for her help in correcting the transcription of Ottoman material in this paperback edition.'For the cover photograph I am indebted to Roni Margulies who gave me the run of his exquisite collection of photographs from the Ottoman era. Also thanks are due to Anna Enayat, my editor, and all those in I.B.Tauris who spent time and effort on the manuscript. O n the material side I owe gratitude to the Fulbright Commission of Turkey for providing me with support during my sabbatical year at Princeton, to the Near Eastern Studies Department of Princeton University for hosting me, and to the staff of Firestone and Gest Libraries. I also owe thanks to Francois Georgeon and Paul Dumont as well as the Ecole des Hautes Etudes en Sciences Sociales for inviting me to give a series of lectures as visiting professor which enabled me to make use of the Biblioteque Nationale. Also the staff of the Basbakamk Archives were very helpful. All that said, it only remains to state clearly that all mistakes, omissions, oversights or exaggerations are entirely my own.

The Provinces of the Ottoman Empire in the fate Nineteenth Century

Introduction

This is a book about a world that is no more. It looks at the last years of the Ottoman Empire and the reign of Sultan Abdiilhamid II (1876-1909), a period which also saw the dawn of the modern Balkans and the Middle East. The Ottoman state was unique in many ways. It was the only Muslim great power. It was the only European Muslim power. It had emerged as the single most setious threat to European Christendom during the period when Europe was expanding as a result of the voyages of discovery and colonization in the late fifteenth and sixteenth centuries. Just as the 'new monarchies' in the emerging European state system were beginning to flex their muscles worldwide, they found themselves mortally threatened by an aggressive and successful power in their very own 'back garden'. Moreover it was a power that vaunted its superiority and seemed poised to conquer the First Rome, just as it had conquered the Second Rome in 1453. The Ottoman armies established a foothold in Otranto in 1479 causing the Pope to flee Rome. 1 Some historians have actually argued that the whole concept of 'Europe' was fashioned around the decline of the Ottoman threat. 2 Even after the days of its greatness were long over, the Sublime State continued to exist as a thorn in the side of a Europe which expected its demise at any moment, and stubbornly refused to die. In the heyday of great power imperialism, at the turn of the nineteenth century, the Ottoman Empire had long been crumbling though, unlike all other ancient empires, it remained a military force strong enough to give the armies of the great powers a distinctly hard time. 3 Even in defeat, it was 'able to compete at a technical level with its European neighbours: Plevna was not a Tel el-Kebir or Omdurman."' This state of affairs has always been treated as something of an anomaly in European culture. It was not, quite simply, supposed to happen. H a d not the Great Mughal ultimately succumbed to the technical and financial power of the gentlemen with the powdered wigs? Had not a handful of conquistadores completely overawed and overcome a native American

1

2

Tfee WelUProtected Domains

population which had been terrified by cannon and called the horses and their riders, 'Gods with six legs and two heads? Had not the mighty Z u lus, the Chinese Son of Heaven, and the peoples of Micronesia suffered the same fate? It was, after all, 'natural' that by the nineteenth century the peoples of the world should be divided into 'ruling' and 'subject' peoples. In Europe, the former were all Christian, all except one. Only the Sublime Porte ruled millions of Christians as a Muslim empire, As the centuries wore on, this situation became more and more galling: Unfortunately for the peace of mankind, it has happened that the Turk is placed in a position where it is impossible to ignore him, and almost equally impossible to endure him; while by his origin, habits, and religion, he is an Asiatic of Asiatics, he is by irony of fate established in a position where his presence is a ceaseless cause of misery ... to millions of Christian people.5 In striking contrast, Said Pasa, who served nine times as Abdulhamid's grand vizier, wrote: 'We find ourselves stuck among Christian princedoms and states ,., The Sublime State has been a member of the European family of states for some fifty years ...' 6 This study is an attempt to put the late Ottoman state in the context of world changes and, as such, is a deliberate attempt to look at the world from Istanbul. I am interested in how the late Ottoman elite reacted to the world around them. 7 1 think it is about time that these men be rescued both from their Kemalist denigrators and their 'fans' on the lunatic fringe of the Turkish Right, as well as from Western perpetrators of the image of the 'Terrible Turk'. Yet, I also hope to 'abandon wishful thinking, assimilate bad news, discard pleasing interpretations that cannot pass elementary tests of evidence and logic ,..' 8 It is not my least intention, for example, to whitewash the Armenian massacres of the 1890s, although the book touches on this issue only tangentially.9 Nor is this a book on legal history; it does not intend to ask whether Weberian concepts of legitimation can be fitted to the Ottoman case.10 It is a book about applied legitimation policies. The emphasis will therefore be on giving a 'voice' to the late Ottoman elite. W h a t was the world view of the Ottoman elite? How did they see the problems of their state and society, and what solutions did they propose? The first thing to be noted is that, like elites everywhere, the Ottoman service elite was not monolithic. Those who served were very differentiated internally when it came to cliques, coteries, cabals, and corruption. Some were more conservative, others more pro-

Introduction

3

.gressive, although hard and fast categories and facile labelling have led to much historically inaccurate stereotyping.11 In this context, Engin Akarii's -words could not be more apt:' [We must] make a conscious effort to engage the voices preserved in the historical data in a genuine conversation, remaining critical of what we hear but also critical of our own position.'12 It may seem strange to the reader to find a discussion in the same book of, say, the Ottoman protest at the Egyptian delegation being given precedence over the Ottomans in the protocol governing Queen Victoria's jubilee celebrations, and an expeditionary force sent into the mountains .of northern Iraq to convert the Yezidi Kurds to Hanefi Islam. Yet both incidents and the Ottoman reaction to them stemmed from the same root cause: the Ottoman elite understood only too well that their world was exposed to mortal danger from within as from without, from the corridors of Windsor Castle to the mountains of Iraq. This obliged them to devise policies which would legitimate their position in the eyes of both their own people, and the outside world. This book is a study of what Carol Gluck has called 'the grammar of ideology by which hegemony was expressed' in a late Ottoman context.13 It is my contention that the Hamidian period represents a critical point in time when, contrary to the standard view that the Ottoman Empire closed in on itself during this period, the only Muslim world empire did in fact succeed in joining the modern community of nations, albeit as a grudgingly accepted poor relation. The period also represents Turkey's response to the challenge to its existence from within and without, a challenge that was also met by other imperial systems, Russian, Austrian, German, and Japanese. In more ways than one, the Ottoman Empire falls between the cracks of the fault lines that determined world politics in the nineteenth century, and this continues to determine the way the history of imperialism is written today. O n the one hand, even at the turn of the century, it was still a force to be reckoned with and, unlike the Indian princely states or African colonized peoples, it was one which could not be pushed aside. O n the other hand, it was the 'Sick man of Europe' whose demise was expected at . any moment. It was heavily penetrated by Western economic interests, and suffered from chronic financial crises. It continued to lose territory in Europe, yet by the mid-nineteenth century it had the third largest navy in the world, and in the last quarter of the century, much to the alarm of the British, it actually succeeded in extending its influence along the Arabian Gulf.14 The central point about this was that the sultan and his staff at the

4

The Well-Protected

Domains

Sublime P o r t e controlled their o w n fate. O p e r a t i n g under severe constraints, to be sure, they were nonetheless able to carve out a critical space for manoeuvre in an increasingly hostile environment. T h i s p r o d u c e d t h e basic dynamic which d e t e r m i n e d the relationship between the O t t o m a n statesman a n d his W e s t e r n colleague. 15

In pursuit o f t h e exotic bird, o r looking beyond t h e turban T h e ' a n o m a l y ' of t h e O t t o m a n p o s i t i o n h a s b e e n reflected in t h e historiography pertaining to t h e O t t o m a n state which has been characterized by a tendency to 'exoticize' this uncomfortable p h e n o m e n o n . T h e singularity of t h e O t t o m a n polity in t e r m s of world power balances h a s led W e s t e r n h i s t o r i a n s to t e r m t h e state a 'slave synarchy'. 1 6 C a r o l i n e Finkel's w a r n i n g against the seduction a n d 'resilience' of this simplistic model, which focuses o n t h e 'static facade' a n d ignores comparable dynamics, is well taken, 1 7 Approaches which find it o p p o r t u n e t o dwell o n the 'peculiar' features of O t t o m a n society overlook t h e civil dimension, ignoring profit motives or any of t h e other impulses t h a t drove people in t h e 'normal' (that is Christian) world. 1 8 W h e r e the 'Turks' dwell is described as, at worst, a bastion of bloodthirsty tyrants or, at best, a decadent fleshpot of O r i e n t a l vice. T h e O t t o m a n polity is t h e alter ego, the ultimate 'other'. Yet it is fascinating to read t h a t n o less an observer than Voltaire h a d a m u c h m o r e realistic assessment of O t t o m a n state a n d society:

This government which has been depicted as so despotic, so arbitrary, seems never to have been such since the reigns of Mahomet II, Soliman I or Selim II, who made all bend to their will.., You see in 1703 that the padishah Mustafa II is legally deposed by the militia and by the citizens of Constantinople. N o r is one of his children chosen to succeed him, but his brother Achmet III. This emperor in turn is condemned in 1730 by the janissaries and the people ... So much for these monarchs who are so absolute ! One imagines that a man may legally be the arbitrary master of a large part of the world, because he may with impunity commit a few crimes in his household, or order the murder of a few slaves; but he cannot persecute his nation, and is often the oppressed rather than the oppressor."

Interestingly the obsession with t h e exotic is not unique to W e s t e r n

Introduction

5

historians. Turkish historians of remarkably recent vintage have also seized upon it to explain away the Ottoman past. Among the Left, who blame the ancien regime for Turkey's underdevelopment, it is fashionable to depict the past as a period when 'pa}as ate magnificent banquets while the people languished on olives'. Taner Timur is actually on record as having said: 'The sad thing is that these politicians did not themselves believe in the "reforms" they had to carry out.' Timur's central argument was that Ottoman statesmen like Ali, Fuad or Resid Pasa acted solely to protect their class interests and then only under severe Western pressure. Their reforms were not 'true' reforms because they were not inspired by European 'Enlightenment thinking'. 'The Ottoman Empire was a different civilisation with its unique characteristics.' Not even Oriental but, 'stagnated', it 'became "Orientalised"' under the unyielding influence of 'the scholastic world view'.20 The Turkish Right, in contrast to the positions held by the Kemalists and the Left, basks in self-adulation and cherishes the illusion that the Ottoman state was somehow sui generis and cannot be compared to anyother polity. Faith in Islam and the organizational genius of the 'Turk' are self-completing and self-evident. These find a particular focus around the personality of Sultan Abdiilhamid II, who has become a political symbol of the Islamic Right. As a writer of one of the 'classics' of this genre describes it: It is high time to declare that Abdiilhamid II is a great saviour whose true nature as the essence and foundation of the Turk has been blackened by certain usurpers ... [He is] the last stand against the unthinking aping of the West, [a ruler] who remained loyal to the root sources of the Turk's spirit.21 O t h e r writers of the same school focus on various aspects of Abdulhamid's reign to claim that he was 'courageous', 'a great diplomat', a 'committed patron of the arts', even liked sports' and'was an accomplished carpenter'. Personal and political issues are seriously confounded in this sentimental discourse.22 The deeply divergent perspectives on Abdiilhamid held by Turkish 'Islamists' and 'Secularists' is but one instance of the 'deep emotional divide [that] exists between these two perspectives. Each camp tends to conceive the other as a representative of a hostile world view, as if they live in separate worlds,'23 Among modern Western historians of the Ottoman Empire' the spectrum of views is somewhat more varied while still reflecting broadly the

6

The Well-Protected Domains

same tendencies. The Ottoman Empire is generally seen as a supine entity which 'constantly had things done to it'. Such a perspective, of course, grows out of the approach that could be termed the 'Eastern Question* paradigm which is itself a legacy of the nineteenth century.24 As Hobsbawm points out'... By the standards of nineteenth century liberalism, [the Ottoman state was] anomalous, obsolete, and doomed by history and progress. The Ottoman Empite was the most obvious evolutionary fossil ...'2S The tendency among the modern heirs of the 'Eastern Question school' has been to see the Ottoman state as an entity which only figured in European calculations as a factor in the 'balance of power'. Indeed the cast of its work is perhaps best exemplified by the fact that entire articles or books continue to be written on 'The Turks' with minimal references to works or to sources in Turkish.26 But the 'Eastern Question school' is at least the result of a bona fide scholarly tradition, if a somewhat outmoded one. At the lower end of the scale Western literature on the Ottoman state abounds with statements such as: 'Muslims do not care too much for the sea'; 'Moslems were virtually incapable of acquiring European languages'; 'the Janissaries were sheep dogs who had to be fed regularly on Christian meat.'27 Or gems such as: 'Until 1918 the Arabian peninsula was ruled by the Ottoman Empire, so called because it had the same amount of intelligence and energy as a footstool.'28 Or, in a more toned down vein we are told that, 'Attempts to attract Moslem students to the new Western-model schools such as Galatasaray proved a failure throughout the 19th century.' In actual fact Galatasaray produced generations of Muslims who became leading figures. In the same source, we read: '[In Islamic polities] capitals have frequently shifted. Even when a single city has remained the capital for a long time, its segmented nature and frequent additions of new palace and garrison quarters dilute its potential for communicating symbols of identity.' It remains unclear why more than one magnificent city should not be a factor reinforcing rather than diluting identity. Where previously writers of general history books used simply to ignore the Ottoman Empire, the recent 'non Eurocentric version' of history aims to redress the balance and in the process lets in gross errors and oversimplifications.29 The harem becomes a major factor in Ottoman history and the Turks are once again portrayed as the 'uncivilized' threat to Europe. Clearly, the earlier versions of Western historiography were infinitely preferable. But even at the upper end of the spectrum, in works written by schol-

Introduction

7

ars who have used Ottoman primary sources, the 'exotic' yet again becomes the focus. Even Carter Findiey's pioneering work on reform in the late Ottoman bureaucracy has as a central figure a dervish-official who inhabits an (unexplained) 'tangled magic garden'.30 The dervish-official Asci Dede is understood to be in some sense oriental because although he takes an interest in the Russo-Japanese War and lives dangerously, taking steamers and riding trams, he believes in oracles and has complete faith in the guidance of his spiritual mentor Sheikh Fehmi. Also, we are told, an Ottoman official in the late nineteenth century inhabited a 'culturally schizoid society', presumably because he had been exposed to a Western education that he really did not (could not?) understand. Abdulhamid's obsession with his personal security automatically labels him as 'paranoid' and the state in his reign as a 'police state'.31 In Findley, the quest for the exotic bird seems to have actually achieved its object. Ottoman diplomats, although they have come a long way, are still a crowd of buffoons because they have to comply with the ruler's whim to procure an 'all-white talking parrot' for the palace menagerie.32 This observation completely ignores the fact that royal menageries were at the time a fashion indulged in by most of Europe's royalty, from Prince Albert to the lowliest princeling in Germany. The London embassy also indulges in suspiciously oriental activities such as keeping track of Ottoman subjects who are known to be opponents of the regime." In a journal article Masami Arai makes the statement: 'From the beginning of the nineteenth century.., most of the Ottoman reformers seem to have intended, consciously or unconsciously, _to_cmi^n^t_a_natipn^st_at,e from the various subjects of their empire.' How one sets about constructing a nation state unconsciously remains a mystery.34 Aral then contradicts his own point: 'The desire to construct a progressive nation-state in the Orient was so keen that no Ottoman intellectuals, whatever ideal they cherished - Westemist, nationalist, and even Islamist, could repress it.'35 Why anyone would want to repress something they wanted so badly is explicable only in Freudian terms. It is interesting to note that some contemporary Western writers who actually lived in the Ottoman lands were aware of the futility of seeking the exotic. In this context, it is worth quoting at length from Leon Cahun: Ask Any random would be traveller in the Orient what he would like to see: he will tell you that he would want to see Jerusalem, Damascus, the cedars of the Lebanon, and if he is ambitious, Baghdad, the City of the Thousand and One Nights. He would also desire to see a serail, some odalisques, a caravan,

8

The Well-Protected

Domains

palm trees, and whirling dervishes.36 Cahun then proceeds deliberately to disabuse the potential traveller of illusions. Jerusalem, he points out, is simply the equivalent of Lourdes. The serail these days amounts to nothing more romantic than 'un batiment de prefecture', odatek'sjid not odalisque are the equivalent of 'femmes de chambre'. There are no more palm trees in northern Anatolia than there are olive groves in Brittany and Normandy, and 'the whirling dervishes of jthe Orient are neither more numerous nor more interesting than the white pilgrims of France,' As for more stereotypes: 'My good potential traveller will not fail to believe that the Christians in the Orient are called rayas and are driven by courbashe wielding pashas.' Cahun points out that the term reaya simply means taxpayer and has been adapted by the British in India as ryot. As for the 'Turkish' character of the empire: The first time I encountered an Osmanh officer ... I told him he looked typically Turkish. He flushed a deep red and heatedly told me that he was from Damascus as were his father and his grandfather before him, and that he had nothing in common with the pouchts from Istanbul. This Arab particularism did not prevent him from being very dedicated to the flag, or drinking the health of the sultan.37 If one attempts to look beyond the turban, as Cahun does, and examine what actually drove Ottoman decision makers in the nineteenth century, it becomes evident that these people were not necessarily unique or 'exotic', In fact, the broad sweep of change that was taking place in European societies as a whole at the same period was to find it paralleled in Ottoman society.38

'Fine tuning' and the legitimacy crisis of the Ottoman state In the history of states there occut periods of crisis during which the es-'' tablished relationship between monarch and people collapses. In Petrine Russia this occured when Peter the Great adopted the role of'Imperator'. J9 • In the Ottoman Empire, this crisis began with Mahmud II (r. 1808-39) continued under his successors and culminated in the Hamidian era.40 The standard wisdom about the decline of the Ottoman state has been that it was a unilinear process which began sometime after the second

JX

'•••*

Introduction

9

.\Cf' '"'

¥

,/\- siege of Vienna in 1683 and lasted until the collapse of the empire in 1918. A'_ Yet this teieological approach is outmoded. There now seems little doubt ' that the empire was actually much more a part of world trends in the seventeenth and eighteenth centuries than has been hitherto acknowledged.41 I would, however, take issue with Salzmann's view that the 'Ottoman state became a victim of its own policies of centralization' 42 1 would likewise entirely disagree with Todorova's rather startling statement: 'That the Ottoman Empire did not create an integrated society is beyond doubt; what some Balkan historians seem unable to understand is t h a t . this empire did not strive to achieve such integration.'4* The Ottoman state was better administered and more powerful after the reforms of the mid-nineteenth century than in the late eighteenth century, and 'the nineteenth-century Ottoman reformer was more conscious [of his mission] than the eighteenth century reformer, at least he was in more of a rush.'44 N ry Nor would it be an exaggeration to say that the modern state as it is ,j understood today - meaning mass schooling, a postal service, railways, ^ \ lighthouses, clock towers, lifeboats, museums, censuses and birth certifi.cates, passports, as well as parliaments, bureaucracies and armies - was lonly constituted in the Ottoman Empire after the Tanzimat reforms of 1839.45 Although Abou El Haj is quite correct to point out that the Tanzimat era is in many ways the outcome of a much longer prehistory than is commonly assumed, I believe he leans too far the other way in minimizing the qualitative changes that took place during this period.46 Concepts such as the legal equality of Muslim and non-Muslim, the rule of law, the state's guarantee to safeguard the lives, property and honour of its subjects, and above all the very fact that the state was publicly undertaking to honour these promises, were a new departure. Yet even while setting but the reasons for the new laws, the Imperial Rescript of the Rose Chamber (Hatt-t §trif-i Giilhane, November 31839) was quite deliberately based on religious dogma, stating as its first principle that, 'it is evident that countries not governed by the §eriat cannot prevail', even though much of what it decreed was indeed in contravention of the $eriat.47 However, just as the state was permeating levels of society it had never \ [ reached before, making unprecedented demands on its people, it created AL | new strains on society, leading to what Jiirgen Habermas has called a 1e\ gitimacy crisis' or 'legitimation deficit'.48 Nor was this legitimacy crisis Iconfined to the relationship of the Ottoman centre with its own society, In the international arena also, the Ottomans found themselves increasingly obliged to assert and reassert their legitimate right to existence as a

10

The Well-Protected Domains

recognized member of the Concert of Europe, as recognised after the Treaty of Paris which ended the Crimean War in 1856.49 In a context of military weakness, diplomacy acquired vital importance, as did the process I shall call 'fine tunirig' as regards the population of the empire.30 Fine \ tuning involved the meticulous inculcation, indoctrination, enticing, frightening, flattering, forbidding, permitting, punishing or rewarding - all in \ precise doses - which had been the stuff of empire since Caeser crossed the limes. In empires such as the late Ottoman, when the state exists in an almost constant state of crisis or emergency, fine tuning acquires/additional importance but also becomes mote difficult. I would even, venture to say that fine tuning is more the characteristic of a state which is constantly on the defensive.51 Not necessarily humane and anodine, it can involve brute force and bloodshed, but only as a last resort. The most important aspect of fine tuning is that it is a process through which the legitimation ideology of the state is promoted and state policy is imposed on society. Legitimation involves the process of making state policy and resembles, 'That which is inscribed into the nature of things and is perennial,' as Ernest Gellner puts it." This clearly is the ideal state of things. The spectrum between the ideal and the actual is the reality as reflected in the historical record. The creation of the 'perennial' and 'natural' state of things which fine tuning sought to accomplish was to be achieved through a reiteration of certain basic formulae, and their incarnation as the official mythology, or mythomotwr, of the late Ottoman state.53 Fine tuning was concerned in the first degree with the power elite, the men who formulated and applied policy. Even as autocratic a sultan as Abdiilhamid II, who was in effect the last real sultan of the empire, had to rely on a staff who fed him information, advised him, indeed influenced him. So the so-called 'Red Sultan' or 'Oriental Despot' of legend, who rarely left his palace, and never left his capital, depended on these men, perhaps more than some of the previous sultans who had actually delegated authority to powerful grand viziers.54 In all societies the legitimating ideology of the state is in many respects the 'no mans land' where a tacit process of bargaining takes place between the state and its people. As Abou El-Haj puts it: 'The credibility if not the very effectiveness of ideology is based in part on the opening up of a meaningful discourse in whose initiation and validation the rulers and at a minimum some if not the majority of the ruled participate.'55 In periods of legitimation crisis this process of bargaining is intensified. The Ottoman Empire was no exception, and the thirty-three years during which

Introduction

11

Abdiilhamid II reigned (1876-1909) are critical because they are years whk|^a.rje_J?oth.Jormative^ destructive. They are formative because of the long-term implications of the various policies ranging from education, to railroads and military.r.efpr.m,.£o.irrigation works, to the first modest beginnings of an industrial infrastructure.56 Yet the same years are disruptive oF much' of the..t.r.adi-, _-,|;V'V) uonal fabrk of society, as the state now came to demand.not passive , -JV 57 "/'•v obedience but conformity to a unilaterally proclaimed normative order. In fact.the Ottoman Empire hedged towards a 'nationally imagined com-_ munity', as Ottoman identity assumed an increasinglyX^tkish character, (even if this identity was. packaged in uniyersalist Islamic term§..5^ Also the redefinition of the basic credos of the state as primarily Islamic implicitly excluded the still sizeable Christian elements who were seen as potentially seditious and subversive. The policies of these years were, therefore, destructive of much of the stability which had allowed the delicate symbiosis of various creeds in less turbulent times.- It must be emphasized here, however, that in the application of various policies the Ottoman centre operated under crippling restraints. The most important of these were lack of financial resources and interference on the part of foreign powers, two increasingly overlapping phenomena. 59

Sources The bulk of this study is based on archival materials found in the Prime Ministry Archives in Istanbul. It has been my aim to understand the mindset of the late Ottoman power elite rather than chronologically study specific events. Therefore the material used is drawn from various collections according to specific themes, such as conversion, control of religious activity in the provinces, surveillance of missionary activity, and so on. I do not pretend in any way to have exhausted the materials on these topics ' and no doubt there is still a mine of information in the Basbakanhk that I have not touched. I have focused mostly, but not exclusively, on the Yildiz collection. The Hamidian. regime was obsessed with information, and as a result the Yddiz Palace Archives occupy a disproportionately large volume among the archival collections of the late empire.60 Many of these have only recently been opened to researchers.61 Some of the major collections that I have drawn on are as follows.

12

The Well-Protected Domains

Yildtz Esas Evrah (YEE)62 This is the main collection of the Yiidiz Palace documents. A noteworthy feature of this collection is that it harbours the papers of most of the high officials who served in the period. These may range from a one-page commentary to extensive reports on specific and general topics. Yildtz Sadaret Hususi Maruzat (Y.A HUS) Correspondence between the palace and the grand vizier's chancery at the Sublime Porte. This collection consists of correspondence on a variety of topics ranging from the status of the yacht of the British consul in Basra (warship or not) to measures to be taken against Armenian partisans in eastern Anatolia. Yildtz Sadaret Resmi Maruzat (Y.A RES) T h e collection is of documents presented by the grand vizier to the palace which provide the background for imperial decrees (trade). Yddtz Mutenevvt Maruzat

(Y.MTV)

Although originally this collection was intended to deal with military matters, it is probably one of the richest collections of the Yildiz archive and includes a wealth of information on a great variety of topics. These include court cases involving religious disputes, tribal uprisings, decorations for foreign dignitaries, police reports on movements of Muslim subjects of foreign powers, etc. Yildtz Perakende Gazeteler Miscellaneous newspaper clippings from foreign and Turkish newspapers. This catalogue comprises thousands of newspaper clippings from the world press dealing with the Ottoman Empire.

Administrative divisions in the Ottoman Empire At this point it would be useful to give some background information on administrative divisions and titles which occur in this book.63 In the postTanzimat period, the administrative divisions of the Ottoman Empire were fundamentally reorganized along the French model of the Code Napoleon. 64 The first Regulation for the Provinces (Vilayet Nizamnamesi) was declared on 7 October 1864. This determined that the old division of

Introduction

13

the provinces according to eyalet would be replaced by the unit of vilayet, clearly modelled on the French departement. This was followed by the General Provincial Regulation (Vilayet-i Umumiye Nizamnamesi) of 1867, and the General Regulation for Provincial Administration (Idare-i Umumiye-i Vilayet Nizamnamesi) of 1871. The administrative divisions would be: vilayet (province), sancak or liva (sub-province), kaza (local district or township), nahiyc (commune) and kur'a or karye (village).65 The vilayet would be governed by a governor (vali). It would also have a provincial administrative council (vilayet idare medht) made up of the vali, leading officials, and representatives of the Muslim and non-Muslim population who assisted the vali in an advisory capacity only. The highest civil authority in the sancak was the mutasamjwho would also be assisted by an administrative council (Uva idare meclisi) reflecting much the same composition as the vilayet. The kaza or liva would be administered by a kaymakam, who would also preside over an advisory council. As centralization increased, authority in the villages went from the imam to the centrally appointed headman (muhtar).66 This structure was to remain largely unaltered during the Hamidian period except for the important fact that the control of the palace over the Ministry of the Interior greatly increased.67

The book This book is made up of two parts. The first deals with the recharged legitimation policies inside the empire, the second with efforts to project the desired image abroad. The first chapter is an attempt to provide an overview of the symbolism of power in the Hamidian period. It focuses on state ceremonies, changes in mosque architecture to suite modern protocol, and the various emblematic manifestations of state power such as coats of arms, decorations, ttc. It also deals with the repeated use of encapsulating phrases and cliches in Ottoman chancery language, which, I argue, are valuable clues helping to understand the world view of Ottoman decision makers. The second chapter will deal with the application of policies which arose as the Hamidian state intensified pressure on its population, thus triggering new tensions which in turn created 'a universal pressure for legitimation in a sphere that was once distinguished precisely for its power of self-legitimation'.68 This brought to the fore a new interpretation of

14

The Well-Protected Domains

religion as the official ideology, in the shape of the 'official faith.',(mezheb-i resmiye), that of the Hanefi school of Islamic jurisprudence. Also the position of the sultan as the Caliph of all Muslims (the claim to worldwide Islamic leadership) and as protector of the holy cities of the Hija2 acquired new content. As will be shown, all these amounted to nothing less than the institution of a secular foundation for state ideology, but through the use of Islamic vocabulary and ideological tools. In previous centuries the sultan's position as caliph had been taken more or less for granted (although there were clearly periods like that of the Ottoman-Safavid wars of the sixteenth century when it was emphasized). Yet by the reign of Abdiilhamid the desire of the state to administer and control with hitherto unprecedented intensity led to a situation where the role of the centre had to be constantly re-defined.69 Chapter three deals with what is perhaps the most striking new departure: the effort made to spread Hanefi orthodoxy through official proselytization. In a self-conscious effort to emulate the activity of the Christian missionaries, the Ottoman centre set out to defend itself by taking the initiative in actively seeking conversions to the Hanefi mezheb. Chapter four deals with the educational policies of the Hamidian state. Unlike the Tanzimat period, when the emphasis had been on elite middle to higher schools, the Hamidian period emphasized primary schooling. Another feature of the times was the increasing emphasis on the 'Islamic' content of the schools curriculum. The documents show us that the Hamidian regime tried to weed out of the educational system what it saw as harmful Western influences. Chapter five focuses on the enemy who was also the example: the missionary. As the nineteenth century neared its end, missionary activity became a major feature of the West's claim to moral as well as physical superiority. Abdulhamid's Ottoman Empire found itself in the forefront of the anti-missionary struggle. Yet, members of the state elite were very aware that the only way to defend themselves against the missionary threat was to adopt similar tactics. Chapter six begins the second part of the book which is devoted to the efforts the Ottomans made to project the desired image abroad: the increasing loneliness of the Empire as the only Muslim great power, and the Ottoman elite's obsessive attempts to defend their image against all slights, insults and slurs. With this aim in mind, the Hamidian regime mobilized what resources it could to counter what it saw as injurious to its prestige in the world press, on the theatre stage or at other public forums.

Introduction

15

Chapter seven deals with the active effort to present the desired image. It was a vital concern of the Sublime Porte to be represented in all congresses and conferences attended by the 'club of civilized powers'. The Porte made a point of being present at all the major world fairs. Reciprocity was acknowledgement of legitimacy. In the Hamidian era, Ottoman statesmen therefore became obsessed with reciprocity which could be procured through representation.70 Abdiilhamid would almost certainly have agreed with Edward Said'that struggle [for empire], is ... not only about soldiers and cannons but also about ideas, about forms, about images and imaginings.'7' T h e Hamidian regime sought to project the image in Europe that 'we are like you'. Just as the kaiser, the Austrian emperor, and the tsar were legitimate autocrats, so was Abdiilhamid. After the defeat of Russia by Japan in 190S, 'when the Sultan's officers congratulated him on the defeat of his old enemy Russia, he replied that he did not by any means consider the result a matter of congratulation, because he and the Czar were the only autocratic monarchs in Europe, and the defeat of the Czar meant a blow to the principle of autocracy.'72

Chapter One Long Live the Sultan! Symbolism and Power in the Hamidian Regime

Yamada Torajiro always made a point of bringing his guests to the ceremony of the Friday prayer, the selamhk. Japanese visitors to Istanbul were a rarity in the 1890s, and he was the city's only long term Japanese resident. On a bright May morning, Yamada and his guests stood spellbound as the Albanian Horseguards trotted up to the palace and took up their stations. The sun glinted on their spears as the band struck up the Hamidiye march. Next came the officers of the Imperial Guard, mounted on splendid Arab horses in their impeccable uniforms, and they too took up their positions inside the Yildiz Palace. Finally the sultan, accompanied by the empress dowager and the reigning empress emerged from the palace in his landau and proceeded to the mosque. As the clear cry of the muezzin sounded the call to prayer, the sultan alighted and all his troops in one voice shouted a loud acclaim. When his guest asked Yamada what they were shouting he replied: 'They shout, "Long Live the Sultan!", just as we shout "Tenno Heika Banzai!" in the presence of our emperor.'1 Ceremony as codified competition between states It was no accident that Yamada and his guests were able to 'read' the ceremony of Friday prayer correctly. The nineteenth century was a period of standardized ceremony, from the Court of St James to the Meiji Palace.2 What the Japanese visitors had witnessed was nothing other than the reinvigoration of the symbolic language of the sultanate and caliphate in a world context where pomp and circumstance had become a form of competition between states. This was particularly important for states like the Ottoman and the Japanese, which were not first-rung powers. As humorously put by John Elliot in his discussion of Spain under Philip IV: I t is

16

Symbolism and Power in the Hamidian Regime

17

as if a form of "Avis Principle" operates in the world of political imagery and propaganda; those who are only second try harder.'3 jThexe...ar.g. indeed many common themeslinking the cult of emperor shared by Ottomans, Austrians, Russians and Japanese.,The Russian tsars from Nicholas I (r. 1825-1855) onwards tried to forge a direct link with their people by using a 'synthesis of Russian myths' in which 'official Orthodoxy serv[ed] purely a state function'.4 This was precisely the way official Islam was seen by Abdiilhamid II. As Nicholas I and Alexander III played up the image of the 'blessed tsarJ Abdiilhamid actually pushed the same title, 'the holy personage', (zat-t akdesA kumayun), further than any of his predecessors. The terms 'holy Russian land' or the 'the holv land Tyrol' found their obvious equivalent in 'the sacred name of the Sublime State', (ism-i kaddes-i devletA aliyye}? It was hoped in both the Russian and Ottoman cases that by forgmgji . link of sacralit^ directly with the people, inconvenient intermediaries like j\\ political parties and parliaments'could be avoided. The same goal is seen in the japan of the Meiji years, particularly in the 1880s and 1890s when there occurred what Carol Gluck has called 'a denaturing of polities', as x ~ Meiji statesmen, all the while preparing for the inevitable emergence of political parties, worked to somehow stigmatize political activity as 'disloyal' to the emperor. 6 The major difference between the Japanese, Russian and Ottoman cases, was that religion played a relatively minor role in the Japanese experience. Yet, in the 'constant repetition' of elements in the 'emperor ideology' {tennosd ideology), the attempt to focus almost reflexive loyalty on the person of the emperor was very similar to the Russian and Ottoman cases.7 In the Austrian example the mythology surrounding the House of Habsburg was the core of the official symbolism of the state, yet it coexisted with a constitutional polity after 1867.8 One of the ways of meeting the challenge of nationalism was by tryingto inculcate a sense of belonging to the 'Imperial Fatherland', particularly by creating an official 'history of the Fatherland' (vaterlandische geschtcbte) to teach the non-Austrian subjects who came to study at the University of Vienna. 9 This is not to deny the differences between the Ottoman polity and its imperial legitimist contemporaries. Russia, Austria, Japan and Prussia/ Germany all had at least a semblance of representational politics by the turn of the century. The Ottoman parliament, though a fairly lively body in its first days of 1876, had to await the end of Hamidian autocracy before it was revived in 1909. A ruler like Abdiilhamid II, who laboured

18

The Well-Protected Domains

under the stigma of the 'Terrible Turk' or the 'Red Sultan', while trying to pose as a modern monarch, suffered the self-imposed handicap of his virtually complete isolation from his own people and the outside world. The following is the story of a desparate rearguard action fought in an effort to overcome this isolation in the international arena.

Levels of meaning in Hamidian ideology As Sultan Abdtilhamid O retreated further and further behind the high walls of the Yddiz Palace he became more and more of a myth. Yet this process of distancing himself from the people created a contradiction at the very core of his conception of state power. On the one hand, the I [Hamidian regime sought to penetrate ever further into the daily life of •k i [Ottoman society, and the Ottoman system had always stressed the perV sonal visibility of the ruler. O n the other hand, the sultan's obsession with his security determined that he was very rarely seen outside the palace walls. This left him open to the criticism, often levelled at him, that he was a 'passive caliph'. The sultan's myth had, as a result, to be 'managed' through a system of symbols which constantly reminded the people of his power and omnipresence. In this context Abdtilhamid II seems to have reverted to the ways of his ancestors before Mahmud II. Rulers like Mahmud, Abdulmecid and Abdiilaziz, played the role of the modern public ruler who went out among his people to give a personal manifestation of state legitimacy. Unlike his immediate predecessors, but somewhat like his ancestors, Abdulhamid's aim was to create 'vibrations of power' without being seen.10 In stark con1 trast to Abdiilaziz, who made the first and only state visit by an Ottoman [sultan to Europe in 1867 to see the World Exposition in Paris, Abduihamid crushed all rumours that he was about to visit Europe.' 1 Communication with his people and the outside world had therefore to be made through a world of symbols. These were based almost entirely ^ on Islamic motifsr.'It was,from Islam that the Muslim Ottomans could • 6 * draw the emotional resonance that could mobilize Jso^h^e_upp_e£_and £ j-p", lower classes. It was Islam that wouH plroviHe the store of symbols which [could compete with the national symbols of the Greeks and the Serbs.'12 t\ iff .^ 1 -C In this process of competition through symbols, it was critical that the 'middle of the message' should get through to the population at large. In this context, Carol Gluck's study of a very similar effort in Meiji japan is

Symbolism and Power in the Hamidtan Regime

19

worth quoting at length: Three kinds of interactions can be identified in the process that produced a universe of shared significance from diverse ideological formulations. The first emerged from the stressed parts of ideological speech, what is called the 'middle of the message'; the second from the unstressed elements that often appeared as 'dependent clauses' of ideological utterance; and the third from rhe unarticulated element, identified as the 'deep social meanings' that made ideological discourse comprehensible to those who participated in it.13 This description can be usefully applied to the late Ottoman case. The first category, or 'middle of the message', was continuous reference to Islam, the sultan as caliph and the protector of the sacred places, etc. The second category, which Gluck also defines as a 'naturalization of meaning', occurred in the 'dependent clauses' of Ottoman civilization which found body in utterances related to nomads and other unorthodox or heretical elements such as the Shi'a or the Kurds: 'the settling and civilizing of nomads (urbamn tamttin ve temeddunleri)'or 'the elimination of the state of savagery and ignorance of the nomads (Urbamn tzate-i cehalet vc vahsetkr't)', or 'they live in a state of heresy and ignorance (bir haU dalalet ve cehalet ictnde yasarlar)V4 The category of'deep social meanings', which remain unarticulated but come to be related to the 'givens' in Ottoman/Turkish society, present in the everyday life of the ordinary person, can be seen in the usage of terms such as 'civilization': 'that toothless monster called civilization' (medmiyet denen tek disi kalmis canavar) in the actual words of the Turkish national anthem. Similarly, the reflexive uses of science (funun) or scholarship (ilim) as central to notions of 'progress' (terakki) inform the thinking of the reform-oriented Ottoman intellectual.15

Intellectual conditioning of the elite W h o were the people instilled with this mentality and expected to translate it into workable policies? What informed the decision making of the typical late-Ottoman bureaucrat-statesman as he sat in his office overlooking the Golden Horn:1 First, he was a man who had been imbued with the ideology of Islam from a very young age. Yet, a very close second, he was a man who had been exposed to the ideas of the Enlightenment, if

20

The Well-Protected

Domains

in a most diffused form.1* The main preoccupation, not to say obsession, IY of theTate Ottoman statesman was the saving of the state. This central cause > •: is addressed by various statesmen in different ways and with emphasis placed on a variety of solutions. A major influence on their thought was Ibn Khaldun, who figures as a sort of touchstone in Ottoman-Islamic statecraft.17 A very good case in point is the late Ottoman statesman-historian A h m e d Cevdet Pasa who, although usually thought of as conservative, was the first major historian to be influenced by nineteenthcentury 'scientific history'.18 Ahmed Cevdet Pasa translated a section of the Mukaddimah in which he paid homage to Ibn Khaldun as a historian who, 'knows only one measure that of verifying and revising what has been recounted.'19 Ibn Khaldun's cyclical conception of state power, by which states grew, achieved maturity and declined, impressed Ottomans like Cevdet Pasa who sought to place their civilization in this context. Ibn Khaldun's view of man as a predator who could only be kept in check by an overarching coercive authority, and his view that there existed only certain peoples who possessed the requisite nature (asabtyya) to constitute this authority, served in particular to legitimate the rule of the Ottoman house.20 Cevdet Pasa is also responsible for the codification of §eriat rulings which took the shape of the Mecelle, which still forms the basis of some aspects of civil law in successor states of the Ottoman empire.21 Yet, ever since the French revolution, the Ottomans had been aware of the new winds blowing in Europe. Particularly after the Tanzimat reform, Europe became a source of emulation to the point of embarrasment. Sadik ' Rifat Pasa, one of the leading men of the Tanzimat, looked to Metternichean Austria and the 'great bureaucrats who had created modern Europe' for examples of the sort of enlightened autocracy he hoped to transplant into an Ottoman context.22 The aims of the Ottoman reformers were very akin to those of the French physiocrats: a contented people engaged in peaceful pursuits which would allow them, and the state, prosperity.2 The unification of Germany and the Risorgimento greatly impressed Ottoman statesmen, although Ahmed Cevdet Pasa had a somewhat jaundiced view of the 'nationalism question': When Napoleon III was fighting Austria over the matter of Italy, he came up with this 'nationality' business ... and this did damage to the system of government which had been practiced for all these years ... It always used to be the case that a legitimate government had the right to punish with force any

Symbolism and Power in the Hamidian Regime

21

rebellious subjects ... However, Napoleon invented this new rule which stated that 'any government which is not wanted by any of its subjects must give up its rule over them ... This has caused great surprise among some states who firmly rejected it.21 T h e pasa went on to comment that Britain had supported this principle whereas Russia had categorically rejected it. This had led to the unification of Italy. Yet Italy was avenged by a unified Germany, and the evil-doer Napoleon got his just deserts.25 Suleyman Hiisnii Paga, a contemporary of Ahmed Cevdet, was also inspired by what he called, 'The European policy ... of integrating all nationalities, languages and religions.' He did, however, admit that 'Even if this were admissible by the §eriat, the present conditions (of the empire) do not allow it... ,m All manifestations of power and social interaction leave a trace to their origin. Translating Clifford Geerts to a late Ottoman context, to read the enlargement of the sultan's private pavilion in a mosque, or the playing of a certain piece of music on certain occasions, or the procession of decorated camels bearing the sultan's gifts to Mecca, or the formulae of power in the everyday language of the policy makers: all this is the stuff of historical texture.27 Broadly speaking, the symbols of power in the Hamidian Ottoman empire fall into four categories. Three relate to the sultan and his palace. There were, first of all, the symbols relating to the sacrality of the person of the sultan/caliph, such as coats of arms on public buildings, official music, ceremonies, and public works which reflected directly the glory and power of the Ottoman state. Secondly, there were the more specific and personal manifestations of imperial munificence such as decorations, specially donated copies of the Qur'an, imperial standards and other ceremonial trappings. Third were the religiously symbolic items acquired by the palace such as calligraphy purported to belong to Islamic great men and other artifacts of similar significance. T h e fourth falls into a somewhat different category. It concerns the symbolism of language in Ottoman official documentation. Although not always directly related to the person of the ruler, certain key phrases and words which frequently recur in official documentation provide us with valuable clues as to how the Hamidian bureaucracy conceptualized such matters as the relationship of the ruler and the ruled, their attitude to the nomadic populations, and relationships between members of the state elite themselves.

22

The Well-Protected

Domains

Public symbolism and its manifestations The immediate predecessors of Abdiilhamid II had always taken great pains to appear as 'modern' monarchs by adopting the European practice of displaying portraits of the monarch in public places. Mahmud II had begun this practice. Various religious sheikhs blessed his portraits before they were placed in government offices and other public places, and a twenty-one gun salute was fired as a guard of honour marched past them. This became a tradition which was continued under Abdulmecid (r. 18391861) and Abdulaziz (r. 1861-1876). 28 Also in 1850, 'three large portraits of Sultan Abdulmecit arrived in Egypt... and were paraded through the city [Cairo] in a great procession.' The portraits were then exhibited at the citadel among great pomp and celebration.29 Abdiilhamid deliberately forbade the display of his likeness in public spaces. It is unclear whether this was out of considerations of Islamic orthodoxy which forbids the depiction of the human image., or an obsession with security. On 15 May 1902, the vilayet of Ankara reported that, 'a portrait of our August Master the Caliph has been seen in a coffee shop in Ankara'. The communication went on to say, 'because the nature of the location was not in keeping with the sacred character of the Imperial Image, [the portrait] has been bought and sent to the palace.'30 Abdiilhamid, apparently in an iconoclastic show of Islamic orthodoxy, replaced the image of the ruler with uniformly embroidered banners bearing the legend 'Long Live the Sultan!' (Padtsahtm cok ya$a\) which served the same purpose. 31 This acclaim, which was shouted by soldiers and civilians, had long been the customary way of expressing loyalty to the ruler. Nevertheless, it also became much more standardized as part of the process of increased international competition in ceremonial displays, very similar to the acclaims of'Long Live the Queen!' in the British Empire or 'Tenno heika banzai!' in Japan.32 Verbal acclaims could also become a symbol of opposition. T h e Young Turk opposition to the sultan focused on the acclaim in a negative fashion: cadets at the Military Academy and the Imperial Medical School would refuse to perform it on public occasions, or mumble a rude version of it.33 Despite his fear of assassination (which turned out to be well founded, as there was an attempt in 1905), Abdiilhamid maintained the tradition of public Friday prayers as a ceremony in which the ruler showed himself to the people. In the nineteenth century, Friday prayers acquired additional ceremonial trappings inspired by European examples.

Symbolism and Power in the Hamtdian Regime

23

A physical manifestation of this shift towards a modern, public persona of the monarch was nineteenth-century mosque architecture. The classical Ottoman mosque was altered to suit the ceremonial protocol of European usage with the addition of a two-storey structure to the main building to serve as ceremonial public space to give a more secular character to the buildings, Aptullah Kuran has pointed out that the space in the imperial mosques designated as the personal prayer chamber of the sultan greatly increased in size from the late eighteenth century: '[The] appearance and evolution of the sultan's prayer platform, or loge, went beyond the prerequisites of architecture... It emerged as a vehicle of pomp and circumstance.'3'' Abdiiihamid's own mosque, the Yildiz, is said to, '[have broken] with the Ottoman architectural tradition altogether/as there was an unprecedented increase in ceremonial space which actually outstripped the prayer space.55 T h e Friday prayer ceremony (cuma selamhgi or simply selamhk) would begin as the royal procession left the Yildiz Palace with great pomp, the Imperial landau escorted by Albanian House Guards in livery, and make its way to the Yildiz mosque. There, after prayers, special officials would collect petitions from the people.36 It also appears that the occasion became something of a tourist attraction, as one contemporary account describes the groups of British, American or Germans whose carriages formed, 'a long line on their way to Yildiz to watch the selamhk ceremony.'37 A sort of dais was built to accommodate foreign visitors, where they were permitted to watch the ceremonies and salute the sultan. They were also told not to make any brusque movements with their hands, as this could be misconstrued as an attempted assassination by the guards, who would react accordingly.38 A frequent spectator was Yamada Torajiro, whose account of the selamhk is particularly interesting because of his extremely detailed rendition. Yamada noted that the sultan would be driven to the mosque in his landau, but on the return journey he would mount a simpler caleche, and take the reins himself. Male members of the dynasty, his palace retinue, leading bureaucrats, and high ranking military, in that order, would then line up behind him. This could well have been a symbolic representation of the ruler taking in hand the reins of the state. He also noted that the 'empress dowager' and the 'empress' would take part in the ceremonial procession and accompany the sultan to the mosque, but in keeping with Islamic custom, where women do not go to Friday prayers, they would remain in their carriages. However, their very presence in the courtyard was a depar-

24

The Wdl-Protected

Domains

cure from Islamic practice. Yamada wrote in his diary that the ceremony was very moving as it 'reflected the former glory of a great empire that has now stumbled.'39 O n one occasion the selamltk provided the setting for a show of personal courage on the part of the sultan. When a bomb exploded at the ceremony in 1905 and the sultan escaped unscathed as the result of an unexpected change in procedure; There ensued a general panic as debris and blood was strewn about. The sultan held up his hands and shouted in his deep voice: 'Don't panic!' He then mounted his carriage, took the reins, and as he passed the foreign dignitaries they all shouted 'Hooray!' in one voice.'10 Another of the rare occasions when the sultan showed himself in public was the ceremonial visit to the Holy Relics at the Topkapi Palace and the shrine of Eyiip on the Golden Horn during Ramadan. Yamada again provides a detailed account. The sultan would set off for the old palace accompanied by 300 carriages bearing his barem and entourage. On this occasion the roads were prepared by 'covering them with a thick white sand'/1' How did these ceremonies appear to the wider population? Hagop Mintzuri, an Armenian baker's apprentice, later wrote in his memoirs that he remembered the sultan's arrival at the Sinan Pasa mosque in Besiktas for ceremonial prayers at the end of the fast of Ramadan; First the Albanian guards, dressed in violet knee-breeches, who were not soldiers or police and did not speak Turkish, would fill the upper part of our market square. Then would come the Arab guards of the sultan, dressed in red falvar and adorned with green turbans. These too, did not speak Turkish and they would fill the road. Finally the Palace Guard of the sultan, chosen exclusively from Turks who were tall, sporting their decorations on their chests, would take up their positions as an inner ring in front of the Albanians and Arabs. It is instructive that even an observer such as Mintztiri should have paid attention to the personnel forming concentric circles of security around the sultan. It is fairly clear from his 'reading' of their ceremonial placing, and the fact that he specifically pointed out that the Albanians and Arabs did not speak Turkish, while the innermost circle were 'exclusively Turks',

Symbolism and Power in the Hamidian Regime

25

that he was sensitized to the gradations of ethnic loyalty projected by the ceremonial guard.42 It is also possible that, as an Armenian, Mintzuri may have been more sensitive to this gradation, as the ceremony described must have been taking place in the late 1890s, not long after the Armenian massacres in the district of Kumkapi in Istanbul in 1895.43 The Turks he was describing were almost certainly the elite Ertugrul regiment, named after the legendary father of Osman, founder of the empire.They were chosen exclusively from the region of Sogiid, the mythical heartland of the Ottoman Turks, where the empire had its origins.44 Mintzuri also records how, as a young boy, he was deeply impressed by the ceremony of the departure of the sultan's gifts to Mecca and Medina, the surrc alayt. These gifts to the Holy Cities were a symbolic statement of the caliph's protection for the most sacred stronghold of Islam. Camels heavily caparisoned in gilded livery would parade though the streets of Istanbul bearing the ceremonial offerings. This too was an occasion for great pomp, and Mintzuri relates how a procession of bands would march through the streets playing the Harnidiye march.45 One of the rare occasions when Abdulhamid actually allowed the dignitaries of state to approach his person was at the Ramadan holiday (bayram) during the yearly ceremony of the kissing of the hem of his robe (etek opmek). In a memoir written by a close aide we get some very interesting details of the actual conduct of the ceremony: The sulran's aids hold up gold embroidered handkerchiefs standing at either side of the royal personage. Rather than the actual hem of the robe, as each dignitary files past he kisses one of these and holds it up to his forehead in a gesture of submission (biat). Only to accept the greetings of the ulema does • the sultan rise.'56 In this adaptation of a very old custom, there was a significant departure from tradition to allow for the change in dress. At these ceremonies the sultan no longer wore a caftan, instead he appeared in dress uniform or a morning coat; thus the embroidered handkerchiefs replaced the hem of his robe. It is also significant that the caliph rose as a gesture of humility and respect towards religious functionaries. In most of these public ceremonies there was a significant blending of old and new, Islamic and Western traditions. The setamhk became an occasion where Islamic tradition and Western-style protocol were combined,

26

The Well-Protected

Domains

with foreign dignitaries and palace ladies present in the same ceremonial space. The sum procession was accompanied by a military band playing Western marches, and the biat accommodated changing dress styles.

Official iconography These events, however, remained very much the exception that proved the rule. It was to be from well within the sanctum of the Yildiz Palace that Abdiilhamid made his declarations in symbols and official iconography, One of the most notable symbols of the renewed emphasis on power and ceremonial in the late nineteenth century was heraldry. T h e Sublime State (devlet'i aliyye) was symbolized by the coat of arms of the House of Osman {arma-i osmant). The design had been commissioned from an Italian artist by Mahmud II. By the time Abdiilhamid II came to sit oh the Ottoman throne, it was such a well established part of Ottoman official symbolism that when the sultan asked for a detailed description of its contents in 1905, as he was apparently upset about the lack of uniformity in its depiction, the bureaucracy was momentarily embarrassed because no official authorized version seemed to be readily available. It was finally dug up and the contents described.47 In a detailed memorandum the sultan was informed that the Ottoman coat of arms consisted of both old and new, Turkish and Islamic, motifs, such as armaments and other symbolic objects. T h e central motif in the shield was 'the exalted crown of the Sultans', topped by the seal or tugra of the regnant ruler. This was flanked by two heavy tomes, one symbolising Islamic law, £eria£,and the other modern law codes (ahkam-t seriyye vt nizamiye'yi cami kitab). Under these appeared a set of scales representing justice. The central motif was surrounded and flanked by symbolic armaments, the old balancing the new: arrow and quiver-infantry rifle and bayonet, old style muzzle-loading cannon a modern field artillery piece, a traditional scimitar, a modern cavalry sabre etc. The coat of arms also included traditional Islamic-Ottoman symbols such as a vase full of blossoming roses and incense, standing for the magnanimity of the state. T h e total design was flanked on the right side by a cluster of red banners and on the left by a cluster of green banners symbolizing the sultanicOttoman as well as the universal Islamic nature of the caliphate. Set under the entire design were the whole array of Ottoman decorations. Thus the central themes of the Ottoman coat of arms revolved around the continu-

Symbolism and Power in the Hamidian Regime

27

ity of the old and the new, the traditional and the modern.''8 Another feature of nineteenth-century commemorative iconography was the commemorative medallion. Perhaps the most interesting among the Ottoman examples of this genre, as a bid for modernity combined with time-honoured historical legitimation, is the medallion struck in 1850, during the reign of Abduimecid (r. 1831-1861). An admirable document on the late Ottoman state of mind, it is emblazoned with the slogan 'Cet Etat subsistera. Dieu le veut'. O n one side it features a fortress battered by heavy seas over which flies the Ottoman banner. On the rim are to be found slogans such as 'Justice egale pour tous', 'Protection des faibles', X'Etat releve', and so on. O n the reverse the motifs include the Central Asian Turkish cap, and engraved in various places the names Mahomet 0 (Mehmed II, the Conqueror of Istanbul in 1453) Solyman I (Sultan Suleyman the Magnificent r. 1494-1566), Reshid (Mustafa Resid Pasa, grand visier at the time and the major figure behind the Tanzimat reforms), Aali (Mehmed Ernin Ali Pasa, together with Resid, a major figure in the reform movement), and Coprulu (Mehmed Kopriiiii and his son Ahmed Kopriiiii, die architects of revived Ottoman power in the second half of the sixteenth century) .'w A similar effort to derive legitimation for the present by using symbols of the past can be observed in the prominent place given to the Ottoman genealogical lineage in the state almanacs (salname). In an almanac prepared for the vilayet of Bursa in 1885, the roots of the Ottoman family ate taken back to the legendary Opxz tribe and from there to Adam and Eve via Noah. The official dynastic myth of how the Selcuk Sultan Aiaeddin Keykiibad protected Osrnan, the founder of the dynasty, is duly recounted, claiming that the House of Osman is 'according to the research of experts one of the oldest in the world, and will last forever.'30 Such manifest official fiction was an ancient tradition in Islamic court panegyrics, but what is interesting here is that it should be featured in a srate almanac which is a creation of bureaucratic modernization and features such mundane data as the names of the various ministers, agricultural produce, and main geographical features of the area. The inclusion of this mythical lineage is all the more interesting as Woodhead tells us that the descent from the Oguz tribe, 'and the speculative genealogy particularly popular during the fifteenth century', was, 'largely discredited' by the late sixteenth century.51 The fact that apocryphal genealogies should have been brought back in during the nineteenth century, when the Ottoman state was beginning to look distinctly shaky, is significant. The aim in this case was to stress that

28

The Well-Protected Domains

the rule of the Ottoman family was a permanent and inevitable feature of the landscape. The salname are in themselves a manifestation of the 'pumping up' of the Ottoman foundation myth. T h e genealogy of the Ottoman sultans does not appear in the state almanacs (devlet satnatnelerf) until 1853 (1270 A H ) although they begin in 1846.52 It then disappears, to emerge again in 1868 (1285 A H ) . In the reign of Abdiilhamid, the genealogy moves up from fifth place in the table of contents to third place. Individual entries under each sultan are also considerably expanded so that where earlier volumes gave only basic data such as dates of birth and death, those in the Hamidian period are much more detailed.53 Just as the Ottomans tried to emphasize preexisting traditions by including them in the symbols of state, they also attempted to curtail the circulation of what were considered 'rival symbols'. A correspondence between the chancery of the grand vizier and the palace, dated 8 June 1892, dealt with the issue of the importation of goods whose packaging bore the coat of arms of a rival power. The matter had come up over a crate of mirrors which were being sent from Greece to Crete. It must be remembered that these were turbulent years leading up to the autonomy of Crete and the Ottoman-Greek war of 1897. The sultan wanted to forbid the entry of such packages, but the grand vizier had to point out that there was no legal means by which the Ottoman customs could keep them out.54 On 2July 1889, the Ministry of the Interior reported that certain 'illustrated plates' (levha), published in Moscow, had been seized in the Pera quarter of Istanbul, 'bearing the images of Byzantine emperors and Russian tsars'. It had come to the minister's attention that these plates had been distributed to the Greek Orthodox Patriarchate in Istanbul, and were accompanied by 'a history book bearing certain harmful information'.55 This episode stands as a striking illustration of the fact that the Ottoman authorities understood only too well the implications of the Russian tsars' claims to the status of protecter of all Greek Orthodox subjects in the Ottoman realm, and of their pretension to being 'descendants of the Byzantine emperors,' with Moscow as the 'Third Rome'. This event occurred in the reign of Alexander III who ascended the throne in 1883. It would have been entirely in keeping with the Moscow (rather than Petersburg) centred state symbolism of Alexander, who also stressed the mystic nature of tsardom. 56 T h e Byzantine past was a sensitive issue as the Ottoman official mythology stressed the position of the Ottoman sultans as the successors of

Symbolism and Power in the Hamidian Regime

29

Rome and Byzantium.57 After the conquest in 1453, the Ottoman imperial tradition came into its own. Necipoglu has confirmed that the Topkapi Palace was deliberately built on the site of the Byzantine acropolis.58 Fletcher has also emphasised that, 'The city itself was symbolic of legitimacy in the Roman imperial tradition so that the Ottoman ruler ... now adorned himself with the symbols of Caesar.'59 The cathedral of Hagia Sophia, converted to a mosque after the conquest, and purportedly the scene of the Ottoman Sultan Selim I's assuming the mantle of the caliphate in 1519, was especially significant. Abdulhamid was to accord particular importance to this mosque as the seat of the caliphate.*'0 Thus, the news that, 'certain Greek and other visitors had been drawing and writing on the walls and galleries' of the mosque seemed particularly untoward. To prevent this sort of behaviour strict new instructions ordered that visitors should be escorted at all times.61 Indeed Istanbul had always occupied a central position in the symbolism aimed at reinforcing the legitimacy of the Ottoman sultans. Giilru Necipoglu has shown that the ceremonial progress by a newly enthroned sultan to the mausoleums of his ancestors in Istanbul was itself a means of declaring his legitimacy: 'These tombs built posthumously by the successors of deceased sultans proclaimed Ottoman dynastic legitimacy architecturally by highlighting the uninterrupted continuity of a proud lineage.'62 The presence of the mausoleums, Hagia Sophia and the Holy Relics of the Prophet in Istanbul all contributed to the city's symbolism. Resid Pasa, possibly the greatest Turkish statesman of the nineteenth century, listed his 'three pillars of the state' (tif rukn-u devlet) as Islam, the sultanate and the caliphate all of which were sustained by the House of Osman which protected Mecca and Medina and the continuity of Istanbul as the capital of the empire. 6 ' The visual confirmation of the sultan's sovereignty took the form of his monogram itugra) which appeared on all public works completed in his time. Clock towers erected all over Anatolia bearing the imperial coat of arms and other reminders of sultanic power became ubiquitous. Some of the clock towers were inaugurated in small Anatolian towns such as Nigde, Adana and Yozgad to commemorate the sultan's silver jubilee in 1901. Finkel has pointed out that 'specifically secular monumental architecture' represented by a clock tower highlighted the confrontation between Qur'anic time punctuated by the call to prayer from the minarets, and conversion to a new economic order 'founded on the conjoining of time to

30

The Well-Protected Domains

labour'.64 Thus, particularly in Anatolia and the Arab provinces, these buildings were intended as physical manifestations of the 'middle of the message' and served as markers of a new concept of time and power. Although clock towers and the like were secular bids for legitimacy, the small village mosques were also made into lieux it memoire, One aspect of the sultan's symbolic representation in the provinces was the building of small uniform mosques bearing commemorative plaques that linked his name with distant Ottoman ancestors. One such order dated 8 September 1892 provided for the composition of chronograms (ebced hesabi) for commemorative plaques to be erected over nine mosques in villages in the Corlu area in Thrace. The sultans chosen, one for each mosque, were: Osman II, Mustafa I, Ahmed I, Mehmed IV, Murad HI, Selim II, Beyazid II, Siileyman I (the Magnificent), and Selim I.65 The records also show that the sultan built three mosques in the villages on the island of Rhodes, dedicated respectively to himself, his mother Tir-i Mujgan Hamm, and the legendary Ertugrul Gazi.66 Orders were also issued that the mosque built by Sultan Yildinm Beyazid, one of the earlier heroic sultans, in the small town of tnegol, not far from Sogud, was to be rebuilt as it was in ruins. This was duly done and sanctification ceremonies took place on the sultan's nineteenth accession anniversary.67 Another symbolic connection between the days of early glory and the Hamidian regime, was the deliberate effort to echo the architecture of the Great Mosque (Ulu Cami) in Bursa in the architecture of the Yddiz mosque. It was specified that 'the pulpit (minber) of the mosque should resemble the pulpit of the Great Mosque in Bursa.'68 Although religious/dynastic legitimation themes were being employed here, the function of these mosques was, if anything, closer to the secular message of the clock towers. It is also significant that both the clock towers and the mosques were built in small places, thus manifesting power at the local level. One example of this 'grass roots' message is the commemorative plaque reported on by the Vali of Baghdad, which was to be erected on an obelisk by the Hindiyye dam in the vilayet. A fairly typical example of the genre, it bore the legend: 'To commemorate the Holy Name of the Caliph and to furnish an ornament to His Eternal Power.' The vali specified that the text would be in Arabic, which indicates that the target audience was the local population, rather than a general statement of power, which would probably have been in Turkish. 69 5erif Mardin has noted that, 'The wide adoption of an imperial name

Symbolism and Power in the Hamidian Regime

31

in Anatolia is a marked feature of [Abdulhamid IFs] reign.'70 In this effort to communicate through symbols, Abdulhamid was in a position very similar to the Russian tsar where the 'synthesis of the Russian myths, in which Orthodoxy could serve as a bridge from the sovereign emperor to the people' was employed to bolster the image of the Russian autocracy.71 In all these efforts it is important to note that the symbolic statements on the buildings, bridges, dams and clock towers were made in a specific historical context: the effort made from the Tanzimat onwards to reform and modernize Ottoman cities. In what Dumont and Georgeon call the 'struggle between state power and the local communities', symbolic manifestations of power such as coats of arms or commemorative plaques played a very critical role in the dynamic tension between state and society.72 Timothy Mitchell has noted the obsession with 'an appearance of order' in Egypt in the late nineteenth century where similar developments were taking place. Ottoman public space was similarly 'ordered', when possible, to suit the symbolic statements the centre wanted to make.73 In nineteenth-century Istanbul, Celik has pointed out that 'the regularizing of the urban fabric'occurred as a result of the desire of Ottoman statesmen to present a 'modern' appearance to the outside world, The 'regularizing' of the urban space, as in the case of Egypt, was not only intended to 'tyater le$ bourgeois, but also represented an exercise of power on the part of the centre. The monumental character of historic buildings such as the Hagia Sophia and the Siileymaniye mosque were emphasized by creating clearings around them.74 Similarly, no expense was spared to restore the tombs of the legendary first two sultans of the House of Osman, Osman and his son Orhan, who were buried in Bursa.75 Bursa was also frequently accorded special honours as, 'the crucible of the Sublime State' (mchd-i zuhur-u saltanat)?6 Indeed, a veritable cult of Ottomania was created around the historical heritage of the Ottoman dynasty, as Abdulhamid focused in an unprecedented fashion on the 'creation myth' of the Ottoman State.77 Part of this was the elaborate commemorative ceremony (ihtijal) staged every year at the tomb of the legendary founder of the Ottoman dynasty, Ertugrul Gazi, the father of Osman Gazi. T h e shrine of Ertugrul in Sogud, a small town in west central Anatolia, was turned into a commemorative mausoleum complex honouring the misty origins of the empire. The tomb of Ertugrul was rebuilt in 1886, and a fountain bearing a chronogram celebrating Abdulhamid as benefactor was inaugurated. The sarcophagus bearing what were reputed to be the remains of Ertugrul was

32

The Well-Protected Domains

refashioned in marble, and a grave reputedly belonging to his wife was also rebuilt and turned into shrine in 1887. Sultan Osman's first grave was also rebuilt next to that of his father.78 Together with this, twenty-five graves belonging to 'comrades in arms of Ertugrul Gazi' received new stones. This activity looks distinctly like the 'invention of tradition' given that the identity of these people is unclear and only a few of the graves actually bear names. Even the historian of the site who eulogized the 'great founder of the Ottoman state' felt obliged to point out that, 'It is difficult to tell how Abdulhamid II established that the grave belonged to Ertugrul Gael's wife. We can only surmise that he relied on reliable hearsay.'79 The Ertugrul Gazi shrine is mentioned frequently in despatches. In 1902 considerable money was spent on the creation of an open square around the Ertugrul Gazi Mosque by the expropriation and demolition of buildings hemming it in.30 T h e complex became the site for annual celebrations when the 'original Ottoman tribe' the Karakecili, would ride into Sogiid dressed as Central Asian nomadic horsemen and stage a parade where they would sing a 'national march' with the refrain: 'We are soldiers of the Ertugrul Regiment .., 'We are ready to die for our Sultan Abdulhamid'. This would be followed by a display of horsemanship and a game of arid, the traditional Central Asian sport.81 It became a custom for the leader of the tribe to telegram the palace every year to the effect that, 'We have fulfilled our annual sacred duty of paying our respects to the shrine of the revered ancestors of His Imperial Majesty.'82 Even the so-called 'mother of Ertugrul Gazi', a Hayme Ana, was to be honoured with a special mausoleum built by imperial order in the village where she was reputedly buried.83 The Hayme Ana mausoleum was maintained and refurbished regularly at considerable expense from the privy purse of the sultan.84 The sultan's renewed emphasis on the 'Turkishness' of the early days of the Ottoman state can also be seen in his treatment of the 'original Turkish dynasties'. He greatly honoured the Ramazanogullan clan, who had been a Turkish beylik in the Adana region, because they claimed to be 'of pure Turkish blood'. Abdulhamid invited Emetullah Hatun, the leading matriarch of the clan, to Istanbul where she and her entourage were lodged at the Yildis Palace and treated as honoured guests. 85 The renewed obsession with the early days of the Ottoman empire in Abdtilhamid's reign can be likened to the same sort of obsession with dynastic legitimation in such times of extreme crisis as after theTimurid

Symbolism and Power in the Hamidian Regime

33

debacle of 1402. Mehmed I, the strongest figure in the interregnum, 'spent some of his precious resources to build a mosque in Sogud'. 86 Like Abdulhamid much later, Mehmed felt the need to issue a statement to the 'grassroots'. From obscure towns and villages of Thrace and Anatolia to the holy cities of Mecca and Medina, the sultan took great pains to ensure that his munificence was acknowledged- His symbolic gifts acquired particular importance in the holy cities. O n 6 April 1889, an imperial gift of several candelabra was delivered to the Ka'ba and presented during a ceremony which was held during Ramadan, ensuring that 'thousands of the faithful intoned prayers for the long life and success of His Imperial Majesty.'87 Also, two ceremonial tents were erected in Mecca every year during the haj on the two hills of Arafat and Mina. These were erected only during the pilgrimage, and were the symbols of the sultan's presence as his annual haj message was read from the tent on Mina. It is significant that the tents were pitched on the two hills that served as the focus of excitement during the performance of the haj rites.88 Further visibility was ensured through the traditional practice of the sultan/caliph providing the holy mantle which covered the sacred stone of the Ka'ba {setre~i fcrif) . On6September 1892, the palace was informed that the new mantle, with the sultan's name embroidered in gold, was ready. It was to replace the old one which still bore the name of Abdiilaziz.89 The visual confirmation of sovereignty was also extended to non-Muslim places of worship. O n 23 October 1885, the grand vizier, Kamil Pasa, reported that the Armenian Catholic church in Buyiikdere in Istanbul had erected a commemorative plaque stating that the church had been constructed 'during the just and glorious reign of Abdulhamid II'. What is interesting is that the initiative seems to have come from the Armenian archbishop, who declared that 'this was being done for the first time in a Christian temple*. In fact the sultan was rather unsure about how appropriate this whole business was, and ordered that 'it be secretly investigated as to what the exact wording on the plaque consists of,' as 'if it is too prominently displayed it might be offensive to Muslim opinion.' Kamil Pasa reported back that it was a harmless display of loyalty, and in any case the plaque was displayed m an inner courtyard where few Muslim eyes would see it.90 It would seem, however, that Abdulhamid soon overcame his shyness and the erecting of official iconography on non-Muslim official buildings became commonplace. An order dated 16 March 1894 declared that the

34

The Well-Protected Domains

request of the Catholic Archbishop of Oskub (Scopje) to display a plaque bearing the imperial monogram (tugra) on the archbishop's residence was to be granted. The decision was based on the precedent which declared that 'since various archbishoprics of other confessions have in the past been thus honoured with the August Symbol', it was appropriate in this case, too. 9 ' The matter ofjust where official iconography could be displayed sometimes led to amusing incidents, one such being the case of Manolaki, a Greek tea house operator in Istanbul who set himself up as the self-styled 'tea maker in chief to the Imperial Palace* (saray-i humayun fayct b
Symbolism and Power in the Hamidian Regime

35

commissioned to take photographs to be sent to the Chicago World Fair of 1893, also applied for permission to 'grace their shop with the royal coat of arms' and the tugra. It is highly unlikely that they were granted permission.95 Louis Rambert, the director of the Imperial Tobacco Regie, recorded a similar event in his diary. The Ottoman customs seized crates containing the products of the Regie, on the grounds that they bore the imperial coat of arms. T h e affair created something of a diplomatic crisis as the ambassadors of France, Austria and Germany became involved. Rambert only solved the problem by producing an official authorization specifying that the Regie's permission to use the arms as part of its logo ran for another five years. 96

Personal manifestations of royal favour The nineteenthcentury was the century of decorations, as royal favour became channelled into precise dosages.97 The Hamidian regime habitually used decorations as a form of investment in the goodwill it hoped they would foster in the recipient. Thus, symbolic manifestation of sultanic munificence had a co-optive aim. Decorating or otherwise rewarding men it could not discipline or control had always been a policy of the Sublime Porte. As real coercive power declined in the nineteenth century this became all the more prevalent. Yet, thedecorations policy of the Ottoman centre must not be seen purely as an alternative to coercion. In a very real sense decorations were a manifestation of the integrative symbolic code and it has been pointed out above that the coat of arms of the Ottoman House was prominently featured on decorations.98 As in the matter of the coat of arms, uniformity of the design of various decorations was a consideration for the sultan, who ordered that precise drawings illustrating all the state decorations be made and presented to him." O n 19 June 1892, the vilayet of Konya reported that certain Greek notables in the town of Isparta had been wearing their official decorations and uniforms to church during the Easter service. The governor proudly reported that he had put a stop to 'this inappropriate practice'. H e was (no doubt much to his surprise) promptly reprimanded and told that 'these people are wearing their decorations as a gesture of pride and loyalty and should not be interfered with.' Evidently the local official was offended by the Christians sporting Islamic symbols such as the star and crescent on

36

The Well-Protected Domains

their chests in the profane space which a church constituted. He was overridden by his superiors who tapped his knuckles for interfering in a practice which they chose to approve of.100 Nor were decorations treated lightly by the recipients or potential recipients. When the Ottoman ambassador to Paris, Munir Pasa, was sent on a tour of the Balkan capitals and distributed decorations, the event caused quite a furore. The Ottoman High Commission in Sofia reported that the decoration of the King of Serbia's daughter with a high ranking medal (ni$an-i all) and his delivery of a Compassionate Order (sejkat nisam) to the Queen of Rumania, had caused heated discussion in the Bulgarian press. It was seen as a move by the Porte to support Serbian and Rumanian interests in the Balkans against those of Bulgaria.101 T h e fact that decorations usually came with an award of money meant that they were often solicited by some rather dubious characters. Such was the case of one Professor Adolphe Strauss, who wrote to Yildiz claiming that his articles in the Hungarian press had created such positive feeling towards the sultan that when they were read aloud in the Hungarian parliament the deputies spontaneously leapt to their feet shouting: 'eljen a saltan? (long live the sultan). The good professor actually went on to recommend that several of his colleagues be awarded specific decorations: 'For Prof. Sigismonde Vajda, the Order of the Osmanieh Third Class would be appropriate since he already holds the Osmanieh Second Class .,.' and so on.103 Together with decorations, the presentation of copies of the Qur'an or ceremonial banners was also part of the symbolic dialogue between the ruler and ruled. One element that the sultan tried to woo were the Kurdish chieftains of eastern Anatolia.103 On 11 September 1891, the vilayet of Trabzon reported that the ceremonial banners and Qur'ans sent to the Kurdish tribes of the Erzurum region had been received, and the proper ceremony in the presence of a military band and a guard of honour had been carried out.104 The banners were manufactured in Istanbul especially for the Kurdish Hamidiye regiments and paid for out of the privy purse.105 Even in the besieged garrison in Medina in 1918, the regimental standards were solemnly decorated. To maintain morale, the garrison commander organized an essay competition which was won by an tract on 'flag protocol'. The writer specified that the flags which bore the coat of arms of the Ottoman state be dipped to salute the sultan, whereas in the case of those bearing suras from the Qur'an the sultan was to offer the salute.106

Symbolism and Power in the Hamidian Regime

37

The crack Ertugrul regiment which served as the personal bodyguard of the sultan also had its sancak renewed in 1892. The privy purse reported that special care had been taken to ensure that 'the prayers, dates and other legends are identical to the one that is kept as a sample in the Imperial Treasury.'107 Robes of honour (hil'at) were another form of the symbolic exercise of sovereignty. In these matters strict protocol was applied and precedent was seen as 'accepted procedure'. When the question of the award of a robe of honour to Sheikh Ibn-Resid came up in 1885, the grand vizier, Said Pasa, advised against it. The vizier's reasons are interesting, for they illustrate the considerations that were weighed in these matters. Said Pasa pointed out that although it was customary to pay Arab sheikhs money to 'provide for the protection of the Hijaz roads', the granting of hil'at was reserved only for the sharifs of Mecca. If any such honour was awarded to Ibn-Resid, it would mean that 'he would be considered an equal of the sharifs and this will imply that the state is dependent on him, thereby increasing his prestige among the Arab sheikhs in a way that would be detrimental to the influence of the state.'108 Particularly in the Arab provinces, decorations and other symbols of imperial favour were employed to 'win the hearts of the local sheikhs and notables'. The long-time Vali of Hijaz and Yemen, Osman Nuri Pasa, was to write: The nomads and sheikhs are lovers ofjustice. The best way to deal with them is to be entirely honest and honour promises made to them. Their men and women should always be given a good reception at state offices, their complaints listened to and the necessary measures taken. The notables among them should be given decorations and a fuss should be made over them as these people are very fond of pomp and circumstance.109

Symbolic objects acquired by the state One means of emphasizing sacrality was for the palace to buy up symbolic objects which it deemed it should have in its safe-keeping. There is also evidence of this practice in the last years of the previous reign. Butrus Abu Manneh has drawn attention to the ceremonial progress, in May 1872, of'the Prophet's sandals' from Hakkari in eastern Anatolia to Istanbul, and the grand ceremony conducted upon their arrival. The press gave reports of the progress every step of the way and offered accounts of

38

The Well-Protected

Domains

miraculous happenings en route.110 Another instance of this was the order to buy what was purportedly a copy of the Prophet's handwriting in the form of a letter written to the ruler of the Ghassanids. Acquired from a certain Monsieur 'Perpinyani' (presumably Perpigniani) in 1875, it was ordered that it should be placed in the Imperial Treasury.111 Objects like this, often of dubious authenticity, bought from people of foreign nationality, continue to come up in dispatches. Another such case was an item of calligraphy, supposedly an example of the handwriting of the Caliph Ali, which the enterprising Perpinyani had presented to the palace in 1877. The grand vizier Said Pasa pointed out that the item had been handed over to the treasury some time ago, and Perpinyani, backed by the French embassy, was now demanding his original fee of 5,000 liras plus interest of 1,000 lira because of the delay in payment and if not the return of the item. Moreover, he was threatening to sell the piece to the British Museum if he was not paid promptly.112 Many years later the matter had still not been settled. O n 13 March 1892, ministers were still discussing the 'appeasement' of Perpinyani who continued to make a thorough nuisance of himself by repeated pressure through the French ambassador. It was estimated that 5,000 lira was a fair price, as the accumulated interest would actually amount to more.113 By August 1893, when the palace was informed that the matter had still not been settled, the ministers were told with evident exasperation that 'even if the authenticity of the item is less than certain it should be bought and placed in the Imperial Treasury where allsuch sacred items belong.' What was being said here in plain language was: 'buy it even though it is probably a fake,' A failure to do so might have resulted in a grievous loss of prestige, particularly if the ubiquitous Perpinyani did suceed in selling it to the British Museum. 114 The news that the palace was prepared to pay good money for such items attracted potential sellers of a similarly dubious nature. One such case was a woman named Fatma, a resident of Makrikoy (Bakirkdy) in Istanbul, who telegraphed that she wanted to present to the sultan what she claimed was the stirrup of the Caliph Ali. This stirrup, she said, had magical qualities: when porters (bamal) had difficulties lifting their burden, she would wash the stirrup and make them drink the water. The porters, thus fortified, would happily swing their burden unto their backs.115 The Minister of Police, Nazim Bey, took a somewhat jaundiced view when he was instructed to summon the item to inspect it. H e reported that the woman had no real evidence beyond what her late father-in-law had told

Symbolism and Power in the Hamidian Regime

39

her, and that the stirrup was somewhat garishly fashioned as a dragon's head: 'although the shape of stirrups at the time is unknown, it is unlikely that the Caliph Ali would have used such a stirrup thus adorned.' Na^im Bey added that the inscription which conveniently adorned the stirrup 'the stirrup of the Caliph Ali' - was 'suspiciously like the writing of our present time'." 6 It is highly unlikely that the palace would have bought the item. In cases where members of the wider population approached the palace on such matters, it is worth bearing in mind that what was going on was a decorous form of giving alms to the poor. It is unlikely that Abdiilhamid II would have been tempted to invest in the occult properties of Ali's stirrup and much more likely that, with the tale of misfortune that went with it (dead husband, son in the army, having travelled all the way from Bursa to make the gift), the petitioner hoped for (and probably received) the sultan's charity. Another such case was that of Tahir Efendi. Like Fatrna, a person of modest means and a minor official in the Ministry of War, Tahir Efendi brought forward what he claimed was a letter of patent (berat) entitling him to certain revenues as the descendant of the chief standard bearer (sersancakdar) of the Prophet Mohammed. The palace ordered that the matter be looked into and that 'it be examined if there is indeed such a hereditary rank'.117 The chances are that Tahir Efendi was sent away with 'a little something' (bir mikdar fey), as the saying always went in the event of such occurances.

T h e symbolism of language in the Hamidian era When one combs through Ottoman archival documentation, one comes across certain words, phrases, or cliches which frequently recur. Usually overlooked as part of chancery officialese, these phrases can provide useful clues to the way the state regarded its subjects, the relationships among the ruling elite themselves, and the way they perceived the basis of their rule.118 This was noticed by a contemporary British observer: For ruling native subjects, the guiding word is 'akilaneb (skilfully), while the brutal and often sanguinary conflicts among the peasantry are described by no fiercer a term than na-saz-hk (impropriety); the correction of the same to be performed in a peaceful mode, is called tarteeb (setting to rights). Voormak (to strike), a word implying resort to force is a word but rarely pronounced,

40

The Weil-Protected

Domains

and then only in a subdued voice, These are specimens of the tatlu dil, the 'sweet tongue' of the Turkish rulers.'19 The first thing that emerges from a close reading of these 'codes' is that the state's view of its people was never negative. The people as a whole were always good, they were occasionally led astray by certain malicious and perfidious elements, but were potentially always capable of loyalty. This was not the result, as it has been argued ad nauseum of the fact that the state considered its people a 'herd' or a 'flock'. Bernard Lewis has pointed out that in the course of the nineteenth century the Ottoman term reaya was replaced by teb'a which was, 'becoming the Ottoman equivalent of the English word "subject"'.120 Particularly in the nineteenth century, the state was in desperate need of a reliable population, it was simply not in a position to dismiss the population as rebellious and to crush insurgency, even if it had the material means to do so, which more than often it did not. In the case of the Yesidis, Iraqi Kurds who were targeted for conversion to Hanefi Islam, the people themselves were seen as 'simple folk who cannot tell good from evil (nik ve bed'i tejrik edemiyen sade-dilan ahali).' They were being led astray by their leaders who were 'fooling and provoking them' (igfalat ve te$vikat).m In another, totally different, context the same words come up. When refugees from Greece who had been settled on the Ottoman side of the border, threatened to go back to Greece because they had not been given the land promised them by the state, they too were termed as 'those who cannot tell good from evil (nik ve bed'i fark etmez kimseler)'a.nd who had been led astray by the Greeks.122 Nor was this attitude confined only to Muslims. When Protestant missionaries became active among the Christian population of the vilayet of Syria, the people were again seen as 'simple people who cannot tell good from ill and are having their beliefs poisoned' by evil elements (nik ve bedi fark ve tetnyize muktedir olmayan $ade~dilan ahali).123 Similarly the Alevi Kizilbas population of Tokad were described as, 'simple village folk (sadedil hr'a halki}' who were to be 'shown the high path of enlightenment' by instructing them in the Hanefi mezheb}24 When American missionaries attempted to open a school in Konya, the view was again that this was' an effort to fool the Armenian simple folk into increasing the influence and number of Protestants in the area (ermeni sade dilamm igfd)'m When the population itself seemed to be involved in untoward activity their leaders were usually to blame and were qualified as 'confused or silly elements

Symbolism and Power in the Hamidian Regime

41

(sebukmagz taktmi}'.116 Nor were Christian missionaries the only source of perfidy and subversion working on the population. The Shi'i missionaries active in the vilayets of Mosul and Basra were also known to be 'perturb' ing the minds (tahdi}4 ejkar)' of the people and inviting them to become Shi'a. 127 This motif of confusing and otherwise troubling minds comes up time and again. Even in mainstream Islam, matters which could provoke controversy were to be treated carefully. W h e n the newspaper Malumat published a rebuttal of some anti-Islamic article published in Egypt it was reprimanded for 'confusing the minds' of the common people (tesvi}4 ezhan) by allowing such debates to appear in the newspaper columns.128 Hasan Kayak has pointed out that this tendency continued into the Young Turk period when, in 1912, 'discussion of political subjects was banned in view of reports from the provinces that religious functionaries who would not be expected to "distinguish good from bad" in political issues were preaching on matters of elections and politics.'139 When Fahrettin Pasa, the defender of Medina in 1918, was betrayed by his subalterns who spread about the rumour that he was deranged, his biographer was to characterize these men as 'those who had water on the brain but nonetheless succeeded in tricking some innocents among us'.130 Another stock phrase particularly when it came to nomadic populations such as the Bedouin Arabs, or the Kurdish tribes, was that they 'live in a state of nomadism and savagery (baU vafyet vc bedeviyetde ya?arlar)..'The Yezldi Kurds who lived in this state were to be 'gradually brought into the fold of civilisation (pey der f>ey daire-i medeniyete idhal)', which was to be done through schooling and the constitution of a municipal authority in their area, Sincar.131 This vocabulary then, was the expression of both the age-old contradiction between the desert and the town, and the 'mission civilizatrice' mentality of the new Ottoman bureaucracy. As Mardin puts it, the Bedouin had to be 'liberated] from the shacldes of community life.'132 The sultan himself once told a European ambassador that in eastern Anatolia there were tribes, 'whose comportment is similar to savage tribes in America'.153 The ultimate aim, of course, was to transform them into reliable members of the 'fundamental elements (unsur-u aslt)'. The obsession with 'bringing civilization and progress to the Arabs {urhamn temeddun ve terakkikri ) and 'transforming them into a settled population (urbanw ttvattinlart), occupied Ottoman officials from the mountains of Iracj to the sands of the Sahara. The main reason why the Ottomans wooed the Senusi

42

The Well-Protected

Domains

sheikhs in the latter area was that they, 'reformed the character (tehzib-i ahlak)' of the Bedouin and, 'abated their savagery (izak-i vah$et)'.m The nomads were usually considered as something akin to wild creatures, who had to be managed lest untoward developments 'provoke their wild nature and hatred (tevahbus ve ncfretlerini mucib olmak)',u5 The effort to create reliable military operatives was the source of another recurring formula: that the people should be incorporated into the ranks 'without lamentation (stzildtsizca)', This word comes up every time there is a question of incorporating otherwise unruly peoples into the armed forces. The Yezidis, the North African desert Bedouins, and the Alevis of Anatolia were ail to be thus treated. The fact was of course that there was no end of lament' on the part of these people,136 When it became a matter of open rebellion then it was necessary to 'punish them and frighten like-minded ones (kendileri tedib ve emsati ierhib)'li7 When it was a question of a Christian minority who had a complaint, the form was usually 'the complaints mixed with gratitude (ftikran ile memzuc sikayetleri)', as in the case of the Armenians of Tokad who complained that they were being deprived of schooling.138 Also there was a constant effort to present a good, or at least a defendable image towards the outside world. This was usually expressed by the stock phrase that something 'would not look good towards friend or foe (enzar-i yar ve agyara kar^u hos gorunmtmeO),' or that it would 'cause loose talk (ferfic-i kil-u kal),'m At a time when the Ottoman state was under constant pressure from the outside world to implement this or that treaty obligation, or to fulfil promises of this or that reform, it was of manifest importance not to provide additional opportunities to exert leverage. Very often these utterances have what $erif Mardin has called 'an incantatory quality', and appear to voice the feelings of a ruling elite that is trying to convince itself of its own legitimate right to existence." 0 The very name of the Ottoman state, memalik-i mahrusa-i sahane, (the well protected domains of His Imperial Majesty')was a testimony to this state of mind, and a monumental irony, because they were anything but well protected. Every time the danger of the disintegration of the state had to be mentioned it was accompanied by huda mgerde (may God forbid), almost as though to voice the very words was to tempt providence,1,11 Another name for the state was 'dcvlet-i ebed-'i muddet-i osmaniyc' the eternal Ottoman state, very akin to Roma aeterna or la France eternelle' and with the same 'halo of perpetuity'.14

Symbolism and Power in the Hamidian Regime

43

Conclusion The most crucial task any project of social and political legitimation must face in an anckn regime state is the need to make itself out to be part of'the natural order of things', 'things as they always have been'. When the population at large accepts the historical inertia of a particular order of things, half the battle has been won; it then becomes a question of maintaining the status quo. The problem arises when the political centre tries to make new and more intensive demands on its population, or the expectations of the people change. Thus, the Hamidian order tried to perpetuate the image that it was what Ernest Gellner called 'the very norm of truth', yet on the other hand it tried to inject new muscle into this image by squeezing society for material and spiritual resources which could only be mobilized at the cost of altering the basic balance.143 The symbols drawn from the familiar Islamic traditions had long since been accepted because, as elegantly put by Edward Shils, 'one of the main reasons why what is given by the past is so widely accepted is that it permits life to move along lines set and anticipated from past experience and thus subtly converts the anticipated into the inevitable and the inevitable into the acceptable.'144 The problem in the Hamidian Ottoman state was that increased Islamic symbolism and reliance on the caliphate as the 'exemplary centre linking the earthly and celestial hierarchies', was an inadequate substitute for real power.145

Chapter Two

The Ottomanization of the §eriat

In the besieged Medina of late June 1918, the Ottoman commander of the garrison, Fahreddin Pasa, issued an order to his troops to eat locusts. Since April 1918 the Medina garrison had been cut off from the outside world by Arab forces who had captured the stations of the Hijass Railway linking Medina and Damascus. The garrison, well equipped in terms of armaments, was soon faced with starvation.1 Yet, even in siege conditions, Fahreddin Pasa set about building a symbolic avenue into the city as a manifestation of the continuity of Ottoman rule. The garrison of Medina, which held out for some three months after the Mondros Armistice of 30 October 1918 when the Ottoman empire surrendered, were fully aware that they were defending the very basis of Ottoman legitimacy, its foundation in bricks and mortar. When asked to surrender after a prolonged siege by Sharinan forces, which lasted until January 1919, ending a three-year last stand since the loss of Mecca in June 1916, the Commander of the Hijaz Expeditionary Force sent the following message to the British command: 'As I am now under the protection of the Prophet and the most High Commander, 1 am busying myself with strengthening of the defences and the building of the roads and squares in Medina. I beg you not to trouble me with useless requests.'* Naci Ksciman, who took part in the defence of the Sacred Precincts (Haram-i Serif), was to express his indignation in his account of the siege, which is something of a crie de coeur of a cosmopolitan late Ottoman whose world was collapsing around him: 'Was this to be the end then? The end of all these centuries of sacred service which it had been our honour to perform since the time of Yavuz (Sultan SelimI)...' 3 As the monarchies of the nineteenth century, the Ottoman Empire included, came to fmd themselves increasingly hard pressed to legitimate their existence towards both their own subjects and the outside world, they felt 'the need to provide a new, ot at least a supplementary, "national" foundation for this institution.' 4 The Ottoman rulers faced the challenge

44

The Ottomanization of the §eriat

45

of nationalism, not only from their Christian subjects and ex-subjects in the Balkans, but also from their own Islamic peoples. As it became objectively necessary to mobilize the Ottoman people along the lines of sotnething coming more and more to resemble an 'Ottoman citizenry', the rules of the game had to be altered accordingly. To use Gellner's phrase, the 'field of tension' between ruler and ruled shifted onto a different plane,5 The Tanzimat Edict (Gulkane Hatt-i £en/f) of 1839 was a critical milestone in Ottoman legal history, for it was nothing less than a public declaration by the sultan that he would respect the rule of law. The term tanzlmat itself is significant as it actually means reordering.6 Yet, although the common wisdom is that the edict was more or less entirely the result of Western pressure and the work of grand vizier Resid Pasa, recent research is turning up evidence that points to an inner dynamic within the Ottoman ruling circles: Butrus Abu Manneh has convincingly shown that the initial formulations which provided the background to the edict emanated from the young Sultan Abdiilmecid, who was nowhere near as passive as has always been rhought, and his close circle of tutors and advisors.7 Subsequent legal reforms included the Penal Code (1858), the institution of secular (nizami) courts in 1869, the empowering of the Ministry of justice to control these courts, as well as the introduction of the principle of advocacy (1879). 8 However, what concerns us here is not the creation of legal infastructure, although that process provided the background for what follows. The object of this study is to examine how Islam was used as a mobilising force in a specific context. At some point after the turn of the nineteenth century Ottoman power holders began increasingly to feel the need to address an appeal to those elements in society who were hitherto only commanded to obey. At that point they found themselves obliged to formulate a common series of reference markers. Ottoman state legitimation became the grey area where these markers were formulated, accepted,'or rejected. Their formulation took place in the space where state power and society confronted one another, leading to a process of implicit negotiation between power holders and subjects. Depending on their interests, relative strengths and weaknesses, and the historical conjecture, the outcome was the late nineteenth to early twentiethth-century Middle East. After the Tanzimat reform of 1839 the Ottoman elite increasingly felt that it needed more than passive compliance with orders, and very gradually shifted the basis of legitimacy accordingly. The outcome was something akin to what Hobsbawm has called 'proto-nationalism'. 9

46

The Well-Protected Domains

The emphasis on the sultan's position as caliph of all Muslims served increasingly as an integrative 'icon' in the new 'religio-ethnic identification': 'The most satisfactory icons from a proto-national point of view are obviously those specifically associated with a ... divinely imbued king or emperor whose realm happens to coincide with a future nation.'10 The Ottoman statesmen of the late nineteenth century thus came to envisage a sort of Imperial supranationalism' in the form of 'Ottomanism'." From being ostensibly supra-religious during the heyday of the Tansimat (1839-76), Ottomanism would undergo a shift in emphasis to become more Islamic in tone and nuance during the reign of Abdulhamid II,

T h e Hamidian Hanefl caliphate Soon after the 1876 'Turkish atrocities in Bulgaria' became the stuff of everyday politics in Britain, the British ambassador to the Sublime Porte, Sir Henry Layard, had a foretaste of what was to come: Layard had never seen the sultan angry before ... 'We are accused in Europe of being savages and fanatics ... [Yet] unlike the Czar, I have abstained till now from stirring up a crusade and profiting from religious fanaticism, but the day may come when I can no longer curb the rights and indignation of my people at seeing their co-religionists butchered in Bulgaria and Armenia .. .'i2 One of the foundation stones of Ottoman legitimation ideology, since the conquest of the Hijas; in the sixteenth century, had been the position of the Ottoman sultan as the 'Caliph of all Muslims on Earth' (Halife-i Muslimin or Haltfe-i JRwi-zemm.)13 As pointed out by Halil Inalcik, the claim to being the universal Islamic ruler gained new content during the reign of Siileyman the Magnificent because 'what was unique in Suleyman's case was that he believed in making all these [claims] a reality through the tremendous power he held.'14 The official myth ran that Sultan Selim I, the conqueror of Egypt, had received the mantle of the caliphate from the last Abbasid caliph in 1517. The myth, however, was by no means uncontested. Arnold points out that the factual basis of the alleged transfer remains somewhat murky,15 Bernard Lewis has stated categorically that the story of the transfer was' [without the] slightest shadow of a doubt apocryphal'.16 tnalcik also refers to the episode as, 'a legend apparently fabricated in the eighteenth century'.17

The Ottomanization of the §eriat

47

Fleischer states that',., Ottoman legitimacy was very weak from the standpoint of both Islamic and nomadic political tradition ... Their success, in fact, owed far more to efficiency than ideology.'18 What was to happen, then, when 'efficiency' radically declined from the eighteenth century onwards? How were the Ottoman sultans to sustain legitimacy in a world context where their position was under threat not only from the outside, but also from their own Muslim population?19 The answer was the following: durir^tjiej^ign^ofAjjdulhamid II (18761909) the Ottomancaliphate launched a majoj^irxkiative.aimed at •toTnrnandingJ^^ This effort was very akin to policies attempted by the other legitimist monarchies of the time such as the Habsburgs and the Romanovs. The policy that Seton-Watson has called 'official nationalism' was an application of 'national' motifs by the ideologically besieged dynasties ruling multi-ethnic empires.20 In its own context, and using its own historical colouring, the Ottoman caliphate attempted to imbue itself with a new mystique, a new self image. Duguid puts this new emphasis of the Hamidian era on the key concepts of 'unity' and 'survival': What is distinctive to the Hamidian period as opposed to that of Selirn III, Mahmud II, or theTanzimat, is that (the policy of reforms and changes) were consistently subordinated to a higher felt need, that for unity among the Muslim population of the Empire. In terms of policies and priorities, 1878 represents a fundamental shift in the Ottoman self view.ZE It is significant that Abdiilharnid's most feared enemies, the British colonial authorities, also referred to '... The new-fangled pretension that the Sultan of Turkey is Khalifeh of Islam ...'n Abdiilhamid was able to capitalize on a general feeling of impending doom and despondency among the people as province after province was torn from the empire.23 The situation led Albert Hourani to comment that: 'The loss of most of the European provinces changed the nature of the empire. Even more than before it appeared to its Muslim citizens, whether Turks or Arabs, as the last manifestation of the political independence of a Muslim world beleaguered by enemies.'24 Islamic legitimation therefore came to be seen as essentially a 'defensive' position. As H u n Islamoglu Inan accurately notes:

48

The WelUProtected Domains Legitimation now became a defensive ideology consciously defined around Islamic motifs ... Against the unjust order imposed by the West, Islam, as a particularistic national ideology, came to stand for justice and equity. It is important to place the institution of the caliphate, and Islamic nationalism as it was consciously fashioned in the Hamidian period, in this context.25

I n addition, since t h e T a n z i m a t Edict of 1839 h a d declared that M u s l i m s a n d C h r i s t i a n s were equal before t h e law, serious r e s e n t m e n t h a d b e e n fostered a m o n g M u s l i m s w h o increasingly felt their position of superiority u n d e r the $eriat was being undermined. 2 6 It was in this setting t h a t the 'new orthodoxy' based o n t h e Hanefi • mezheh as the official ideology was to take shape. T h e Hanefi school of Islamic j u r i s p r u d e n c e h a d always been the closest to t h e hearts of t h e O t t o m a n rulers as t h e 'official belief (mezheb-i resmiye).27 T h e reason for this preference was the Hanefi interpretation of t h e caliphate, whereby a s t r o n g a n d able ruler was to be recognized as the legitimate sovereign of all M u s l i m s o n the condition that he protected Islam a n d upheld the $eriat even if he was n o t from the original sacred A r a b clan of Qureish. 2 8 Fleischer has p o i n t e d o u t that the official title of the O t t o m a n sultan as t h e 'Shadow of G o d o n E a r t h ' (zill alkhfi'l arz) was a late invention based o n t h e need to reconcile the steppe tradition of t h e Turks w i t h the reality of aspirations to high Islamic universality: 'Universal sovereignty ... is t h e highest form of kingship, t h e legitimacy of which cannot b e denied because such striking success gives tangible evidence of G o d ' s will' 2 9

r

D u r i n g t h e reign of A b d u l h a m i d I I there occurred a self-conscious att e m p t o n t h e p a r t of O t t o m a n bureaucrat/intellectuals to recharge a n d redefine basic Islamic institutions, namely t h e §eriat a n d the caliphate, as t h e basis of t h e q u e s t for a n e w I m p e r i a l / n a t i o n a l identity. N o less a n __ observer t h a n the famous D u t c h Orientalist, S n o u c k H u r g r o n j e was to comment:

The concession made by the Turkish Government in the period of their victories ... recognized in the conquered provinces that, along with the Hanafite, a judge belonging to the native rite should be appointed, (this) has nowadays [1885-86] come to be considered unnecessary ... T h e Turkish Government has suppressed them all... except the Hanafite who became the sole judge in the Law of Religion.30

The Ottomanizatwn

of the §eriat

49

Siileyman Htisnii Pasa, to whom we have already referred, had very interesting views on the subject. From political exile in Iraq he was to pen an extremely detailed memoranduwm, dated 7 April 1892, relating to measures to be undertaken by the state to ensure the integration of heterodox and heretical elements into the official belief.31 After offering a rather sophisticated breakdown of the complex ethnic mosaic of Iraq which included Turks, Kurds, Arabs, Chaldeans, Nestorians, Armenians and Jews distributed across various sects and sub-sects of Islam and Christianity he commented: 'As can be seen from the above the elements belonging to the official faith and language of the state are in a clear minority whereas the majority falls to the hordes of the opposition.'32 This situation was to be remedied by systematic propaganda and the 'correction of the beliefs (tashthn akaid)' of the 'heretics' or 'deviants' (ftrak* i dallt). (The term 'correction of beliefs' would become something of a watchword in the Hamidian regime.) In order to accomplish this the Ottoman state should, he proposed, sponsor the writing of a Book of Beliefs (Kitab'ul Akaid), to consist of some fifteen chapters, each dealing with one unorthodox element. The pasa was rather generous in his wisdom and the projected chapters of the book he suggested ranged from obvious targets such as the Shi'a through Christianity Judaism to 'the pagan practices of Indo-China'. 33 It is also significant that one of the heretical beliefs was positivism referred to as 'the new philosophy (fehefe-t cedide}'. For this reason it was recommended that the book also be translated into French.34 Those who took up their pens in defence of the 'official faith' did so with the knowledge that the written word held a power to reach a far broader public than ever before through the world-wide distribution of newspapers and books. As Mardin has pointed out, what can only be described as social mobilization literature of an officially approved kind, and 'pocket libraries', reached even hitherto inaccessible areas such as Van in eastern Anatolia in the 1890s.3S Much of this literature was deliberately written in relatively simple language with a view to reaching a popular audience. The 'official faith' was also disseminated far and wide through the press.36 Similarly/The tone and trend of Turkish papers is to intensify the hold of the Sovereign and Khalif on the imagination of the 'true believers', especially in the lower classes, even in the outlying districts of his extensive dominions.'37 This attitude was also reflected in social mobilization literature as exemplified by the work of an obscure religious functionary from a small Anatolian town which included the comment, '... Thanks to the newspapers the brilliant rays of the One True Faith

50

The Well-Protected

Domains

have even reached into Europe where it is now recognized that Islam is a scientific great religion (fenni bit d'm-i all) ...'3S

T h e Ottomantzation of the §eriat Ottoman legitimation also came to rely on a process that can only be described as the 'Ottomanizatton of the $eriat'. T h e very fact that the Ottoman ruling elite felt the need to standardize and regularize ;er'i rulings is indicative of their desire to rationalize and 'naturalize' Islamic practice. This attitude is best seen in the works of Ahmed Cevdet Pasa, and particularly in his codification offtkh rulings, the Mecel/e.39 In the introduction to this monumental work, Cevdet Pasa points out that 'civilized and advanced countries (medem ve muterakki milletler)' have constitutions that deal with civil law. However, he contends, because of changing times and the institution of civil law and civil courts (nizamiye mabkemeleri), judges were no longer trained in religious law, and yet they still had to refer to it; 'As indeed it has become difficult these days to find those talented in the }er'i science, it is equally difficult to find judges of the civil courts who are versed in the §eriat. Even more striking is the lack of properly qualified kadts, to staff the religious courts'.''0 The pasa also bemoaned the fact that Hanefi rulings were not systematically codified and existed 'in a state of disarray' {dagtmk hir halde olup), something that had now come to be deemed undesirable.41 Cevdet Pasa took pains to point out that his effort at codification was a response to the, 'need to bring the science offikb up to date, and render it applicable to the needs of modern times.' The collection of Hanefi jikb rulings in one source would, therefore, 'be a source of reference to all officials who would thus be able to accord their practice with the $er'tat'.'>2 The $eriat would, as Brinkley Messick has pointed out, thus lose its 'open ended' nature whereby it had put the emphasis on the individual jurist, who had been given considerable leeway in interpretation. The aim was a standardization of rulings which would also be disseminated to the far corners of the empire and as such 'the legal text also foreshadowed the Western notion of the responsibility to the law of the ordinary individual, the citizen.'43 Although Cevdet Pasa and his collaborators sought to standardize legal practice, in outlying distant provinces such as Iraq and Yemen the attempt to forge closer ideological effective links with the centre often met with great difficulties, particularly in Yemen which was in an almost

The Ottomanization of the §eriat

51

constant srate of rebellion. A report from the Yemen's provincial capital, Sana, dated 6 April 1884, referred to a petition of complaint prepared by one Kadi Huseyin (gagman, a leading jurist of the area. Among the kadi's many complaints was the striking appeal that Istanbul reduce the number of civil courts (ntzamiye mabkemekri) in the Yemen and leave the population to seek justice in the }er'i courts. T h e petition and the accompanying letter from San'a both stated that,'... Since the people of the Yemen be~ came Muslim some thirteen hundred years ago they have been seeking justice in ser'i courts, it is therefore injurious to the interests of both the local people and the state to insist that they use civil courts ..." It was, moreover, advised that the state take pains to ensure that it appoint to the province,"... Only those officials whose religious probity is beyond reproach.'44 It would appear, however, that even late in the century, the Ottomans were experiencing difficulties staffing government posts in Yemen. A memorandum by the vali, dated 31 October 1898, complained bitterly that although he had tried to fill lower level posts with Yemenis, this had not proved workable, "because those who speak the official language of the state are so few'.45 Yet in Yemen, while a modus vivendi was established between the Hanefi and Shafi'i mezhebs and the Ottoman administration received the active support of the Sha'afi regions, in upper Yemen they met with constant hostility from the Zaidi imams who considered them 'foreigners'. Messick has pointed out that the ideological fault lines between the Turks and the Yemeni population were determined by their respective interpretations of the §eriat: 'Refusing to allow the imams the discursive high ground, the Ottomans stoutly defended the empire's long-standing commitment to uphold the shari'a and mobilized a locally tailored sharxa politics of their own to counter that of the imams."16 In ah extensive report written by Osman Nuri Pasa, one of the most famous of Ottoman provincial administrators of the late nineteenth century, the emphasis is clearly placed on the 'civilizing' influence of the religious courts under Ottoman administration. 47 In a conversation with an Arab sheikh of the Hijaz, Osman Nuri Pasa urges him to, 'abandon his tribal habits and laws and come into the civilizing fold of the §eriat.' There ensues a very interesting exchange in which the men discuss the taking of the lives of two young lovers who have eloped without the blessing of their elders. Upon the insistence of the sheikh the exasperated pasa entreats him to see reason: 'O Sheikh! W h a t manner of law is this Arab law

ODTO KO**BAW*

52

The Well-Protected Domains

of yours to take precedence over the sacred law of the fcriatl'46 Osman Nuri Pasa was to write in 1888 that the secular courts were 'drawing the hatred and provoking the fear and timidity of the local population,' who had always sought justice in ser'i courts. It was therefore imperative that the secular courts be given up in Yemen, otherwise 'it will be impossible to fulfil the aim of the state which is to win the hearts of the people to the projects of reform.'49 In another detailed memorandum Osman Nuri Pasa recounted how he had abrogated the secular courts in the Hijaz and instituted ser'i courts in their place.50 He also gave a detailed account of his efforts to win the hearts of the local population: 'Because this (institution of the religious courts) made it possible to put a stop to the numerous blood feuds in the area they were very grateful.' The pasa also emphasized that it was advisable to use the local influence of the sheikhs and sharifs as this influence used to favour the interests of the state meant that the desired objectives could be achieved without spilling blood.' T h e primacy of the 'Ottomanized §eriat' occurs throughout the documentation of the period. On 27 June 1882 the grand vizier's office prepared a memorandum stating that all care should be taken to ensure that proper procedure and decorum be observed in the legal procedures in the Hijaz. It was determined that all cases be tried according to high Islam and very few be left over to tribal customs, A Council of Inquiry (Divan-i Tedkiki Ahkam) was to be established which was to be presided over by the kadi, and its members were to be chosen from among local experts in Islamic jurisprudence. Istanbul was to ensure that only the most qualified kadts were to be sent to the Hijag and it was specifically stated that, 'All matters should be judged according to the sacred precepts of the $eriaL'5X T h e §eriat as the embodiment of order contrasted with local custom; the irony was that the same attitude was to be reflected in the policy of colonial regimes which replaced the Ottomans in the region: 'From the point of view of the colonisers, custom had to be either standardised or abolished altogether in favour of a unified legal system. Although the Shari'a was considered disorderly relative to Western law, when compared to 'custom' it appeared orderly.'52

The Ottomanization

of the §eriat

53

T h e effort to monopolize official sacrality Another striking feature of late Ottoman legitimation is the effort to monopolize and control official sacrality. This is best illustrated by the effort to control the printing and importation of the Holy Qur'an. O n 15 December 1897, the Minister of Education Ziihdu P a p sent a memorandum to the office of the seyhulislam reporting that Muslim subjects of Iran and Russia had made official applications to for the printing and sale of the Qur'an. 53 The minister reiterated that the law of 1276 (1858-60) specifically forbade the importation and sale of Qur'ans coming from Iran: I t is well known that the Iranians have been bringing in copies of the Holy Qur'an into Ottoman dominions, particularly to the Seat of the Holy Caliphate (dar-ul biUfet-i altyye). Here they secretly print them and circulate them. It is quite unnecessary to remind you that this practice is strictly forbidden,' It was accordingly determined that all Qur'ans would henceforth be printed at the official press.5'' Although it is understandable that the Sunni Ottoman caliphate should be wary of Qur'ans produced by Shi'i hands, the same interdict applied to Qur'ans emanating from Sunni Kazan, and even the seat of high Islam, the al-Azhar medrese in Egypt: 'The importation of Qur'ans ... coming from Egypt is likewise forbidden according to long established practice.'55 The historically specific character of the Ottoman practice comes out further on in the memorandum: ... Although it seems inauspicious to forbid the printing of Qur'ans to one who is of the sunna, if we open this door it will mean that we will be opening it to any Muslim from Kazan or India or Algeria ... This will mean unforeseeable dangers for the Holy Word which has survived untarnished for some thirteen hundred years. Particularly since these are troubled times in which the foreigner's calumnious views regarding the Holy Text multiply ... The matter may go well beyond the printing of the Qur'an and, God forbid, create untold complications for the Sublime State. It was therefore determined that a commission for the Inspection of Qur'ans (Tedkik-i Mushaf-i §erif Komisyonu) should be established, and all publication of the Qur'an be undertaken only with its approval.56 The repetition of the orders over the years to prevent the printing of Qur'ans by Iranians and Russians ' indicates that the matter was never resolved and continued to be a source of worry. O n 18 October 1901, the

54

The Well-Protected

Domains

Ministry of Education pointed out in a memorandum that it had been 'preoccupied for some time with these Iranians and Russians, and if they are granted permission, this will cause a very unfortunate precedent, actually becoming a matter of state.' The Commission for Inspection of Qur'ans was charged in no uncertain terms to: ... pay close attention to this matter of extreme importance. If we open this door, the Iranian, Algerian, and Russian foreigners, even though they are Sunnis, will cause a very dangerous precedent, leading, God forbid! to the Holy Text suffering various falsifications (envai tabrifat) in these croubled times when conflicting opinions and the agitations of the foreigners are on the increase ... (ejkar-t mutenevvia vt ecnebiyenin $u zaman-t gahyamnda..)s? In referring to 'foreigners ' or 'non-Ottoman Muslims', the minister was clearly making a point which runs throughout the Ottoman documentation of this period: that the legitimacy of the Ottoman caliphate was under threat, and to a considerable extent the threat emanated from Muslims living under Christian rule who could be manipulated to challenge Ottoman legitimacy. O n e might recall at this juncture that this document was prepared at approximately the same time that talk of the 'Arab caliphate' wajs gaining currency in Western orientalist/colonial circles. The reference to 'troubled times' and loose talk about Islam almost certainly reflect the degree of anxiety caused by these stirrings of ominous portent with which the Ottomans were only too familiar. The obsession with foreign intrigue surrounding the Qur'an could be quite naive. At one point the palace actually ordered the Ottoman embassy in Stockholm to buy up any copies of the Qur'an in Sweden and Norway that they could lay their hands on, as 'it is improper for the Holy Book to be in foreign hands'. The ambassador replied in a somewhat embarrassed tone that the only copies were rare collectors' items in public libraries which would not consider selling them, and that 'public opinion here is not very well disposed to the Sublime State, making this a very delicate matter.'58 A well known opponent of the Hamidian regime, the philosopher and Young Turk Rtza Tevfik, also drew attention to the special pains the government took to control the circulation and content of books dealing with Islamic jurisprudence, noting in a somewhat ironic tone that more obviously 'harmful' publications such as the works of Buchner, Darwin, Spencer and Mill were sold freely in the Istanbul bookshops. He claims in a par-

The Qttomanization

of the $eriat

55

ticularly scathing way that even the Mizrakh llmihal, the basic school text dealing with the caliphate and legitimate government in Islam, was 'altered by that sham of a government, for which the British severely criticized them.' It is significant here that Tevfik should refer to the British, indicating that the fears of the government were not unfounded.59 The campaign against the Ottoman caliphate, and the alternative based on the Arab caliphate, gained momentum after Abdulhamid ascended the throne. Sir James Redhouse, the famous lexicographer, was to speak out in support of the Ottoman claim: The title is generally known and adopted by the orthodox Muslim world, by the Sunni church universal... Their general acceptance has probably resulted in part from a consideration of the power of the Ottoman empire, and in part from its possession of the two holy cities of Makka and Madina... one of (the sultan's) most valued titles being Khadimu-1 Haramayni-'sh-sharifayn - Servant of the two Sacred Precincts.60 Thus, like the §eriat, the conception of the caliphate became much more political than religious. It was no longer enough to be a Muslim, or indeed a Sunni. Nothing illustrates this better than the case of the Algerians seeking asylum in Ottoman dominions after the French invasion of Algeria. The issue which caused the Ottomans considerable headache was the legal status of Algerians who had immigrated to the province of Syria, ostensibly escaping from the dar ul harh. Some of these elements worked to live in Ottoman dominions while conserving their French passports, thus benefiting from special privileges as French subjects living in Ottoman lands. O n 20 November 1889, the Sublime Porte issued a decree stating that they would be required to choose within two years of their arrival whether to remain French citizens and leave, or to be automatically considered Ottoman citizens and stay. T h e Algerians of French allegiance would be forbidden to many Ottoman women, and any Algerian contravening this regulation 'would be treated according to the regulations pertaining to Iranians' and would be forced to leave Ottoman soil.61 A Frenchman from Algeria, Gervais Courtellemont, who went on the baj in 1890 posing as a convert, noted that everywhere he and his guide were followed and treated with suspicion. Upon landing in Jeddah, 'From the very first outing we were treated as suspects, and questioned as if by chance in the shops where we made our purchases ... Not a soul invited

56

The Well-Protected Domains

us in for a meal ... a bad sign in an Arab land! Yes, decidedly, I was a suspect!'62 Another critical issue in this concext was the acquisition of land and property in the Hijaz by non-Ottoman Muslims. O n 7 April 1882 the Council of State prepared a memorandum where it reiterated the ban on the acquisition of property by Indian, Algerian and Russian Muslims. The reason stated was fear of potential fifth column activities: If we remain indifferent to the accumulation of property by devious means in the hands of foreign Muslims, with the passage of time we mayfindthat much of the Holy Lands have been acquired by the subjects of foreign powers. Then, the foreigners, as is their wont, after lying in waiting for some time, will suddenly be upon us at the slightest opportunity and excuse and will proceed to make the most preposterous of claims.63 The long-time governor of the Hijaz, Osman Nuri Pasa, who served in the Yemen and Hijaz vilayets in the 1880s, was also obsessed with what he considered to be slack behaviour on the part of the government on the question of the accumulation of property in the hands of foreign Muslims. The pasa lamented the fact that 'most of the productive resources in this province (the Hijaz) are passing into foreign hands and this is having the shameful consequence of our own people working for foreigners.'64 H e also complained that Javanese and British Indian subjects were staying far beyond the time necessary for the pilgrimage, and benefiting from exemption from taxation while they studied in state schools, thus using scarce resources while not paying for them. 'It is therefore obvious that all those living in these lands should abide by the laws of the state and fulfil their tax obligations.'65 Courteilemont also noted that, 'Presently all the trades (sale of cloth, tools, manufactured objects) are in the hands of Indians and Javanese established in Jeddah and in Mecca. These people trade with the Dutch Indies and British India and no doubt, turn a considerable profit.'66 Another factor noted by Courteilemont was 'the numerous printing presses owned by Indians who are serious competition for the local presses.'67

The Ottomanization

of the §eriat

57

Ottoman versus non-Ottoman Muslims in the Hijas; The confrontation between Ottoman and 'foreign' Muslims can also be traced through the archives of the Government of India, It is ironic that Queen Victoria's title as Empress of India, and Abdulhamid's title as Hadem ul-Hartmeyn for the Hijaz used similar symbolic imagery. In the Ottoman context, this found symbolic emphasis in the official formula, 'the jewel in the crown of the exalted caliphate' (gevherA ikliU hihfet-i seniyye), which was the official title of the Vilayet of Hijaz.68 The interesting aspect of this title used in conjunction with the Hijaz, is that it is almost an echo of the 'jewel in the crown' title referring to Queen Victoria's crown in her capacity as Empress of India.69 The British consul in Jeddah wrote on 7 May 1882 that the Ottoman ban on foreign Muslims' acquisition of property in the Hijaz was causing considerable discomfort among British Indian subjects. T h e consul noted that this new state of affairs might well cause poorer Indians to take up Ottoman nationality, although 'the better to do will probably be unwilling to sacrifice British protection, and considerable annoyance is caused among them.'70 Even in the matter of charitable gifts to the holy places the sultan showed the same jealousy. The British consul at jeddah reported in a somewhat grieved tone that a silver ladder for the door of the Ka'ba 'worth 45,000 rupees' sent by the Nawab of Rampur had been turned down because ' the Sultan ... objected to a foreign subject [making such gifts] and sent orders that no such a one should have the privilege,'71 The reflection of the issue in Ottoman documentation leaves no doubt about the basis for the sultan's exclusivism: 'All such gifts can only be made by the Exalted Personage of the Caliph who alone holds the august title of Protector of the Holy Places. N o foreign ruler has the right to partake of this glory.'72 A major concern was that many of the hajis overstayed their welcome, becoming a drain on already scarce resources and, in the case of rich Indians who attempted to settle, a disturbing alternative focus for loyalty. At the end of the haj rites, town criers would announce the departure dates of caravans, encouraging people to leave. In the case of a prominent Indian nawab: 'He had brought with him great treasures with the intention of settling permanently in the Holy City. One evening after prayers in the Haram, he was approached by Abd El Montaleb [sic] who asked in very severe language why he was thus acting against practice by prolonging his stay.'73

58

The Well-Protected Domains

The same jealousy and suspicion was in evidence towards Russian Muslims. T h e initial debate over the printing of Qur'ans became con' fused with a debate about who was qualified to do so when a Muslim Russian subject had applied to print Qur'ans. 74 Like the British, the Russian government also sought to protect what it saw as the interests of its subjects. In May 1894, a Haci Mirza Mehmed arrived in Istanbul accom' panying a crate which had been sent by the Emir of Bukhara to the Hijaz. The Russian embassy intervened with the Ottoman customs authorities to allow the crate transit without opening it for inspection, as ' . . . This would break the hearts of the Muslims of Bukhara.175 The Ottoman government had to content itself with a rather lame face-saving solution: Mr Maksimoif, the dragoman of the Russian embassy, swore on his honour that the crate only contained fine cloths and 'that the crate did not contain any harmful goods (efyayt muztrra)'.76 Rulers of independent Muslim states were also a problem. O n 21 March 1891 it was reported that the Crown Prince of Morocco intended to go on the haj with a large entourage and that h e ' . . . would be bearing many gifts for the notables of the Hijaz,' Although it was admitted that it was his religious right to go on the pilgrimage, the fact that he would be bearing so many gifts 'gives rise to the suspicion that he has political intentions.' It was also stated that this suspicion arose because the ruler of Morocco had 'previously sent an envoy to Istanbul asking that he be permitted to buy property in the Hijaz and establish a pious foundation (vakf) there.' It is also significant that on this issue the Ottoman government did not hesitate to ask the assistance of a Christian power, Spain, which had designs on Morocco. The Ottoman ambassador in Madrid was instructed t o ' . . . make discreet inquiries of the Spanish Foreign Ministry to determine the truth of the matter.'77 The unstated reason for this hesitancy bordering on fear was the fact that the sultans of Morocco had always claimed the caliphate as their own.78 Egypt was another problem. Although it was still officially apart of the Ottoman Empire ruled by a special governor, since the midle of the century Egypt had become a de facto independent state.79 Furthermore, the Porte had never formally recognized the British occupation of Egypt in 1882, which put an end to its autonomy, and it was increasingly worried by British claims to defend the interests of Khedive Abbas Hilmi Pasa, who began to parade himself as an alternative caliph and protector of the holy cities. Using the young khedive as a stalking horse, the British consul, the domineering Lord Cromer, began to harass Ottoman interests in the

The Ottomanization

of the §eriat

59

Hijaz and on the Ottoman-Egyptian border in Ottoman Cyraneica (Libya).This was precisely the sort of fifth-column activity that had caused Abdiilhamid's nightmares. 80 From time to time, the palace would receive intelligence to the effect that, 'The Khedive of Egypt is sending agents into the provinces of Syria and Yemen in order to cause uprisings leading to the annexation of these provinces by Egypt.' Repeated orders would be isssued to take the necessary measures to 'stop this intrigue'.81 In this context, any development emanating from Egypt which could even in the most remote sense be injurious to the sultan's prestige in the Hijaz, was to be nipped in the bud if at all possible. In the winter of 1886 a bad harvest in Egvpt created the possibility diat the country would not be able to send the Hijaz the traditional yearly stock of provisions, and the suggestion was made that the Holy Lands should be supplied from India as an alternative. The sultan himself took a hand in insisting that, 'The provisioning of the Hijas; from Bombay smacks of British intrigues as it will inevitably mean the increase of British prestige in the area.'82 The response to threats, from colonial powers also subsumed a reappraisal of the Ottoman centre's relationship to this hitherto privileged province. A long memorandum written by Osman N u n Pasa is worth looking into in some detail at this point. 85 Dated 18 July 1885, the memorandum dwells at length on reform measures to be taken in the Hijaz and other Arab provinces. T h e pasa states that the majority of the armies stationed in the Hijaz, Bingazi and Yemen were made up of'the fundamental elements' {unsur-i asti) of the empire who were, 'the Turks and Anatolian peoples', who paid 'the blood tax', that is to say, fought in the armies. It was therefore imperative that the Hijaz be brought into line as a regular province of the empire and its population be made to contribute to its defence.84 Osman Nuri Pasa was very evidently imbued with the mobilizing ethic of the nineteenth century which informed so much of late Ottoman statecraft:

Although ir is possible to transform all of the Muslim population into a fundamental .element, events have shown that the time is not yet ripe. Even if it were possible today to blend all the Muslim tribes and nations together by causing them to lose their special characteristics through the application of rigorous policies, they would still be no more than rhe boughs and branches of the tree whose trunk would still be constituted by the Turks.85

60

The Well-Protected Domains

Indeed, the over-representation of the Turks in the armies stationed in the Arab provinces meant that they were being withdrawn from the productive labour force in their own regions: 'those who are conversant in the science of economics (iltn-i tedbtr-i servet) will know that this causes grievous loss to state revenues.' Osman Nuri Pasa also emphasized the danger of property accumulation in the hands of Javanese, Indian and Russian Muslims. 35 This 'Physiocratic' emphasis on productivity is thus combined with 'Ibn-Khaldunian' imagery and a proto-nationalistic appreciation of the nature of the late Ottoman state. 87 If the Ottoman sultan were to have credibility in the eyes of Muslims as the 'Protector of the Faithful' he had to be seen to be providing for their well being. This meant above all guaranteeing the security of the pilgrimage to the holy cities.88 It also meant that he had to ensure a just and fair basis for economic interchange during the haj season. The Hijazi natives, from the Bedouin who raided and pillaged caravans, to the local mutawwaj or guides, were notorious for overcharging pilgrims.89 Every year Istanbul would issue orders and injunctions forbidding the charging of exorbitant rates for camel hire and guidance fees'.90 The very fact of their repetition indicates, however, that the abuses continued. An additional embarrassment was the fact that non-Ottoman Muslims would appeal to the protection of Christian governments, a move that was regarded as damaging to Ottoman prestige as the protector of pilgrims.91 T h e state had to be seen to be performing its duty and due care was taken to publicize the construction of official buildings such as hostels in the local press.92 It would appear that the symbolic communication worked both ways; sometimes the Hijazis themselves would build and consecrate a kiosk, I n the honour of the descendent of the Protector of the Shari'a, our Benefactor unto whom obedience and loyalty are due.' Such an occasion was reported on 14 September 1892, on the occasion of the sixteenth accession anniversary of the sultan.93

Efforts to defend the Hijas As talk of the 'Arab caliphate' spread from the early 1880s onwards, the Sublime Porte felt the need to give special emphasis to the defence of the seat of its legitimating ideology. The most concrete symbol of this preoccupation was the Hijaz Railway.94 Begun in the 1890s, it became something of a pet scheme for Abdulhamid, who hoped to combine symbolic achieve-

The Ottomanization of the $eriat

61

ment with logistical advantage. H e took great pains to ensure that the project not be dependent on foreign capital and technology, and that each stage of its completion be well publicized in the Turkish and world press.95 Another measure considered was the creation of a special 'Hijaz Fleet' to defend the coasts of the vilayet. An order was to be placed with European shipyards for specially constructed coast-guard vessels which would be cheap to run (using both sail and steam), and would be used for policing the Red Sea and Persian Gulf coasts. 96 Although Abdulhamid deliberately set aside the large and costly fleet his uncle Abdulaziz had built up, he did invest in coastal defence and cheap but efficient coastal defence craft: 'Even otherwise critical observers of the sultan's naval policy conceded that the torpedo-boat flotilla" was kept up to scratch.'97 By the 1880s the primary external threat to the unity of the empire was seen to emanate from that ever-present bite noire of late Ottoman politics - 'British intrigue'. Sometimes imaginary, sometimes all too real, this factor was to plague Abdulhamid for all his reign. The Hijaz, distant and vulnerable, was the object of the sultan's worst fears, the most recurring theme in them being the spectre of an 'Arab caliphate'. Indeed, the Ottoman press mounted a pre-emptive propaganda campaign against any designs the British may have had in the area. In the early 1880s there were frequent references in papers like the Arabic semi-official al-Cawaib to the position of the sultan as Caliph of Islam receiving recognition in Zanzibar and Oman. This caused something of a stir in British circles. The British consul in Zanzibar reported on several occasions that he had received the personal assurance of the Sultan of Zanzibar that he had no special regard for the Ottoman sultan: 'Neither Zanzibar nor the mother province of Oman ever rendered allegiance to the Ottoman sultans ... The surreptitious attempts now being made by some of the papers (in Istanbul) are evidently designed to serve a poiiticai purpose - namely that of representing the whole of the Moslem world as subject to the nominal Khalifare of Islam.'98 The British consul in Muscat reported that he had confronted the Emir of Muscat with the al-Cawaib article and that the latter had denied any correspondence with the sultan." The poiiticai resident in the Persian Gulf also reported that the people of Oman were mostly of the Ibadi sect, and that some even inclined to Wahhabism, and that they therefore totally rejected Turkish claims.100 The Anglo-Turkish struggle was also joined at the other extremity of the Arabian Peninsula, in the borderlands between British-held Aden and

62

The Well-Protected

Domains

Ottoman Yemen.101 O n 4 October 1880, the Residency at Aden reported that the Ottoman commander of San'a had 'appropriated' several villages belonging to the Emir of Zhali, Sheikh Ali Mokhbil, a British client ruler. The resident suggested that 'the yearly subsidy' of L.50 paid to the emir should be slightly increased as his territory formed the border between British and Turkish possessions.102 A few months later the matter flared up again, and on 14 February 1881 the Aden Residency reported that 'the border issue' was reaching critical proportions as the Turks had been 'taking possession of different villages as (the) occasion offers'. This brought with it the danger that if they were not stopped, 'There is no saying to what lengths their rule might extend.'103 Of course, as far as the 'Turks' were concerned, they were doing no more than taking back what was theirs. The resident enclosed a letter from the Ottoman governor general of Yemen and musir of the Seventh Army Corps, Ismail Hakki Pasa. The letter was a petition from the local people: 'On our oath we say that even if we are ruined... and none of us remain alive, still we do not agree to be disobedient to the Porte Government, who serves the two Harams .., We will never bend our course to the English ... N o one of us shall alter the Mussulman's way and will enter the English way.' The Viceroy of India, acting on this information, decided to ratify the agreement with Ali Mokhbil and increased his subsidy from 50 to 100 pounds sterling per annum. 104 The Government of India were particularly worried about what they called, 'The long cherished plan for the extension of Turkish authority along the Arabian coast and the Persian Gulf...' This would lead to a situation where, 'Very important British interests in the Persian Gulf.,. would at once become affected.'105 T h e somewhat 'cloak and dagger' undertakings of the British in the area tended to justify the sultan's suspicions. On 21 December 1887, the grand vizier Kamil Pasa was ordered to look into the truth of the rumour that the 'Sudanese rebel army under Osman Dikna is preparing to invest Suakin and thence to attack the Hijau ...' The sultan had received information that this was ' . . . a British plot to use the Sudanese rebels to ... transfer the caliphate to an Arab government.'1"6 Kamil Pasa responded that the rumour was somewhat off the mark as the British and Sudanese had just been bitter enemies and there was no indication that this enmity had abated. The sultan does not seem to have been satisfied, because on the same day Kamil Pasa again wrote that the rumour lacked substance as the Sudanese rebels had no means of sea transport. Even if they comman-

The Ottotnanizatton of the $eriat

63

deered fishing vessels, the Ottoman gunship stationed in Jeddah was more than capable of dealing with them. Moreover, there was no indication that the rebels had any support among the Hijazi population.' 07 Some six months later the sultan's fears had still not abated. Kamil Pasa again pointed out that the rebels were not united and were not capable of mounting an amphibious operation. The paga stated that the reai danger to the Hijaz was not from the. Red Sea but from the east, from Muscat. The grand vizier was very firm in pointing out that,'... For some time now the British have been fostering good relations with the Emir of Muscat whom we have neglected. There is every indication that the government of India had been supplying him with arms.'108 Set against the information available from British sources it emerges that Kamil Pasa was certainly justified in his warnings against British activity in Muscat.

Militant mysticism Towards the end of the nineteenth century the 'legitimacy crisis' of the Ottoman empire seems to be compounded by a split in the ranks of the legitimizers, the ulema, Mardin has noted that Abdiilhamid, wary of the de-legitimizing power of the high ulema, as shown in their sanction for the deposition of his uncle Abdiilaziz, let them -sink into a morass'.109 This meant that the lower ulema Found a new avenue of mobility through the tekkes and the officially sanctioned media.110 While there was a primary emphasis on Sunni orthodoxy in the reign of Abdulhamid, one of the sultan's major propaganda weapons were the Sufi sheikhs. During the 1877-78 Russo-Turkish war, the Nak^ibendi sheikh Haci Fehim Efendi from Erzincan had personally fought in the Caucasus campaign.111 Another Naksibendi, Ubeydullah Efendi, had organized the local Kurds as an irregular force, some 2, 000 strong. Also Sheikh Ahmed Ziyauddin Giimushanevi had fought at the Batum front. These men were also active in 'promoting unity among the population through the establishment of schools.112 Yet, as the absolute rule of the sultan became established, firebrand sheikhs wandering about uncontrolled came to be looked upon with disfavour. O n 21 June 1887, it was reported from the vilayet of Ankara that a certain Dagistani, Ahmed Efendi, had established himself near Ankara and had gathered about him some 2,000 followers. The report remarked: 'This number being likely to increase ... it is quite clear that this is not

64

The Well-Protected Domains

desirable.'It was therefore determined that the sheikh be given a stipend and removed to Damascus." 3 Religious fervour and seal were acceptable as long as they did not involve a sheikh establishing himself as an independent focus of allegiance. O n 1 December 1890, it was reported that the famous Naksibendi sheikh, Giimiishanevi Ahmed Efendi had been investigated at the order of the sultan. It had been established that the worthy sheikh had something of a murky past. In the 1860s and 1870s Ahmed Giimiishanevi had established himself at the Fatma Sultan Mosque in the vicinity of the Sublime Porte, where he had formed a circle of followers. This group had become exclusively devoted to Sheikh Ahmed, to the extent of beating up other religious figures who disagreed with him. What is more, they had established themselves in the Fatma Sultan mosque to the point of: 'dwelling permanently therein and treating the Caliph's chamber in the said mosque as a sort of bachelors' lair (bekar odasi)\nA There had been repeated efforts in the past to remove Giimushanevi from the capital, but he had been saved by the intervention of various highly placed ladies to whom he had been a source of 'spiritual solace'. Giimushanevi had collected one thousand lira from a prominent lady who had been told he was going to build a tekke, yet he had proceeded to construct a sixteen-room konak opposite the Sublime Porte and filled it with his followers. Even after the building was taken away from him because he had constructed it under false pretences, his followers had gradually moved back in and re-established themselves. The report deemed it very undesirable that this sort of thing should go on. 115 Yet, despite this report, Giimushanevi remained close to palace circles and continued to preach in his tekke where he died in 1894. As a sign of special favour he was buried in the courtyard of the Suleymaniye Mosque. This special favour may well have been the result of the fact that the events discussed in the report occurred in the previous reign, and Giimushanevi was known for his opposition to the Tan;zimat reformers, AH Pasa and Fuad Pasa. Abu Manneh points out that Giimushanevi's teaching put a special emphasis on particular hadith which enjoined 'absolute obedience to the authority and total pacifism'.116 However, the more activist Khalidi Naksibendi order, which was immensely powerful, was seen by the palace as volatile and dangerous. Sheikh Irbili Mehmed Es'ad was exiled by the sultan from Istanbul in 1897 and could not return as long as Abdulhamid was in power.117 The sultan's effort to ensure that the public be exposed only to the 'approved' religion is illustrated by the instructions sent to the seyhulislam

The Ottomanization of the §criat

65

on 6 July 1883 warning him that the palace had been informed that 'impostors' were teaching in the mosques of the capital: 'These men, posing as Muslim preachers, simply don a turban and venture beyond religious subjects, some even talk about political matters.'118 It was also reported as an extremely undesirable development that 'women of good family had been frequenting Bektasi tekkes and imbibing alcoholic beverages.'119 Sufism, although not orthodox or mainstream, was a useful way of galvanizing grass-root support and reinforcing legitimacy, provided it was controlled from the centre.120 As Abu Manneh pointed out in his seminal article on the subject; 'Sultan Abd.ttlham.id seems to have been of the belief that his two predecessors ... the sultans of the Tanzimat period, neglected to create links with the common people especially in his Arab provinces,' m Abdiilhamid's 'official faith' was propagated through such ideologues as Abulhuda al-Sayyadi, Sheikh Hamza Zafir, and Ahmad As'ad. A native of Aleppo, Abulhuda was of the Rifai order and had been close to the sultan since 1876, receiving the office of head of the Sufi Sheikhs (§eyk ul-mesa'th). Hamza Zafir, a member of the Shadhili-Madani order from North Africa, had become acquainted with the sultan before his accession to the throne. Ahmad As'ad was Anatolian by origin, but had become established as a dignitary of the Hijaz. He had been instrumental as the sultan's unofficial envoy to the Urabists during the events in Egypt in 1881-82. 122 Abulhuda's main message was aimed at Syria, and there were 'some 212 titles of books and booklets attributed (to him)'. The continuous message in his major work, Dai al-ra$had li sabbii al'ittihad wa'l'inkiyad, was the legitimacy of the rule of Abdulhamid as caliph of Islam, H e deliberately focused on the issue of the legitimacy of absolute rule, denying allegations to the effect that it was a Turco-Mongol import. 123 In this capacity Abulhuda was a weapon aimed directly against the notion of the Arab caliphate. As such he earned himself few friends in British-occupied Egypt, Kamil Pasa was to state in 1899 that, 'All the perfidious statements in the Egyptian press are a reaction to the recent anti-British policy of the Sublime State, and are a result of the vengeful sentiments of the Egyptian and Arab population towards Ebuihuda Efendi.'12"1 Sheikh Hamza Zafir's main target seems to have been Africa. Trimingham has pointed out that his work, al-Nur al-sat't (the brilliant light), published in Istanbul in 1884, provided the basis for much of the pan-Islamic movement: 'From the Madani tekkes went out propaganda seeking to influence sheikhs of various orders. Emissaries, protected

66

The Well-Protected Domains .

through the imperial power, won recruits among Algerians employed by the French.. .'125 What is interesting to note here is that Abdiilhamid was attempting to do precisely what he feared the British and French would do to him that is to use Muslims of French or British allegiance as a potential fifth column. In the eyes of foreigners, the Sufi mystics had quasi-occult functions and their influence on the sultan was seen as very sinister. One of the principal biographers of Abdiilhamid has referred to, 'the sultan's superstitious belief in his [Ebulhuda's] prophetic powers ... .'n& In a contemporary report the 'religious advisors' of the sultan are described as 'unscrupulous adventurers', who 'do their utmost to terrorize him into sanctioning the present maladministration ... 'm Even an established authority such as Trimingham notes that Ebulhuda maintained his influence on Abdiilhamid through 'astrological' or 'divinatory' powers.128

Conclusion In sum, the Ottoman ruling elite was trying to foster a sense of Ottoman identity both among its own members and among the wider population. Anthony Smith remarks that, 'Aristocratic ethnic cultures persist exactly because their identities form part of their status situation; culture and superiority fuse to create a sense of distinct mission.'129 Although the elite circle itself was small, its message reached a far broader stratum. Mardin has noted that Said Nursi, a Kurdish mystic of humble origin from rural eastern Anatolia, was able to present the sultan with a petition in which he argued for the institution of secular Ottoman schools in the eastern provinces, as this would 'eliminate internal factional strife among the tribes and make them into good Ottoman citizens.'130 This is very comparable to the ideas of Stileyman Husnu Pasa, a figure much higher in the social scale who had similar plans for the vilayets of Baghdad, Mosul and Basra.151 Even in an obscure and poor vilayet such as Bitlis in 1889 the 'civilizational motif was present in spite of the fact that the place was actually seething with unrest.132 At the level of 'official belief the Ottomans made an attempt to fuse pre-existing values such as the Hanefi mezheb with new energy, and project it as the social cement for their increasingly intense relations with their subjects/citizens. In this sense Ahmet Yasar Ocak's observation sums up the matter in a nutshell:

The Ottomanization of the §eriat

67

The Islarnism of the Hamidian period ... was entirely a reaction against classical Ottoman Islam. As such, essentially it was a modernist movement. Despite all its anti-Western posturing, because it favoured of modernization, it must be considered together with other modernist movements in Turkish history.1" In this sense too the Ottoman experience is very similar to the experi' ences of other nineteenth-century powers who tried to 'reinvo[ke] the values and customs of the past to serve an ever more complicated present.'15'1 The Ottoman state was in tune with world trends where, one after the other, empires borrowed the weapons of the enemy, the nationalists. What held true for Paris in the 1870s, also held true for Istanbul: the transformation. of 'peasants into Frenchmen' paralleled the 'civilizing' or 'Ottomanizing' of the nomad.135

Chapter Three 'To Enjoin the Good and to Forbid Evil': Conversion and Ideological Reinforcement x

As the Ottoman 'punitive expedition' of October 1892 reached the first Yezidi villages on the lower terraces of Mount Sincar in the mountain fastness of northern Iraq, some Yezidi chiefs came forward with the promise rhac they would embrace Islam and guide the soldiers to the lairs of other Yezidis further up the mountain, if their villages were spared. The commander Asim Bey agreed. Soon after they set out, the Ottoman column passed through a narrow gorge, where they found themselves in a hellish crossfire from Yezidi tribesmen stationed on either side. Their 'guides' who had led them into the ambush quicidy changed sides. A hundred Ottoman soldiers were killed and many more wounded. The news was heard in Mosul on 4 November and the military music in the government compound stopped.2 The Yezidi Kurds were one of the many groups that the Hamidian state attempted to integrate into the mainstream population and expose to the Ottomanized $eriat. Thus imbued, these elements would also be more reliable as soldiers. Yet it was not enough to simply press people into service. In the short term this might prove effective, but the new ideological requirements of the state meant that, in the long term, the population had to be convinced as well as coerced.' In other words, the people had to be made to believe in, or at least acquiesce to, the legitimation ideology of the ruling power. This was done through a systematic policy of conversion to the 'official faith', the Sunni Hanefi mezheb. This policy was nothing less than a long-term bid to integrate elements which had been 'nominal Muslims' like the Bedouin, or 'heretical sects' such as the Yezidi Kurds, into mainstream ideology. Although officially Christians were not forced to convert, documentation indicates that Christian conversions increased in the period. Just as post-revolutionary France had worked to make 'peas-

68

Conversion and Ideological Reinforcement

69

ants into Frenchmen', the Ottoman elite wouid attempt to transfotm 'nomads into Ottomans' while not neglecting to buttress the orthodoxy of the sedentary population.'' As the Ottoman Empire found itself increasingly short of usable manpower, the urgent requirements of defence meant that the sultan had to squeeze the population for the last reserves of fighting men. This chapter will examine how the state attempted to instill the new orthodoxy in various strata of its population.

T h e Yezidis: a case study of Ottoman conversion policy The campaign carried out in 1891-92 to convert the Yezidis to the Hanefi mezheb is in many ways a microcosm of Ottoman policy towards elements which had hitherto been considered marginal and left more or less to their own devices. Standard practice usually followed a set pattern. First, an 'advisory commission' (heyet-i nasiha, or heyet-i tejbimiye) would be sent, usually including local notables or respected religious leaders. These were expected to get results through what would today be called the 'carrot method', a system of persuasion helped by decorations, bribes, sometimes invitations to leaders to come to Istanbul where, if they were lucky, they would be kept under comfortable house arrest. This method of employing 'lenient and moderate means' (vesait-'t leyywe vc mutedite) was preferred because it was cheaper and far less destabilizing than the military option. If this did not achieve the desired result, only then would the Ottoman centre resort to sending a 'punitive expedition' (heyet'i tedibiyz)? The obvious problem with this second method was that the centre often had very little control over the amount of violence used, and this often resulted in massacre and wanton destruction. The Yezidi Kurds had inhabited the mountainous regions of eastern Anatolia and upper Mesopotamia from time immemorial. Their religion, attracting some attention in the late nineteenth and early twentieth centuries as 'devil worship', with its obvious exoticism, held great appeal for travellers to the Orient. 6 As far as the Ottomans were concerned, Yezidis were a 'heretical sect' (Jtrak-t dalle) who could be called to account through Hanefization: 'The Yezidis, Kurds but not Muslims ... represented an anomaly in the mind of the pious Sultan.' 7 Accordingly plans were made to assimilate them in the Cossack-inspired 'Hamidiye' regiments of light auxiliary cavalry then being formed in eastern Anatolia.3 In 1885 their exemption from military

70

The Well-Protected

Domains

service was lifted and they were put on the conscription registers. Yet their attempted full integration into the Ottoman scheme of things seems to have awaited the fateful spring and summer of 1891. O n 18 May 1891, the 'advisory commission' (heyet-i nasiba), headed by a Major Abdelkadir, which had been sent to the §eyhan region to 'convince' the Yezidis to perform military service, reported that they had gathered together all the Yezidi chiefs and imparted to them the decision of Istanbul, 'in a language that they would understand'. 9 But, these worthies wrote, although the attitude of the local population was favourable, little could be accomplished with their leaders, 'who have always led their people astray and have manipulated them to serve their selfish ends'. It was therefore recommended that, if these people, who numbered some 150,000, were to be called to account, their leaders had to be banished as soon as possible to areas of the empire where there were no Yezidis.10 When the Yezidi leaders telegrammed Istanbul complaining that Major Abdelkadir was using brutal methods, the major wrote that this was proof of their 'perfidious nature'. H e stated that the Yezidi leaders had given him a written guarantee that they would fulfil military obligations, but were now attempting to renege on it. It was reiterated that 'the correction of the beliefs of these people can only be successful if their leaders are

exiled.'11 One month after this the palace was informed that a particular trouble maker was the mir (chief) of §eyhan, Mir Mirza, 'without whose removal the aim of correcting their beliefs (ta$hib~i akaid) is a hopeless task'.12 O n 25 June, the sultan's A D C , $akir Pasa, evaluated the situation for Abdulhamid, stating that military service was a sacred duty incumbent on all Muslim subjects, and that it was unheard of that any group of people should b e ' extended a special invitation and be asked to give written reassurances'.13 T h e pasa also stated that military service should be given first priority, the 'correction of the beliefs' of the Yezidis could be undertaken once they had been included in the ranks.14 Yet it seems that, in the eyes of the sultan, religious orthodoxy was indeed more important than military service: the Yezidis must convert. The Muslim Yezidis in the area of Hakkari did indeed volunteer for service in the Hamidiye regiments. The view of the palace, however, was that as 'these regiments are made up of elements who are of the Sha'afi sect and whose religious resolve and loyalty is known, it is out of the question that the [heretical] Yezidi should be considered on equal footing with them.' 15

Conversion and Ideological Reinforcement

71

I n July 1891, an ambitious O t t o m a n general, O m e r Vehbi Pasa, was posted to the vilayet of M o s u l as t h e ' c o m m a n d e r of the reformatory force' (firka-i islahiye kumandani). Yet still the 'soft option' was favoured. O n 13 O c t o b e r 1891 a Yildtz Palace m e m o r a n d u m stated t h a t 13,000 kuru} h a d been spent on the construction of a m o s q u e a n d a school in a Yezidi village called P a t r a k w h e r e t h e majority h a d converted t o Islam. 16 However, by the s u m m e r of 1892 O m e r Vehbi Pasa h a d already decided o n more direct m e t h o d s . In J u l y - A u g u s t he began a p r o g r a m m e of intimidation which involved terrorising Yezidi villages in t h e $ e y h a n area, some of which h a d resisted the call to Islam. T h e s e were raided a n d severed heads were b r o u g h t back to Mosul. 1 7 T h i s intimidation b r o u g h t a delegation of s o m e forty Yezidi chiefs to t h e town. O n 19 August, t h e pasa reported t h a t he h a d gathered t h e local administrative council (meclisi idare) a n d in a public ceremony called on the chiefs to embrace Islam. S o m e refused a n d were beaten, at least one died of his wounds. 1 8 T h e remainder, headed by the M i r All Bey, converted in a ceremony t h a t was duly reported to Istanbul by t h e pasa as a great success:

After repeated unsuccessful attempts through the centuries to bring them back to the true path, eighty villages of the Yezidi and thirty villages of the Shi'a have acceded to the honour of the True Faith. Yesterday, their leaders, with total freedom of conscience, accepted my invitation to come to Mosul and become Muslim.This morning, as the military band played the Hamidiye march, and rank upon rank of ulema intoned the holiest of prayers proclaiming the One True God, a great crowd of notables and military personnel gathered around the municipality offices. As a guard of honour stood to military salute, the miiftu asked each one in turn if he accepted Islam of his own free will. Upon each confirmation the crowd shouted, 'Long Live the Sultan! (gadifahim fokyapa).19 Despite his brutal measures, O m e r Vehbi Pasa also intended to p u r s u e education as a means of securing the long-term loyalty of the Yezidis. T h e day after h e reported the conversion ceremony (merasim-i telkiniye) he wrote to Istanbul t h a t he intended to build six mosques and seven schools in Yezidi villages a n d five schools a n d five mosques in the villages of the Sebekli, all of w h o m h a d converted to Islam. 20 T h e pasa also wrote t h a t the Yezidi Sanctuary at Lalis should be converted t o a M u s l i m medrese a n d twenty pupils should be given scholarships to study there u n d e r a w o r t h y S u n n i scholar who would also be funded by the state. It was i m p o r t a n t t h a t local ulema should be employed in the

72

The Weil-Protected Domains

schools of the area because they would be familiar with the language and customs of the people.21 The vilayet council, probably at the pasa's prompting, telegraphed Istanbul in much the same vein, pointing out that it was advisable to use local ulema as they would be 'familiar with the character and language of the local people, well versed in political subtleties as well as religious dogma ' It is worth pointing out that this formula, or close versions of it 'emzke ve elsine-i mahalliyeye afina ve dekaytk't siyasiyeye dana ve hakaytk'i diniyeye vaktj..,' is often repeated in Ottoman documentation dealing with measures against heterodoxy.22 It was deemed at this point appropriate not to disregard the soft option, and on 26 August the Ottoman Council of Ministers (Meclis-i Viikela) discussed a telegram received from Omer Vehbi Pasa, in which the latter suggested that leading Yezidis be decorated for their praiseworthy conduct in accepting Islam. T h e pasa had claimed that if these chiefs were decorated and given a pension, 'they would promise to bring the Yezidis of the entire empire as well as those of Iran and Russia into the sacred fold of Islam.'23 In mid-September Omer Vehbi Pasa reported that the teachers and prayer leaders assigned to the Yezidi villages had taken up their posts.2'1 Yet things were not progressing to the satisfaction of the ambitious general. O n 24 September, he reported that Mir AH Bey and a few other 'troublemakers and traitors' had gone back to their 'hateful state of ignorance' and had reneged on their vows to Islam. These 'evil men who had been used to milking the villagers for years by parading bronze peacocks and other such craven idols among the villages and thus collecting large sums for their own use' had to be put down.25 At this point the Yezidi leaders further complicated matters by contacting the French consul in Mosul and telling him that the community was prepared to embrace Christianity if France could protect them against Omer Vehbi Pasa.26 This caused an outraged pasa to cable Istanbul to report that some of the Yezidi leaders who had been 'confounded and frustrated' by the state's efforts to bring civilization and education, 'thus preventing them from swindling and fooling the local population', had now appealed to the French consulate for protection. 'The effort being made to bring them back to the true path from which they have strayed through ignorance and heresy, is being presented to the foreigners in a different light, thus enabling them to continue their evil ways.'27 Acting on this information, in Istanbul the Council of Ministers recommended on 7 October that the Yezidi chief, Ali Bey, and others who

Conversion and Ideological Reinforcement

73

had 'misled' their people should be exiled to Ottoman Tripoli.28 Bv now the alternating 'carrot-and-stick' policy applied by Omer Vehbi Pasa was having some effect. At the end of September it was reported that eight villages of 'heretics' from a sect called the 'Harh' in the sancak of Zor, and the Mosul area had presented themselves in Mosul 'of their own free will', and announced 'that they had become Muslim',29 The Council of Ministers decided to approve the vilayet's suggestion that the leaders of these villages, together with the leaders of the Sebekli who had already converted, should be given low level decorations and be paid a pension of 1,000 kuru}. It was clearly stated that all of these 'achievements' were the result of 'auspicious developments of the era of our caliph' (mubassenat-i asriyye-i bilajetpenahi)',i0 On 11 October, the Vali of Mosul, Osman Pasa, asked for: '300 copies of the Qur'an (kelam-i kadim), 700 primary school readers and 700 pamphlets teaching Islamic beliefs, for the instruction of the children of the Yezidi and Sebekli who had converted to the Hanefi belief.' It was said that this was the 'fruits of the exalted pious path (asar celile-i takva}iarikr i)' taken by the sultan. 31 At around the same time, in mid-October, Omer Vehbi Pasa ordered a punitive expedition into Yezidi territory. The commander of the force was the son of the pasa, Asim Bey, a hotheaded young officer. The force cut its way through Yezidi villages in the Seyhan leaving death and destruction in its wake.32 T h e sanctuary at Lalis was sacked and the ritual artefacts were removed. It was to be transformed into a medrese. This was followed by the expedition to the Sincar mountain where the Ottoman force was ambushed and suffered heavy casualties at the hands of the Yezidi hillsmen." By the end of October 1893, the new Vali of Mosul, Aziz Pasa, telegraphed that he had carried out his orders to inform Istanbul of the circumstances of the 1892 defeat. H e reported to his superiors that the tactical error which had been committed by Omer Vehbi Pasa was the use of violence against the Yezidi who had in fact been peaceful. This had caused them to combine forces with their more militant tribesmen on M t Sincar. The new governor also suggested the application of the old methods of first sending in an advisory commission 'made up of locally influential men*. Only if this did not yield results would it be necessary to 'punish [them] in an exemplary manner'." The importance of the 'advisory commission' was particularly stressed and its participants told that the sultan would reward them generously if

74

Tfee Well-Protected

Domains

they were successful in bringing about the surrender of the rebellious chiefs without bloodshed.35 O n 1 December the governor confirmed that most of the leaders of the seventeen Yezidi villages in rebellion in the Sincar area could be made to repent through advisory commissions.36 Although an uneasy truce was restored to the Yezidi areas, it was fairly obvious that the Ottoman centre did not envisage anything like a return to the status quo ante. T h e long term hope was to keep the region quiet, while inculcating Hanefi orthodoxy through a network of schools. In the last days of 1893, the Yezidi leader, Mir Mirza Bey, applied to the vilayet of Mosul for the return of the sacred relics which had been removed from the Sheikh Adi sanctuary during the raids of 1892. H e was told in no uncertain terms that it was out of the question that these should be given back to him, or to any other Yezidi chief, and that the sanctuary had now definitively become a mectrese}7 Meanwhile, complaints regarding the excessive brutality of Omer Vehbi Pasa, and his bullying of the local notables, had been reaching the sultan throughout 1892 and 1893. A commission of inquiry was duly established and it reported from Mosul on 24 November 1892 that Omer Vehbi Pasa had 'undertaken actions totally contrary to imperial interest and ran the risk of plunging an area as important and sensitive as Iraq into a state of chaos.' He had exceeded his orders and had 'indulged in excesses beyond his authority (ijrat vt tefrite kacttgi).'18 O n 28 November, the chiefs of staff commented that Imperial instructions were to attempt to disperse the rebels without spilling blood, and only to resort to force if the [Ottoman troops] were fired upon ... .'39 A few weeks later, the commission had a remarkable report to submit: some twenty Yezidis had presented themselves before it with the grisly evidence of seven severed heads, which they claimed had belonged to men slaughtered by Asim Bey's forces.The commission felt that this was scandalous behaviour and 'that such acts present an ugly spectacle to friend and foe.MQ Extremely damning evidence was brought to light by the commission in Mosul who reported on 6 February that the so-called 'conversion ceremony' reported by Omer Vehbi Pasa was a total sham: 'not one of the Yezidi leaders in fact converted sincerely, and those who did, did so to avoid the public beating and insult which was meted out to their peers ... Therefore the pasa's report that these people had, one by one, converted by word and heart is totally without foundation.' The commission also reported that the commander's forces had 'undertaken such actions totally contrary to the imperial wishes as beatings and torture of villagers in

Conversion and Ideological Reinforcement

75

$eyhan and Sincar in order to secure their conversion.'111 O n 19 August, the palace ordered that the commission of inquiry dismiss Omer Vehbi Pasa from his post and that he be ordered back to Istanbul to face a court martial.'12 It seems that the Yezidis had still not been included in the ranks of the Hamidiye regimenrs by December 1893. At this time the vilayets of Mosul and Van were ordered to carry out a census of Yezidi villages and to continue the efforts to educate their children.43 The ideological seesaw between the Ottoman centre and the Yezidis was to continue. As late as January 1914 they had still not been subjugated. O n 25 January 1914, the vilayet of Mosul was to report that the Yezidi chiefs had accepted that they would be issued with Ottoman identity cards, but on the condition that the section declaring religion should read 'Yezidi'.44 The Yezidi story remains a curious one. In many ways it is a microcosm of the Ottoman attitude to marginal sects whom they wanted to convert ideologically so that they could be utilized as reliable members of the population. The 'carrot and stick' policy was obviously subject to abuse by the men on the spot who received mixed signals from the centre. O n the one hand, they would be told 'not to spill blood' and to settle the matter at hand through the good offices of'advisory commissions'. On the other, a strong willed general or officer would be ordered in no uncertain terms to procure the submission and conversion of unruly elements. Added to this there were the logistical problems of communications and deployment of troops in hostile and inaccessible terrain. In such circumstances considerable initiative fell to the commanding officer, yet he was told that he must ask for instructions before undertaking any decisive action. This confusion lies behind much of the massacre and counter-massacre that went on in the eastern provinces of the Ottoman state during these years.

T h e correction of the beliefs' of Muslims and marginals It became the practice in Abdtilhamid's reign for the corps of religious teachers to be used as a sort of religious secret police-cum-missionary organization whose religious training and credentials would be determined by the centre. 'Freelance' preaching was decidedly frowned upon and seen as dangerous. By 1892 it was established that preachers and 'missionaries' (da'iyan) would only be allowed to preach after they had graduated from four years

76

The Well-Protected Domains

of study and received a diploma (riius) from the seyhulislam.The preachers in question were to include the sons of prominent religious functionaries (zadegtm) who, upon the successful completion of their studies, would be given the highest diploma, the Istanbul ruusu.*6 It would appear that this policy was implemented throughout the Hamidian era. As late in the reign as 1902, the seyhiilislam's office was to write to the palace, reporting that the latest crop of daiyan had graduated from their special medrescs, and were now 'ready to take up teaching posts in the provinces'. The letter expressed special thanks to the sultan for his award of one thousand lira out of his privy purse to be apportioned among the teachers.47 T h e enforcement of orthodoxy went together with conversion of heresy. Even elements who were nominally Muslim were not beyond the searching gaze of Istanbul. O n 2 December 1884, the Council of Ministers, acting on the sultan's orders, arranged to send twenty religious teachers to the vilayet of Syria, to inculcate the 'true' Islam into the local population. This was deemed necessary as 'the Islam of the local population, sedentary and nomad alike, consists of nothing more than something they inherited from their ancestors, indeed, their Islam does not go much beyond their having Islamic names."18 It was necessary that such unthinking practice be replaced by 'a consciousness of the honour of being Muslim'. This would prevent them being 'duped by missionaries and Jesuits who set out to deceive them.' It had been determined by a special commission in Damascus that special teachers, familiar with the local vernaculars and customs should be sent out to travel with the nomads and bring them back to the true path. In settled areas schools and mosques would be established. The Council of Ministers declared:'This practice should also be extended to Yemen and Baghdad, thus teaching the people the rules and practices of Islam and ensuring their obedience to the caliphate'.'19 Yemen in particular continued to be a problem area where Hanefi 'official belief was besieged by what Istanbul perceived as the heretical beliefs of the Zaidi, a Shi'i sect.50 This made it all the more remarkable that the Vali of Yemen should write on 2 November 1898, that some of the leading !Zaidi dims had asked for a Hanefi medrese to be built in San'a. The vali pointed out that this request was somewhat 'unexpected' (hilaf-i memut) coming from these people, 'who have not been renowned for their loyalty'. Yet the vali recommended that the medrese be opened as the Zaidi of the towns were 'relatively more moderate' and susceptible to reason than were the 'savage tribes'. Education, he pointed out, achieved 'gradual but lasting results'.51

Conversion and Ideological Reinforcement

11

A report from the same province dated a year later left little doubt as to the aims of Ottoman primary education in Yemen: 'The people of Yemen are ignorant of Islamic history, particularly of the great works and deeds of our August Master, the Caliph. The programme of primary schools in the region should be of a nature to reinforce the Hanefi mezhcb, and instill the feeling of obedience towards the sultan ... .' The same report stated, however, that where the Sha'afi mezbeb was predominant, school religious textbooks should accommodate Sha'afi doctrine.52 This practice was to continue to the end of the century and well into the 1900s. On 20 July 1899, the Yildiz Palace recorded that an imperial order had been issued for the sending of twelve qualified teachers to the Ma'an valley in the southern part of the vilayet of Syria, to work among the Bedouin. It was declared that this was necessary because 'these people are entirely bereft of the light of Islam, and have to be taught the religious sciences and the rules of Islam.'53 It is particularly interesting, in the light of this order, that when the Bedouin of the Karak region in the Ma'an valley revolted against the Ottomans in 1911, they selectively targeted the Hanefi schools and mosques for destruction, leaving such obvious targets as the Christian and Jewish quarters relatively undamaged.54 Nor was the orthodoxy of the Turkish heartland in Anatolia taken for granted. In May 1899, the $eyhiilislam's office in Istanbul ordered preachers (va'iz) to be sent to the district of Mihalicik near Ankara because 'It has come to his Imperial Majesty's attention that the people of this place are completely ignorant of the §eriat and the Sunna.' In the same vein it was decided to send preachers to Trabzon, as 'the people of Trabzon and its environs have strayed from the true Islam and have ventured down the path of ignorance.' In both Ankara and Trabzon the assignment of the preachers was the same: 'to teach religion and to correct beliefs' (ta'alim-i diyanetk tashih-i akaidkri),'55 It would appear that something of a religious crisis occurred in Anatolia at this time. In the late 1890s there was a very real fear that in many areas the population might convert to Christianity as a result of missionary activity. As it turned out this was a totally unfounded panic. But the question greatly worried Ottoman officials at the time because m many of these areas the people had converted from Christianity in the late eighteenth to early nineteenth century. The crisis also mtermeshed with the Armenian massacres which brought intense pressure from the West for the Ottoman government to fulfil its pledges of reform.56 The position of 'General Inspector for Anatolia' {vtlayat-i sabane mufettis-i umumisi) was ere-

78

The Well-Protected Domains

ated in some measure as a safety valve against Western pressure. T h e first general inspector was none other than the most trusted of the sultan's ADCs, Ahmed $akir Pasa.57 O n 16 January 1898, the vilayet of Trabzon reported to $akir Pasa that, '[it] had heard that some people in the region had the intention of converting back to Christianity (irtidad) as a result of 'the deception of the priest.' An investigation had been launched in the districts concerned but 'because of the great size of the monasteries in the area and the numerous priests' it was advisable that this investigation keep a somewhat low profile as 'the priests might otherwise show things in a different light, and they are not above complaining to foreign officials.'56 A year and a half later the vilayet of Trabzon was to report that the rumours had come to nothing. Yet, considerable furore had been caused. 59 N o r was this furore unfounded. After the Ottomans officially admitted freedom of religion with the fertnan of 1856, there were indeed cases among recent coverts where attempts were made to return to Christianity. A striking example is that of the Istavri or the Stavriotae. The Istavri were Greeks of the Black Sea regions of Trabzon and Gumushane who had converted to Islam at various periods of persecution since the end of the seventeenth century, but had remained crypto-Christians, openly declaring their faith only after the guarantee of religious freedom granted by the Hat-i Hiimayun Reform Edict of 1856,60 Some had resettled in the kaza of Akdag Madeni near the central Anatolian town of Yozgad in the 1830s. A contemporary Western account referred to them as: Crypto-Chrisdans proper, belonging to the Greek rite and Greek by speech, [who] also existed till recent years in the neighbourhood of Trebizond: they were known as the 'Stavriotae' from a village Stavra in the ecclesiastical district of Gumushane. They are said at one time to have numbered 20,000 in the vilayets of Sivas, Angora, and Trebizond: now all have returned to the open profession of their faith.61 When $akir Pasa personally inspected the situation in the summer of 1897 he referred to the Istavri as 'a people of two sets of religious practices (iki ayin icra eder ahali)',62 As such they represented the height of anomaly in the Ottoman system.65 In earlier periods, they had simply been bypassed; but now, when the empire had to squeeze every last resource in order to survive, like the Yezidis and other hitherto marginal groups, they found themselves the focus of the unwelcome attentions of the state. They amounted to a considerable population numbering some 'eighty four

Conversion and Ideological Reinforcement

79

households' in Akdag Madeni itself and were also to be found in the surrounding villages amounting to a total of 'some one hundred and forty households'. They had taken Muslim names and done military service, thus 'appearing to be Muslims while secretly remaining Christian'. Their community leaders had maintained contact with the Greek Patriarchate in Trabzon, to the point where one of them, a Mahmud Efendi, was clandestinely referred to as 'the representative of the patriarch'. In 1879 they had openly applied to return to Christianity, thus becoming exempt from military service; but permission had been denied, causing $akir Pasa to remark: 'this then led them into devilish machinations '(bir mecrayt seytamye siiluk fie)',64 These 'machinations' consisted of refusing to register their births, deaths and marriages in the population registers, and burying their dead in the Greek cemetery, They had also taken to openly giving their children Christian names, and marrying into the Greek community. Sakir Pasa felt that all this was very dangerous as 'such an example can cause confusion in the minds (tag$i$~i ezhan) of other simple Muslims'. Yet the inspector general of Anatolia recommended nothing worse than the temporary exile of the community leaders and the reprimanding of the Patriarch of Trabzon. H e added that 'reliable imams should be sent to their villages, they should be severely adjoined to send their children to school and give them Muslim names.'65 However, it appears that the istavri continued their errant ways. Some months after Sakir Pasa left the area, the mutasamf of Yozgad was to report that the Istavri continued to marry among themselves and bring up their children in the Istavri tradition. 66 The Istavri became something of a personal obsession for Ahmed Sakir Pasa. More than a year after his visit to Akdag Madeni, he still wrote lengthy reports on 'this heretical people (erbab-t datalet halk)', H e pointed out that although they were relatively few in the region of Yozgad, in the area from which they originally came, the kaza of Turul in Gumiishane, ehey were equal in number to the population of an entire township (nakiye), Their hope was to avoid military service and 'procure the friendship and protection of Europe'. 67 Instructions had been given to the authorities in the area that 'they be told that the only result of their changing religion would be that they would be cursed by their neighbours', and would still not be exempted from military service. The pasa noted that it was a waste of time to try to preach to those 'whose minds were already poisoned', but the target should be their children, 'whose beliefs should be corrected (tashih-i akaid)'. The

80

Tfee Well-Protected

Domains

papa's methods for forcing the parents to send their children to school were somewhat insidious and smacked of blackmail: On the face of it the ostensible reason is that their children should be educated. If the [parents] persist in returning to Christianity, they should be told in an unoffkal manner that their children liable for military service will be compelled to serve in the military units of Yemen, Trablus (Libya), and Hijaz. This method is sure to do away with any such inclination and they will become as zealous in their beliefs as the Pomaks of Bulgaria.68 T h e pasa was also conscious of Western pressure. H e pointed out that it was necessary to build and operate schools not only in the Istavri villages but also in the whole region so that 'the complaints of the foreigners can be met with the answer that the compulsion to send their children to school applies equally to all.'69 T h e Istavri continue to come up in despatches well towards the end of the Hamidian period. The vilayet of Ankara reported on 10 May that an Orthodox priest by the nameof Kirilus had been active in fomenting sedition among the Istavri. The report claimed that the said Kirilus posed as an Ottoman subject was but actually a Greek from the island of Corfu. He had been actively collecting money for the Greek nationalist organization, the Philiki Etheria, from among the Istavri community and had, it was claimed, actually caused the conversion to Christianity of some Muslims.70 One of these, a certain Sofracioglu Omer, had married a Greek girl after his conversion, On hearing this, the authorities had conscripted him and sent him off to serve in the army. Yet unredeemed, upon return he moved in with his wife's family and began attending church. All this was enough to cause the Vali of Ankara considerable grief: 'It is absolutely without doubt that the continuation of this state of affairs will engender countless dangers for the Sublime State and the Holy §er'tat.The despicable acts of these people are like a wound oozing pus into the healthy flesh of the Muslim population.*71 T h e year 1905 saw a concerted effort to solve the Istavri problem. The vilayet of Ankara was ordered in no uncertain terms to register the Istavri as Muslims. The minister of the interior, Mahmud Memduh Pasa, together with §akir Pasa made this into something of a pet project. T h e protestations of the patriarchate and the foreign embassies proved to be of no avail. In several instances, the vilayet of Ankara and the mutasarnf of Yozgad actually declared that they had granted freedom of religion. This proved, however, to be a subterfuge, designed to force more Istavri

Conversion and Ideological Reinforcement

81

into the open. The priest Kyrillos Caratzas and several Istavri leaders were marched to Ankara where many died in prison. Freedom of worship only finally came to the Istavri after the revolution of 1908 and the deposition

ofAbdiilhamid.72 . The Balkan Muslim population, greatly reduced since the empire was shorn of most of its Balkan provinces, was also a focus for 'correction'. Towards the end of 1889 it was reported that 1,600 people in the district {nabiyt) of Elbasan, in Ottoman Albania, had declared that they were Christian, and therefore not eligible for military service. Istanbul ordered that a local prefect (mudur) be appointed to the area and that these people 'be brought into the circle of salvation of Islam' through the appointment of teachers and the construction of schools. The state would thus 'establish contact with the population', who were to be expected to supply soldiers as they had in the past.73 A group which had always been suspect in the Balkans were the donme, the Jewish converts to Islam. O n the 24 April 1892, the Ottoman Council of Ministers discussed the case of a donme girl, Rabia, who had fallen in love with her tutor, a Haci Feyzullah Efendi. It was stated in the minutes of the council that, 'Rabia, a girl of some eighteen or twenty years, having developed a relationship with Haci Feyzullah has run away to his dwelling and there declared that she has converted to Islam and abandoned the avdeti faith.'74 What is immediately striking here is that the Ottoman official documentation refers to the matter as conversion (thtida), although the donme were officially Muslims, The girl took refuge with the municipality mayor and her family demanded her return. The assessment of the council was that 'although the Avdeti are called Muslims, and should therefore not object [to the marriage) they have always abstained from intermarriage with Muslims.' In the end it was recommended that the couple be brought to Istanbul by the first boat, where Feyzullah would be provided with some form of employment in a government office.75 Serif Mardin has also drawn attention to the 'quite novel proselytizing in the Balkans' by the Melami and Naksibendi Sufi orders as the borders of the Ottoman Empire shrank rapidly in the nineteenth century.76 The Kurds who served as auxiliary forces in the eastern Anatolian provinces, the Hamidiye regiments, were also to be instructed in the 'correct' religion by the institution of schools and mosques at the expense of the sultan's privy purse.77 The Kurds who were deemed to 'have strayed from the true path' were to be put back on the straight and narrow by the printing and distribution of'religious guides' (ilm-i hal), which would be printed

82

The Well-Protected

Domains

in Turkish and Kurdish and would be distributed among them. 78 A group of unorthodox people who had always drawn the ire of the Ottoman ruling class were the Kizilbas, or the 'Red Heads', so described for their traditional head covering.79 Known as a problematic branch of the Shi'i faith, these elements could not fail to attract the attention of the Hamidian centre at a time when emphasis on orthodoxy was so pronounced. O n 4 January 1890, the Ministry of Education was instructed to send preachers and ttm-i hal to the Kizilba$ of the vilayet of Sivas.This was necessary because 'the number of Kizilbas in the area, while once quite small, has recently increased day by day as a result of their ignorance.'80 The mutasarrif of Tokad in east-central Anatolia was ordered in 1891 to undertake a census of the Kizilbas. population in his area. H e duly reported that 33,865 Kizilbas lived there. His instructions were that these people were to be 'rescued from their ignorance and shown the high path of enlightenment' through the appointment of preachers and the distribution of religious tracts. In the meantime, the imams in these villages were to be summoned to the provincial capital where they were to be 'trained' in the local rusdiye secondary school. Also, 'advisors' (nasih) were to be sent to the villages for extended stays, as 'if they are left in the villages for some time they can be mote effective in saving these poor pagans who have not had their share of salvation.'81 O n 2 August 1891, the Governor of Sivas wrote a long memorandum dealing with the Kizilbas. H e stated that owing to the neglect of the previous reign and its officers (an opening that was sure to get Abdulhamid's attention), 'the numbers of the ignorant and deviant have increased greatly'. H e then went into detail on their 'pagan' religious practices such as 'worshipping dry twigs which they esteem greatly as a branch of the tree of paradise.' 82 T h e pa§a had a much more serious lament, however: the Kizilbas also believed that to die in battle did not automatically open the door to paradise, as was the credo in the orthodox Muslim faith. This made it obvious that 'as together with everyone else, they serve in the regular forces, this superstition, which has possessed them from an early age, will make them highly unreliable in war.' Therefore all effort was being made, in keeping with official instructions, to establish the 'new style' schools in their villages and to train village imams as instructors as 'it is more likely that we will be able to rescue the young from the pit of sin and educate them into abandoning their fathers' beliefs than those who have reached the age of thirty or forty whose eyes have been darkened by heresy.'83 This dichotomy between 'darkness' and 'illumination' is indeed a Qur'anic motif,

Conversion and Ideological Reinforcement

83

as was noted by Hasluck who devoted considerable attention to the problem of conversion to Islam: 'A convert to Islam is not unnaturally regarded as a person specially illuminated by God, being thus enabled to see the true faith in spite of the errors of his upbringing.'84 Another people who had been victims of 'errors in their upbringing', and became a target of the conversion policy were the Nuseyri. A sect found in northern Syria, the Nuseyri in the region of Latakia attracted the attention of Istanbul in 1889. A payments register for the Ministry of Education records that, 'because the efforts of the Christian missionaries have not been lacking among them, it is established that seven primary schools should be established in the area in order to invite them into the fold of Islam,85 Indeed, Abdulhamid II appointed an imam for the Nusayris in order to lead them in orthodox prayer.B6 By March of the next year, school construction had advanced and AH Resid Efendi from Trablus (Tripoli) 'renowned for his learning and loyalty' was appointed as inspector of the schools in the area,87 Throughout these years the Nuseyri continue to come up in dispatches. In June of 1891 the mutasamf of Lagkiye reported that the Nuseyri notables of the Sahyun district had called on him and presented him with a petition declaring that they had voluntarily accepted Haneft Islam, and now wanted schools and mosques, 'to confirm thern in their faith and to instruct them'.88 Inspector Sheikh Ah Efendi had been present and had recommended that fifteen schools and as many mosques were needed in the area otherwise, 'if these people are left in their forlorn state of ignorance, this will butter the bread of the missionaries who will then be able to tell them, "you see, your government cannot help you".'89 Although the Ottoman Government considered the Nuseyri as heretical or deviant Muslims, the missionaries considered them a 'pagan sect' and protested against the closure of missionary schools in their area.90 The Nuseyri were also to attract the attention of the indefatigable Ahmed Sakir Pasa. During a tour of inspection of the vilayet of Aleppo in mid-1898, the Nuseyri of the kaza of Antakya (Antioch) had approached him with petitions complaining that their conversion to Hanefl Islam had not been accepted by the notables of the area, who would not allow them in the mosques.91 Sakir Pasa had looked into the matter and had come to the conclusion that the attitude of the notables was due to the fact 'they use [the Nuseyri] as a slave-like labour force and if they become Muslim they will no longer be able to do so.' He deemed it absolutely necessary that the notables be compelled to accept the conversion of the Nuseyri:

84

The Well-Protected Domains

'[If this is not done] a tremendous opportunity will be wasted. I have recently seen mention in the European press that thirty thousand Niiseyri have miraculously converted to the True Faith at the merest gesture of His Sacred Majesty. T h e political value of this development cannot be squandered in the narrow interest of a few landowners.'92 It is interesting that the Pasa put the onus on the international propaganda value of the sultan's position as Caliph of all Muslims. Syria continued to be the focus of Ottoman missionary zeal. Another obscure sect in the Lazkiye district who were deemed 'in need of having their religious practice and belief reinforced' were the Hidaiye. O n 19 July 1896, the Sublime Porte recorded that schools had been built in their area and able teachers had been appointed. It was also decided that the primary schools in the districts of Markab and Cebele were to be upgraded to secondary school (rusdtye) status, and a total of some forty schools in the area were to be regularly supervised by an inspector.93

Conversion of Christians The process of conversion from Christianity to Islam is one of the most vexed issues in Ottoman studies. It seems from the evidence that forced and arbitrary conversion was officially scorned, while it probably went on informally. Even according to an unsympathetic Western account: 'Under the Ottoman Turks at least there is very little historical evidence of conversion on a large scale in Asia Minor. So long as the rayahs were not dangerous, they could be 'milked' better than true believers, and conversion en masse was to no one s interest. During the earlier part of Sultan Abdulhamid's reign, the possibility of Christian subjects serving in the ranks was seriously considered. A memorandum prepared by Grand Vizier Said Pasa, on 26 February 1880, pointed out that the Greek Metropolitan in Istanbul had suggested that Christian and Muslims serve in separate units. This was dismissed as dangerous by the Council of Ministers who pointed out that 'in the Cossack and Dragoon units [of Russia] Christian and Muslim serve together in brotherly fashion.'95 It was also pointed out that in the anti-Austrian resistance during 1875-76 'Latins and Muslims fought side by side'. The sultan, however, became increasingly suspicious of arming his Christian subjects, on the grounds of security and also because the inclusion of Christians in the ranks would mean the loss of the considerable revenue gained from the military exemption tax (bedel-i askeri) levied on them.96

Conversion and Ideological Reinforcement

85

However, the evidence from the Ottoman archives indicates that Christian conversion did become more frequent during the last quarter of the nineteenth century. Indeed it is possible to speak of a late Ottoman conversion policy regarding Christians, the basis of which was that conversion was only admissible if carried out through 'proper channels' and followed 'proper procedure'. As in the cases of the §er'iat and the Haaefi mezheb, what we have here is a 'bureaucratization' of the conversion procedure. This procedure is spelled out clearly in the Ministry of Interior's frequent circulars to the provinces dealing with complaints about conversions (ibtida) which had taken place 'against accepted practice' (usul-u car'tye kilaftnda). The central idea informing decision-making on this issue was to show that the conversions had been voluntary. This point did not escape the attention of a British traveller who noted:

The whole procedure that is prescribed jn cases of conversion to Mohammedanism from any form of religion is judicious, moderate, and calculated to distinguish between real and forced conversion, and to give the former coreligionists of the convert every opportunity of satisfying themselves that the conversion is voluntary.97 The accepted practices had some striking features. One of these was that, 'during the act of conversion the highest ranking local priest belonging to the confession of the convert [has] to be present'.98 It would appear that what was happening here was a 'bureaucratization' of an earlier practice. Cypriot sources from the seventeenth century testify to the fact that in Cyprus when a Greek converted to Islam, the practice was that a priest should be present to ascertain that there was no coercion, and to 'strike the convert off his list'.99 According to the nineteenth century-regulation, together with the priest, the convert's parents or next of kin should be in attendance. The documents testifying to the act of legitimate conversion were to be signed and sealed by Muslim and Christian officials alike. The procedure should not be hurried, and if a few days delay was required for the priest or the next of kin to arrive, the conversion ceremony was to be put off. Only those children who had reached the age of puberty were allowed to convert. Also, in the case of girls who came to the ceremony veiled, 'the veil should be lifted to ascertain identity'. If the convert was above the age of puberty, they should not be forced to go back to the home of their parents, and should be settled in 'suitable Christian or Muslim homes'.100 The careful

86

The Well-Protected

Domains

legalese of the circular report can, however, be read in another way. This and many other circulars repeat the same thing: frequent cases of illegal conversion were being reported in the Greek and Armenian communities. The Greek Patriarchate frequently complained that members of its congregation had approached it with tales of woe involving underage children converting as a result of being abducted.101 The abduction of young girls seems to have been particularly rife. The legal advisors of the Porte were to note that, 'often these children are kept in the imam's house where they are made promises that if they encourage others like them [to convert] this will be in their interest,' What seems to have been happening was that convert children were being used as decoys for others to follow in their footsteps (emsalintn celhi i$in kendiknne icrayi mevaid ve traiye-ifevaid).101 This was seen as truly reprehensible behaviour and the legal advisors bemoaned the fact that provincial administrators were lax in the punishment of those who abducted young girls. It was also seen as reprehensible that in the case of children who were under age, 'these children are hurried off to Istanbul and this causes loose talk and complaints.' Another problem was the case of separated parents. If one of the parents accepted Islam the children would be free to choose which religion they were to belong to after the age of fifteen, 'according to the usage of the past forty years'.103 In the matter of deciding the age of young converts a clash occurred between the Greek Patriarchate and the Ottoman authorities. T h e Patriarchate demanded that the baptism certificates should be the basis of determining whether the convert was of age, whereas the Ottoman authorities maintained that the legal basis for such procedure were the Ottoman birth certificates.104 By 1913 the minimum age for conversion was increased from fifteen to twenty.105 The Armenian crisis in Anatolia provided the background for many of the incidences of conversion. O n 3 March 1895, it was reported that a sizeable population (some 200 households) of Armenians had converted to Islam in Birecik, in central Anatolia. As this was at the heart of the 'Armenian question', the British and Russian embassies had taken a close interest in the matter. The Ottoman government had therefore consented to send a mixed commission to the area consisting of two officials from the vilayet of Aleppo, and the dragoman of the British Embassy, Fitzmaurice. The commission had interviewed leaders of the convert community who were named as, 'ex-Gregorians Hacik Efendi, now called

Conversion and Ideological Reinforcement

87

Mehmed §akir Efendi, and Abos Efendi, now called $eyh Muslim Efendi, ex-Catholic Hacikebuzan Efendi now called Mehmed Nevri Efendi, a Protestant notable Abraham Aga, now called Ibrahim Efendi.'106 The leaders of the community openly stated that 'the recent events had caused them to fear for their lives and that was why they became Muslim.' They also promised that they would not convert back to Christianity once the danger was over: 'also the conversion of the church which they had made into a mosque was done entirely at their own expense ... and they had no intention whatever of converting it back into a church.' Indeed, if any of their number wanted to return to Christianity, they declared, those people would be refunded the sum they had spent on the church/mosque and told to go on their way. The Armenians interviewed were then asked to sign the report written by the commission. There was apparently some disagreement between the Turkish members and Fitzmaurice as to the wording; the Turkish side wanted the wording: 'and no matter what happens even when they are completely safe or have gone to a safe place they (promise) not to abandon Islam.' Fitzmaurice objected stating that this was against the freedom to change religion guaranteed by the law and was mortgaging these peoples' future. The Turkish side consented and the final report read: 'for the present time they do not intend to abandon Islam.' The community representatives, the report recorded, also stated that 'they had not been subjected to any pressure or force in accepting Islam but had only acted out of fear of recent events, being obviously in a state of great distress and poverty after the recent calamities and the sacking of their property.' The report ends here and is marked 'conforms to original'. Yet there is something remarkable in this document. At the bottom of the page, under the seals of the signatories, a scrawled sentence reads: '18 §ubat 1311. Judging from the demeanour of some of them it seems that they intend to return to their old religion,' One can only surmise that once Fitzmaurice's back was turned after he had signed it, a Turkish member of the commission hurriedly penned the addition.107 O n 14 May, the sultan was presented the report of the commission with an accompanying note from the grand vizier stating that Fitzmaurice had 'been given permission to join the commission'.108 O n 28 May the Vali of Aleppo was asked for his opinion on the matter. The Ottoman official reiterated that the Armenian converts had converted 'only because they feared for their lives' (strf muhafaza-i hayat mahadma musteniddir)}09 H e pointed out that the state had to choose one of two options. If it wanted to allow the Armenians to remain where they were, and revert to Christi-

88

The Well-Protected

Domains'

anity, it was necessary to 'execute some among those of the Muslims who had taken part in the killing and looting, in order to frighten the [rest of the population]' and to counter popular outrage at re-conversion. This would only work, however, for the urban areas. Those who lived in the countryside would still not be able to circulate among the Kurdish tribes who were their neighbours, 'which is in fact the very reason for their insistence on the sincerity of their conversion'.110 T h e other option was for the state 'not to officially acknowledge their conversion but to leave things to time, as in their present state they can go about their business in peace'. Eventually, those among the converts who 'wanted to return to their original religion can, with the state's permission, be moved somewhere far away from Birecik where they can live as members of their original faith'.111 This document is remarkable on two counts, Firstly, the Ottoman official is quite prepared to consider the exemplary punishment of Muslims. Secondly, the state's official position towards the converts is very far removed from the classical Islamic position where it is permissible to kill someone who abandons Islam. The vali clearly stated that the Armenian converts insisted on the sincerity of their conversion in order to be able to live among their Muslim neighbours, 'which they would not be able to do if they reverted back to their old faith/ presumably meaning that they would be killed according to the classic precepts of the feriat. The vali was quite prepared to admit that the Armenians could be allowed to return to Christianity simply if they removed themselves bodily to some place where they would not be known. 112 By early July it seems that the situation had once again become tense. T h e vali reported on 16 July that, 'The Birecik Christians, including members of the administrative and civil courts, have taken off their turbans and reverted to their old faith.' They had also taken back the key to their former church. Tension was mounting in the kaza: 'Those among the people who are zealous Muslims are already grumbling that the conversion of these people was not acceptable, as there is also evidence of foreign intrigue it is to be feared that the smallest incident might spark off very untoward events.' T h e vali's tone was worried and he called for regular troops to be sent up from Adana, 'with the shortest possible delay'.113 T h e vali was right to point to foreign interest in the matter. O n 12 May 1896, the Russian embassy expressed close concern for the fate of the Birecik converts, and declared that if they were allowed to return to their original religion the news would be well received in Europe.114 About a year later the matter still evoked enough interest in London for Lord Salis-

Conversion and Ideological Reinforcement

89

bury to be asked a question in the British parliament on the issue of'forced conversion'. Salisbury answered that the British, in consultation with all the other embassies in Istanbul, had urged the Ottoman government to allow the Armenians to return to their original faith.115 A similar case was reported from the vilayet of Ankara on 30 May 1891. An Armenian woman convert, having gone back to her native Yozgad in central Anatolia to see her child 'whom she had left there before her conversion', was waylaid by the local priest 'who tricked her into going to his dwelling'. There, members of the local Armenian community put pressure on her to convert back, and threatened to kidnap and remove her to the town of Sis. A police official was sent from Ankara, but he abstained from marching in to the priest's quarters, 'in order to avoid giving the impression of a forcible removal.' He had instead arranged for the local headman (muhtar) to intervene and had the woman brought to the residence of the local muftu. It was decided in Istanbul that 'it is in no way permissible to hand over someone who insists on her loyalty to Islam.' As in the case of the donme girl, it was recommended that the woman and her husband, who was a policeman, be brought to Istanbul.116 Naturally, conversion was taken quite seriously by the Armenian authorities, who kept close watch over such cases. On 3 July 1872, it was reported from Mus, that a woman, Peros;, had two husbands, one in Bitlis and one in Mus. Having run into difficulties with her first husband, who announced that she might no longer consider herself his wife, she decided to become Muslim. 'We tried to change her mind at the government palace for two hours but could not convince her and now the Dajigs want to marry her off to a third person, but we forbade them, saying she already had two husbands. Now we hear that she wants to return to Christianity. We ask you (the Patriarchate in Istanbul) to petition the Sublime Porte ... so that the woman may escape the claws of the Muslims.'117 The Ottoman records also mention cases such as that of a certain Mehmed, an Armenian convert from Gumushane. The remarkable thing about Mehmed, was the fact that the sultan personally paid for his circumcision, and arranged for him to be employed as a cleaner at the hospital where he was circumcised and to receive a salary of 300 kuru$.m A legendary case of abduction and forced conversion is the story of Armenouhie Kevonian, a young Armenian girl from the Mus. region who was abducted by a Kurdish Sheikh, Musa Bey, in 1889. Forced to convert by the Kurds, young Armenouhie secretly retained her faith, and international pressure forced the Ottoman government to bring Musa Bey to court in Istanbul.

90

The Well-Protected Domains

Although Musa was actually acquitted, causing considerable outrage in Armenian and European public opinion, the girl was restored to the Armenian community.119 As the tensions between Muslims and Armenians mounted in the 1890s, the Ottoman centre became more attentive to the smallest detail that could influence inter-community relations. On 3 September 1890, the Vali of Sivas reported that the Armenians of Sivas had applied to build a church and a school on property belonging to them. H e recommended that they be allowed to do so on the following grounds: 'Because Muslim and non-Muslim live in mixed neighbourhoods in Sivas, there has been no opposition to the building among the Muslims. Because the Armenians generally tend to live in separate neighbourhoods, the fact that they ate mixed in with Muslims here, makes for good relations (hiisnu hiltat) and the maintenance of law and order is thereby facilitated.'120 Presumably the pasa was aware of the fact that inter-communal violence occurred more readily when communities were segregated, making it easy to stigmatize the 'other'. T h e matter of conversion 'by due process' continues to be the dominant theme throughout the period. O n 4 February 1903, it was reported that a number of Christians from Alkus, in the vilayet of Mosul, had been rounded up by a Sheikh Muhammed Nur, marched into Mosul, and there converted to Islam. It was stated that 'although they publicly signed a petition asking to convert, some of them secretly made it known that they were not sincere in their conversion and had only undertaken such an act out of fear of Sheikh Muhammed.' The sheikh had also 'pressured some of the local ulema to expedite the regular process, and a priest who should have been present, was not sent by the Patriarchate.'121 A few days later, it was to be reported that the local Islamic population of Mosul were in a great state of excitement because 'the due process of law had not been carried out during the conversions, and the local Patriarchate did not send a priest,'122 O n 14 February, the Vali of Mosul reported that the local Muslim population had attacked the government building. It was said that this was because: 'the vali had accepted the conversion of the Alkus Christians who had converted against their will, without the official procedure being observed (muamelat-i resmiye ifa olunmakstztn).' There is a very distinct air of foreboding in the document, and the Ministry of Interior declared that it feared for the vali's life.123 On 3 November 1891, a group of Christians from Antioch complained that 'they had been denied the honour of Islam' by the officials of the

Conversion and Ideological Reinforcement 91 town. Istanbul ordered that the matter be investigated, particularly in relation to what denomination the petitioners belonged to, as ' there is freedom of religion'. Strict instructions were given that, 'they be treated according to accepted practice and regulations were also given.'124

Conclusion In examining Ottoman conversion policy and ideological control in the Hamidian era, the interesting aspect of this rather murky story lies in the different ways in which the Ottoman elite perceived the various minorities and specific cases. Their reaction to the recalcitrant Yezidi was very different from their reaction to the dbnme girl who could have complicated matters if she was to remain in her native Salonika. The Armenian converts and potential 're-converts' were a much more serious problem and one which achieved international proportions. The Kizilbas. had always been a thorn in the side of the Ottomans, but the response of Selim I in the sixteenth century had been very different to that of Abdulhamid II. The former could afford to contemplate the massacre of large numbers of Kizilbas, whereas the latter's main concern was the question of their reliability in the army. It is particularly significant that even sections of the population who were ostensibly 'mainstream', such as the people of Ankara and Trabzon, were seen as being in need of 'ideological rectification' in a way that was not wholly dissimilar from that imposed on the Bedouin of the Ma'an valley, Another fact which emerges from the evidence is that the $eriat ruling making it admissible to execute Muslims leaving the faith (murted), was no longer officially practised. After the ferman of 1856 officially allowed everyone the freedom of religion, the Ottomans had to officially admit Christian missionary activity. This led to a fear among the Ottoman ruling classes that the religious base of state legitimation ideology was slipping from under them. Ahmed Sakir Pa§a was to state unequivocally that 'As the basis of power {us-u saltanat) is the Muslim population, the power of the state will increase in proportion to the number of its Muslim subjects ... Therefore we must do our utmost to prevent the defection of those such as the Istavri and to accept the conversion of those such as the Nuseyri.'125 The evident desire of the Ottoman centre was to stabilize the religiousethnic status quo. There was nothing new in this policy. W h a t was particular to the times was the constant affirmation and re-affirmation of

92

The Well-Protected Domains

what should have been the norm anyway. In this context the impression derived from the documentary evidence is one of worry -cum-panic among the administrators. In the immediate aftermath of the Armenian massacres and general upheaval in Anatolia in the late 1890s, the large number of orphans was a great cause of concern to the Porte. It was feared that both Muslim and Armenian orphans would be the target of Protestant missionary activity. An imperial order, dated 1 August 1897, reflects this attitude quite clearly as it provides for the constitution of a mixed commission whose specific brief was 'to prevent the spread of the influence of foreigners and to ensure that each remains in his own confession, and to prevent the conversion of His Imperial Majesty's subjects to Protestantism as well as other foreign creeds.' The commission was to be made up of three Muslim officials, (one each from the seyhulislam's office, the Ministry of Education, and the Ministry of the Interior), two Armenians, and one Greek Orthodox. 126 In the event, Muslim conversion or re-conversion to Christianity remained minimal, but this was not something that could have been taken for granted at the time. A memorandum prepared by the sultan for Ahmed Sakir Pasa reflected this state of mind: 'Although the Sublime State cannot force anyone to accept Islam, we can never tolerate the conversion of Muslims to Christianity.'127 It is now time to turn to the issue of how the desired legitimating principles were to be instilled in the subject/citizen population through education.

Chapter Four Education: The Answer to all Evil?

The wiry little man looked distinctly out of place in the imperial audience chamber of the Yildiz Palace, like a small bird of prey that had accidentally flown in through the window, His Imperial Majesty warily eyed the slight figure clad in Kurdish hillsman's garb. He had been told that the young man before him, already a sheikh with a considerable following, was a potential asset for Islamic mobilisation in Anatolia, a matter close to his heart. Yet, something bothered him in the fellow's manner, particularly his way of looking the Commander of the Faithful straight in the eye. The sultan's disrurbance was to turn to shock as the man, introduced as Sheikh Said-i Nursi from Bitlis, addressed him in the familiar form (sen) and went on: 'You have been a passive caliph. Anatolia cries out for schools. Van is overrun with missionaries who have provided excellent schooling for the Armenians. Why is rhere not a Kurdish university on the shores of Lake Van?' Clearly the man had taken leave of his senses. He was immediately set upon by the chamberlains and bustled our of the imperial presence. The dignitary who had introduced him hastened to make profuse apologies and point out that the young Kurd must be deranged, Yet that evening, when the excitement of the day had died down, the sultan reflected on the event, and in his heart he knew the Kurd was right.

In the second half of the nineteenth century the Ottoman Empire came into its own as an'educator state' with a systematic programme of education/indoctrination for subjects it intended to mould into citizens. 1 Together with the Russian, Austrian, French, British, German and Japanese empires, the Ottoman Empire set about creating what Hobsbawm has called 'a captive audience available for indoctrination in the education system', in a 'citizen mobilizing and citizen influencing state'. 2 Education "* had always been an integral part of the Ottoman statesman's 'mission»

93

94

The Well-Protected Domains

^civilisatrice' since the Tanzimat reforms early in the century; but in the iHamidian era mass education was extended to the primary school level.3 As in other imperial states, the main aim was to produce a population which was obedient, but also trained into espousing the values of the centre as its own. In this sense legitimist monarchies were definitely adopting the road taken by their ideological enemy, the French Revolution. As Eugene Weber pointed out in his pathbreaking book 'teaching the people French was an important way of "ciyjlizing''„them'.4 Count Uvarov, the Russian'm'inister foFeducation responsible for the policy of'Russification' of non-Russian peoples in the empire, could not have agreed more when he told Nicholas I (1825-55) that the whole basis of his rule was 'Orthodoxy, autocracy, and nationality'5 Indeed, the Russian example was very much on the minds of Ottoman statesmen like Ahmet Sakir Pasa, who advocated mixed schooling in primary schools for Muslim and non-Muslim, using as a precedent Russian educational policy in Poland, Tataristan and the Caucasus. Sakir Pasa also advocated the exclusive use of Turkish as the language of instruction, as 'linguistic unity was basic to national unity'.6 In an Ottoman context this translated itself into the administration of ever increasing doses of the approved religion through the school system in an effort to induce conformity. It also meant combating rival educational systems such as that of the Christian minorities and the missionary schools.

Ottoman education policy in application O n 2 March 1887, the seyhulislam's office presented a memorandum to the Yildtz Palace dealing with an imperial order to revise the curricula of secondary and higher schools. The memo stated that, 'for a nation to securely establish the basis for its religion the matter should be tackled at its source/ 7 The seyhulislam's view was that religious education was sorely lacking in primary (ibttdatye) and secondary (rii}diye) schools, as was knowledge of Arabic, 'the pillar of the Ottoman language'.8 Higher secondary/middle schools (idadi) and the Imperial Civil Service School (Mekteb'i Mulkiye'i §ahane) were also to have their curricula revised and additional hours of religious instruction inserted. The same was true for the Imperial Military Academy (Harbiye), the School of Medicine (Tibbiye), and the School of Engineering (Miihendishane). In these at least the seyhulislam was prepared to admit, 'although we are not

Education

95

qualified to judge their scientific teaching matter it should be sufficient to fortify religious belief by the addition of reading materials of a religious nature.'9 The Galatasaray Lycee, an elite high school on the French model, and with a heavy emphasis on French language, was singled out by the seyhiilislam because of'its special nature'. The students of this school he stated, 'had to be kept away from Western books harmful to the interests of the Sublime State.' Also the fact that some Muslim schools were teaching in foreign languages was a decidedly bad thing and, 'should be avoided altogether'. Only approved books would be translated into Ottoman. 10 The director of the Galatasaray Lycee added that, 'Latin and Philosophy lessons should be removed, nothing should be taught relating to the lives of European philosophers.' Religious instruction should be increased and entrusted to teachers whose loyalty was beyond question and who would instruct the students on the life and deeds of the Prophet Muhammed and his followers." The memorandum also included a detailed breakdown of the lessons to be taught in the five-year rufdiye. In the first year, religion and Arabic accounted for eight out of twelve hours of teaching per week. The Turkish grammar lesson was also to be based on a 'book of Islamic morals'. Geography and 'the globe showing European countries' was only to be introduced in the third year and that for only one hour out of twelve. The whole of the five years' instruction was to be very heavily weighted in favour of classical Islamic learning. Geography was to be taught only in the third and fourth years for one and two hours respectively. At each Ramadan the students were to memorize by rote the works of Abu Hanifa, and before each recess prayers would be said for the long life and health of the sultan.12 Maps of the world were also carefully inspected to ensure that no official claims to territory on the part of the Ottoman State were omitted. On 7 April 1898, it came to the palace's attention that, 'maps of the Asian continent which were being used in primary schools included certain errors as to the Asiatic frontiers of the Sublime State.'13 It was duly ordered that the maps be confiscated, and new maps drawn. These latter were only to be distributed to schools after they had been submitted to the palace for inspection.14 This parochialism was reflected in a denunciation of the history master of Galatasaray, a certain Midhat Bey, who was roundly criticized for 'going beyond the set book and talking about ancient Egypt, Greece and Rome'. It was stressed that his behaviour was particularly offensive as 'the

96

The Well-Protected Domains

first duty of every instructor should be the correction of the beliefs (tashihi akaid) of his students.' The instructor was also accused of having what would today be called an 'attitude problem': 'he has been known to make light of the four sacred schools of Islamic jurisprudence by mentioning them in the same breath as the idolatrous gods of ancient Greece. He also makes fun of the other instructors as well as the Christian religion.'15 The man was probably very popular with his students. The emphasis on training a loyal and competent state elite, which would be thoroughly imbued with the values of the centre, was a basic consideration of the Ottoman higher educational establishment. As stated in a circular directive from the Ministry of Education to all the higher schools, students graduating from these institutions were expected to be 'of good character and breeding, ready to serve their state and country unwaveringly'.56 At the Civil Service School (the Mtilkiye) in particular, all instructors were ordered to 'take every opportunity to instill the sacredness of duty to the state, which must be the first thing in the consciousness of every civil servant.' All instructors were ordered to stress religious lessons and Islamic jurisprudence (Jikh). As in the case of the Galatasaray teacher mentioned above, all were specifically and sternly enjoined, 'not to venture into topics beyond the curriculum'. 17 Even topics which were not religious were submitted to strict censorship. The Ministry for Education was ordered to inspect the international law textbooks used at the Mulkiye, and 'to delete any harmful materials'. A commission was duly established to inspect the course textbooks used in the school, and to report on those which were found objectionable.18 In another elite institution, the Imperial Naval Academy (Mektcb-i Bahriye)' cbe religious probity of the student body became a matter of close concern for the sultan. The school authorities were asked to provide the names of the books used in religious instruction classes, and both the names and the titles of the instructors of religion. The palace also wanted to know whether or not prayers were being said regularly and where the students attended Friday prayers. The school authorities assured the palace that the utmost attention was devoted to religious probity and that signs which outlined daily religious obligations were posted everywhere in the school under the title 'General Behaviour' (harekat-t umumiye).19 A general circular was sent out from the palace to all military secondary schools demanding to be informed about the nature of religious instruction enforced. T h e Minister for Military Schools, Mustafa Zeki Pasa, replied on 23 January 1898 that students who missed one prayer

Education

97

were punished by being confined to grounds on their days of home leave. Students who missed their fasts during Ramadan were also punished by being imprisoned in the school.20 Another way of attempting to prevent the spread of 'harmful' ideologies among the young was to prevent them from studying abroad. In the last years of the Hamidian regime this inclination became more acute. Particularly in Salonika, where the crypto-jewish dbnme elite had accumulated considerable capital, and could afford to send their sons to Europe, this was seen as a problem. The inspectorate for the vilayets of Rumelia was to note that: '[These young people] go to Switzerland and France to study and acquire experience in commerce. Failing that they attend the schools of commerce in Salonica belonging to the French, Italians, Greeks or Rumanians. Either way this is detrimental to their Islamic and Ottoman upbringing.' T h e inspector suggested that additional classes in commerce and finance be added to the curriculum of the elite Terakki and Fevziye schools in Salonika, which most of sons of the local leading families attended, in this way'preventing the pollution of their Islamic morals by study abroad'-2E The intensification of religious instruction was to spread to all the elite educational establishments by the turn of the century. The Ministry for Military Schools was to attribute to this policy, '[The fact that] the survival of the Sublime State depends on the preservation of the Islamic faith,'22 The most immediate instrument for the 'survival of the Sublime State', namely, the army, also became a focus for religious education. It was confirmed on several occasions that 'the rank and file are having the approved religious texts read to them at regular intervals', to instill in them the 'sacredness of their duty'.2* As far as the 'sacredness of duty' was concerned, Abdulhamid's efforts do not seem to have been in vain. As Carter Findley has expressed it: 'To the extent that an effort to train a new type of civil official had occurred under Abdul-Hamid, however, the effort was his, made chiefly through the School of Civil Administration.'2,5 Findley points to the beginnings of 'a new professionalism' in the Ottoman administrative cadres.25 Another extensive memorandum prepared some eight years later was even more explicit about instilling a sense of its own mission in the Ottoman ruling class: 'It is His Imperial Majesty's wish that men of great firmness of religious faith and national zeal be trained to believe that their happiness is one with the eternal Sublime State.'26 The memorandum declared that 'everyone should be required to have an education while pre-

98

The Well-Protected Domains

serving his religion and traditions.... Also all forms of progress in a state are contingent upon the training of students and in the preservation of their religion and national honour.'27 Therefore utmost attention had to be paid to teachers, so that the wrong kind of teacher did not poison the minds of the future generations. Despite the conservative tenor of these instructions and directives, it would appear that the actual programmes that were formulated were more realistic. Although a table specifying the curricula for schools from primary to Miilkiye, was heavily weighted in favour of religious studies, Qur'an reading (tecvid) always having a preponderant place from primary school to the highest level, more practical considerations were also at work. In the provincial idadi middle schools the programme was also to include courses like cosmography, general and Ottoman geography, Economics, geometry and agricultural science. Moreover, in some areas provision was made that some of these could be taught in Arabic, Armenian, Bulgarian or Greek. In the Arab provinces, particularly in intellectual centres like Damascus, the government schools continued to be centres of learning, and were greatly influential in the formation of the Arab elites who would later go on to become the first leaders of independent Arab successor states. 28 French language and translation also seems to have continued to figure in the curricula of Galatasaray (despite the best efforts of Recai Bey) and in the Miilkiye curriculum, in addition to training in Arabic, Greek, Bulgarian and Armenian. Another aspect of the instructions was the strict supervision of students 'lest they show signs of wayward and unreliable behaviour.'29 The extensive reports and memoranda written by Ottoman officials stationed at various outposts of the empire shed very useful light on how these men perceived 'national' education. Osman Nuri Pasa, the longtime governor of the Hijas and Yemen vilayets, left several such memoranda. 30 The pasa's view of education was in general directly influenced by Enlightenment thinking: 'The people of a country without education are like so many lifeless corpses, not benefiting humanity in any way (maarifsiz memleket ahalisi cemiyet-i btferiyyt'ye neftyyi olmayan ecsad-i mejluce gihidir)! The pa§a maintained that although religious schooling was 'the only basis for the protection and reinforcement of Islamic zeal (asabiyyet-i hlamiye)', rational and political learning acquired through 'modern education' was also essential for the continuing strength of Islam. The medreses of the Holy Lands were sorely lacking on this score, and were in any case 'full of

Education

99

Indians, Javanese and Turks, which means that the learning dispensed here only benefits foreigners.'31 The Dutch Javanese Muslims, British Indian Muslims, or Turkish subjects of the Russian empire, who filled these medreses in the holy land, like those non-Ottomans in the Hijaz, are perceived very clearly as foreigners, interloping on scanty Ottoman educational resources.52 In matters of taxation Osman N u n Pasa bemoaned the fact that the foreigners in the Hijaz paid no taxes, complaining: 'It is a source of great sorrow that the foreigners, who already control much of the land and real estate of the Holy Lands, benefit from exemption [from taxes] even more than the local inhabitants.' H e advocated the taxation of the foreigners and recommended that the proceeds be used to lighten the load of Ottoman taxpayers elsewhere.33 The local Arab tribes should be 'gradually civilized' through education and 'thus by their own hands brought to destroy their [savage] customs.' Also, the sons of local Arab notables such as sheikhs, sharifs and other highly placed religious functionaries should not be allowed to inherit their fathers' positions without first passing through the Ottoman schooling system and acquiring the appropriate diplomas. The teaching of Turkish was of extreme importance as, 'one of the primary means of blending in the various tribes and peoples of a state is the unification of language (tevhtd*i lisan).m Bismarck and Mazsini would have fully agreed. A more cynical view of educational policy was put forward by Ahmet Sakir Pasa during one of his tours of inspection in Anatolia. The pasa strongly advocated giving priority to the establishment of primary schools in regions whose loyalty was doubtful.35 Another target of Ottoman 'special schooling' was the Shi'i population of largely Shi'i vilayets such as Mosul, Basra and Baghdad, The attempt to indoctrinate Shi'is involved children of modest backgrounds. A letter from the governor of Baghdad, dated 30 December 1891, stated that in keeping with the sultan's instructions, ten Shi'i children were being sent to Istanbul from Baghdad and Kerbela. Yet the governor had seen fit to include two Sunni boys, 'to set the minds of the Shi'a at rest and show that these children were going to Istanbul to study as the result of die Sultan's generosity.' As the families who volunteered were poor, six of the boys were to have clothes made at the state's expense and the sultan paid their travel expenses and the fees of their travel chaperones. It is worth quoting from the document at length as it openly declares that the aim of the exercise was to train future propagandists/missionaries:

100

The Well-Protected Domains

Since so much money has been spent [on these children] it is important that the necessary benefits be derived from their education. The training of those among them who are Shi'i should ensure that they abandon this sect and become Hanefi, in order to enable them to convert their countrymen to the Hanefi sect upon their return.*6 The Iraqi provinces of Baghdad, Basra and Mosul come up very frequently in dispatches relating to education. Geographically remote, and religiously suspect because of the heavy Shi'i presence and the proximity of Iran, the Ottoman valis in these provinces adopted something of a siege mentality when it came to educational policies. In this vein, the Vali of Mosul, Ismail Nuri Pasa, reported to Istanbul on 11 November 1892 that 'Mosul constitutes a strategic barrier between the Shi'is in Baghdad and Iran.' Its population was mostly Hanefi or Shaseafi and was therefore deemed reliable. The effort to convert the Yezidis and the Sebekli had been a success. T h e Yezidi shrine of Sheikh Adi in Seyhan had been converted into a teacher training school for Hanefi scholars. Education in Turkish should be emphasized even more as this was a vital matter. The most revealing aspect of the dispatch lies in the vali's use of military imagery in educational matters: If the light of education is made to shine in these provinces, this will provide a line of defence and a screen of resistance for our frontiers against the covetous glances of the foreigner... This will block the roads against the illegitimate sect of Shi'ism by erecting educational defence works in its path.37 Clearly, the ibtidaiye and rupdiye schools that the imperial centre was encouraged to institute in the region were envisioned as so many fortresses against foreign penetration. The military imagery went hand in hand with the mission civilizatrice, as the vali did not neglect to stress that the tribes, 'who live in a state of nomadism, savagery and ignorance', had to be 'brought into the community of civilized peoples'.38 W h e n it was a matter of this or that element of the Muslim population misbehaving, education was always seen as the first cure, as an alternative to the military option. When the vilayet of Mosul reported that a Kurdish tribe, the Pesderli, had been ransacking the countryside, and asked permission to send in troops, the instructions from Istanbul were quite explicit: 'Before troops are sent against them, in order to avoid spilling Muslim blood ... their leaders should be advised and admonished ... moderate

Education

101

methods (vesait'i leyyine) aimed at winning them over ... like schools ... should be made use of,'39 That education was the alternative to 'the stick' if only by a hair's breadth, is brought out in the following statement by an Ottoman local official addressing the population of a village who were resisting the establishment of a gendarme post: 'With people like you only two things are possible. One, schools in which to educate you to see the necessity of law and order ... the other is the stick. Now, schools take fifteen years to produce a man such as I want; the stick is a matter of five minutes ...'40

The Tribal School Much of what officials like §akir Pasa or Osman Nuri Pasa advocated was taken up by Abdiilhamid II. One such initiative was the establishment of the Tribal School or Mekteb-i Agiret. As a vehicle of inculcating Ottoman patriotism in Arab and Kurdish boys, a special school was established by imperial order {irade) and opened its doors on 3 October 1892.41 The school was initially founded for the purpose of training the children of leading Arab sheikhs and notables to be good Ottomans and to counter increasing British influence around the Arabian peninsula. The idea was then extended to include the sons of Kurdish sheikhs and some Albanians. Starting as a two-year boarding school, eventually building up to five, it was stated that a major aim was 'to bring civilisation' to the Arab tribes and to ensure their loyalty to the Ottoman state. The curriculum stressed intensive Turkish, although at its earliest stage, given that the boys would speak no Turkish, Arabic and Turkish would be used together.42 To start with four students between the ages of twelve and sixteen would be sent from each of the vilayets of Aleppo, Syria, Baghdad, Basra, Mosul, Diyarbekir, and Tripoli; and four from the sancaks of Bingazi, Jerusalem and Zor. T h a t they sent five students each was evidence of the special importance accorded to the Yemen and the Hijaz vilayets.43 Although originally intended for the sons of leading Arabs, demands from the Kurdish sheikhs, particularly those who were in command positions in the irregular Hamidiye cavalry, soon meant that their sons, too, were (somewhat grudgingly) accepted into the fold. They in turn were joined by sons of Albanian notables who likewise put pressure on Istanbul to have their sons admitted. 44

102

The Well-Protected

Domains

There was indeed enthusiasm among the Arab and Kurdish sheikhs for their sons to be admitted. The Ministry of the Interior reported on 20 September that Arab sheikhs, 'who normally do not even let their children venture far out of their tents, have been happily bringing their children with their own hands, because they know that their ignorance and rudeness (cebl ve nadanlyct) will thus be replaced by the light of civilization.'45 T h e minister also reported that sheikiis were even competing with each other: a Sheikh Ali al-Necri from the sancak of Zor, when told that the year's quota was full, bitterly protested that if his son was not taken he would lose face, and that he was even ready to pay for his travel expenses.'16 O n 3 October 1892, the Vali of Mamuretula^iz (modern Elazig) reported that the orders to send three boys from his area had been made known to the local notables and respected leaders' of the Kurdish tribes.''7 This had caused great enthusiasm among them and six boys rather than the stipulated three were 'extremely eager to make the journey'. T h e vali pointed out that given the 'sensitive nature of the Dersim area' it would be inadvisable to turn back the extra three, and recommended that all six be admitted to the Tribal School. 48 The prestige of the school increased to the point where powerful families were using their influence with local authorities to pass their children off as members of tribes to have them admitted too. T h e provincial authorities were instructed to 'ascertain whether or not the prospective student was of the tribal element,' and turn back those who were not.49 T h e demand for education on the part of the Kurdish population is brought out in §erif Mardin's study of the legendary Sheikh Said-i Nursi. When actually granted an audience by Abdiilhamid, Nursi, then an impetuous young sheikh, spoke out very plainly in favour of a Kurdish university which he demanded be established in Van. Nursi told the sultan, in no uncertain terms, that if he hoped to compete with the Protestant"' missionaries, modern Islamic education was his only hope.50 The Tribal School was established in an old palace in Besiktas which had once belonged to the Ottoman family. The first principal was the ubiquitous Recai Efendi who also remained assistant principal of the Miilkiye. Life at the school could not have been easy for these bewildered boys, a long way from home and not speaking the language. The regime was very strict and they were kept under virtual house arrest, only being allowed out occasionally and then under strict supervision. The discussion of the issue of confinement to the school grounds is very interesting as it sheds light on the mentality of the Ottoman educa-

Education

103

tors. The students had first been permitted to spend weekly holidays with relatives and friends in the city. Soon it became evident that in the early days of the school this was not a very good idea; those who had relatives and could go out were viewed with envy by those who had no such contacts. The minister of education, Ziihdu Pasa pointed out that, 'because many of the students are still in their state of nomadism and savagery, they have yet to understand the blessings of civilization.'51 Part of this 'savage' behaviour consisted of'jumping over the school wall in an attempt to escape.' It was therefore much mote preferable for the students to be strictly supervised and confined to the grounds, 'until they appreciate their surroundings.' O n Fridays they were to be taken around the great mosques of the city 'to give them a chance to see the splendour of the spectacle/ which they would doubtless appreciate as 'they are people of the desert and have not had much contact with civilized folk.'52 The list of'directives for the administration' stipulated that all the boys' letters to their families were first to be censored by their teachers, and anything untoward be 'corrected'.55 A weekly register of events at the school was to be kept and 'no erasings were to be permitted: any change in wording can only be done by drawing a line through the mistake, still leaving it clearly legible.' At the end of each day the boys were to be assembled and instructed on the glory of the Islamic faith and the duty incumbent on each Muslim to obey the sultan who was caliph of all Muslims. 54 It would appear that Abdulhamid assumed something of the role of a paterfamilias with his young Arab and Kurdish charges. The sultan was regularly presented with a table showing the yearly marks each student had achieved in each subject. On one particular occasion the Palace made it known to the school administration that, '[It has] come to HIM's attention that some of the students have been behaving badly, it is most undesirable that these students should be sent back to their homes with evil in their characters (mesavi-i ahlak sahtbi).' The school administration was strictly enjoined to ensure the proper behaviour of the boys/lest they' be held responsible'.55 About one year later 'proper behaviour' seemed fairly thin on the ground as it was reported that 'Arab and Kurdish students have been involved in a fight using stones, shoes and fists, resulting in light injuries to four Kurds and six Arabs.' N o less a personage than the minister of education, Ziihdu Pasa, immediately descended on the school with a military escort. The minister took pains to report that 'the matter appears to have been the result of a small argument, and does not involve any internal or external

104

The Well-Protected

Domains

provocation.' The comment clearly implied fear of possible ethnic strife between the Kurds and the Arabs, who were supposedly learning to be brothers. 56 After four (later five) years at the school the students would be expected to attend the Mekteb-i Sultani and then the Mulkiye, eventually returning to their home provinces as teachers and officials.57 A memorandum from the Imperial Arsenal (Tophane-i Amire) dated 22 September 1897, drew attention to the fact that the first fifty graduates of the Tribal School had completed their studies, and had been sent on to the Military Academy (Harbiye) and the Mulkiye. There the Kurdish boys had been given one year's training in cavalry tactics, and the Arabs trained as civil servants. The commander of all military schools, Zeki Pasa, pointed out that a certain delay had occurred in their being sent back to their home provinces, and that this was causing gossip. As fledgling Ottoman military and civilian officials the graduates were to be given third, fourth and fifth rank position (sdise, rabia, batnse) in the Ottoman bureaucratic system and sent back forthwith.58 After a period of further training in their home region, the graduate of the Mekteb-i Astret who had completed his studies in the Military Academy or the Civil Service School would be expected to serve in hardship posts such as Yemen, Bingazi, or Mosul. This caused considerable problems because the young graduates of these elite institutions looked down on appointments to areas which they regarded as backwaters, and wanted to stay on in Istanbul or, at worst, to be sent back to the provincial capital of their own regions.59 The Tribal School lasted until 1907 when it was suddenly closed down, ostensibly because of a food riot among the students. One can only surmise that the student population was stirred up by something more serious than flies in their soup: nascent Arab nationalism perhaps, or the spread of Young Turk propaganda since the Young Turk 'Revolution' was to follow one year later.60

T h e Ottoman attitude to non-Muslim education Contemporary discussions about the education of both the general population and the elite are pervaded by a sense of impending danger. In them, the empire is envisioned as a besieged fortress where the utmost care must be taken to counter the subversive activities of a potential fifth column.

Education

105

For that reason non-Muslim educational establishments were kept under the closest supervision and were always regarded with extreme suspicion. A circular from the sultan to his ministers clearly drew the following parallel: 'While the Sublime State's Greek and Armenian subjects are fewer in number compared to Islamic subjects, even in villages of five or ten houses we see the construction of big churches and schools... while many mosques languish in a state of disrepair and many medreses have become almost vacant.'61 There was a distinct shift of emphasis in the Hamidian era as the relatively tolerant atmosphere of the Tanzimat period evaporated.62 As early as 1858, under the Law on Education, all non-Muslim schools were made subject to license and their activities allowed only if they conformed to the regulations of the Ministry of Education.63 It must be said, however, that although these laws were passed and stringent regulations made, the fact that they were incessantly reconfirmed in subsequent years indicates that the Ottoman authorities had very real difficulties applying them. The 1869 Law on Education authorized the inspection of the curricula of non-Muslim schools. In 1880 this was reinforced, and local educational commissions were assigned the duty of supervising the schoolbooks and syllabuses of such establishments. 64 Turkish language teaching was made obligatory in non-Muslim schools in 1894.65 Turkish teachers were assigned to non-Muslim schools and their salaries paid by the Ottoman government.66 In 1887 the inspectorate of non-Muslim and foreign schools' was established for the first time.67 Interestingly enough, one of the most stringent memoranda on the need to curtail non-Muslim education, came from a Christian teacher, an Armenian by the name of Mihran Boyaciyan Efendi, teacher in the Beirut idadi. In 1891 he was to warn of the danger of Christian schools in Beirut, and particularly the danger of Muslim children attending them. He also stressed the importance of Turkish language and Ottoman history teaching in non-Muslim schools.68 The Ottoman archives are full of orders to close down this or that school ostensibly for lacking a license. O n 20 March 1890, the Ministry of Education pointed out that the Wallachian (Utah) and Greek schools in the vilayet of Manastir (present-day Bitola in Macedonia) were all without licenses and should be closed down.69 Teaching the Armenian language in government schools in the provinces was also a problematic issue. T h e Ministry of Education was firmly instructed not to appoint a teacher of Armenian to the idadi of Erzurum

106

The Well-Protected

Domains

as 'it is sufficient to confine the teaching of Armenian to the schools in Istanbul where officials are trained.'70 The teaching of Armenian in provincial government schools was deemed 'not unlikely to have a bad effect on the population of Erzurum, the majority of whom speak Armenian.' 71 This information is paralleled by H . B. Lyncti s account, a travelogue, where he recorded that an Armenian private secondary college in Erzurum, the famous Sanasarean College was being harassed in 1894 by the local authorities. School textbooks on Armenian church history and Armenian culture had been confiscated,72 This indicates very clearly that the Ottoman authorities understood only too well the connection between language and nationalist ferment. The issue becomes all the more significant when we learn from Andreas Tietze that 'significant portions of the Greek as well as of the Armenian communities in Constantinople were Turkophone and had little or no knowledge of the language of their own ancestors.'73 T h e extent of this is shown by the case of the Jesuits of Galata who, in 1702, obtained the Armenian patriarch's agreement to preach in Turkish in Armenian schools.7'1 Given that Turkish was so widely spoken among Armenian and Greek communities, the emphasis placed by Armenian and Greek nationalists on the use of their own language represented a very direct challenge to the Ottoman system. So deeply was this challenge felt that the state considered suspect even such seemingly innocent facets of school life as schoolchildren's songs in Armenian. O n 6 May 1890, the vilayet of Bursa was instructed to 'punish in an exemplary fashion' the teachers and director of an Armenian school in the Ayna Gol district, as the examination of the song books used in the school had revealed 'their subversive nature (bazt ejkar-tfaside)'.75 Similarly, on 26 August 1890, the Armenian Patriarchate complained to the Ottoman authorities that 'some of their prayer books which were a thousand years old have been banned as subversive.' The Ottoman authorities recommended that the genuine old prayer books should be permitted but the interdict should apply to 'the newly written [books] which include symbols and abbreviations (rumuzlu) which are potentially harmful.'76 In the same year, the Ministry of Education was informed that correspondence which had been intercepted between the Greek embassy in Istanbul and the Greek foreign ministry had revealed that Greek schools in the Ottoman Empire were being actively encouraged by the Greek government to teach 'in accordance with Greek educational practices'.77 A novel by Kozmas Politis, which has recently been translated into Turkish,

Education

107

gives some indication that on this score the Ottoman authorities were right to suspect the worst. It gives a charming description of schoolmasters of a Greek school in Izmir (Smyrna) in the late 1890s taking their boys on a field trip to the countryside around the city. The boys visit some Greek ruins on a mountain top where they are made to stand in silence for one minute 'in honour of the ancient Hellenic past'. They are then quizzed on the Greek names of mountains and rivers they see before them, at which point the schoolmaster in charge tearfully declares, 'all this was Hellenic once - it will always be Hellenic!178

Conclusion The project of Ottoman education policy was to reinforce the ideological legitimacy of a social order which felt increasingly threatened by changing world conditions. This could only be done through the education and indoctrination of the people, 'which ... raised unprecedented problems [for the state] of how to maintain or even establish the obedience, loyalty and cooperation of its subjects and members, or its own legitimacy in their eyes.'79 There were very real constraints on the establishment of an empirewide educational system. The most basic was money. Although all local administrations were expected to contribute a share of their revenues as the 'education budget' (maarif bissesi), very often this money was not forthcoming, and schools were not built and teachers left unpaid. A representative complaint on these lines came from Yemen in 1890 when the teachers of the primary school in San'a protested that they had not been paid for some time.80 Similarly, the vilayet of Baghdad complained in 1890 that it had attempted to raise funds for the local ru$diye secondary school by seeking donations from 'charitable persons', but had been unsuccessful.81 In 1899, the vilayet of Kastamonu was to propose the payment of village primary school teachers in kind, the parents of the pupils providing a certain proportion of their agricultural yield at each harvest.82 The other major problem was the low level of literacy. Carter Findley has estimated that literacy went from 'perhaps 1 percent in 1800 to 5-10 percent for the empire in 1900.' Although the educational reforms started from a very low base, by the Young Turk era rtifdiye middle schools were to be found throughout the empire. By the end of the empire, the areas that are now Syria and Iraq had '570 Ottoman government schools with 28,400

108

The Well-Protected

Domains

pupils in elementary and 2,100 in secondary grades,' from which many went to higher schools in Istanbul.83 Modest village primary schools made a favourable impression on foreign travellers who noted the presence of girls: I n the village of Topakh-Keui, in the school we noted with great surprise that many girls were seated next to the boys. This is a symptom of the great moral progress recorded in Turkey, and is a definite indicator of the coming salvation of women in that country.'34 Official propagandists were also directed to write about the progress in education as an indicator of successful reform. Princess Annie de Lusignan visited the 'Sani, Turkish school for girls' and came away suitably moved by its clean and orderly appearance: 'I should have to write several books in order to do justice to the efforts of the Sultan on behalf of national education, or to give any adequate idea of his ceaseless activity in that great cause.'85 Abdulhamid also made a point of showing foreign visitors around his 'showcase schools' designed to impress them. A major feature in the photograph albums sent to leading foreign libraries as the 'Imperial Self Portrait' were picture after picture of grave little girls clutching their diplomas.86 As the monarchies of the late nineteenth century *sidle[d] towards a beckoning national identification', the Ottoman Empire found itself in the same boat as the Russian, Japanese, Austrian, German, and even Chinese empires.87 Almost at the same time as Count Uvarov advocated 'autocracy (samoderzhank), Orthodoxy (Pravoslavie) and nationality (narodnost)' as the foundation of the Russian state, Resid Pasa, the moving force behind the Tanzimat reforms, envisaged the 'basic pillars of the Ottoman State' as 'Islam, the House of Osman, the protectorate of Mecca and Medina, and Istanbul as the capital of empire.'68 To foster the feeling of'belonging' among-their people was the basic dilemma of all imperial education systems and nowhere more so than in the cases of Russia and the Ottoman Empire, where the sheer scale of the country and the variety of their peoples meant that 'the emerging popular conception of national identity included only a rudimentary concept of citizenship.. .'89 Nor was there any 'supra-ethnic concept of nation or empire to which diverse peoples could be attracted with a modicum of voluntarism,' 9 0 In this context the gap was filled by a recharged conceptualisation of religion and/or direct attachment to the quasi-sacred person of the emperor/sultan. This attachment was to be inculcated through mass schooling. It is interesting to note that efforts to standardize education were made in the Russian and Ottoman empires at very

Education

109

similar times. The Russian statute on primary schools is dated 1864, whereas the Ottoman Ministry for Education was established in 1857, and the general Law on Education was passed in 1869. In both educational systems religious instruction had a central role,9* In 1897 religion and church Slavonic accounted for 9 out of 24 hours of instruction per week in Russia, and in Ottoman schools we have already noted the renewed emphasis on Qur'anic study (tecvid) from primary to higher schooling.92 In both the Ottoman and Russian cases heroic figures from history and an image of a glorious past was used to instill a sense of common identity in schoolchildren.93 For instance, the romantic poet/author Namik Kemal, and in particular his early work, became standard reading for schoolchildren.91 Another similarity between the Russian and Ottoman cases was the attempt to assimilate non-Turkish and non-Russian elements in the value system of the centre, just as the Ottomans tried to instill Ottomanism in Arab notables' children in the Tribal School, the Russians attempted 'Russification' of non-Russian peoples.95 Sheikh Cemaleddin, the son of the legendary leader of the Dagistani people of the Caucasus, Sheikh §arnil, was taken hostage during the twenty-seven-year (1834-61) campaign to subdue the area, brought up as a member of the Russian elite and made a member of the crack imperial guard.96 In similar fashion, the sons of leading notables would be appointed to the personal guard of Abdulhamid. 97 Another similarity in the Russian and Ottoman cases is the low level of literacy and available schooling.98 Like Thomas Babington Macauley, the president of the Committee of Public Instruction in India in 1834, whose aim was to create an indigenous elite who would be 'Indian in blood and colour, but English intaste^ inopjlmon, in morals and intellect? anctUke the Russians who educated Cemaleddin in Petersburg with the aim of making him 'Russian',99 the aim of the Tribal School was to create Arabs, Kurds and Albanians who would be Ottoman. The main thrust of Japanese education policies in the late Meiji period to was the effort to ^^A^^Y^^^I^^^^^^^Y^Y ^K.'s&SiQJHL. 100 ^polity' (kokumin), Although religion per sc in the Japanese context of Confucianism and Buddhism clearly meant something very different from Islam or Orthodoxy, the attempt to build an emperor cult as the focus of loyalty for a citizenry, and the role of education in the process, bears certain striking similarities. In the first place, the dialectic between Western and native themes in the struggle between progressive and conservative

110

The Well-Protected

Domains

elements is observable in all three of the Russian, Ottoman and Japanese cases. Just as there was a see-saw effect between Tanzimat liberalism and Hamidian conservatism there was a similar tension between Japanese liberal reformers and palace circles who felt that Westernization was detrimental to Japanese native values.101 The Japanese Education Ordinance of 1879 was revised in 1880, and the Ministry of Education compiled a list of unacceptable texts, while moving ethics instruction to the top of the list of subjects to be taught in elementary schools.103 All this rings a bell when we recall that instructors were reprimanded for teaching Greek and Roman civilization in Ottoman schools in precisely the same years that Western philosophy was removed from the curriculum, and that religious instruction was moved to centre stage.103 The basis of Ottoman education policy was therefore in keeping with contemporary world trends, an effort to gradually 'civilize' subject populations into espousing the value system of the centre. The words of Osman Nuri Pasa serve as an eloquent testimony to this aim: 'It would be against reason to construct a mechanism of government which would be totally in contradiction to the customs and beliefs of the nomadic population ... T h e spread and progress of modern education in a country is best served by the [reconciliation] of the needs of the country and the customs and habits of the population.'104 It is clear that in the documents cited above religion and nationality are being used interchangeably. In 'religious tradition' and 'national morals' (adab'i diniye vt ahlak-i milliye) the order could just as easily be reversed as 'national traditions' (adab-t tnilliye), and 'religious morals' (ahlak-i diniye). The word hamtyet' meaning 'zeal' was often used as 'religious zeal' (hamiyeti diniye) or national zeal/patriotism (hamiyti-i tnilliye). That the 'national zeal' which the power holders were seeking to instil emerged with a vengeance in the opposition was the ultimate irony. There is no escaping the fact that the increasing dosage of 'correct' ideological conditioning in the elite schools was in direct proportion to the spread of the influence of the Committee of Union and Progress in these institutions. The organization that was to seal the fate of the Ottoman Empire was founded in 1889 as a student organization in the Imperial School of Medicine.105 This was soon followed by the establishment of C U P branches in the Imperial Military Academy, the Imperial Civil Service School, and the School of Law. There was a general purge of suspected C U P sympathizers from the Military Academy in May 1895. In the same year, C U P

Education

111

members clandestinely made contact with the British embassy.106 C U P membership rapidly increased in the elite schools. This caused a great deal of concern, a government report was to comment, in 18 December 1895: 'It is particularly distressing that these elements should be found among the students of the imperial schools whose aim is to train men of solid character and morals, loyal to the Sublime State*.. This would indicate that stricter control should be exercised over the teachers and students of these schools.'107 But the Ottoman ruling elite would not have the field to themselves in attempting to instill their values. Together with the opposition of the CUP they would have to contend with the opposition of foreign missionaries. It is to this challenge that we now turn.

Chapter Five

'They Confuse and Excite the Mind': The Missionary Problem

Reverend Blakeney Davis peered out into the gloom of the Anatolian plain. Dusk was falling and the huge sheep dogs had begun their nightly uproar. Soon it would not be safe to go out in the dark; the brutes did not yet know him and somehow particularly objected to his presence. So this was Sivas, he sighed, where he had come to do the Lord's good work, a long way from Freeport Maine. The local Armenian population were suspicious and indifferent at first, but now a few of the poorest boys had begun to attend the mud hut he grandly called his 'Academy for Central Anatolia'. T h e local kaymakam had called yesterday, a sullen-faced man who kept poking his riding crop into their personal effects and asking if they had brought explosives with them from America.

N o n e o f t h e challenges t o t h e legitimacy of t h e O t t o m a n state, a n d all t h a t it s t o o d for, was more dangerous In t h e long t e r m than t h a t posed by missionary activity. T h e t h r e a t posed by t h e soldier, t h e d i p l o m a t , t h e merchant, all h a d to d o with the here a n d n o w ; the missionaries, t h r o u g h their schools, constituted a danger for the future. T o a large extent, the measures discussed in t h e previous chapters, to convert or reinforce the o r t h o d o x y of marginal elements, were designed as a reaction to t h e missionary presence. T h r o u g h o u t t h e world the missionary appeared as the representative of a superior culture, the 'white man's b u r d e n ' personified. T h e O t t o m a n missionary struggle t h u s also serves to p u t events in the O t t o m a n empire into a world perspective. Missionary activity was proving to b e t h e t h o r n i est of q u e s t i o n s from L a t i n A m e r i c a to Africa to C h i n a . O t t o m a n diplomats, w h o w o u l d closely m o n i t o r missionary activity in an obscure

112

The Missionary Problem

113

corner of the globe such as Hawaii, were well aware of the implications of this global reach.1 Missionary activity in Japan had been the reason for the anti-Christian stance of some Meiji bureaucrats in their fight to stigmatize Christianity as 'incompatible with loyalty'.2 Although the Japanese and Ottoman attitude to religion was very different, and the Meiji Restoration became much more secular in the 1870s, Christians were stigmatized as disloyal to the state.5 In reaction, the Buddhists 'emulated the practice of their Christian rivals, expanding charitable institutions in Japan and undertaking missionary work in Taiwan and Korea."5 At around the same time the turn of the century, the Ottomans began implementing their policy of Hanefi missionary activity. The situation in China was more comparable to the Ottoman experience, Even weaker than the Ottomans in their capacity to resist Western encroachment, the Chinese focused much of their resentment and anger on the missionaries and their Chinese converts. The Empress Dowager Tzu Hsi, a contemporary of Abdulhamid, declared that: 'These Christians are the worst people in China ...They rob the poor country people of their land and property, and the missionaries, of course, always protect them, in order to get a share for themselves.'5 N o doubt Abdulhamid would have fully agreed. Edward Said comments that: 'Even the legendary American missionaries to the Near East during the nineteenth and twentieth centuries took their role as set not so much by God as their God, their culture, and their destiny.'6 Not only did the missionaries undermine the efforts the Ottomans were making to legitimize the basis of their rule at home, they also proved influential in creating adverse conditions abroad by feeding the Western press with anti-Turkish sentiment: 'Many missionaries and Western journalists proceeded upon the confident assumption that the Terrible Turk belonged to a retrograde race of Devil-worshippers.'7 Particularly after the British occupation of Egypt in 1882, the Protestant missions felt that the way was open for the conversion of Muslims to Christianity. Together with the Cyprus Convention of 1879, which established a British protectorate over the island, these developments were seen as the thin end of the wedge, making possible the triumphant advance of Christianity in the Middle East. In Tibawi's words: ... by 1882 Protestant missionaries ... had made clear their attitude towards the legitimate Government and territorial integriry of a state in which they resided and worked. Not only did they publicly declare rheir intention of sub-

114

The Well-Protected

Domains

verting its established religion, not only did they openly pray for the extinction of the state and the absorption of its territories by their own Governments, but pending the achievement of these ambitions they claimed special privileges and exemptions, and with these very claims they accused the Ottoman authorities of intolerance, fanaticism and bigotry.8 T h e last ditch ideological defences erected during the H a m i d i a n era, t h e increasing emphasis o n schooling a n d t h e a t t e m p t t o enforce Hanefi orthodoxy were perhaps best u n d e r s t o o d for what they were by t h e m i s ' sionaries. T h e secretary to t h e American Mission in Syria, H e n r y Jessup, h a d to confess surprise t h a t 'the T u r k s are n o t after all on their last legs.' 9 I n t h e 1 8 8 0 s J e s s u p a d m i t t e d t h a t t h e y faced a s i t u a t i o n w h i c h , ' d i s c o u n t e d ] previous predictions of Islamic decay, a n d cite[d] as evidence of revival t h r o u g h o u t the O t t o m a n E m p i r e the opening of schools for boys a n d girls ,.. a n d t h e building of n e w mosques.' 1 0 Missionary activity gathered m o m e n t u m during the 1880s a n d 1890s w i t h British, American, Russian, a n d French missionaries parcelling out spheres of their 'work'. T h i s led to a situation in which, according t o Jeremy Salt, 'the relationship t h a t developed between t h e missionaries a n d the O t t o m a n government was o n e of m u t u a l suspicion a n d m u t u a l dislike.' 11 Indeed, by the 1890s t h e missionaries h a d come to be regarded by t h e sultan as 'the m o s t dangerous enemies to social order', a m o n g all t h e foreigners living in h i s domains. 1 2 Before w e g o any further i n t o this tense relationship, it is w o r t h looking at h o w A b d u l h a m i d II himself saw t h e situation. O n 25 M a y 1892, h e dictated the following to his private secretary:

In England, Russia, and France, there exist Bible Societies which become exceedingly rich through the donations of rich and fanatical Christians who bequeath all their wealth to them in their wills ... Although the English, Russian and French governments seem not to be involved in their activities, they secretly aid and abet them in sending missionaries into darkest Africa. In this way they spread their beliefs among the local population. By increasing the numbers of their followers this religious influence is then transformed into political leverage ... Although it is obviously desirable to take firm measures against them, if open opposition is brought to play, the Sublime Porte will suffer the vexing interventions of the three powers' ambassadors. Thus the only way to fight against them is to increase the Islamic population and spread the belief in the Holiest of Faiths.13

The Missionary Problem

115

It is interesting to note how close the sultan's description of the ambivalent relationship of the missionaries with their governments comes to the assessment of an eminent modern historian: 'Missionary effort was by no means an agency of imperialist policies. Often it was opposed to the colonial authorities ... Yet the success of the Lord was a function of inv perialist advance.'14 We are, therefore, when we speak of the struggle between the missionaries and the Ottoman government, dealing with nothing less than ideological war, a war that challenged the very basis of Ottoman legitimacy among Christian and Muslim.

Ottoman defence against missionary incursion The legal basis for proselytizing in Ottoman domains, the missionaries claimed, stemmed from the Tans;imat rescript of 1839 and the Reform rescript (Islahat Fermant) of 1856, which together guaranteed the freedom of all Ottoman subjects, Muslim and Christian alike, to choose whatever religion they wanted to belong to.15 Although the principle of these two documents was never called into question, the Ottomans interpreted 'freedom of religion' as 'the freedom to defend their religion'. The Foreign Minister Ali Pasa was to instruct his ambassador in London to ask Earl Russell the following questions: Can it be supposed that whilst condemning religious persecutions, the Sublime Porte has consented to permit offence and insult to any creed whatever? That at the same time as she was proclaiming liberty to all non-Mussulman creeds, she had given them arms against Islamism? That she had, in fine, destroyed at the same stroke the guarantees with which she surrounded the liberty of religious convictions?'6 This defensive attitude was actually written into Article 62 of the Treaty of Berlin in 1878, an interpretation which actually amounted to 'an official recognition by Europe of the Turkish definition of religious liberty. It is so worded as to allow the Turkish Government to take any measures which it may deem proper to prevent Mohammedans from changing their faith.'17 The fact was that this measure had very litde to do with religion, and everything to do with the Hamidian regime's new political definition of religious loyalty: '[The] Turkish Government was never more deter-

116

The Well-Protected

Domains

mined than it is now to prevent all defection from Islam ... It is now a political rather than a religious principle, designed to maintain the strictly Mohammedan character of the Turkish Government, and to retain all political power in the hands of the Turks.' 18 Although the precept in the feriat allowing the killing of any convert from Islam was formally disallowed, the spirit of its draconian ethic still lived in the popular psyche.19 Sultan Abdulhamid himself was to tell Layard that although in principle any Muslim was free to embrace Christianity, 'be he the Sheikh ul Islam himself, the reality on the ground was something else; 'there were parts of the empire where the population was not only very fanatical, but frequently almost beyond the control of the local functionaries when their fanaticism was aroused.'20 T h e Ottoman definition of religious freedom is clearly expressed in a communication from the grand vizier's chancery, dated 22 March 1896. In this document the grand vizier of the time, Rifat Pasa, reported that the Council of Ministers had discussed the issue of the expulsion of missionaries who had been 'involved in activities injurious to public order'. H e pointed out that, 'the activities of these missionaries such as the effort to cause people to change their religion, is totally contrary to the present laws guaranteeing freedom of religion (serbesti-i idyan)'21 The Ottoman government had already taken measures to ensure the control of Istanbul over missionary teaching and other activities by measures such as the Law of Education (1869) which stipulated that all foreign schools had to submit their curricula and teachers to public inspection.22 Yet, the evidence shows that these restrictions were never effectively enforced and missionary schools filled the vacuum left by the insufficiency of Ottoman education. T h e Ottoman sources indicate that the local officials were briefed to do their utmost to curtail the spread of missionary schools. O n 21 January 1892, the governor of the vilayet of Syria reported that he had compiled a list of foreign schools 'constituted by devious means' such as converting dwellings to schoolhouses. H e had established that there were 159 such establishments in his area. The vali noted that although the state had been making great effort to increase state primary schools, these were still insufficient and this meant that 'Jesuit and Protestant schools therefore accept non-Muslim children free of charge, clothe and feed them and even pay subsidies to their parents.' 23 The presence of these schools was also seen as the thin end of the wedge as far as the Muslim population was concerned. T h e vali continued: 'It is therefore necessary that in the approaching holy month of Ramadan special ulema should be sent to preach

The Missionary Problem

117

secretly to the Muslim population about the ills that will accrue to them if they send their children to Christian schools.'24 The emphasis on secrecy is significant, as any overt activity against the foreign schools was likely to draw the wrath of the foreign consuls. Throughout the 1880s and 1890s Istanbul constantly renewed orders to the provinces to prevent the proliferation of unlicensed foreign educational and religious establishments. The fact that the matter came up so frequently suggests that the problem was never solved. O n 12 January 1898, the vilayet of Aydm received very strict instructions to prevent the admission of Muslim children into foreign schools. The reason was very clearly the political implications of such schooling, 'the attendance of foreign schools by Muslim children will lead to such evil results as the damaging of their national and religious training (diyanet ve terbiyeyi milliye) and, God forbid, may even result in their abandoning their religion.'25 The Vali of Aydm was to report on several occasions on the matter of conversion, stating that he 'had no knowledge of any Muslim who had converted'.26 O n 28 February 1892, the vilayet of Syria was told that a team of local ulema was to be charged with the task of 'imparting to the local population the ills that will result from their children attending Christian schools'. It was determined that Ris;a, Muhsin, Cemal, Resid, Mustafa, Ahmed and Mehmed Efendis should be sent on a tour of the vilayet with a total salary of 14,000 kuru;.27 By 25 June these 'official ulema' had reported back to Istanbul on the measures needed to protect Muslim children from missionary schools. T h e measures suggested, however, did not go beyond the education of Muslim children by village imams. For this each village should be equipped with adequate reading primers. 28 Yet the effort to enforce the licensing of foreign schools continued to be plagued by foreign intervention. On 28 March 1892, the Sublime Porte issued a circular stating that the deadline given to foreign schools and churches for them to procure licences or be closed down had expired. The circular lamely concluded that the deadline was to be extended yet again for a further three months. 29 This is not to say that the Ottoman officials did not realize what had to be done to control what they called the 'spreading of foreign influence behind the shield of education'. On 28 October 1898, a Special Commission on Education reported that more teeth had to be put into the current laws regulating foreign religious establishments.30 The commission had no illusions about the fact that 'these schools established by foreign priests and missionaries ... seek to convert the population into the Protestant and Catholic faiths

118

The Well-Protected Domains

and to confuse and excite their minds.'51 The 'confusion of minds' was a recurrent theme in the official documentation. This confusion even went to the extent of causing 'some youths to end up marrying foreign women (yabanct kardanyla istihsal-i rabtta-i izdivac)' and thus 'daring to act against the teachings of Islam'. O n 26 June 1906, a memorandum prepared by the Council of Ministers declared that this was caused by the fact that Muslim schools were lacking in 'proper training in religious morals ... which would fortify youth [against such waywardness] P2 The Ottoman government made every effort to prevent the spread and circulation of missionary literature. On 29 March 1892, the decision was taken to apply the regulations concerning the circulation of religious literature to 'travelling booksellers', usually a euphemism for people distributing Bibles under the auspices of the Bible Society.33 But this was often counterproductive. O n 9 March 1896, the translation bureau of the Porte translated an article from the London Daily News claiming that the Ottoman authorities were infringing on the provisions of the reform edicts, and were confiscating Bibles and religious books found on foreign travellers. It was alleged by the Bible Society in Istanbul that this had been done on several occasions, followed by a public burning of the confiscated materials. The Porte made haste to deny this allegation by arranging for the publication of articles in the Western press.34 Ottoman officials were well aware that rules and regulations were losing their clout each time the state backed down before foreign protest. The Military Reform Commission prepared a memorandum, on 27 December 1891, dealing with the matter of unlicensed schools and churches, These, it said, should be closed down forthwith and the clamour of protest not be listened to. In response to any protest, the foreigners were to be told: 'Yes, there is freedom of religion, but this freedom must be exercised within the laws and regulations.35 Nor did the officials of the Porte nourish any illusions about the stiffness of the competition they were up against. O n 2 March 1892, the Porte sent instructions to the Ministry of Education, to appoint a teacher of French to the Ottoman ru^diye in Nablus in Palestine. The local population were sending their children to 'Latin schools' so that they might learn French 'even before they had learned their religious beliefs'. The French teacher appointed had to be good enough to compete with the French tutors in the foreign schools, as 'only then will the population send their children'.36 The other interesting aspect of this document is that it illustrates that the officials understood that it was not enough simply to forbid

The Missionary Problem

119

people to send their children to the schools; an alternative service of more or less equal quality had to be provided. Another case of a similar nature found its way into the records from the other end of the empire. O n 1 October 1889, the vilayet of Iskodra in Albania reported that the local idadi lycee had been closed for lack of funds. This was playing into the hands of the Italian consul, who was promoting the local Catholic schools as alternatives, and it was inevitable that the local population would indeed send their children to those schools if Istanbul failed to make adequate provision.37 The missionaries who were most prominent in the Ottoman Empire in the last quarter of the nineteenth century were the French, British and Americans. It is proposed here to treat them under their national categories, although in the case of the British and Americans, in particular, there was considerable overlap in their activities.

French missionary activity The French missionaries were the group who were active in Ottoman dominions for the longest time. In 1622 Pope Gregory XV had established the institution of Propaganda Fide whose object was to convert the heathen and bring back into the Catholic fold the ancient Christian communities of the Orient. Catholic missionary work in the Near East had been assigned by Propaganda Fide to the French Capucin, Carmelite, and Jesuit orders.38 In the early 1890s French diplomatic-missionary activity picked up in the eastern provinces of the empire, T h e French missions were particularly active among the Nestorian population of Mosul. O n 23 May 1892, the Vali of Mosul reported that several 'special missions' had been sent by the French in the last four years, 'which leaves no doubt as to the existence of a French master spy (ser-hafiye).' T h e vali reported further that the Nestorian religious leader, the Maresh-Maun, had complained bitterly about Catholic poaching among his flock. T h e Chaldean patriarch, supported by the French, was bringing constant pressure to bear on the Nestorian leader. T h e Maresh-Maun bemoaned this fact which, he feared, would mean that very soon the entire Nestorian community would become Catholic. 39 Indeed, the sultan himself felt strongly enough about missionary poaching among his Christian subjects to send them a circular warning them to be on their guard.''0

120

The Well-Protected

Domains

The French presence also comes up in dispatches during the Yezidi conversion campaign of 1892-93. At the height of the campaign, Yezidi leaders approached the French consul in Mosul with the offer that the whole community would become Christian (and presumably Catholic) if the French supported them against the Ottomans. 41 This was seen as the height of treason by the Ottoman authorities whose worst fears of missionary meddling seemed to be realized.42 Lebanon was another area where French missionary influence seemed to be undermining the legitimacy of the state.43 In an unsigned letter in Arabic, this danger was clearly noted. The letter stated that 'since Napoleon [Bonaparte's] invasion of Egypt and Syria ... France has not renounced her aims in the region.' Through the efforts of the Jesuits, France had established very real presence in the region among the Maronite Christians, 'through the education of thousands of their children.'44 This reached such an extent that 'ten years ago during the war between France and Prussia some fifteen thousand Maronites volunteered to fight in the French ranks. Let it be noted that in the late war with Russia [1877-78] it did not occur to them to form even one company of volunteers to fight for their country.'45 A report written by one of Abdiilhamid's major grand vizier's, Kamil Pasa, also pointed to the continuing interest of France in the Lebanon, 'since the time of Napoleon Bonaparte'.46 He did not fail to mention that 'the high quality schools that the French have established in Beirut have produced thousands of Ottoman subjects who feel indebted to France.'47 Kamil Pasa's report seems to have been provoked by a very interesting letter written by an Arab Catholic, Yusuf Zeki Efendi, which sheds further light on the activities of the French-inspired religious functionaries. Yusuf Zeki took part in an Easter service in Damascus where the bishop's sermon amounted to nothing less than a propaganda speech. H e reported that the bishop harangued his flock in the following terms: 'O Syrians! See how you are progressing thanks to France! Thanks to the French you are being educated in the sciences! You are protected by this most learned and enlightened of nations! Your children are protected by such ajust and compassionate state which is above all other nations!'48 Yusuf Zeki wrote that the Syrians had been favourably inclined to France until the British invasion of Egypt, but when they had witnessed what the British did in Egypt they had become wary of France. Now France was trying to reestablish her ascendancy through members of the clergy such as the bishop whose sermon he had heard.49

The Missionary Problem

121

What was the extent of Ottoman control over the construction and maintenance of Christian places of worship and schools? What were the criteria for the granting or denial of permission for such establishments? It is fairly clear that in places like Beirut or Damascus where French and other influence was already well established, Istanbul ran into serious difficulties. But in the Ottoman heartland of Anatolia their control was more effective. One such case was reported from the vilayet of Mamuret-ula.ziz (present day Elazig), where the Capucin priests had applied to establish a church which would also serve as a school by converting a dwelling.50 The local authorities reported that the projected construction was a modest affair capable of accommodating some thirty students. It was pointed out that the location was surrounded by high walls and was some distance from the barracks and the military training ground. T h e priests had erected a bell tower 'but ha[d] not yet rung their bell'. The Catholic Armenian population of the area amounted to no more than 'some eight or ten households including the priests' own servants.' The only disadvantage was that it was situated in the Muslim quarter. But this was compensated for by the high wall. In this case, the local authorities recommended that permission for this church be granted, 'so that they will not have recourse to the intermediary of the French embassy.'51 The ringing of church bells, in particular, appears to have had a symbolic significance for the Ottoman authorities. On 22 June 1897, it was reported that a Jesuit from Izmir, Father Dominique, had erected a belfry over his dwelling which served as a school. The bell was particularly objectionable: 'the ringing of this bell under the pretext of calling children to school and other such mistakes must be prevented and not allowed to recur in future.'52 French official support for these institutions was never lacking, and as much as the Ottomans tried to impress upon the French that this was their domestic affair, they were unable to prevent such interference entirely. O n 15 June 1896, the vilayets of Trabzon and Erzurum were instructed to provide all the necessary support for a Jesuit inspector of schools from France who was to tour the schools in the area. This 'was the special request of the French embassy.'53 The inspection of schools was indeed a matter of great delicacy; it revealed in microcosm the clash of two sovereignties. The Ottomans claimed that they had the right to inspect all Christian schools in their domains, whereas the French and other authorities tended to view their schools as ex-territorial. On 11 March, Grand Vizier Kamil Pasa wrote to the Min-

122

The Well-Protected

Domains

istry of Education reiterating that, by imperial order (trade), no Muslim children should attend Christian schools and that this should be checked through the inspection of various French schools. The schools in question, however, replied that they would only allow inspection through the intermediary of the French consuls. This provoked Kamil Pasa's ire: 'It is also necessary to know what ts being taught to the Christian children of the fatherland.'54 W h e n some Jesuits were expelled from France in 1902, the Ottoman Ministry for Foreign Affairs and the provincial authorities were instructed that, 'these people are not to be admitted into Ottoman domains. A country which is expelling certain elements has no right to prevail upon another to accept them."55 Although Catholic missionary activity was a thorn in the side of the Ottomans, it was nonetheless a familiar 'evil'. Compared to the rigour and aggressiveness of the newly arrived Protestants, the Catholics appeared positively soft. Also, the Catholic presence had a much longer history. Indeed part of the effort to appear as a member of the club of Great Powers centred around the Porte's diplomatic recognition of the Papacy. Relations with the Holy See had been improving steadily since the 1880s.56 In 1894 the Catholic Patriarch of Istanbul was to write to the palace that the pope had received in audience the Nestorian and Malachite patriarchs, together with 'other well wishers of the Sublime State'. T h e outcome of the meeting had been, 'a document in Latin and French', wherein it was stated that, 'from now on the Roman clergy (ejrenc rahipleri) will be strictly forbidden from interfering in the affairs of the Ottoman Catholics who will be responsible solely to their local Patriarchates.* This would mean that all mission schools would be run only by Ottoman Catholics and the papacy would defer to the local patriarchates in all matters. T h e Pope had also stressed the importance of the Eastern patriarchates in matters of state security: 'They are to take every opportunity to enjoin their flocks not to associate with subversive and corrupting elements, and to remain loyal to their legitimate sovereign'.57 In 1898, when it was decided to send an Ottoman ambassador to the Vatican, the Caliph finally recognized the Pope, who was seen by the West as his opposite number. It was also thought that improved relations with the Vatican would enable the Potte to counter Protestant missionary pressure, and keep better control of its Catholic population. 58 T h e effort to move closer to the Catholic world was reflected in the Porte's continued protection for the Armenian Catholic monastery in

The Missionary Problem

123

Venice. This was part of the policy of attempting to separate the Catholic Armenians from the Gregorians, the latter being the target of most major massacres. The Armenian Catholic Patriarchate in Istanbul noted that the monastery, whose students bore Ottoman passports had, 'remained loyal subjects and benefited from Ottoman protection for the last two hundred years.'59

British missionary activity Although British and American missionary activity overlapped in many instances, for the sake of clarity it is proposed to treat them separately here. The relationship between the British and American missionaries and the Porte was often much more acrimonious than that of their more established French counterparts. Indeed, the implication in the documents is that the Catholics, on the principle of 'the evil you know', were preferred. The British occupation of Cyprus in 1878, and the supposedly 'temporary' British administration of the island, was to provide the first clue as to the nature of British religious politics.60 Karnil Pasa wrote, on 30 August 1884, that 'under the guise of rationalization of the education system' the British authorities were discontinuing the salaries of some Muslim primary school teachers and combining several village schools. They had also established a school in the Muslim quarter of Nicosia which was administered by 'a Protestant priest' who was charged with teaching the children English. The Orthodox schools were in good shape and were not being interfered with, but the British measures were causing a radical decrease in the number of Muslim village primary schools.61 Kamil Pasa feared that this was the first step to their being replaced by Protestant schools.62 On 23 June 1884, the British Embassy complained that the attempt to build a Protestant school in Aleppo was being hindered by the local authorities. The British consul in Aleppo had reported that this was all the more objectionable as the Catholic schools and churches in the city were not being interfered with in any way. This 'preferential behaviour [runs counter to] official Ottoman guarantees that all foreign schools in the empire were being dealt with on a basis of equality.'63 Even matters of extreme detail came to be dealt with at the highest reaches of state. Such a case was the affair of the Protestant cemetery in

124

The Well'Protected Domains

Basra. On 16 August 1907, the Vali of Basra reported that there had been continuous correspondence with the local British consul over this cemetery which had become unusable due to frequent flooding. The consul had demanded that the Protestant community be given an alternative location. The matter was discussed at length and the affair was closed when the Ottoman authorities agreed to make the necessary payment for the construction of a protective dike.64 O n 26 January 1892, it was reported that 'English priests' had been seen in the vicinity of Kevar, a district on the Ottoman-Iranian border. These priests, it was reported, had been distributing books and pamphlets among the Nestorian population. One of them had been apprehended and an investigation launched. The sultan decreed that they be 'chased away in the firmest possible manner (suret-i hakimanede omlardan dej'leri).'65 The fear of 'English priests' seems to have been voiced from far-flung corners of the empire. The vilayet of Yemen reported, on 3 April 1891, that 'missionaries from Aden' had appeared on the Yemeni coast and had been s&en distributing literature in Arabic to the locals. Istanbul sent firm instructions that this be stopped and 'it be established just what sort of book they are circulating and if it is the Bible, which Muslims do not need, they should be confiscated.'66 Yet it appears that not all the contact between the Ottoman authorities and British missionaries was of an antagonistic nature. The Alliance Evangeiique, 'a very influential and respected religious group' according to Grand Vizier Cevad Pasa, was thanked by the sultan for 'its recent statements indicating that the excitement of British public opinion regarding the Armenian affair was quieting down.' The London embassy was instructed to thank Mr Arnold, the secretary of the organization, in the name of the sultan for what was regarded as an intercession favourable to the Porte.67 Another worrying development for the Ottoman authorities was the tendency for the American and British diplomats in the empire to work together on the missionary issue and support one another's nationals. This was reported by the Ottoman embassy in Washington, which gave details on how the British consul in Trabzon had informed the American minister in Istanbul about the details of the burning of the American mission school in Mersifon by the local Ottoman police.68

The Missionary Problem

125

American missionaries Perhaps the biggest headache for the Ottomans were the American missionaries.69 The latest arrivals on the Ottoman scene, this particular brand of New England zealot was to constitute an enigma for Istanbul which only saw their efforts to educate minorities as an attempt to undermine state legitimacy,70 T h e problem lay in the fact that the two sides could not possibly speak the same language. The missionaries, very often at odds with their own diplomatic representatives, somewhat naively thought that what was good for the Armenian or Bulgarian Ottomans would somehow end up being good for the Muslim majority. T h e O t t o m a n s authorities, from the sultan down, felt that they were establishing nests of sedition and training revolutionaries. Most disturbing was the fact that the Protestant subjects of the Ottoman empire unlike other Christian minorities, were not a collectively recognized member of the millet system. A palace memorandum stressed this fact, pointing out that 'this makes it possible for the Protestant governments to pose as the spiritual champions of the protestant Ottomans.' 71 It was therefore necessary to bring their status into line with that of other Christian communities 'according to the precepts of the Reform Edict (1856]' by establishing a body which would be responsible to the Ottoman government.72 The relationship between Istanbul and the missionaries became increasingly characterized by mutual accusations and bad blood. Things were made all the more complicated by the Armenian massacres of the 1890s. Ottoman representatives in the United States closely monitored missionary publications such as the Missionary Herald, and translations from this paper often turn up in Ottoman records. One example among many was the despatch sent by the Ottoman minister in Washington which included an article from the Herald reassuring the missionaries' backers at home that, despite the troubled state of the Ottoman interior, their work was going on. The article waxed particularly poetic on the achievements of evangelical missionary schools in Kayseri, Kara Hisar, Odemis, and Trabzon. 73 The legation was to report, on 4 March 1894, that the American missionary societies had reconfirmed their commitment not to support 'Armenian intrigues' and as a gesture they had stopped funding an Armenian church in Worcester Massachusetts, 7,1 Any anti-missionary views which appeared in the American press were avidly clipped and sent back to Istanbul. O n 26 September 1893, the Ot-

126

The Well-Protected Domains

toman legation in Washington reported on a congress of world religions which had been held in Chicago. At this meeting of religious dignitaries from all over the world, the Buddhist delegate had taken the American audience to task telling them, 'How dare you judge us? How many in this audience have read anything to do with the Buddha's life?'. When only five people raised their hands, the Buddhist delegate told them, 'So only five of you know about the belief of four hundred and seventy five million people. You complain that you have not been able to convert people of other faiths ... this is the height of selfishness.'75 The Japanese delegate then took his turn to tell the audience that the aim of the missionaries that came to Japan was to prepare that country for foreign annexation. He went on to say,' I am proud to say that I was the first to start a society to work against Christianity in japan.' The delegates also roundly condemned missionary activity in China which was causing suffering and bloodshed.76 O n 26 October, the legation reported that no less a paper than the Washington Post had spoken out strongly against missionary meddling in other countries, and the paper's view was that they might be more useful at home. It was suggested that the Ottoman press publish a translation of the article.77 An altogether remarkable document illustrating just how wary the Ottomans were of the American missionaries, is a report from the Ottoman-consul general in New York, Munci Bey, dealing with the history of missionary penetration of the Hawaiian islands.78 T h e consul began his report by saying: As is well known, these missionaries are an infamous band who use religion to achieve political power and advance their material interest ... When they first arrived on these islands whose mild climate is very suitable for the growth of rare and beautiful plants, and whose people are of a soft and accommodating disposition, ir was as if they had fallen upon a free banquet the likes of which they had never seen before. He went into great detail on how money was collected in public meetings describing a mass hysteria verging on that of the televized evangelists of modern times: 'In a state of mind that approaches a trance, they strip off their rings, necklaces, gold watches and chains, and throw them into the collection box.'79 Evidently having made a detailed study of the matter, Munci Bey went on to describe in detail the establishment of the Dole Fruit Company, the leper colony on Molokai, and the agitation on the

The Missionary Problem

127

part of the missionaries for the protectorate status of the islands. H e concluded on an ominous note: 'As these missionaries are also active in our August Master's well protected domains, and their malicious work is observable daily, I judged it advisable to give an example of what manner of evil they are capable of.'80 The Ottoman diplomatic representatives continued throughout the 1890s to give their government up-to-the-minute information on Ottoman matters as reflected in the American press. On 14 April 1896, the Ottoman minister in Washington gave detailed translations relating to the matter of the arresr and deportation of George Perkins Knapp from the Bitlis Boys Academy, for alleged incitement of Armenians to revolt. H e reported that the American press had compared conditions for missionary work in the Ottoman empire favourably with the conditions obtaining in Russia. But the arrest of Dr Knapp had provoked a hail of protest. 81 The general gist of the reporting was that the Ottoman government was upset that the missionaries were informing the world about the massacres of Armenians. 82 The legation was instructed to inform the American authorities that there was no general plan to expel all American missionaries as had been reported in the American press.83 It would indeed have been impossible for the Ottoman authorities to expel all missionaries given the degree of international pressure, although no doubt they would have wanted to. But they did make every effort to keep track of their activities. On 6 December 1899, the Foreign Ministry reported that several American and English women had arrived in Erzurum intending to go to Van on a 'humanitarian mission'. The American Legation had intervened on their behalf, pointing out that if the Ottoman authorities gave these women all the help they could (such as providing them with an armed escort) this would have a very good effect on European and American public opinion. If they were hindered in their movements quite the opposite effect would be achieved.84 The vilayet was instructed to speed them on their wa.y and make sure they did not dally in Erzurum. 85 ' The various vilayets were constantly warned to be on their guard for missionaries in their areas. On 23 September 1897, the vilayet of Aleppo warned that a Dr Michael Westerly was seen 'going here and there' in the region, and was being kept under surveillance. The vilayet was instructed that '[this person] be given his passport and sent on his way forthwith'.86 Surveillance seemed to be the key word from one end of the empire to the other. On 24 May 1895, the matter of travelling American Bible Society

128

Tfce Well-Protected Domains

representatives was brought up at the Council of State. The vilayet of Manastir (in present-day northern Greece) had been approached by these representatives who had complained that the obligation placed on them by the law regulating the permits of travelling booksellers was too onerous because it included such 'unreasonable' requirements as the yearly renewal of such permits, together with a fee 'which was causing much expense and hardship'. 87 They were demanding a flat rate and a permit without a time limit which would be valid throughout the empire. The Council reiterated that it was required that 'such travelling salesmen be regulated by law' and refused to make any changes.88 The preoccupation with bad press abroad is evident throughout the Ottoman documents. O n 2 August 1891, the vilayet of Sivas complained that the Armenian community was itself responsible for unfair practices in tax collection. Various taxes were levied on the community as a whole and it was their responsibility to apportion shares to individual households. But because the self-serving rich apportioned equal shares on rich and poor alike, this meant the destitution of the poor as the local authorities were then obliged by law to foreclose on their property. In the foreign press this became another instance of Turkish oppression, The missionaries then stepped in and told the Christians, 'you see, we who have come all this way are here to help you when your own government is unable to do so.'89 There was also considerable exaggeration of grievances on the part of Christian minorities.90 Although the missionary record denies any help to Armenian revolutionary organizations, the Ottomans thought otherwise. O n 29 December 1896, the Ottoman consulate in Hoy (northern Iran) reported that a Mr Howard and a Mr Elliot had been in the town for some two months. They had been followed by members of Ottoman intelligence services who had established that they had been in touch with Armenians who had been active in the recent events across the border in Van. They had also been promising Armenians money if they emigrated to the United States via Russia. Some of the money they were distributing came from the Armenian revolutionary organization, the Dashnaktsutiun. Elliot had then moved on to Van and Howard had made his way to England via Russia.91 There was deep-seated mistrust bordering on hatred on both sides. An American clergyman, George H . Hepworth, who was sent to eastern Turkey as part of an official inspection team sent there by the sultan to investigate the attacks on the Armenian population, painted a very bleak

The Missionary Problem

129

picture: 'To educate an Armenian is regarded by a Turk, who obstinately refuses to be educated, as very close to a crime; and both the educator and the educated are denounced.'92 Even if one leaves aside the proto-racist comments about Turks who 'refused to be educated', it is clear that the missionaries were either hopelessly naive about what result their education was going to have, or worse, deeply cynical. As the educational facilities provided by the American Board were often far better than those provided by the Ottoman state, this could only have the effect of further intensifying the racial hatred that was brewing in the region, and undermining the legitimacy of the government, which was not helped by its own weakness. The naive aspect of their aims is brought out very clearly by the most up-to-date apologist for the missionary cause: '[The] American Board sponsors never intended to educate only the youth of one millet, but rather hoped that it would have an impact on all of the ethnic communities in south-eastern Turkey.'93 This was exactly what Istanbul feared. The analogy drawn by Gi-abill is perhaps exaggerated, but it does make the point: 'The American Protestants did not imagine how they might have behaved if for several decades in their homeland a foreign educational system directed by Muslims had devoted itself to, say, Afro-Americans, with the result that the black Islamic minority became more proficient than the majority of white Americans'.94 The fact was that the massacres that occured throughout the 1890s were a severe embarrassment for Istanbul, all the more so as local officials were evidently involved in events such as the burning of the American school in Merzifon in 1893, for which the sultan paid $2,200 damages.95 T h e constant pressure and bad press, which culminated in Abdiilhamid II being termed as 'the Red Sultan' was exactly what they did not need, This is best born out by the effort made to establish the exact number and location of Armenian orphans, in order to ensure that the state was seen to be taking care of its own. T h e Ministry of the Interior was ordered, on 4 April 1899, to ascertain just how many Armenian orphans were being taken care of by foreign missionaries and provide an alternative. The directive given to the ministry stated: 'Because the places set aside for the care and education of Armenian orphans... have been closed down, and this is a humanitarian matter, if the SubHme State takes care of these children without leaving this important duty to the foreigners nothing can be said ,..'96 The key phrase here is 'nothing can be said' which alludes to the constant foreign pressure on the government. Another important aspect of the Ottoman-Missionary struggle was

130

The Well-Protected Domains

the element of anti-Catholicism in the Protestant missions. Cyrus Hamlin, the founder of Robert College, the American college which began life inl863 as the Bebek Seminary on the banks of the Bosphorous, saw this struggle as the 'survival of the fittest' and referred to the Jesuit schools as the 'Roman peril'.97 This intense sectarian competition between missionary organizations sometimes reached the stage where it caused 'the transformation of religion into a sport and trade'.98 The Ottomans were able to tap into the rivalry quite successfully. Yet it meant that sometimes they came across some fairly dubious characters. A remarkable example of this was their relationship with Revd Edward Randall Knowles, a Catholic clergyman from West Sussex, Massachusetts. On July 1897, the Ottoman •-legation in Washington filed an interesting despatch. It had been contacted by Knowles who had demanded that he be decorated with the Mecidiye Medal third class for his services to the Ottoman State. 99 These consisted of various articles favourable to Ottoman interests, published in fairly obscure New England missionary papers. Knowles sent the Ottoman legation in Washington a sample of cuttings from these papers. H e wrote that Armenian revolutionaries were earning publicity at the cost of thousands of innocent lives, and he spoke of them as '[revolutionaries] whose unprincipled agitations at a safe distance have made victims of thousands of innocent souls.'100 Knowles also held forth on the commendable qualities of Islam, such as abstinence and inheritance rights of women.101 In appallingly bad French, Knowles offered his services: 'Je desire vous offir mes services gratis et sans prix comme un agent de presse ... a ma propre depense et charge absolument! Je vous at envoye plusieurs livres se regardant moi [sic]. En recompense ne pouvez vous moi donner [sic] l'Ordre de Mectdieh 3eme classe ...' The good reverend also appears to have been something of a quack telepathist, who was involved in, 'transmission of thought, without electricity or wire ... in a sympathetic unity of consciousness between the persons communicating ...' im N o doubt Abdiilhamid took a rather jaundiced view of this last item. At a more serious level, there can be little doubt that the thorniest question in the Ottoman-missionary confrontation was the issue of schooling. Through these institutions the missionaries undermined Ottoman legitimacy in their own country. The evidence indicates that this confrontation became more severe at the dawn of the twentieth century, and became more and more embroiled with the Armenian crisis. O n 16 January 1906, the vilayet of Konya was ordered to prevent the establishment of a Protestant school for girls by a certain Mademoiselle Maria Garber who 'by using

The Missionary Problem

131

the [foreign] press and by preaching has worked towards the confusion of the minds of Armenian simple folk, aiming by this to increase Protestant influence in the region.' It was also pointed out that this was a pattern whereby the school would be established without a license, thus confronting the state with a.fait accompli.103 It is also clear from the evidence that the Ottoman authorities realised that as they could not close down all of these schools, the only way to counter them was by competing with them on their own ground, that is to say, by improving the quality of Muslim schools. On 6 June 1905, the vilayet of Beirut wrote that the only way to, 'ensure that Muslim children are saved from the harmful clutches of the Jesuits and Protestants [in this city] is for there to be established a network of modern schools capable of competing with them.'104 The general feeling was that, as it was impossible to close these schools down, it was at least desirable that they be brought under license. A Council of Ministers report dated 18 October 1903 established that there were nearly 200 American educational establishments in the empire. Of these only four American schools who had received imperial fermans from the beginning. Seventy-five were accorded this permission after the fact, and the remainder did not hold valid licenses. The record of the council meet' ing made it clear that as of late the American minister had increased his pressure on the Porte to grant the additional licenses. It was also made clear that a list of teachers would be required to make sure that 'men of ill will working against the interests of the Sublime State were not employed'.105 The American Legation kept up the pressure demanding that American establishments enjoy the same rights accorded to French schools, such as freedom from taxation. This demand was seen as particularly dangerous as '[it] would have the undesirable result of increasing the number and influence of Protestant establishments.'106 Although the American Legation disclaimed official endorsement of missionary establishments, in the last years of Abdiilhamid's reign it actually stepped up the pressure to have the Porte grant official licenses to their schools. As American self-confidence increased after the Spanish American war and the conquest of the Philippines, this was reflected in their relationship with the Porte, Early in 1906, the Porte attempted to negotiate with the American Legation by telling them that they would grant licenses to American missionary establishments if they formally undertook not to admit Muslim children.The US representative told Foreign Minister Tevfik Pasa that, 'the American Government cannot accept that

132

The Well-Protected Domains

its nationals be treated any differently than the French, Russians, or the British.' It was even more troubling to hear the American add: 'The United States also has millions of Muslim subjects like the British, the French and the Russians/ The implication was very clear, Tevfik Pasa had to conclude that 'because America has become a Great Power like the others, we are forced to accept their views.'107

Conclusion The fact was that the missionaries were the best placed Westerners at the time to correctly assess the Hamidian regime's efforts at ideological regeneration. The missionary problem must be seen as one in a whole series of threats to the state's efforts to mould what it had increasingly come to regard as an Ottoman proto-citizenry. The conversion of Muslims to Christianity was something the American missionaries always strenuously denied. This official line was repeated throughout the period by the missionaries themselves and by their sympathisers: 'In imparting education they did not proselytize. The converts were not made in their colleges and schools. They sought, first and foremost, to organize an improved system of education for a people already Christian, but deplorably ill-educated and debarred by the Turkish policy for many centuries from receiving any proper education.'108 T h e Muslim majority may not have been an immediate target but it definitely 'stood on the broader horizon'.109 T h e Armenians, particularly in Anatolia, had been ill-educated. But then again so had the Muslims. Yet, it is true that the educational efforts of the state were directed particularly at Muslims in their efforts to create a politically reliable citizenry. T h e irony of the situation lies in the fact that the Ottoman aims were exactly the same as those of the missionaries, to the point where the words of a contemporary observer could apply equally well to both camps: 'The work of the American missionaries has been to produce an educated middle class in Turkish lands/ 110 The 'educated middle class' that the Hamidian regime was aiming to create was to be predominantly, but not exclusively, Muslim. The fact was that the missionaries achieved precious little success among the Muslim middle classes, as Karl Barbir has succinctly put it: 'The role of the Christian missionaries and the schools they created, while important in stimulating Westernization, pale in comparison to the Ottoman empire's much more

The Missionary Problem

133

pervasive modernizing educational efforts during the period of the Tanzimat.' m T h e other irony was that in the matter of schooling as in the matter of conversion, the Ottomans and the missionaries competed with each other. Just as Suleyman Hiismi Pasa advocated the creation of a Muslim missionary society, and Hoca Tahsin Efendi formed a 'Society for the Study of the Geography of Muslim Lands' whose purpose was to be a 'duplication of the efforts of Christian missionaries', the Hamidian governors of the 'well protected domains' understood only too well that the only way to compete against high-quality education was to create a viable mirror image." 2 The tragedy was that it was a question of too little too late. Not only in their relationship with the missionaries, but also vis-a-vis the Ottoman non-Muslims, the Ottoman efforts to evangelize and their attempts at creating a new ideology of Ottoman nationalism must be seen against the background of the 'nationalizing' of the Greek Orthodox churches.513 In the case of the Armenian and Greek schools, Ottoman intellectuals and officials clearly appreciated that modern schooling could achieve much in terms of national identity formation, and were inevitably influenced.114 The Islamic intellectual militant, Sheikh Said-i Nursi, would in fact draw specific parallels between his 'Kurdish academy', the medrese that he sought to establish on the shores of Lake Van, 'to educate tribesmen into becoming fully-fledged Ottoman citizens, and the missionary establishments in the same region." 5 The fact was that some Muslims were only too willing to send their children to missionary schools, and a few even helped their foundation. The land for the campus of Central Turkey College in Antep was donated by a Turk and a haji at that 'a large trace of land for the campus had been donated by an Aintab Turk, Kethiidazade Haci Gogiis Efendi.'516 Yet, for those involved in the struggle between Islam and Christianity, there was never any doubt as to what the stakes were: the survival of one or the other in an ideological war reminiscent of the late Cold War. A pamphlet by Hafiz Mehmed Sadik, printed in Istanbul, made this very clear. After bemoaning the fact that 365 million Muslims were living under the Christian yoke, he put the blame squarely on the missionaries: [The defence of these millions of Muslim brethren] must be primarily carried out by the intellectuals of the Islamic world. There should be no doubt that the main aim of the missionary movement today is the destruction of our religion and social morals. For this reason it is imperative that the intellectuals of Islam should persevere in their religious struggle (mucahede) against

134

The Well-Protected

Domains

the united forces of the missionary high command (misyoner erkan*i harbiyesi). T h e military imagery is significant here, all the more so as the writer appealed to Muslim intellectuals to wage a 'spiritual Holy War' (manevi cihad) against the perfidious foe.117 T h e same sentiments are clearly reflected in a pamphlet by Dr Muhleisen Arnold, a figure well known in nineteenth-century missionary circles,: 'Not only does the creed of Mohammed spread over nearly a fifth portion of the globe, but Moslem delusion and despotism extend their baneful sway over Palestine, Syria, Arabia, Egypt, and Asia Minor ... this abomination of desolation has been left standing where it ought not ...' u a Arnold also spoke of 'reclaiming] the souls of Muslims from the power of Satan.'119 The two gentlemen would have understood each other. W h e n all was said and done, however, the undeniable truth was that missionaries made very little headway among Muslims and jews. Although their later claims that they did not try are highly suspect, and smack of ex post facto justification, the fact remains that, relatively speaking, very few Muslims converted to Christianity. At the turn of the twentieth century a missionary who had worked in Turkey was to deplore the fact that: 'All our work is practically destroyed; not a single church of Moslem converts in existence in all the Turkish area after a hundred years of foreign missions.'120 Even the most radical zealot like Henry Jessup in Syria, who continued to the end to entertain hopes of converting Muslims, had to keep secret the number of converts he made and smuggle them out of the country.121 The greatest irony of all had to wait over a century to materialize. The Arab Protestants, most of whom had been converted from Greek Orthodoxy, were told by the American and European synods in the mid-1980s that it had all been a huge mistake, and that 'now they should go back' to Greek Orthodoxy. The American Principals of the missions had decided that Eastern Christianity had always been in effect 'really constituted by the Greek Orthodox Church'. 122 N o doubt His Imperial Majesty would have permitted himself a smile.

Chapter Six Ottoman Image Management and Damage Control

Colonel Selaheddin Bey looked up at the grey English sky. A steady drizzle was falling and it dripped off his epauletted dress uniform. W h a t a change after the burning sands of Yemen! T h e mourners had begun to gather around the graveside where General Christopher Teasdale OBE, V C was to be laid to rest. General Teasdale had once been Miralay Teasdale of the Ottoman Fourth Army charged with the defence of Kars during the Crimean War. Knight of the Garter and Master of the Horse to the Prince of Wales, Teasdale had fought heroically in the service of the sultan. Now it was time for the Ottoman state to show its appreciation by sending the younger military attache of the London embassy to his funeral. As the vicar intoned 'ashes to ashes, dust to dust', the reporter from the obituaries column of the Vail Mall Gazette noted that the handsome young Turk cut a fine figure in his gold braid, even if he did look distinctly uncomfortable. As their world s h r a n k a r o u n d t h e m , t h e O t t o m a n s realized that a vital aspect of survival was the projection of a positive image abroad. In a world w h e r e t h e r e was increasingly less space for t h e 'unspeakable Turk', in

Gladstonian parlance, this was more often that not a question of damage control as Ottoman statesmen tried desperately to make the case that they were a Great Power recognized by the Treaty of Paris of 1856, with a legitimate right to exist. Their effort centred around two major areas. First was the attempt to contain the damage done by incessant pejorative publications in the international media, and in other forums such as the theatre, which sought to project the Ottoman state as a degenerate nest of bloodthirsty tyrants at worst, or a decaying fleshpot of 'Oriental' vice at best. Second came the presentation of a positive image, in the course of which any opportunity to appear in the mainstream of world events was seized upon. Such opportunities ranged from financial aid for medical purposes to both the Russian and Japanese sides in the Russo-Japanese war, to the

135

136

The Well-Protected

Domains

presentation of 300 liras to the victims of forest fire in the United States. T h e aim in all of this was to buttress the Ottoman state's shaky claim to be a member of the civilized world and the Concert of Europe. Just how shaky this position was is brought out very clearly in the assessment of a contemporary British writer: They are Mahometans on Christian ground .., They are exceptional Mahometans in a Christian system. The difficulties created by their position are great; but for four hundred years they have not been great enough to prevent the Christians doing political business with them ,.. As Mahometans they are ... impracticable members of the European system.' Ahmed Cevdet Pasa was to note the irony of the Ottoman position in similar terms: 'While Rumelia, the most precious of European lands was under Ottoman controlj the Europeans tefused to consider the Sublime State as European. After the Crimean War, the Sublime State was included in the European state system.'2 The implication here is very clear once much of Rumelia, the richest of the Ottoman lands, was lost, the Europeans would be only too happy to consider the Ottoman state as 'European'. The Ottomans' deep concern with their image predated the Hamidian reign. Thus, a leading statesman/poet of the Tanzimat era, 'Ziya. Pasa, had pointed out that Catherine O of Russia conducted a much more effective public relations campaign by winning over men like Voltaire and Diderot. Ziya Pasa had also reacted to the neo-Hellenism of the fledgling Greek state, and protested that the Ottomans did not do enough to refute Greek claims that, while Greece was the land of Socrates, the Ottomans were habitual tyrants.* It appears very clearly from a close reading of the Ottoman archival records that the Ottoman decision makers would have well understood Edward Said: 'That struggle [of imperialism] is complex and interesting because it is not only about soldiers and cannons but also about ideas, about forms, about images and imaginings.'4

General damage control and image management It is no accident that one of the major collections in the Yildiz archives consists of newspaper cuttings from over 100 newspapers ranging from The Times and the Debats to obscure Serbian or Bulgarian publications. 5

Ottoman Image Management

137

These were combed daily by the Foreign Ministry Foreign Press Service (Hariciye Nezareti Dts Matbuat MiidurlUgu) for material injurious or complimentary to the Ottoman state. It was a thankless and Sisyphean task, as at the time of the Armenian massacres and the general liberal wave in Europe, the 'Grand old Turk' of Crimean War fame was long forgotten. The best the Ottoman diplomats and other officials could do was, therefore, indeed damage control, or at the very best 'image management'. The Porte was only too aware of the importance of European public opinion and attempted to influence it by sponsoring officially inspired Ottoman and foreign writers, or resorting to downright bribery. This often put them in a weak position as European and other journalists were very often not above blackmail. The difficult nature of the task is well reflected in a telegram dated 7 April 1885 from Musurus Pasa, the Ottoman ambassador in London, who was given instructions to work on The Times with a view to making it less anti-Ottoman. The ambassador reported back that he was not at all sanguine about rhe prospects of working such a transformation, as the editorship of The Times had recently changed. He somewhat lamely concluded, 'nonetheless I will do my best to make The Times change its language.'6 It is not surprising to see that a newspaper as influential as The Times should feature quite prominently in the nightmares of Ottoman Foreign Ministry officials. Again it was TheTimes that published a letter forwarded by its Paris correspondent on 25 July 1885, claiming that Europe would know no peace until the power of the sultan as caliph was reduced to that of the pope, with responsibility to spiritual matters only, and Istanbul put under the protection of a mixed commission representing all the Great Powers. The London Embassy was again instructed to write an official riposte.7 A relative of Musurus Pasa, Musurus Ghikis Bey, was also to present a favourable impression of the Ottoman state as the 'bulwark of civilization on the shores of rhe Aegean Sea*. In 1901 he published an article in the influential French journal Questions Diplomatiques et Coloniales, claiming that the Ottoman Empire 'could become something else rather than a mass of peasants labouring for the fisc ... and can indeed become the committed advocate of reform in the Orient'. 8 Publications could even cause diplomatic incidents. One such event occurred over the publication of an article in the exclusive opinion-making journal, The Nineteenth Century, by the famous ex-ambassador to the

138

The Well-Protected Domains

Porte, Sir Henry Elliot.9 In this article the ambassador argued, in sum, that the Turkish reform movement, not being given the support it deserved by the Liberal government, was vanquished by Abdulhamid II, who then proceeded to do away with its principal author, Midhad Rasa: 'Abdul Hamid was enabled to recover despotic power, unchecked by Parhamen' tary or other control... the hopes of an improved government vanished into thin air, and now the prospects of Turkey are more gloomy than at any previous time in her history.' 10 There ensued a rather muddled exchange of telegrams between Istanbul and London, the gist of which was that the sultan was assured that the Nineteenth Century article had caused only limited excitement, having been picked up only by the Pall Mall Gazette and there 'only inserted as a small item among the advertisements'.11 The Ottoman ambassador reported a few days later that Sir Henry Elliot had approached him at a reception in an effort to explain himself, but Musurus told him 'that I cut off all relations with him and made plain the anger and hurt caused by [the article]'. H e further claimed that Elliot had later written to the Foreign Office expressing his regrets. n Elliot got his answer when a work was commissioned a year later from Anne Marie de Lusignan, a Cypriot living in Istanbul under the sultan's protection. She claimed that, 'under the reign of H M Abdul Hamid II the condition of the Ottoman Empire might be much better described as one of robust convalescence than of sickness and ruin.' i3 Lusignan entered into a direct polemic with Elliot's Nineteenth Century piece, stating openly that Abdulaziz had been murdered and 'this was the first time in the history of Turkey that men accused of a serious political crime were tried by ordinary law, and in the presence of the public, the representatives of the foreign powers and the foreign press.'1'' Yet not all publications were anti-Ottoman, nor were all pro-Ottoman publications officially inspired. On 2 March 1889, the London embassy reported that an article written by a 'Hayd Clarck' had appeared in a journal called Diplomatic Flashes, extolling the fairness of the Ottoman administration towards Armenians who, he maintained, were much better off than the Armenians in the Russian empire. The writer, it was reported, had lived many years in the Ottoman Empire and had been a close friend of the late Ali Pasa. 15 A good example of image building is the draft article written by Selim Melhame, an Arab and one of the sultan's closest aides, 'for publication in the most influential Paris newspapers'. In this article there is clear emphasis on the sultan's position as 'legitimate autocrat' similar to Russian or

Ottoman Image Management

139

German rulers. After giving extensive coverage to all the beneficial and constructive aspects of Sultan Abdiilhamid's reign, Melhame's article continued: Before his Imperial Highness' reign the contacts between European rulers and the Ottoman empire were confined to the Sublime Porte; thus in the absence of direct communications relations became cold and distant. In previous practice the Porte was almost entirely under the influence of the Great Powers' ambassadors who easily imposed their wishes ... But now His Imperial Majesty can deal directly with European rulers and thus settle all the most difficult problems. The bothersome intermediaries were thus being dispensed with. Melhame also went into great detail on the importance the sultan accorded to restoring the solvency and credit worthiness of the state. Nor was it accidental that he should dwell at some length on the sultan's legitimate right to remove any official failing in his duties. The late Midhad Pasa still cast a long shadow.16 Annie de Lusignan also spoke out against intermediary power holders, in this case parliamentary democracy; I t was said long ago by a distinguished oriental that in a multitude of counsellors there is confusion. From such a confusion and its terrible results Abdul Hamid has saved his people....' 1 7 Abdulhamid exerted considerable effort to appear as one of the club of European autocrats. Part of his early diplomacy had included an approach to Bismarck proposing a German, Austrian, Ottoman alliance similar to the Three Emperor's League. Although this was designed primarily against Russia, relations with Russia improved considerably after 1880.!B Blackmail was another problem that the image-obsessed Abdulhamid often found himself confronted with. A striking case was that concerning the famous Hungarian orientalist, Arminius Varnbery. A well known figure in Hamidian Istanbul, he was allegedly many things, including confidant of the sultan, as well as double agent for the British and the Ottomans. The word got out in mid-October 1888 that Varnbery was about to publish a booklet mocking the person of the sultan and his Yildiz Palace coterie.19 Despite the intercession of the Ottoman Embassy in Vienna at the Austrian Foreign Ministry, and the efforts of an Ottoman diplomat who was sent to see him in Budapest, Varnbery insisted that he would publish in Hungarian and in English. However, Varnbery first de-

140

The Well-Protected Domains

nied that he had written such a booklet, but admitted that he had been insulted by the cold reception he had received in Istanbul during his last visit, which he blamed not on the 'Exalted Personage' himself, but on the palace clique. 20 As the matter unfolded, it became apparent that Vambery had indeed admitted that he had written such a piece, but was willing to negotiate. His price for witholding publication was two hundred gold Kras plus 'the gift of a catalogue in the Imperial Library listing three hundred and ninety four manuscripts'. 21 Foreign Minister Said Pasa recommended payment from a fund known as 'the fund for secret foreign press' (matbuat-i ecnehiyyei hajiyye tertibt). Evidently relations with Vambery improved after that. On 18 February 1889, the Vienna embassy reported that Vambery had called on them and expressed his thanks for the gracious welcome he, and several members of the Hungarian Academy, had received at Ytldi2, including a visit to the Imperial Library. Apparently Vambery got rather carried away in his rendition, claiming that when the academy members reported to their fellows back in Budapest, every time the sultan's name was mentioned the room full of Hungarian academics cheered, "Long Live Sultan Abdiilhamid Khan!'22 The ubiquitous Vambery continued to come up in dispatches. The Ottoman embassy in St Petersburg reported on 14 May 1889 that a conference given by Vambery in London, where he had 'extolled the accomplishments of the Ottoman Empire in the last thirty years', had drawn the attention of the Russian newspaper, the Noya Vremya. The Russian paper, however, attacked the orientalist whom it referred to as a 'Hungarian vagabond' whose views were outdated. The paper claimed that the Russian empire had no designs on the Ottoman empire, and in fact preferred the sultan to rule the straits rather than a rival power like Britain. 23 The presentation of the Ottoman Empire as a savage place, where travellers were not safe, was a theme frequently echoed in the foreign press. N o doubt brigandage and highway robbery were realities in Ottoman dominions, nonetheless the cover of the Petit Journal published in Paris, depicting bandits attacking a train in the east of Anatolia, drew vigorous protests from the Porte. The pictures went against the civilized image they desired to project,2'1 Another incident that smacked of blackmail involved a Hungarian news service called the Informateur Hongrois', and concerned the news item published in several Hungarian newspapers to the effect that Abdiilhamid

Ottoman Image Management

141

was p l a n n i n g t o visit E u r o p e . T h e palace reacted most virulently to this event, calling it a 'malicious a n d agitating r u m o u r ' . T h e O t t o m a n a m b a s sador in V i e n n a m a d e contact w i t h t h e owners of t h e news service w h o told h i m openly t h a t 'the regular p a y m e n t s t h a t were being m a d e to their firm from t h e O t t o m a n G o v e r n m e n t h a d b e e n discontinued; if such paym e n t s were r e s u m e d they w o u l d pay more attention in the future to news items regarding the S u b l i m e State.' 2 5 Even seemingly innocuous publications c o u l d cause major diplomatic incidents: when a booklet concerning the 1 8 7 7 - 7 8 R u s s o - O t t o m a n war appeared in Vienna, allegedly 'full of lies a n d enmity', the palace t o o k a personal interest in a t t e m p t i n g t o b a n its circulation. 2 6 T h e foreign press was regularly scrutinized for articles o n religious q u e s t i o n s . W h e n t h e French j o u r n a l Dix-Neuvicme Steele ran a piece claiming t h a t t h e seyhulislam h a d p r o n o u n c e d t h a t the haj was not compulsory, the Paris embassy was instructed to publish a rebuttal, stressing the absolute obligation of every M u s l i m with the material means to visit t h e H o l y Precincts. T h e implication of t h e French article went directly c o u n t e r to one of t h e m o s t basic claims of legitimacy of the O t t o m a n state, a n d may well have b e e n motivated by fears of pan-Islamism spreading a m o n g hajis w h o were French subjects. 2 7 Indeed, t h e O t t o m a n s seem to have h a d a rather cynical view of t h e freedom of the press. O n 28 N o v e m b e r 1895, the O t t o m a n m i n i s t e r to W a s h i n g t o n wrote;

Although it is undeniable that the United States of America has recorded great progress in science, technology and industry, the daily press is full of torrid nonsense, libel, and downright lies. Any group of rabble or scum (baldm ciplakhr) can get togerher and publish any blasphemy or invective. There is no respect for religion or rulers, and they do not hesitate to attack even their own president in the most scandalous language, to the extent of calling President Cleveland a drunk. It is useless to bring libel suits against such people because the publicity only serves to increase their fame. Good families are ruined every day by such malicious attacks. Therefore it would be correct to say that this is a savage country. 28

142

The Well-Protected

Domains

Presenting a positive image: the press and the theatre Apart from newspapers and journals, another sphere where the Ottomans strove for image management was the theatre. In the heyday of popular orientalism, it was fashionable for European companies to put on productions t h a t emphasized the exotic. W h a t Said has called 'Orientalizing the Oriental' could take the form of a genuinely noteworthy academic treatise or, equally, that of a vulgar cabaret. Another of the tasks of Ottoman representatives abroad was to prevent the staging of such productions which they saw as demeaning and insulting.29 Ottoman diplomats usually worked through the good offices of European statesmen, sometimes at the highest level. The French president Sadi Carnot was awarded a decoration, the Nifctn-t imtiyaz (The Ottoman equivalent of the Legion D'Honneur), for 'his extraordinary goodwill and help' in securing the banning of a play on the Prophet Mohammed. 30 Some nine years later it was reported that Mohammed would be the subject of another play in Paris, staged by no lesser a troupe than the Comedie Francaise. Again, the Ottoman embassy interceded to prevent the event,31 The following year notices appeared in the London press that a play on Mohammed would grace the London stage. The Ottoman ambassador immediately saw Lord Roseberry who set his mind at rest, telling him he had heard of no such project, and in any case Britain would also be wary of giving offence to her millions of Muslim subjects.32 The Porte had less luck with an opera concerning Mehmed II, the conqueror of Istanbul, which was due to open in Milan in December 1893. The Ottoman embassy in Rome interceded but were told that the Italian Foreign Ministry could do nothing. The embassy then approached the author of the opera who assured them 'no disrespect or injury to national honour' would result from the performance. The production, he promised them, had nothing to do with two previous versions of the opera due to be staged in London and Paris, 'which were banned because of the intercession of the Porte'.33 T h e authors of offending plays or pamphlets were often anonymous, only surfacing to extort money from the Porte. O n 19 July 1890, the Ottoman ambassador in London reported that he had managed to get his hands on the anonymous script of a play 'viciously lampooning the Sublime State and the August Person'.34 Upon investigation it was discovered that the author was Osman Millingen, a colourful figure and the son of Dr Millingen, Lord Byron's doctor. An interesting discussion ensued be-

Ottoman Image Management

143

tween the ambassador and Sir Philip Currie who told him that in Britain the freedom of the press meant that British politicians had to put up with the most vulgar lampooning. T h e ambassador replied 'public morals are seen very differently in the Ottoman state and attacks on the August Person are very dangerous.'35 Some of the cases where the Ottomans felt their honour was being injured bordered on the ridiculous. One such was the little skit acted out in a minor cafe chantant in Amsterdam that caused a furore. The Ottoman resident at the Hague was ordered to look into the matter, as the action of the skit evidently centred around 'a harem' and a 'sultan'. The Ottoman representative, Karaca Pasa, assured the palace that 'these cafe's are places of little consequence only frequented by the meanest sort of people.' 36 Karaca Pasa, did not omit to include a brief sketch of the 'play': Dutch peasant boy and girl as well as gentleman are shipwrecked off a strange Oriental shore. T h e boy is made a eunuch, the girl becomes the favourite of the sultan. Gentleman (who is sweet on the girl) attempts a rescue, but after pleading with the sultan, at which point 'on execute quelques danses', he ends up becoming the 'commissaire de police du Harem'. 37 Even seemingly innocent items could be seized upon as harbouring negative symbolism. Louis Rambert, director of the Ottoman Tobacco Regie, was one day approached by the Ministry of the Interior over a poster the Regie was using to sell cigars: 'The poster depicted in vivid colours a tall Englishman and lady smoking cigars astride donkeys, posing in front of the Pyramids; next to them is a beturbanned Muslim smoking his water pipe, In the background are the mosques and skyline of Cairo.' Rambert, who was Swiss, found it most comical that the Ottoman government should object to the poster. He obviously did not realise that the admission of the poster would very clearly have amounted to an aclcnowledgment of the British occupation of Egypt, a fait accompli that the Ottomans never recognized.38

Protocol and the Ottoman Empire as a'world power' Actions that could enhance the image of the Ottoman state were purposefully sought. When the United States suffered from exceptionally heavy forest fires the sultan made a point of sending 300 gold lira as succour for the victims. The Foreign Ministry reported that this gesture had received very favourable publicity in the American press.39

144

The Weil-Protected Domains

During the Russo-Japanese war of 1905, the Council of Ministers fitst considered sending a Red Crescent field hospital to both sides, but the cost of a such a venture proved prohibitive. It was thus decided that the Ottoman government would send financial aid for medical purposes to both belligerents as 'this was in keeping with humanitarian practice undertaken by other states.' T h e decision was based on the fact that the Russian government had sent such a field hospital to Turkey during the Greco-Ottoman war of 1897,40 Abdulharnid also made a special effort to be part of the inter-monarchical protocol of the crowned heads of Europe. On the occasion of Queen Victoria's jubilee celebrating the fiftieth anniversary of her accession to the throne, the Ottoman ambassador to London, Musurus Pasa, was most upset that Nubar Pasa, the 'foreign minister' of the khedive of Egypt, 'should pose as the envoy of an independent state' when Egypt was still nominally part of the Ottoman Empire. He interceded with Lord Salisbury to ensure that Nubar attended the service at Westminster as part of the Ottoman delegation. It must be recalled that, since 1882, when Britain had occupied Egypt, the country had in effect become a British protectorate. T h e Ottomans never recognised this and until the Great War the British went along with the polite fiction that Egypt was still part of the empire. What seems ironic now would have appeared quite normal in those days - that Musurus Pasa, a Greek, should defend the honour of his state against the transgressions of Nubar, an Armenian, representing another Muslim power.41 The Ottoman delegation had already survived a crisis when it became evident that the special envoy, Ali Nisiami Pasa, sent to convey the sultan's congratulations, did not hold the proper diplomatic credentials. This would have meant that 'the Rumanian and Serbian ambassadors would be received in audience ahead of the sultan's representative.' It was hurriedly arranged that Nizami Pasa receive a telegram clearly endowing him with the necessary credentials.42 The Ottoman Embassy in London continuously had trouble with those it considered 'Egyptians who did not know their proper station'. When the sons of Khedive Tevfik applied directly for an audience with the Prince of Wales, the ambassador was livid at 'the rudeness of these men' who only got their audience in the company of the Ottoman ambassador.43 This was not just empty posturing, as it must be recalled that at precisely this period, the Khedive of Egypt, Abbas Hilmi Pasa, was involved in a show of independence from the Porte. T h e jubilee celebrations indeed coincided with the khedive's visit to London and he made a point of boy-

Ottoman Image Management 145 cotting them. This was considered churlish behaviour by Istanbul,1''' On the occasion of Queen Victoria's sixtieth anniversary the Porte was more careful. It was determined that the Minister of Protocol (Te}rifat Nazin), Miinir Pasa, should be sent 'because of his considerable experience in such matters'. It was also determined that he would be accompanied by an Englishman in Ottoman service, Admiral Woods Pasa.45 O n 25th January 1892, Queen Victoria wrote to'My Good Brother the Sultan of Turkey', advising him that, it has pleased T h e Almighty to call from this World my much beloved Grandson, His Royal Highness Prince Albert Victor.' The letter ended, 'With earnest good wishes that your Imperial Majesty may long be preserved from all similar causes of sorrow!"'6 The sultan replied, expressing his condolences,'naturally We are affected by all that affects You, and are much saddened by this tragic event. There is no other solace than to surrender to the will of God."17 Sultan Abdiilhamid's twenty-fifth coronation anniversary demonstrated that to some extent the sultan had been successful in his efforts to be included in European royal protocol. The very fact that the celebrations were held is significant since twenty-five is not a particularly auspicious number in Islam. Here too, as in other matters, European protocol patterns were followed. Even Louis Rambert, the director of the Ottoman Bank, was to lament:

The celebrations have taken place and the Red Sultan of a few years ago can now congratulate himself that the whole of Europe is at his feet... Ail the powers have sent special emissaries, Russia and England their most vaunted admirals, the French Republic a general of cavalry, Italy yet another admiral, Germany an illustrious personality... And as for the ex subjects, they are in gyrations of loyalty ... Bulgaria sent its entire cabinet ...48

Even the most modest oppottunky was seized upon to present a fa- vourable image. The Ottoman Embassy made a point of sending a military attache, in dress uniform, to the funeral of General Teasdale, the former miralay because '[the General] had been decorated by the sultan during the Crimean War for his bravery in the defence of Kars.' The incongruous sight of a young Ottoman officer, standing to attention as the rain dripped over his finery in an obscure English village was said to have had 'a very good effect on military and political circles here'. Although Teasdale had

146

The Well-Protected Domains

been in retirement he had maintained close relations with the Ottoman Embassy 'in his capacity as a former comrade in arms and the Master of Horse to the Prince of Wales'.49 Foreign newspaper reporters who actually began to understand the difficulties facing the Ottomans could end up being quite sympathetic in their reporting of what they saw. Charles Williams, a freelance correspondent attached to the staff of Gazi Ahmed Muhtar Pasa, the commander of the eastern front during the 1877-78 war with Russia, wrote: 'I have seen much which has endeared to me for ever the Ottoman nation ... But any stick is good enough, in the opinion of some Christians, to beat a Turk with and the rule which they make so elastic in the case of the Muscovite is very rigid indeed when it is brought to bear on the Osmanli.. ,'50 T h e Ottoman-Russian comparison was quite current among intellectuals who pontificated on the two 'barbaric' empires of late nineteenth century Europe. R. G. Latham, a Cambridge don, was to write: The creed of the Sultan is, at least, the equal to the Czar as the representative of a grear section of a great creed ...That the Ottoman Empire is a mere encampment, that the Ottoman Turks are mere squatters, that the Ottomans hold Constantinople by sufferance, that they should be driven back to the deserts and steppes of their original Asia, are flowers of rhetoric which may be found in writings of influential authors ... As far as the question of time is concerned, the title of the Ottomans to Constantinople is what that of the English was to London in the reign of Edgar ...51 Of course much of this grudging acceptance and admiration centred around the fear of Russia, and the accepted wisdom that the Ottoman Empire and Britain had a special relationship against a common enemy. T h e rhetoric of the Liberals had still not entirely effaced this even during the Armenian massacres of the 1890s: 'O'Connor and Sykes shared an attachment to Abdul Hamid's regime ... as well as a belief that a special relationship existed between the British and Turkish Empires ... this had survived the attack of London Liberals who pictured the Ottoman Empire as a slaughter-house where the 'Terrible Turk' butchered helpless Bulgarians and Armenians. 52 The sultan knew who to cultivate. When the influential dilettante Orientalist, Mark Sykes, and his wife visited Istanbul on their honeymoon, they were given the 'image tour' which included, 'a hospital for small children recently built by Abdul Hamid in memory of one of his daughters,

Ottoman Image. Management

147

and a new school the Sultan had founded for sons of Kurdish aghas and Arab sheikhs.'53 Sykes' admiration of the Ottomans as a 'ruling people' was based on his youthful travels in Turkey. While travelling in eastern Anatolia in the company of a Turkish military detachment, they had a brush with bandits. The officer in charge, to whom Sykes had just offered a cigarette, 'calmly lit his cigarette [while bullets buzzed around his head] then ordered the infantry to divide into two detachments and chase the would-be robbers away.' Sykes marvelled at his coolness; like 'an English officer on field duty' he seemed to think no more of the affair 'than a farm boy would of crow scaring',3,1 The similarity of the Turks and the British as imperial powers also drew the attention of other writers. Karl Blind, in 1896, was to severely criticize Gladstone's comment that the Turks were 'the one great antihuman species of humanity' with the remark: 'How if he had been reminded by a member of the anti-human race that there are some Irish Home Rulers and Secessionists who, in United Ireland, speak of England, on account of her rule of the sister Isle and her many polyglot dominions, as the "Anglo-Saxon Grand Turk".'55 Indeed the Ottomans gloated over Gladstone's troubles with the Irish Question. On 23 May 1883, the Ottoman embassy in London reported that 'the Irish issue has caused no end of troubles for Mr Gladstone.'56 Three years later, the embassy wrote that the 'Queen has no affection whatever for Mr Gladstone' whose Irish policy was 'looked upon with great dislike by classes of good breeding.'57 The Irish and Armenian questions were linked together by Annie de Lusignan: 'As a matter of fact there are more Christians in high office in Asia Minor than there are Catholic nationalist magistrates in Ireland, and religion is less of a bar for advancement under Sultan Abdul Hamid than under Queen Victoria.'58 It would appear that 'people of good breeding' in England did indeed have some sympathy for Abdulhamid. As Constance Sutcliffe wrote in 1896: 'It is, for instance, as intolerable to Ali Baba that an Armenian should sit by the wayside and mix Rahat-Lakoum as it would be for the Duke of Allshire that a Hottentot should place himself on the crimson benches of the House of Lords and take part in debate.'59 On the Ottoman side, it was felt that foreigners had no idea of what life in Turkey was like. Fatma Aliye, Cevdet Pasa's daughter and one of the first Ottoman women to have intellectual aspirations, wrote in 1891 that foreigners had no idea of what true Muslim Turks were like as they only came into contact with 'Westernized' Turks. She felt, therefore, that it was necessary to invite vis-

148

The Well-Protected

Domains

iting dignitaries into 'true Muslim homes' and show them true Turkish

life. 60 The Ottoman self-conceptualisation as an imperial power like all the others was reflected in a work assessing the colonial penetration of Africa. Mehmed tzzed, a palace translator, spoke of a 'dark continent' where 'civilized' powers sent colonists. The Ottoman state was encouraged to do the same and spread 'the light of Islam' into 'savage' regions. T h e general tone of the book is very much in keeping with the 'white man's burden' approach of late nineteenth-century colonialism. Izzed's very definition of colonialism is worth quoting at length; The concept that we express in our language with the terms, mustemlekat or mustemir&t is called a 'colony' in European languages. The practice of colonialism is the process whereby civilized countries send surplus populations to continents whose people are in a state of nomadism and savagery (hal-i vatyet ve bedeviyette bulunan), thereby rendering them prosperous. This also means that [die civilized] country thus acquires new markets and adds the colony to its territories.61 It is interesting that the terms the Ottomans usually used to refer to their own nomads and tribes - living in a 'state of nomadism and savagery' ~ should be used in a colonial context here. Cleatly, the Ottoman notion of 'civilizing nomads' appears here as a migrant concept in a colonial setting.

Conclusion Why was the ftn-desieck Ottoman elite obsessed with its image abroad? How does this fit into the world context? The truth of the matter was that, in the late century, the whole world was discovering the power of the popular press, and the Ottomans were no exception. The Ottoman archives are full of newspaper cuttings referring to various aspects of the Dreyfus affair. The scope and implications of the matter were very well understood by the Porte and the palace, down to Zola's famous declaration, 'J'accuse'.62 W h e n Gladstone's pamphlet 'The Bulgarian Horrors' caused mass demonstrations in London, it became very evident that something had to be done to counter the negative propaganda. One of the answers was of course counter-propaganda, particularly counter-propaganda aimed at the

Ottoman Image Management

149

Muslim world. O n 15 April 1880, the British ambassador in Istanbul, Henry Layard, wrote to Salisbury that the newspaper Vakit had announced the expected arrival of Mehmed Emir Ali Khan, an eminent Indian Muslim, 'to gather subscriptions for the Holy War'.63 The Ottoman-financed Paik-i Islam, an Urdu and Arabic publication printed on the official presses in Istanbul, was aimed at the Muslims of India. This newspaper was seen as a serious threat by Sir Louis Malet, the undersecretary of state for India: I f a printing press were found in the Vatican from which were issued by the Pope's connivance journals for circulation among the Irish Catholics, stirring them up against British rule it would hardly be treated so lightly.'64 In another India OfBce memorandum it was openly stated that; 'There have been of late many proofs of the fact that Mahometans in general very are widely impressed with the belief that their interests as a religious community are closely dependent on the maintenance of Osmanli power.' This situation was largely the result of, 'the greatly improved facilities of correspondence, of travel, of the interchange of thoughts and ideas'.65 The contrapuntal character of the Muslim international press was well understood by India Office officials. When the Ottoman papers published an appeal for public collection of subscriptions for the Muslims displaced by Russian armies, the India Office view was that, 'It is very much in tone the sort of appeal that would be made in England if it were prepared to collect subscriptions for distressed Christians in Syria for example.'66 T h e effort to present their views in the press was also to involve the Ottoman ruling elite in propaganda for internal consumption. A very interesting example of this is the magazine ServctA Vunun, aTurkish equivalent of ^Illustration or the London- Chronicle. The magazine featured extensive illustrations covering what would today be called 'human interest' .stories. Some of the themes that frequently appear are 'cruelty to blacks in the United States', 'a tram robbery in the middle of Paris', 'appalling living conditions of Chinese coolies in San Francisco', or 'British captain throws his mistress overboard on the high seas'. All of these stories come with vivid illustrations depicting a hapless black person being burned at the stake, or a pretty young woman clinging to the boots of the cruel mariner as she dangles over the side. T h e connecting theme appears to be the degenerate and violent nature of the Western world.67

Chapter Seven The Ottoman 'Self Portrait'

T h e dervish zikr on Fifth Avenue was coming to a close. The five 'dancing dervishes from Arabia' twisted and turned to the accompaniment of flutes and drums. As the tempo quickened in the late summer heat, and sweat beaded the brows of the dervishes under their tall ktilahs, several street urchins started to laugh and mimic the bizarre strangers. 'I remember seeing something rather similar among the Chipewwa,' remarked a tall well dressed lady to her companion in the Panama hat. 'Well blow me down, don't they ever get dizzy!' shouted a man rather the worse for drink. The crowd grew larger as the my built up to its reedy falsetto climax and the drum beat louder. A few policemen on horses appeared to investigate the commotion. Suddenly, the music stopped and a man stepped forward wearing a smile and a/ez and holding a collection box.

A t t e m p t s t o m i n i m i z e ' t h e exotic' T h e O t t o m a n ' s obsession with their image was m o s t certainly a reaction t o w h a t w o u l d t o d a y b e called O r i e n t a l i s m . E d w a r d S a i d ' s i d e a t h a t ' O r i e n t a l i s m is fundamentally a political doctrine willed over t h e O r i e n t because the O r i e n t was weaker t h a n t h e W e s t / would have struck a reson a n t c h o r d a m o n g many O t t o m a n s in the late nineteenth century. 1 T h i s was t h e state of m i n d which informed the actions of t h e O t t o m a n m i n i ster t o W a s h i n g t o n D C w h e n h e intervened to stop t h e ' d a n c i n g dervishes of Egypt' from performing their ztkr in the streets of N e w York. 2 T h e m i n i s t e r , Mavroyeni Bey, h a d h a d it b r o u g h t to his attention t h a t , ' s o m e thirty m e n from Egypt h a d been b r o u g h t to the U n i t e d States by a C h r i s tian n a m e d M a l l u h from Syria w h o h a d p r o m i s e d t h e m money a n d fame.' A l t h o u g h a handful of t h e 'performers' were b o n a fide dervishes, m o s t were 'people of the m e a n e s t sort w h o h a d been bribed (esafil-i nasAan ve

150

The Ottoman 'Self Portrait'

151

cehelcden}.' Having become destitute, and being obliged to continue performing in order to eat, their presence in the streets of New York was 'causing insult and injury to Islam'. The minister had instructed the Ottoman consul in New York to arrange for their removal. H e also wrote to the State Department urging that they put a stop to the insult. There ensued a very lively exchange of correspondence with the State Department telling the minister that 'they could not interfere in the internal business of the States,' and Mavroyeni retorting that 'he knew the American constitution as well as they did and how would they like it if the Shakers were to perform on the streets of Istanbul?'3 The 'dervishes' were rapidly shipped off home. The struggle against 'the exotic' could indeed take on the proportions of a diplomatic incident. As in the case of the dervishes, a 'Panorama of the delights of the Bosporous' was to provide further headache for the Porte. In December 1893, representatives of a British firm arrived in Istanbul declaring that they proposed to stage a 'live panorama depicting the delights of Istanbul' in London. The object of their visit was to procure the services of the various 'components' of the display, sleek rowing boats, brawny oarsmen and lissome dancing girls were to be rounded up and transported to London. When the Yildiz Palace got wind of this the reaction was instantaneous. The grand vizier was ordered to stop 'such mockery that will damage national honour and propriety... (ahlak ve aclati milliyeyi tezyif)'* T h e sultan personally objected that, 'certain gypsy and Jewish women should be displayed as the so-called specimens of Oriental peoples ... all such display is demeaning and uncalled for.' Although the firm actually pleaded that 'they were not depicting real people but were only creating an illusion', the palace remained unmoved. The British Embassy became involved, arguing that 'this was against freedom of commerce and the arrangements had already been made.' As a solution the firm was told that they could buy the rowing boats but that no 'live displays' were to be permitted. 5 Attempting to escape the designation 'exotic', Ottoman statesmen sought to capitalize on aspects of their society and civilization which were attuned to the mainstream of world trends. By emphasizing symbols which had come to denote modernity, the Ottoman state was staking its claim to the right to exist. One very telling testament to this effort is the collection of photograph albums that the sultan arranged to be presented as gifts to the Library of Congress in 1893 and the British Museum in 1894.6

152

The Well-Protected Domains

T h e themes of the albums provide very useful clues to the 'imperial self portrait'. The photographs cover four main categories. 'Category F consists of 'Views, Buildings, Monuments and Antiquities'; fairly basic scenic photography such as views of the Bosphorous and the Golden Horn, the imperial mosques, and the various palaces.7 'Category IF features, 'Military, Naval, Rescue, Ancillary Services, Military and Industrial Establishments'. 8 This is a much more interesting selection which depicts scenes such as 'The launching ramps for rescue boats at Riva', featuring strapping coastguardsmen standing by well-found rescue craft, or 'Divers of the Imperial Naval Arsenal' gazing complacently at the camera wearing full paraphernalia. Also featured in this section are 'Cannon drill on the Imperial Frigate Mahmudiye', a n d ' Imperial firemen at drill', the latter a particular irony in a city notoriously plagued by disastrous fires.9 'Category III' is perhaps the most interesting and features 'Educational institutions'. Here it is plain that the Hamidian state considered not only its elite educational establishments but also relatively humble schools as showpieces of its claim to modernity. It is particularly significant that girls of 'The Uskfidar School of Trades and Crafts' or the 'Emirgan Primary School for Girls' are shown with their heads uncovered, sporting hand' some diplomas. T h e 'Asiret School' for the children of tribes and nomads is particularly well represented. On 4 November 1892, the Ministry of War reported that, according to decree, it had arranged for the photographs to be taken of'those students in the Mekteb-i Asiret who are well proportioned of body and limb (mutenasib-iil aza)\ The pictures were to be taken in groups of two and were to be of uniform size. T h e famous photographers, Abdullah Freres, were to be commissioned.10 The results of their labours feature in the albums as Arab and Kurdish boys staring at the camera wearing somewhat puzzled expressions and 'full local costume".11 'Category I V is entitled 'Horses, Imperial Stables, and Yachts'. This category is significant as it serves as a message to the outside world that the sultan understood the 'engine casings of power'.12 Photographs in this series feature 'Asil, a white horse', 'Gazelle, a bay mare born at the Imperial Stud Farm', 'The Imperial Yacht Sultaniye', 'Sleeping quarters aboard the Sultaniye'.13 The emphasis on yachts and horses was part of Sultan Abdulhamid's effort to be part of what Benedict Anderson has called 'the semi-standardized style' of 'civilised' monarchy in the late nineteenth century.14

The Ottoman 'Self Portrait'

153

International congresses I fAnother way of 'minimising the exotic' and appearing as a member of theV. \\civilized family of nations was through participation in international con- J• gresses. The nineteenth century was the century.._o£..the international congress with subjects tanging From the preservation of wild species in Africa,_tq improvements iri'tKe'con'ditionrd'F'tKe'Blind, to world health . problems. In relation to the various congresses of Orientalists in p a r t i e s "Tar the Ottomans found themselves in a delicate position. O n the one hand they wanted to be included in such prestigious world events, on the other they feared being the objects rather than subjects of 'study'.15 Therefore, when Dr G. W. Leitner, the organizing secretary of the ninth congress to be held in London in September 1891, called on the Ottoman sultan to be the official protector of the congress this was to cause some serious soul searching. Leitner made a point of stressing that the previous congresses had been protected by various crowned heads of Europe and that the Maharajah of Jahore was to be on the organizing committee. As there was likely to be "discussion of the various aspects of Islam', b u t ' these were not to be directed to any religious or political aim and were to be of a purely scholarly nature/ it was decided to send Ottoman observers to the conference. There is, however, no record that the sultan accepted the trusteeship. 16 The Ottomans were also represented at the Thirteenth World Congress of Orientalists m Hamburg, but rather than sending a specialist they settled on a diplomat of moderate rank from the Berlin embassy.17 An international congress on archeology provided another forum where science combined with national prestige. At the meeting of the World Archaeology Congress in Moscow in 1892, the organizing committee made it known to the Ottoman representative that they would like to hold the next congress in Istanbul. The ensuing debate in the Council of Ministers centred around the fact thatjajljdvilized nations take_part in the_congressjland that if it were held in the capital of the Sublime State, 'this would serve as an opportunity for the whole world to witness the progress in education and science that has taken place [in the sultans' reign].' It was also pointed out by the ministers that since the discussions of these congresses had nothing to do with politics and were only concerned with science, they should not be objectionable.18 The desire to be included in even fairly ephemeral events in order to present a respectable profile is discernible throughout. One such event

154

The WelhProtected

Domains

was the occasion of the meeting in London of representatives of those powers with African possessions. The topic was the preservation of the wild animals, birds and fish of the continent. The Council of Ministers decided that since the Ottoman state had such possessions in Africa, it should be represented.19 In the same vein it was decided that the Ottoman state should be represented at the World Congress to Improve the Conditions of the Blind, to be held in Brussels in 1902. T h e discussion related to the topic stressed the point that 'many of His Imperial Majesty's subjects suffer from this condition.' 20

T h e Hamidian state and world fairs:'The whole world is watching!'21 fThe world fairs presented one of the most spectacular arenas for the disp l a y of pomp and power in the heyday of imperialism.22 The Ottoman [ state had made a point of being present at nearly all these fairs since the I first World Exposition in London, in 1850.23 In the Hamidian period, jwith the increased emphasis on the presentation of a 'civilized profile', attendance at the fairs acquired a new importance. As Burton Benedict put it: 'A world's fair can be seen as one of a series of mammoth rituals in which all sorts of power relations, both existing and wished for are being expressed.'34 In a world ruled by giants such as Britain, France, and Germany, the Ottoman Empire was inevitably relegated to the position of an 'also ran'. Yet, the very fact that it felt obliged to participate, and spent money it did not have on 'being there', shows that for the Ottoman Empire this was not just a matter of picayune posturing. Its 'fairs policy' consisted of two main elements. First, there was the aim of presenting the , / Ottoman Empire as the leader of the Islamic world yet a modern member ^ \ of the civilized community of nations. Second, constant vigilance aimed to repell any slight or insult to the Sublime State's prestige. This obsession with prestige was the principal motivation for Hasan Tarhan Pasa, the Ottoman ambassador in Spain, when he reported that in the Barcelona fair of 1888 a 'panorama of the Battle of Plevna' had been on display. T h e Battle of Plevna in Bulgaria was the critical turning point of the Ottoman-Russian war of 1877-78. The heroic defence of the position by Ottoman forces led by Gazi Osman Pasa had captured the world's imagination and sympathy at the time. T h e ambassador reported that, as far as he could see, there was nothing objectionable in the display, but he was sending a photograph of the panorama to Istanbul. 25

Tfcie Ottoman 'Self Portrait1

155

The occasion of the opening of the 1889 Paris Exposition on the centenary of the French revolution was something of an embarassment for the powers still ruled by monarchies. The Ottoman ambassador in Paris, Esad Pasa, somewhat anxious as to the appropriate course of action, telegrammed that the British, Austrian, Russian and German ambassadors had made it clear to him that they would absent themselves from Paris on the occasion. This would mean that he would be the 'only ambassador of a Great Power to attend the ceremony'. He was instructed by the Porte to absent himself'from so insalubrious an event so damaging to the idea of monarchic sovereignty.'26 T h e Chicago Columbian Exhibition of 1893 is most interesting from the Ottoman perspective, as the considerable material relating to it in the Ottoman archives invites the following questions; How did the Ottomans see their position as the only Muslim Great Power? What was the rationale informing them in their selection of what to display? What were the perceived threats to the image they wanted to project? The usual arrangement when the Porte decided to take part in a fair was for it to contract out the whole affair to a commercial firm. The firm in question for the Chicago event was a certain Ilya Suharni Saadullah 8C Co. On 24 May 1892, the Ministry of Trade and Public Works forwarded to the Porte a copy of the contract it had concluded with Saadullah 8C Co. The contract provided for the building of an 'Ottoman Bazaar' in the shape of the Sultan Ahmed fountain at the entrance of the Topkapi Palace.27 The building was to be quite a modest affair of 400 square meters. Yet, the feeling among the ministers was that a definite location should be settled as soon as possible with the fair administrative committee 'lest delay oblige us to accept a dishonourable location'. It is thus very clear that the Ottomans were very alive to the matter of prestigious location in an exhibition.28 Saadullah & Co. were also to make plaster casts of the Egyptian obelisks in Sultanahmet Square, as part of the Ottoman exhibit would consist of replicas of these monuments. T h e contract also provided for the construction of a mosque which would be open to prayer by all Muslims at the fair, with the critical proviso that 'proper Muslim etiquette (adab) should be observed at all times and visitors to the fair be admitted [into the mosque] only at the discretion of the [Ottoman] representatives.'Part of the exhibit was to be a theatre, and here, too, stringent conditions were laid down: 'no plays injurious to the honour and modesty of Muslim women or damaging to national honour and prestige (haystyet ve adab-i mcmlekete mugayyir) are to be performed'29 in close proximity to a 'mosque',

156

The Well -Protected Domains

as seen in the Egyptian exhibit at the Paris Exhibition.30 When the Ottoman minister to Washington, Alexander Mavroyeni, reported on the opening of the Chicago fair he gave a realistic assessment of the 'Turkish village'. His view was that, 'although it cannot hope to compete with the exhibits of countries who spend thousands of francs, given our means it is a success.' He, too, reported on the mosque which was part of the Ottoman display and was being used by Muslims at the fair. Mavroyeni, however, had one disturbing development to add; some 'Nestorians' had built a mosque outside the perimeter of the fair and were charging money to display 'Muslim devotional practices'. H e had intervened with the Chicago city authorities to have the 'mosque' closed down. 3 '

o^> ;>

The Ottoman's were objecting to what Edward Said has termed 'representations as representations, not as "natural" depictions of the Orient.' 32 The effort to depict themselves as 'modern' or even 'normal' clashed head on with the West's relentless quest for the 'unchanging Orient'. It was this contradiction that we encounter in the matter of the photographs to be sent to the Chicago fair. As the sultan himself dictated to his private secretary: 'Most of the photographs taken [by European photographers] for sale in Europe vilify and mock O u r Well-Protected Domains. It is imperative that the photographs to be taken in this instance do not insult Islamic peoples by showing them in a vulgar and demeaning light.' Yildiz ordered that all photographs taken were to be vetted by the palace before ' they were sent to Chicago. 33 Once the exhibit opened, the Ottomans made every effort to appear as second to none, even in matters of minor protocol. When the fair administration decided that each nation represented at Chicago would have a special day as a 'national day', the Ottomans were initially a bit resistant. However, they rapidly recovered and informed the fair administration that the anniversary of the sultan's accession to the throne was their 'national day'.34 It was also determined that the band should play the Hamidiye March on the occassion, and that specially made gas lights be displayed reading 'Long Live the Sultan! (Padisabtm Qok Yasa'.)' when lit. There was also to be a fireworks display.35 To coincide with the Ottoman exhibit at Chicago, the publication of a newspaper in Turkish, Arabic and English was planned. The editor was to be the Syrian intellectual Suleyman al-Bustani. It was specified that publication in Chicago could take place providing 'that nothing about politics is mentioned and only news relating to the Ottoman exhibit, and the progress seen in all things in T h e Well Protected Domains is featured.'36

The Ottoman 'Self Portrait'

157

The most fascinating aspect of Ottoman involvement in the world fairs phenomenon was the way decisions were reached as to what to display and how to display it. The process of selection of those facets of their society and civilization that the Ottomans wanted to show the world provides a kind of code matrix to the Ottoman self-image. When a certain Raci Bey from Acre proposed the creation of an 'Ottoman Hippodrome' [Osmanli At Meydam) for the Chicago fair, his view was that 'it is well known that among the most coveted and popular of Ottoman products are fine Arab horses and camels which are renowned throughout the world.' Raci Bey proposed to export some forty horses and riders as well as several camels. This, he stated, 'would serve to increase the glory of the Sublime State through the presentation of its natural resources and subjects, by drawing the admiring regards of foreigners'.37 In this letter Raci Bey enclosed a 'Programme for the Ottoman Hippodrome at the Chicago Fair'. In six articles he set out his ideas for the display. In article one, under 'Horses', it was stated that 'forty horses of fine breed will be piclced and each shall be given a special name.' In article two, under 'Camels', Raci continued: 'Six of the most robust Arabian beasts shall be chosen, the spectacle of a whole train of these mighty animals being led by one small Arab boy will cause much amazement.' In article three under 'Riders', Raci really warmed to his subject: o n j j ^ J ^ ^ o m e ( AjrabsjQ£^p donal skills of horsemanship and javelin throwing (arid)' These men must also be 'of honourable character (efeJ-i irzguruhundan}'.3* Because the traditional Arab horsemanship games were 'irregular and chaotic', and in order to 'prevent any occurrence which might not look good to foreigners', Raci proposed that the riders in question be put through a 'period of training' before embarkation which would involve a sort of dressage with the riders sporting Ottoman and American flags.39 It seems that Raci Bey's proposal was accepted by Istanbul as an order was issued to send to the fair two Ottoman regular cavalry officers 'skilled in the arts of horsemanship' who would train the Arab riders-'10 A display along these lines was indeed taken to Chicago and appeared in the fair as i 'Bedouins in their encamprnent'.'51^,; From tKe above it would appear that the Ottomans had internalized much of the West's perception of the 'Orient', even as they were striving for equality. The horses had to have 'special names' because this was how things were done in polite society. T h e 'little Arab boy' leading away a whole string of fabulous beasts, any one of which could have trounced

158

The Well-Protected

Domains

him, was also a suitably exotic motif. The proviso that the horses and their riders be 'handsome', that they refrain from coursing about in a babble of noise and dust (one can almost hear a pasa in Istanbul add 'in the Arab manner'), and ride around in an orderly fashion, shows the extent to which the Ottoman missiontivtlizatriceinformed their thinking. The fact that the original suggestion came from an Arab subject who was seeking to please Istanbul (doubtless also to make some money), gives the story an additional twist. The same intertwining of self perception and perception by the outside word can be traced in the debate over the sending of janissary models to the Chicago fair. The suggestion came from a certain Nuri Bey, a notable of Gemlik on the Aegean shore, who proposed that an 'Ottoman Museum' be set up at the fair, featuring models of janissary troops wearing their traditional garb. This proposal was opposed by the Sublime Porte who felt that 'this would evoke unpleasant memories among the Christians'. Nuri, however, appealed to the Ottoman commissioner at the Chicago fair, Hakki Bey, pointing out that, 'today the Ottoman nation has reached the level of European civilisation and the [present] appearance of the military and civilian personnel of the Sublime state is proof enough [of this] .'42 Hakki Bey's view was that all this fuss was unneccessary as life-size models of janissaries were already on view in Istanbul, and he saw no reason why they should not be displayed at the fair given that 'in America and in Europe there exist such museums where the most bizarre of old military costumes are proudly put on display.' If a good impression was to be conveyed, said Halcla Bey, 'the jannissaries should be displayed side by side with the modern Ottoman officer in full dress uniform, which will show the progress that has been recorded since the Tansimat.' 43 Hakki Bey regularly reported on the progress of the Ottoman Pavilion at the Chicago fair. Evidently, he had a clear conception of the basics of showmanship and representational protocol. On 13 February, he was to telegram that, 'the German commissioner to the fair has given a lavish banquet at the Hotel Lexington at which all the flags of participating states were displayed and the Ottoman flag had pride of place.' As these banquets were likely to go on, it was 'only in keeping with the glory and splendour of the Sublime State' that the Ottoman commissioner be given funds, too, to enable him to reciprocate.'14 Hakki Bey also requested that he be sent two guides who spoke English and were capable of giving information on the various displays in the Ottoman exhibit. H e was duly granted both requests.45

The Ottoman 'Self Portrait'

159

By 15 May he was writing that the fair was getting off to a slow start, with the Russian and Belgian pavilions only just being completed. He noted that the fair would not get into full swing for another few weeks. The goods scheduled to arrive for the Ottoman pavilion had also been late and many needed repair. Most worryingly, the Navy display, consisting of the sultan's ceremonial state rowing boats, had gone missing. Hakki Bey also gave news about the 'Ottoman Hippodrome' project. Raci Bey and his Bedouin riders, as well as the horses and camels, had indeed arrived safely but, he lamented, 'the field put aside for the hipprodrome is at present a morass of mud'. The Bedouin, who were in severe danger of catching pneumonia, had been found suitable quarters. The two Ottoman cavalry officers, Tevfik Bey and Sabit Bey, had been accommodated in the building occupied by Hakki Bey himself. But things did not augur well for the project: 'the affairs of this firm are distinctly lax.""5 Apparently everything was downhill from then on as things did not go well for the Ottoman cavalrymen. The original arrangement meant that Raci Bey's firm was to pay their expenses, but the firm rapidly went bankrupt, leaving Hakki Bey responsible for the officers and the Bedouin. The exasperated commissioner was to write: 'as it is not quite suitable for the Imperial government to take legal measures against a firm which is not doing very well, and as it is not in keeping with the glory of the Sublime State for its officers to be destitute [arrangements should be made for them to be sent back].' The officers were duly shipped back at the earliest possible date.47 Financial considerations indeed placed a constraint on Hakki Bey's efforts at a proper presentation of the Ottoman self portrait. He was to lament that, 'the Americans do not provide any service gratis, and we are presented with a bill for even the smallest service.' Although the Porte had budgeted 7,500 lira for the fair, a few weeks after the opening it was becoming clear that additional funds would be necessary. On 15 May, Hakki Bey gave a detailed breakdown of the monies he had spent and asked for an additional 2,500 lira.48 A major consideration was always the keeping up of appearances. Hakki Bey stated that as all the countries' commissioners had office premises in addition to their pavilions, it was unthinkable that the Sublime State should not do the same. Accordingly an office building 'of most pleasing appearance' had been built and furnished with carpets and precious cloths from the state carpet factory at Hereke. It was also necessary to give banquets at the opening and on the occasion of the birthday of the sultan. To justify these expenses, Hakki

160

The Well-Protected Domains

Bey emphasized the importance of the Ottoman presence in the fair: As this [fair] is an event at which all the civilized nations of the world are represented, and even obscure states such as the Kingdom ofjohore from the Malaca peninsula, and some small Central American republics whose very names are unknown, make great sacrifices ro show themselves, it would be unthinkable for the Sublime State not to do the same ... 49 These words from the Ottoman commissioner at the Chicago fair almost echo the statement made by a contemporary literary critic, J. Snider, who called the Chicago Midway Plaisance the 'sliding scale of humanity'. Nearest to the central point of the white city were the Teutonic and Celtic races as represented by the German and Irish villages the Islamic World was situated somewhere down the middle of the Midway.50 According to this scale, the Ottomans were in a respectable neighbourhood, the 'Turkish village' was in the same block with the 'German village', 'Panorama of the Bernese Alps', and the 'Dutch village', and just down the road was 'the Murano Crystal exhibition'.51 Hakki Bey got what he wanted as Ottoman goods continued to arrive in Chicago. In August the Ottoman exhibit was in full swing as large samples of Ottoman handicrafts and jewellery arrived and were placed in 'four large glass cases shaped as crescents around a case fashioned as a star'. These were 'exquisite examples of Ottoman porcelain called the Hamidian style worked with jewels and with some depicting historic scenes' which, when placed in the Womens' Building, 'found great favour with the ladies as they rushed to see them and marvel at the workmanship.' Also displayed was a scale model of the Sultan Ahmed fountain at the entrance of the Topkapi Palace. It was further reported that the Ottoman exhibit had 'eclipsed the displays of all other nations', 52 T h e evident desire to compete successfully as a modern civiliggfiUaemo ber of the club of Great Powers also expressed itselfmjrhe^ffgttjo^grevent ^objectionable portrayals of things 'oriental'. The Ottomans particularly objected to the display of'dancing girls and dervishes'. During the Anvers fair in 1894, the Sublime Porte instructed its embassy at Brussels to 'prevent the display of anything which can touch pur national honour and religious beliefs'. They specified that 'shameful displays of Islamic women or dervish life are to be prevented at all cost.'53 It appears that their fears had some foundation. A certain Agob Balyan was reported to be acting as the agent of a

The Ottoman 'Self Portrait'

161

Belgian firm which intended to stage such a display, and was 'going about Beyoglu and Sulukule contacting gypsy women and buying objects supposedly used during Mevlevi dervish ritual',54 The Ottoman embassy in Brussels duly appealed to the administration of the fair and was assured that any stand or display 'injurious to the modesty of Muslim women or damaging to national pride' would be immediately closed down.55 A few years later, the Ottomans were approached with an invitation to take part in a fair that was to be held in Brussels. The Ottoman honorary consul in Brussels, Monsieur C. H . Sudre, appealed to the Porte to take part, 'in order to show the world the progress accomplished by the Ottoman state in the glorious reign of His Imperial Majesty.' If the sultan were to provide the Ottoman stand with his patronage and protection, 'as had the king of the Belgians and other sovereigns of Europe', it would be properly represented.56 The sensitivity to becoming the 'object' of display was not exaggerated or misplaced. At the Budapest fair of 1896 the Ottoman representative discovered to his horror that paying visitors were being admitted to the mosque at their pavilion, 'to watch Muslims praying'. Measures were immediately taken to have the mosque closed down.57 The fact that such a 'display' took place was all the more surprising as the Ottoman exhibit was run by the ubiquitous Armimus Vambery who would be expected to know better. The display included agricultural and artisanal produce as well as 'things of historic value which had been taken from Hungary.'5" It is interesting to note that in many instances the Porte's sensitivity focused on what would today be called orientalist depictions of Islamic women and dervishes.59 This sensitivity comes out very clearly in the proposal of the famous publicist, Ebuziyya Tevfik, to supervise the creation of a 'Museum of Ottoman costume' for the Paris Exposition Universelle of 1900. Tevfik pointed out that as he was in partnership with a Frenchman, 'whose only concern is to make money', it was absolutely necessary that he be present to supervise the propriety of the exhibit: If this exhibit is not supervised by someone who has the greatest regard for his national honour, the Europeans, who always exaggerate, will no doubt render our costumes and civilization unrecognizable. They will depict outhistorical costume in the most garish fashion, making it virtually unrecognizable, and most of all, display our women in the most scandalous Fashion, totally contrary to the precepts of the $eriat and the values of Ottoman honour.60

162

The Well-Protected

Domains

Tevfik went on at some length about the total acceptability of the depiction of old Ottoman military costume yet, as in the case of the Chicago Fair of 1893, he too specified that the military costumes 'would not be janissaries'. If properly supervised, said Tevfik, the exhibit would 'act as a very good countermeasure for all the unfavourable publicity that is propagated against our nation in Europe...' 61 The Paris Fair of 1900 was co be an event of considerable importance for the Ottomans. As early as the summer of 1896, measures were taken by the Ministry of Trade to begin a selection of the goods to be sent, and to arrange for the purchase of a suitable location on the fair premises. It was also determined to send an Ottoman representative to oversee construction. It was recommended that Yusuf Razi Bey from the ministry's railroad department be sent, as he 'had been educated in Europe'.62 Yet, it appears that the Sublime Porte was none too keen on its own subjects visiting the Paris fair. The Ministry of the Interior mentioned that 'since in the previous fair in Paris (1889) many Ottoman subjects who went became destitute, placing a great burden on the embassy in repatriation expenses ...', it was now deemed desirable to ask for a cash bond from those who wanted to go. It was also stated that a thorough inspection of the identities of those intending to travel to Paris was to be carried out. 63 A few months later the Ministry of Trade reiterated that it was of the utmost urgency to send a representative as 'the other participating states have already done so.'64 Yet, by 2 February 1897, no representative had been sent and the Paris Exposition committee was beginning to pressure the Ottoman embassy which recommended that a military attache at the embassy, Leon Bey Karakyan, be appointed as temporary commissioner.65 It appears that, by the end of 1897, the Porte had still not appointed its official representative, or purchased its lot. The Ottoman ambassador wrote that he had been approached by the chief commissioner for the fair, Monsieur Picard, who had told him that if the Porte did not reply soon it would not be admitted. The ambassador evinced some alarm that the best locations for the 'Grands Palais' had already been taken. 66 The French kept up the pressure, and on 21 June 1898 the head of the Istanbul Municipality reported that the French embassy was complaining that the plans put forward by their architect, Vallacery, for the Ottoman pavilion were not being considered with due speed.67 Some months later, by 8 September 1898, the Ministry of Trade had determined that some 50,000 lira were needed for the Paris fair. A com-

Tke Ottoman 'Self Portrait'

163

mission had been established, headed by Hakki Bey from the Sublime Porte Legal Councillors' Office, 'because of his experience during the Chicago fair', and including Saadullah Efendi, the contractor who had undertaken the official Ottoman exhibit in Chicago. It was also determined that the Ottoman exhibit would cover 'over 500 square meters'. Each vilayet was to be ordered to inform the Ministry of Trade as soon as possible as to what it proposed to exhibit.68 Things seem to have speeded up by the end of the month. The Ottoman ambassador wrote to the Porte that the total projected expenditure for their exhibit was to be in the region of 300,000 francs, of which 10,000 was immediately needed for the building of a wall around the Ottoman lot, and as contribution to the fair's railway fees. The ambassador specified that 'special accommodations have to be provided for important guests to our exhibit in order to entertain them in a manner fit for the glory of His Imperial Majesty.' Specially woven Gordes and U$ak carpets were to be shipped forthwith as, 'we are behind the other countries in our preparations and this means we have to double our efforts.'69 It would appear that the Ottoman exhibit at Paris, when it actually materialized, was quite a success. Horses were again prominent, with the Imperial stud winning nine prices. The prize-winning horses were to be photographed and an album presented to the sultan.70 Also, the Ottoman pavilion ended up being built on the prestigious Rue des Nations, between the pavilions of the United States and Italy. It was the only Islamic power to get such treatment; Persia's much smaller pavilion was in the back row, between Peru and Luxembourg, and Egypt was treated as a British colony.71 Private entrepreneurs, who invested a great deal of money in displaying some aspect of the 'exotic Orient, also put pressure on the sultan. On 30 December 1903, a banker, Monsieur Konta, wrote to the Sublime Porte asking that it co-operate in the putting together of an exhibit at the St Louis fair. Jerusalem would be featured and an Ottoman pavilion included. Konta promised to undertake to deliver safely all the components of the Ottoman exhibit and to return them. The centrepiece of the exhibit would be an exact replica of the Mosque of the Caliph Omar: 'As His Imperial Majesty Abduihamid Khan is the Caliph of all the Muslims who will visit the fair, from China to Sumatra to Java and Africa, it is only appropriate that he appoint the imam and the muezzin/ The organizers also promised to collect donations for the Hijaz Railroad. The Ottoman cruiser, Mecidiye, which was being built at an American shipyard, would be

164

The Well-Protected Domains

launched for the occasion, with the entire crew being shipped out at the firm's expense to man her. The proposal seemed too good to be true, and the bite came at the end: 'If for some reason which we are unable to fathom the Porte refuses our request, we will be obliged to apply to the good offices of the Egyptian, Tunisian, Algerian and Moroccan governments ... As our sole aim is to celebrate the glory of His Imperial Majesty, we would only resort to such a measure with the utmost reluctance.'72 Evidently Monsieur Konta had been well briefed about Abdulhamid's obsession with his claim to being the sole Defender of the Faith and Protector of the Holy Places.73 T h e Foreign Ministry took pains to point out that the conditions offered were most advantageous, and that to bring in the governments of rival Muslim powers, some of whom were under Christian control, would be a grievous loss of prestige.74 It would appear that some of the conditions at least were met as an order was issued absolving all goods destined to the St Louis fair and returning from it from customs duties.75 Another matter which was seen as a loss of prestige at the St Louis fair was the display staged at the Bulgarian pavilion of burned out villages in Ottoman Rumelia. At the height of the Balkan crisis, this was just the sort of publicity the Ottomans did not want.76 The agricultural fair that the Bulgarians had organized in Plovdiv in 1892 had already been snubbed by Istanbul. It was felt that it would be inappropriate to participate in a prestige event in a country which had only recently been an Ottoman province, and was still technically under Ottoman suzerainty,77

Conclusion It will be recalled that when the British firm which proposed to launch a 'panorama of the Bosphorous' was prevented from collecting the live components of its display, it retorted that it was 'not depicting real people but only creating an illusion'. Similarly, when the Ottomans planned their 'Ottoman Hippodrome' for the Chicago fair, they too created an illusion when they objected to 'unruly Arabs' and specified that the Bedouin horsemen be trained by regular army cavalry officers. However, the two illusions pulled in opposite directions. The frustrated'panorama of the Bosphorous' pushed the exotica of the Orient, whereas the Ottoman efforts were precisely to play down the exotic and present a 'civilized' image of their Arab subjects. Yet, in a strange paradox, the Ottomans were viewing 'their' Ar-

The Ottoman 'Self Portrait'

165

abs through the very same prism through which the Europeans viewed them. Inadvertently, the Ottoman self-image adopted much the same value system as that of the West: 'Yet, European paradigms were not simplistically appropriated; they were filtered through a corrective process, which reshaped them according to self visions and aspirations."78 In a sense, therefore, the Arabs in the Ottoman Hippodrome were to some extent 'people as trophies', with the 'corrective' elements of regular cavalry drill.79 As world fairs were 'mammoth rituals' involving'blatant efforts to manu~ facture tradition and to impose legitimacy', it was imperative for the threatened Hamidian state to make a good showing.80 In this process of 'communication, discussion, and mutual recognition among ... unequal partners [Islam and the West]' the Ottomans assigned themselves the thankless task aiming for a recognition as the only Islamic Great Power which wanted to participate in modern civilization.81 In this sense, Zeynep Cpelik is incorrect in her assertion that: 'The Turks who wrote about their visits to the international fairs ... did not analyse the differences between their culture and the representations of that culture abroad... The Egyptians, in contrast, were deeply concerned about the image of their country.'82 The cases discussed in this chapter show that the Turks were concerned to the point of obsession with their image. What was being undertaken by the Ottoman state was nothing less than a symbolic statement of its right to exist in a world which was constantly trying to relegate it to his: tory.

V

-av 4N o>P SP > o
!

^

Chapter Eight Conclusion

T h e ba^sic contention of this study has been that a legitimacy crisis' took place in the Ottoman empire in the second half of the nineteenth century. This was a crisis that had both external and internal dimensions. The external dimension was the uphill struggle to secure the acceptance of the Ottoman state as a legitimate polity in the international system. The internal dimension was the struggle to overcome the 'legitimation deficit' that accrued as the state permeated society physically and ideologically to an unprecedented extent. 1 A further contention has been that the Ottoman empire in the period 1876-1909 was undergoing an experience that was broadly similar to that which was being experienced in the world at large in the same period. In the Ottoman case a background of crisis lay behind it. I have argued that the Ottoman experience represents a case of imperial adjustment to the challenges of the times, comparable in varying degrees to that seen in other multi-ethnic legitimist systems. Finally, a / major theme has been the effort made to create a modern secular state w\using traditional religious motifs and vocabulary.

T h e 'tacit knowledge' of a service elite In a situation of constant crisis it was imperative that the staff of the empire, from grand visier to the lowliest kaymakam in the provinces be imbued with the correct ideology. Ideally, for a member of the service elite, the sultan's authority should constitute the 'natural order of things'. Then and only then could he set about inculcating the same values in the population at large. The framework to support such a system was to be created through elite education, service in the field, shared experience, and the system of rewards and punishments intended to instill respect (or, failing that, fear) and loyalty.

166 i

.1

Conclusion

167

The intention was to induce that phenomenon which Edward Shils has called the repository of 'tacit knowledge' in a service elite: 'Some of [scientific knowledge] remains inarticulated, but that too is presented and received. The transmission of the articulated part of the rational, scientific tradition is made effective by the reception and mastery of its inarticulated part. The mind of the recipient is formed by this reception of both the articulated and the unarticulated.' 2 The way the universe of 'tacit knowledge' changed in the Ottoman Empire is observed by $erif Mardin: 'The gradualness with which the Turkish world view changed may be seen in that by the end of the reign of Mahmud [II] (1839) the net increment of his rule, in this respect, had been purely and simply the establishment of the respectability of change? The fact that the new sultan, Abdulrnecid I, should use migrant notions from the vocabulary of Young Ottoman progressives, like 'fatherland' and 'liberty', in his accession speech is further proof of 'genuine change in the political athmosphere'. 4 Although the change was indeed slow and incremental, if one examines the century between the reigns of Mahmud II and Abdulhamid II (1808-1909) it becomes apparent that the political context changed agreat deal. It was explicitly stated in the Tanzimat Edict of 1839 that people had rights and liberties which could not be arbitrarily taken away from them. To understand how this would become 'tacit knowledge' in Shils' sense, and figure in the calculation of fine tuning that informed the rule of Abdulhamid, let us take the case of Abdtilhamid's highly unpopular measure of exiling and executing the great reformer Midhat Pa$a. It was before then accepted as part of the normal order of things that grand viziers could be summarily executed. Mahmud II for example had executed the powerful Pertev Pasa. Carter Findley indicates that Pertev had indeed enjoyed such power as the favourite of the sultan that the populace knew him as the 'uncrowned king (tugsuzpad'tsab)'• When he fell from power on 12 September 1837, and was ignominiously executed by poisoning and strangling, his rival Akif Pasa secured a pithy announcement of his death in the official gazette, stating that he had 'died suddenly (fue'eten)'. It is striking, too, that Pertev's son-in-law, Vassaf Efendi, a secretary at the palace and also a favourite, and Pertev's brother, Emin Efendi, should both suffer judicial torture before their execution.5 However, by 1881, when Midhat was exiled to Taif in the vilayet of Yemen., and there strangled, the legal norms guaranteeing freedom and rights had also led to an accumulation of tacit knowledge that made this

168

The Well-Protected Domains

act reprehensible, even though the infamous 'exile clause' in the 1876 constitution was evoked as justification. Even if the Ytldiz Tribunal, as it was grandly called, was little more than a kangaroo court, the very fact that Midhat Pasa was tried at all is indication enough of just how much things had changed. 6 Moreover, the palace never officially admitted that it had ordered the execution and claimed that he had died from 'natural causes'. In fact Uzun^arsih, who wrote the major monograph on the issue, states that Abdulhamid deliberately put about the rumour that Midhat and some of his companions had escaped, in order to 'dispell suspicions that they had been killed'. The sorry tale of Midhat's exile and execution is veiled in furtiveness and the awareness that a shameful act was being committed pervades. The guardians of the prison where Midhat was kept attempted several times to poison him to give the impression that his death had indeed been natural. After the execution, Midhat Pasa's servant was imprisoned for three years to prevent him from telling the true story,7 The odium surrounding Abdulhamid's name has been carried over to the present day in the historical literature which views him as an ogre and Midhat as a 'martyr'. 8 Mahmud II, on the other hand, is seen as a courageous reformer, despite the fact that it was he who ordered the ruthless massacre of thousands of janissaries, the so-called'auspicious event' (wk'ayt hayriye)'? It was this shift in the tacit knowledge of the elite which enabled them to provide the impetus for secular reforms within the 'cocoon' of a religious vocabulary. Starting with the Tanzimat Edict of 1839 itself, which gave the §eriat first priority, but went on to make critical breaks from it, the mission c'wilisatrice of the Tanzimat man was carried forward into the Hamidian era. Talking about the Tanzimat, Cevdet Pasa stated: 'During the reign of Abdulmecid Khan the auspicious Tanzimat was declared, and all subjects were declared equal before the law. This was what the $eriat demanded (icab-i $er'isi bu idi).' O n the matter of reform, the pasa was to state: 'Because changing times require that even some rulings of the serial change, it is no longer possible to continue the ways of a century ago.'10 In a similar tone, Kuciik Said Pasa, referring to the §criat as the 'basic law of the state' referred in the same breath to 'the need to suit measures to changing times as is ordained by the §eriat'.n The $eriat therefore became a fluid notion and was used in conjunction with, indeed as the ultimate justification for, change. As Kantorowicz pointed out in the case of European jurists seeking to legitimate change:'... the jurists custom of borrowing from ecclesiastical

Conclusion

169

language for secular purposes h a d its o w n tradition of long standing, for it was a practice as legitimate as it was old to draw conclusions dt simtlibus ad similia.'12 T h e s t a n d a r d historiography on t h e nineteenth-century O t t o m a n E m pire proceeds o n the spurious assumption of the split between t h e so-called 'Westernizing', a n d 'revolutionary' reforms of the T a n z i m a t a n d the 'conservative' H a m i d i a n period. However, Ortayh's character sketch of t h e 'Tanzimat m a n ' is equally valid for the ' H a m i d i a n ' :

The Tanzimat administrators combined pragmatic reformism and conservatism in their personalities, and as such, their world views, modes of behaviour, policies, are all typically representative of the new man in nineteenth-century Ottoman society. Yet it is also evident that this new Ottoman would consciously continue the world view and life style of the old masters of society. n A r e p o r t prepared by A h m e d Cevdet Pasa, one of the m o s t p r o m i n e n t statesmen of t h e nineteenth century, provides a striking insider's view of this world, a n d is therefore w o r t h citing at some l e n g t h . " I n remarkably straightforward language, Cevdet talks a b o u t critical issues such as t h e foundation of t h e legitimation of t h e state, Islam as a unifying factor, O t t o m a n relations w i t h the W e s t e r n powers, a n d t h e position of C h r i s t i a n s i n t h e empire, a n d as such this d o c u m e n t represents a valuable 'repository of tacit knowledge' of a leading O t t o m a n : Since the time of Yavuz Sultan Selim [Selim I] the Sublime State has held the caliphate and it is thus a great state founded on religion. However, because those who founded the state before that were Turks, in reality it is a Turkish state (bir devlct-'t Turkiye'dir). And since it was the House of Osman that constituted the state this means that the Sublime State rests on four principles. T h a t is to say, the ruler is Ottoman, the government is Turkish, the religion is Islam, and'the capiral is Istanbul, If any of these four principles were to be weakened, this would mean a weakening of one of the four pillars of the state structure. 15

It is striking t h a t one of the m o s t c o m p e t e n t a n d oft-quoted of O t t o m a n s t a t e s m e n s h o u l d explicitly say that t h e O t t o m a n state was, in the first instance, T u r k i s h . A l t h o u g h other Muslims have their place, the T u r k s m u s t always c o m e first:

170

The Well-Protected Domains

The Sublime State is a great structure made up of various peoples (akvam) and strata (sinuf), all of these constituent elements are held together by the sacred power of the Caliphate. Because the only thing uniting Arab, Kurd, Albanian, and Bosnian is the unity of Islam. Yet, the real strength of the Sublime State lies with the Turks. It is an obligation of their national character (kavmiyyet) and religion to sacrifice their lives for the House of Osman until the last one is destroyed. Therefore it is natural that they be accorded more worth than other peoples of the Sublime State.16 Yet, the pa§?a took care to point out that the Arabs were 'those who spoke the language of our religion', and should therefore be respected. Unfortunately, he added, this fact seemed to escape the notice of, 'some state officials in Arabia who still insult the Arabs by calling them fellah. Naturally the Arabs therefore hate the Turks.' If these errors were corrected, and responsible officials appointed to the Arab provinces, 'The Arabs who have been under the rule of the House of Osman for all these centuries, will not seek to go their own way, or those who do will be so few as not to matter. Nor were the Christians to be neglected: Some people of conservative bent argue that it is not advisable to employ Christians in important positions. This is the greatest of errors. Ever since the Abbasid Caliphate, in commercial and financial matters, Christians were employed. The very founder of the Ottoman state, Osman Ga*;i, was accompanied by his comrade in arms, Mihal Bey, and history is unclear about whether Mihal ever converted to Islam.18 Also, in his famous Tarih, Cevdet was to praise highly the efforts of Ferah Ali Pasa, 'the spiritual conqueror of the Circassians and the Abhaz'. The pasa receives honourable mention for 'spreading Islam (in the Caucasus) and teaching the Circassian and Abhaz tribes what state and nation is (devlet ve millet nt oUugunu ogretmif) and binding them for ever to their rightful sovereign, the Caliph of Islam'. Although the events Cevdet recounts took place in the late eighteenth century (1195-99 A H ) , his whole treatment of the subject is much more in keeping with the nineteenthcentury mobiligational ethic that stirs in his other writings, such as the report mentioned above.19

Conclusion

171

T h e Ottoman Empire in world perspective It is often forgotten that, at the turn of the century, the majority of the world's countries were still monarchies. The institution of monarchy had suffered a mortal blow as a result of the French Revolution, but it took the First World War to finally dispatch it, In the interim, not only did monarchies survive but, in cases such as the Japanese, actually revived themselves. T h e Ottoman sultan, the Meiji emperor, T h e Russian tsar, the Habsburg emperor were all drawn towards the twentieth century at different tempos, but down broadly similar paths. All invested in a recharged state mythology, which they sought to inculcate through mass education. All invested to varying degrees in the inevitable technical trappings of modernity: railways, the telegraph, factories, censuses, passports, steamships, world fairs, clock towers, and art-deco palaces. All looked to each other to see how their peers were playing the role of'civilised monarchy'. This process of emulation among the crowned heads of the world had a definite competitive bent, but the model was inevitably the concept of modernity as seen in the core cultures of Western Europe, perhaps even mote specifically, the France of the Second Empire. The Ottoman Empire's story, as told in this book, has been largely the story of the participation of the only sovereign Muslim world power in this process of adaptation. T h e attempts of the Ottoman sultan to reinforce his ideological position at home overlapped with his efforts to resist the ever-increasing pressure of rival empires. The HijaK and the haj became the arena of confrontation between the Ottoman centre and Muslims from India, the Dutch East Indies, Russian Central Asia, and Egypt, all of whom the Ottomans suspected of serving as stalking horses for the imperialist powers. The use of the term 'non-Ottoman Muslim' or simply 'foreigner' for these elements was indicative of a fundamental recasting of Ottoman identity, and a corresponding shift in attitude to their own people. What was occuring was nothing less than a move towards conceiving a loyal population as a proto-citizenry. It was no longer enough to be any Sunni Muslim what mattered was the passport one held. The Ottoman saw himself as an equal participant in the zero-sum games of world politics, and demanded to be treated as such. T h e European saw him as an anomaly, a master who should really be servant, a ruler who should really be a subject. It was this dichotomy which produced the Ottoman obsession with image and a determination to defend it against all

172

The Well-Protected

Domains

slights, insults and slurs. Even worse, of course, was the possibility of being ignored. This was the rationale that informed the Ottoman statesmen who understood full well the importance of what Edward Said has termed 'representations'.20 Thus a good part of the Ottoman legitimation effort was directed towards aiming the right 'representations' at the outside world. Nor was this simply a question of decorum, of pretty posturing or the polite niceties of diplomatic move and counter move. The Ottomans were aware that it was precisely decorum and ability in diplomatic subtleties that determined who remained standing and who fell, whose summer palaces were burned by foreign troops on the rampage, and who maintained their dignity even when bloodied in war.21 Ahmed Cevdet Pasa was to note the importance of reciprocation in matters of protocol: One of the new usages of the rimes is that on the occasion of the accession anniversary and birthdays of foreign rulers guns are fired in salute, their ships are decorated, and their ambassadors receive the congratulations of other missions. Although these ceremonies were carried out in Istanbul and the sultan also sent his congratulations, the foreign powers did not reciprocate. For the official days of celebration of the Ottoman empire were religious holidays. Now since the Ottoman state has joined the European family of nations, Fuad Pasa has set aside the birthday and accession day of the sultan and made them official holidays.This obliged the European states to reciprocate ...22 T h e humiliation of 1876, when the representatives of the foreign powers had met in Istanbul to discuss Turkish reforms, without a Turkish delegation, still rankled.23 The Hamidian state made a point of participating in all major and many minor international conferences, ranging from the 1884 Berlin Africa conference which was to determine spheres of influence in the continent, to the conference on the protection of flora and fauna in the same continent. The claim of the Sublime Porte in both instances was the same: 'we have the right to be present because we also have posessions in Africa.'24 Another world arena where Istanbul made a point of being present, were the world fairs, or expositions unwerselles. Abdiilaziz had actually been the first Ottoman sultan to travel outside his empire for peaceful purposes when he visited the Paris Expo of 1867. Although Abdiilhamid stamped on all rumours that he was going to visit Europe, he made a special effort to see that the Ottoman state was represented at all major fairs.

Conclusion

173

'De-legitimation' T h e Ottoman caliphate was abolished on 3 March 1924 by something as banal as an act of parliament.. The crowds did not surge into the streets, and the cobbles did not run red with blood. An institution that had been recognized as the highest authority in Islam vanished overnight. How did this happen? Kantorowicz refers to Richard II's ritual disrobing of the trappings of monarchy as 'a coronation in reverse'?5 In this context, the scene depicting the Young Turk delegation delivering thefetwa of deposition to Abdulhamid also bears important clues as to the process of de-legitimation. T h e parliamentary delegation was to tell the sultan: 'We come representing the parliament. The nation has deposed you. But your life is secure.'36 The phrasing here, with its reference to the 'nation' having deposed the monarch, is very evocative of the Jacobin attitude to 'Louis Capet' (Louis XVI) , The delegation states that the 'nation' has deposed 'you' (using the familiar sen, or tu„ form of you).27 But in the next breath, the speaker, Esat Toptani, goes on to speak as a man conditioned by the ancien regime: 'but your life is secure' and here the 'your' is the respectful formsi'z (like) vous.28 The wording of the actual/etu^ of deposition itself is also interesting because it provides vital clues as to what constituted valid grounds for de-legitimation. The very first reason given for the deposition is: 'If a person (zeyd) who is the Imam el Muslimin removes and deletes certain important matters relating to the serial from the holy books and causes the said books to be banned or burned ...*29 Here reference was being made to the attempr to monopolize the printing of the Qur'an and its importation from non-Ottoman Muslim lands. It is ironic that a measure which Abdulhamid hoped would reinforce orthodoxy and bolster his position as caliph, was in fact used as the major accusation against him. 30 The second accusation brought against the sultan was, 'the killing, imprisoning, and exile of subjects without ser'i justification'. This was a clear reference to the exile and execution of Midhat Pasa, and as such it is very significant that a practice that had been a frequent occurence, the exile and execution of a grand vizier, now had to be justified . A clearet manifestation of the shift in the 'tacit knowledge' informing Ottoman state tradition could not be wished for. A further dimension of the same shift was the fact that the fetwa was accompanied by an act of the national parliament. 31

174

The Well-Protected Domains

In fact the Ottoman Hanefi Caliphate ended with the deposition of Sultan Abdulhamid II. The sultan/caliphs who followed him were more or less the puppets of the Young Turk junta. Whatever mystique remained in the office of the sultanate/caliphate was completely hollowed out by deposition, defeat in war, the attempt to militarily oppose the Kemalists in Anatolia, and the escape of the last sultan, Mehmed Vahdettin, on board a British warship. This latest episode shows how far things had moved. The news of Vahdettin's escape created something of a shock in the Grand National Assembly and actually caused one of the members, and a baji at that, to comment: 1 told his Excellency the Pasa [Mustafa Kemal] that [the British] were up to something... I actually said they will take the fellow and go ..."The use of the familiar term 'fellow' is striking here, all the more so as the word used in Turkish, hcrif, is in fact quite rude. 32 The period from the abolition of the sultanate (1 November 1922) to the declaration of the Republic (23 April 1923) to the actual abolition of the caliphate on 3 March 1924, is in many ways the 'coronation in reverse' of the last Ottoman caliph/sultans, as the caliphate is first shorn off from the sultanate and then abolished altogether. During this period as a means of discrediting the House of Osman, Mustafa Kemal embarked on a systematic campaign of vilification." Halil Inalcik has pointed out how the 'Caliphate question' became a symbolic issue between Kemal and his opponents, both sides using religious arguments.34 Meanwhile, attention was given to how the affair was to be presented to the world Islamic community. In the course of 1923, when Kemal was preparing the ground in Turkey for his coming move, the Grand National Assembly in Ankara issued an official statement. 35 The official position was that the caliphate was not being abolished, but in being separated from the sultanate was reverting to its original form. In fact the whole institution of the caliphate was very deliberately played down and de-emphasized. The declaration stated that this was purely an administrative matter, and was not even a theological issue: 'il resort de ceci que la question du califat n'est pas, comme on le croit a la base de la religion.'36 T h e sublime irony was that the selfsame Hanefi ftkh which had for centuries been used to legitimate the Ottoman caliphate was now used to de-legitimize it. In the document there are repeated references to the famous Hanefi alims, including Abu Hanifa himself who is cited as never having legitimated the rule of the Ommayad and Abbasid caliphs.The Ottoman sultans therefore, were only given the title, 'par suite d'une simple habitude'. In fact the whole issue of who should inherit the caliphate had been deliberately left vague by the

Conclusion

175

Prophet, and the office was actually a purely temporal authority, 'comme un President de Republique'.37 The document set the trend for future discussions of the 'caliphate question', as Hamid Enayat has noted: 'The document is also significant because of its pioneering value in modern discussions of the caliphate: nearly all critics and supporters of the caliphate after its abolition seem to have done no more than develop its broad propositions.'' 8 In the historiography of the Arab lands, the Ottoman period has consistendy been seen as a period of decline and degeneration; yet a revisionist approach is developing. T h e writers in this school point out that the history of the region cannot be understood, as Youssef Choueiri points out 'unless the Ottoman option is restored to its centrality in the Arab World .. .This restoration can no longer be the familiar drawn-out decline, but as a response to European penetration and domination.' Choueiri also draws attention to the need to 'grasp Ottomanism in its secular dimensions and implications'. 39 In the case of mid-nineteenth century Egypt, Ehud Toledano warns against 'the imminent loss of the Ottoman historical context' which has led to a distortion of terms and their relation to the past. Toledano echoes Albert Hourani's warning of losing the 'Ottoman background of the Middle East'.110 In the Syrian context Philip Khoury points out that the French mandate system that replaced the Ottomans ran into serious difficulties because:

There was a significant difference in the nature of the new imperial authority it was illegitimate and thus was unstable. France was not recognized to be a legitimate overlord as the Sultan-Caliph of the Ottoman Empire had been. The Ottoman Empire had behind it four centuries of rule and the very important component of a common religious tradition.41 It must also be recalled that the majority of the Arab subjects of the Ottoman Empire stayed loyal right to the bitter end. British expectations of massive desertions from the Ottoman army during the First World War proved futile, despite brutal measures undertaken by the Young Turk regime/52 So, was everything in vain? Did the last ditch defence of the Ottoman centre fail completely? Were the Kemalists a total break with the Hamidian past? The answer is no to all three questions. In fact a direct thread can be drawn from the gilded antechambers of the Sublime Porte to the ramshackle parliament building in Ankara. Cevdet Pasa and Mustafa Kemal,

176

The Well-Protected Domains

not to say Abdulhamid and Mustafa Kemal, would have found that they had much in common. T h e 'young officials' who 'were propelled into the realm of abstract possibilities', and who 'transcended the present and soared into the future', in soaring took much of their Ottoman intellectual baggage with them. 43 Ankara, in fact, took Cevdet Pasa's offering of a 'land of the Turks', but set aside the religious cocoon. In effect, Cevdet Pasa's vision was to become history. The generation which took Turkey into the twentieth century cut its ideological teeth on a certain corpus of intellectual raw material. This was the 'tacit knowledge' which the Kemalist cadres soaked up in Abdiilhamid's schools and which was frequently quite distinct from the approved curriculum. T h e interpenetration of the two worlds, Western and Eastern, since the late eighteenth century, and with accelerated tempo since the Tanzimat reforms, had produced a society where the religious could express itself in secular terms, just as the secular could use religious motifs. The Kemalists were in many ways the personification of this cross-civilizational synthesis, a graft that took. T h e countless memoranda in the Ottoman archives, the reports, telegrams, pleas for help, calls for money, troops, the anguish behind the cold officialese, the ruthlessness behind the clinical orders, the subtle equilibrium of weighed possibilities, all represent aspects of the 'fine tuning' process. Of course we now know that the empire collapsed despite all the efforts detailed above. But that should not blind us into thinking that the failure of all these policies was self evident or inevitable, The decision makers and the policy formulators of the time sincerely believed that they could 'save the state'. Ahmed Cevdet Pasa is a representative voice: 'All these dangers demand that the Sublime State carry out the neccessary reforms. If a serious and effective policy [of reform] is carried out, the Sublime State will, in a matter of ten or fifteen years, be strong enough to resist all danger.'44 The true 'tragedy' in the Greek sense of seeing what is coming, of knowing what to do to avoid it and yet of being unable to resist the march of events, is largely the story of the preceding chapters. In more ways than one, as liber Ortayh has eloquently expressed it, this was really 'the longest century of the empire', when everything hung in the balance and the Ottoman lengthened his stride.45

Notes

Introduction 1. Halil Inalcik, The Ottoman Empire in the Classical Age: 1300-1600. (London 1973) p. 30. The Ottoman garrison only surrendered in 1481 after the death of Mehmed the Conqueror. Albert Hourani noted the specificity of the Ottomans when he called them, 'the Romans of the Muslim world'. See Albert Hourani 'How Should we Write the History of the Middle East?' International Journal of Middle East Studies 23 (1991) pp. 125-36. 2. Hugh Seton Watson, 'On Trying to be a Historian of Eastern Europe' in Harry Hanalc (ed.) Historians as Nation Builders. (London 1988) p. 7. 3. Eric Hobsbawm, The Age of Empire (London 1991) p. 283. 4. E A. K. Yasamee, Ottoman Diplomacy. Abdiilhamid H and the Great Powers 1S78-18S8. (Istanbul 1996) p. 4. The legendary siege of Plevna took place between July and December 1877 when an Ottoman army, much against Western expectations, held off the Russian forces in a heroic defence, T h e Ottoman commander, Osman Pasa, surrendered with full military honours. For an eye witness account of the battle see William von Herbert, The Defence of Plevna (1877). Tel el-Kebir was the battle in the Egyptian delta where the nationalist forces of Ahmed Urabi were cut to pieces by a British expeditionary force in September 1882. Ahmed Urabi was deliberately humiliated in a show trial and exiled to Ceylon. O n this see, Alexander Scholch, Egyptfor the Egyptians. The Socio-Politkal Crisis in Egypt 1878-1882 (London 1981). At the battle of Omdurman in the Sudan, the Sudanese Mahdi's forces were routed by the British. On this see Mathew Anderson, The Ascendancy of Europe 1815-1914 (London 1972). pp. 227-8. 5- Mr Owen Davis, 'Those Dear Turks. A Lecture at the Congregational Church, Lee, on 1st November 1876/ Papers on the Eastern Question. (London 1876) p. 15. 6. Prime Ministry Archives Istanbul (Basbakanhk Arsivi) hereafter BBA. Yildiz Esas Evraki (hereafter YEE) 31/1950 Mukerrer/45/83. Said Pasa was referring to the Ottoman Empire's inclusion in the European Con-

177

178

The Well-Protected Domains

cert with the Treaty of Paris of 1856. (Duvel ve emaret'i Hiristiyane icine caktluh kalmtstz). 7. I refer here to the state power elite, the people involved in decision making in the establishment, not the opposition. Although the Young Turk opposition to Abdulhamid saw itself very much as part of the elite, their views will not be alluded to here. For an excellent recent source on the Young Turk opposition see, Sukrli Hanioglu, The Young Turks in Opposition. (Oxford 1995). 8. T h o m a s L. Haskell, 'Objectivity is not Neutrality' History and Theory vol. 29, (1990) pp. 129-57. 9. This is by design and not by accident.The 'Armenian Question' as it has been somewhat insultingly called, is a major issue which merits scholarly study in its own right. For two excellent recent studies of Armenian nationalism see R. Grigor Suny, Looking Towards Ararat. Armenia in Modern History (Bloomington and Indianapolis 1993) and Claire Mouradian, Armenie, Que sa'tsje ( Paris 1995). Also the works of Anahide Ter Minassian and Ara Sarafian. 10. For a very competent example of legal history on the Ottoman Empire see, H u r i Islamoglu-Inan, 'Koyluler, Ticarile^me Hareketi ve Devlet Gucuniin Mesrulasmasi/ Topium ve Bittm 43/44 (1988-1989) pp. 7 - 3 1 . 11. Engin Deniz Akarli, 'The Problems of External Pressures, Power Struggles, and Budgetary Deficits in Ottoman Politics under Abdulhamid II (1876-1909): Origins and Solutions/ Princeton University P h D Dissertation 1976. Particularly pp. 1-9, and pp. 7 7 - 9 8 . 12. Engin Akarh, 'Abdulhamid II Between East and West,' Paper presented at the colloquium: 'Education, Nation Building and Identity in the Period of Abdulhamid' Bad Homburg, Germany July 12-15 1993, p. 23. 13. Carol Gluck, Japans Modern Myths. (Princeton 1985) p. 4 1 . 14. L. Carl Brown, Jnternational Politics and the Middle East. Old Ruks Dangerous Game. (Princeton 1984). Particularly p. 67; 'The overwhelming majority of the Ottoman Muslims did, however, identify with the Ottoman Empire. It was the last great Muslim state holding out against European encroachment'. 15. Some more recent work in the old British diplomatic history tradition does however indicate a refreshing revision of classic positions. See, David Gillard, 'Salisbury/ in Keith M . Wilson (ed.) British Foreign Secretaries and Foreign Policy. (London 1987) p. 133: 'Ironically the most intelligent of the rulers in this region [the Middle East] during the Salisbury era faced the same problems as he did, namely how to defend their interests in a dangerous world with inadequate resources, and were just as skilled as he was in the economical and realistic use of power in pursuit of strictly limited goals.'.

Notes

179

16. Perry Anderson, Lineages of the Absolutist State (London 1974). Particularly statements like that on p. 365: 'The economic bedrock of the Osmanli despotism was the virtually complete absence of private property in land.' When? Where?. 17. Caroline Finkel, The Administration of Warfare: the Ottoman Military Campaigns in Hungary. 1593-1606. (Vienna 1988) particularly pp. ix and 313, See also, Finkel, 'French Mercenaries in the Habsburg-Ottoman war of 1593-1606: the Desertion of the Papa Garrison to the Ottomans in 1600, SOAS Bulletin, vol. LV part 3 (1992) pp. 452-71. 18. Rifa'at 'Alt Abou-El-Haj sounds the clarion cty against such an approach. See, Formation of the Modern State. The Ottoman Empire Sixteenth to Eighteenth Centuries, (Albany 1991) p. 62: 'First of all, it is necessary to reaffirm a simple truism which has been consistently denied in the scholarly literature: Ottoman society, like all human societies throughout history, was fluid and dynamic'. 19. Francois-Marie Voltaire, Oeuvres completes de Voltaire. Essai sur les moeurs. et I'Esprit des nations. (Paris 1858) vol. 3 bis. p. 271. My translation. Voltaire may have meant Selim I rather than Selirn II. 20. Taner Timur, Osmanli calismalan (Ankara 1989) pp. 85-6, also Timur, Osmanli Kimligi (Istanbul 1986) p. 36. Also, Timur contends, on p. 71: 'Documents of 'reform' such as the Tanzimat Rescript (1839), the Reform Edict (1856), and the Constitution (1876) were not the product of serious liberal and secular thought, but were much more the result of the diplomatic crisis known as the 'Eastern Question,' Also in Osmanli Kimli gi p. 162 : 'Let us underline the fact that the Ottoman ruling classes were always in the course of history completely at one with the established system with which their interests always overlapped. Let us not forget that in the 19th century as the state continuously got poorer, the Ottoman ruling elite lived in a state of luxury seldom seen by their western counterparts.' T h e sad thing is that these ahistoric statements are made on the basis of considerable research in mainly Western sources. Timur uses no Ottoman archival material. 21. Necip Fazil Kisakiirek, Uk Hakan 11. Ahdiilhamid Han. (Istanbul 1981) p. 11. 22. See for example, Mustafa Muftiioglu, Ahdiilhamid. Kiztl Sultan mi? (Istanbul 1985) pp. 6,181,151, 210. There is a veritable Abdiilhamid library which keeps growing as more publications of the same genre continue to appear. Most are repetitive of each other, and contribute little if anything to furthering the debate. For a more recent example see, Ihsan Siireyya Sirma, Abdulhamid'in Islam Birligi Siyaseti (Istanbul 1983). 23. Akarh, 'Abdiilhamid II Between East and West', p. 11. 24. The classic example of the genre is M. S. Anderson, Tfcc Eastern Question. 1774-1923: A Study in International Relations (London 1966). For a good

180

25. 26.

27.

28.

29.

30.

31. 32. 33. 34. 35. 36.

The Well-Protected Domains revision of the Eastern Question paradigm see Brown, International Politics p. 39: 'Giving due attention to Ottoman History as seen from within provides a corrective to the narrowly Eurocentric perspective of the classical Eastern Question.'. Eric Hobsbawm, Nations and Nationalism Since 1780 (Cambridge 1990) p. 38. For example, Ulrich Trampener in his 24-page bibliography of his Germany and the Ottoman Empire 1914-1918. (Princeton 1968) cites only six; sources in Turkish. In an article he wrote sixteen years later on the same topic Trumpener actually reduces the number of his sources in Turkish to one. See, Ulrich Trumpener, 'Germany and the End of the Ottoman Empire' in Marian Kent (ed.) The Great Powers and the End of The Ottoman Empire (London 1984) pp. 111-40, For notable exceptions to this state of affairs see works by f eroze Ahmad, Caroline Finkle, Feroze Yasam.ee et al. j . Armstrong, Nations before Nationalism (Charlottsville 1982) pp. 295,249. O n the culinary delights of the Janissaries see Joel Carmichaei, The Shaping of the Arabs (New York 1967) pp. 261-2: 'The essence of the Ottoman slave system was the training of the sheep dogs that tan the human cattle of the Ottoman Empire'; and Bryan S. Turner Weber and Islam (London 1974) p. 128: 'Continuing the metaphor, the principle of ruling through trained slaves was successful until, in the sixteenth and seventeenth centuries, the Imperial Household ran out of adequate supplies of dog-meat.' P. j . O'Rourke, 'Give War a Chance' (New York Atlantic Monthly 1992) as quoted in Karl K. Barbir, 'Memory, Heritage, History: Ottomans and Arabs' in L. Carl Brown (ed.) Imperial Legacy. The Ottoman Imprint on thz Balkans and Middle East p, 100. Armstrong, Nations Before Nationalism, pp. 64 and particularly p. 77: 'A Moslem historian notes that huge harems and entourages were more characteristic of Ottoman and Chinese monarchs than they had been of relatively modest Arab caliphs,' T h e source cited is Halil GanemXes Sultans Ottomans. (Paris 1901, 1902) who happens to be a well known anti-Ottoman polemicist and who also happens to be a Christian. Carter Findley, Ottoman Civil Officialdom. A Social History (Princeton 1989) p. 184. See also Bureaucratic Reform in the Ottoman Empire. The Sublime Porte 1789-1922 (Princeton 1980). Ibid., pp. 135. Ibid., p. 226. Ibid., pp. 226-7. Masami Arai, 'An Imagined Nation: T h e Idea of the Ottoman Nation as a Key to Modern Ottoman History,' ORIENT vol. 27 (1991) pp. 1-2. Ibid., p. 4. Leon Cahun, Excursions sur les Bords dt I'Euphrate. (Paris 1885) pp. I - I L

Notes

181

37. Ibid. My translation, Cahun footnotes the term poucbt and defines it rather coyly as: 'A term I deliberately refrain from translating. Let the readers who do not know Turkish substitute the worst insult that they know.' T h e term means catamite. 38. Serif Mardin, Religion and Social Change in Modern Turkey. The Case of Bediiizzman Said Nursi (New York 1989) p. 202. 39. Michael Cherniavsky, Tsar and People. Studies in Russian Myths (New Haven and London 1961) p. 115-16. 40. Serif Mardin, 'Centre Periphery Relations: A Key to Turkish Politics?,' Daedalus (1973) pp. 169-90. 41. Ariel Salzmann, 'An Ancien Regime Revisited: "Privatisation" and Political Economy in the Eighteenth-Century Ottoman Empire,' Politics & Society vol. 21 (1993) pp. 393-423. See also Abou El Haj, Tfce Formation of the Modern State. 42. Ibid p. 411. 43. Maria Todorova, 'The Ottoman Legacy in the Balkans' in Carl Brown (ed.), Imperial Legacy, the Ottoman Imprint on the Balkans and the Middle East (New York 1996) p. 48. 44. Ilber Ottayh Imparatorlugttn En Uzun Yiizydt (The Longest Century of the Empire) (Istanbul 1983) p. 11. 45. For a good sampling of some recent research on the Tanzimat see: Tanzimat'tan Cumhuriyet'e Turk'tye Ansiklopedisi. (Encyclopaedia of Turkey from the Tanzimat to the Republic). 46. Abou El Haj, Formation of the Modern State, p. 62. 47. See Dustur (Register of Ottoman Laws) l.Tertip Istanbul Matbaa-i Amire 1289. (Istanbul Imperial Press 1872) p. 4. See also, j . C. Hurewitz, The Middle East and Africa in World Politics (New Haven 1975) pp. 269-71 for an English translation of the Rescript. 48. Jurgen Habermas, Legitimation Crisis (Boston 1973) p. 71. Habermas defines the 'legitimacy crisis' as: T h e structural dissimilarity between areas of administrative action, and areas of cultural tradition, [which] constitutes, then a systematic limit to attempts to compensate for legitimation deficits, through conscious manipulation.'. 49. See, Selim Deringil, 'Legitimacy Structures in the Ottoman State: T h e Reign of Abdulhamid II (1876-1909)' International journal of Middle Eastern Studies, vol. 23 (1991) pp. 345-59. 50. I owe this term to my colleague Professor Faruk Birtek of the Department of Sociology, Bogazici University. O n the incorporation of the O t t o m a n Empire into the European diplomatic network see, j , C Hurewkz, 'Ottoman Diplomacy and the European State System' Middle East journal 15 (1961) pp. 141-52. 51. For a masterful study of how fine tuning worked in a precise context see: Engin Deniz Akarh, The LongPeace: Ottoman Lebanon 1861-1920 (Berkeley,

182

The Well-Protected Domains Los Angeles, London 1993). This is a meticulous study of the Ottoman administration in Lebanon in the nineteenth century. In the same context see: Leila Fawaz, Occasion for War. Civil Conflict in Lebanon and Damascus in 1860 (London 1994).

52. 53. 54. 55.

Ernest Gellner, Nations and Nationalism (Oxford 1983) p. 11. Anthony Smith, The Ethnic Origin of Nations (Oxford 1986) p. 24. Ali Fuad Tiirkgeldi, MesaiU Muhimme-i Siyasiye (Ankara 1966). Rifa'at Abou-El-Haj, 'Aspects of the Legitimation of Ottoman Rule as Reflected in the Preambles of the Two Early Liva Kanunnamekr,' Turcica vol. 2 1 - 3 (1991) p. 371. Abou-El-Haj points out that although the legitimation of Ottoman rule in Syria, in Arabic, appears in the preamble of the first kanunname of 1519 when Syria was incorporated into the empire, in similar preambles in kanunname formulated sixty years later the legitimation tract does not appear, but is mentioned in passing in the body of the register itself. Moreover it is now in Turkish.

56. O n educational reform in the Hamidian period see, Bayram Kodaman, Abdiilhamid Donemi Egitim Sistemi. 57. See Deringil, 'Legitimacy Structures' p. 346. 58. Benedict Anderson, Imagined Communities (London 1991) the term 'official nationalism' belongs to Hugh Seton Watson. See also, Selim Deringil, 'The Invention of Tradition in the Ottoman Empire 1808-1908'. Comparative Studies in Society and History, vol. 35 (1993) pp. 3-29, 59. Sevket Pamuk, The Ottoman Empire and World Capitalism (Cambridge 1987); Roger Owen, The Middle East in the World Economy 1800-1914. (London 1987). Zafer Toprak, Tiirkiyede Milti tktisat (Istanbul 1983). Feroze Yasamee 'The Ottoman Defence Problem'. 60. O t h e r sources I have used but have not included here for reasons of space will be presented in the notes. For a discussion of the nature of late Ottoman archival documentation see also, Akarh, The Long Peace, pp. 200-4. 61. T h e Prime Ministry Archives have recently published a guide to the archives. See: Basbakanlik Osmanh Arfivi Rehbtri (Ankara 1992). 62. T h e coding used here is the actual coding used in the catalogues. 6 3 . T h e aim here is merely to provide information and terminology. T h e detailed study of administrative reform is beyond the scope of this study. 64. Ilber Ortayh, Tamimattan Soma Hahalli tiarekr (1&40-1876) (Ankara 1974). This is still the best book on local administration specifically, and is unique in that it ties in reform in local administration with the overall picture of foreign policy, legal reform etc. T h e other standard reference on this topic is Findley, Bureaucratic Reform. 65. Ibid., p. 43. Ortayh points out that the status of the Nahiye was never very clear. Originally intended as a unit halfway between Kur'a and Kaza, often grouping several villages, its application proved problematic. 66. Ibid., pp. 54-69; 9 8 - 1 0 1 .

Notes

183

67. Findley, Bureaucratic Reform, pp. 250-1. 68. Habermas, Legitimation Crisis, p. 7 1 . 69. As noted by Suraiya Faroqhi in her recent work based on primary source research in the Ottoman archives: '[In the first two hundred years of their rule in the Hijaz] the Ottoman Sultan's activities and responsibiKt.es in the Hejaz are discussed in official documents with some frequency, but practically no texts have come to my attention which link the Sultan's role in this matter to his responsibilities as caliph ...'. See Suraiya Faroqhi, Pilgrims and Sultans. The Hajj under the Ottomans. (London 1994) p. 8. 70. For a very similar treatment of the problem to mine see: Timothy Mitchell, Colonising Egypt (Berkeley 1991). 71. Edward Said, Culture and Imperialism (London 1994) p. 6. 72. Sir Charles Eliot, Turkey in Europe (London 1965 reprint of original published in 1907) p. 426.

Chapter O n e 1. The scene above is my fictional reconstruction based on Selc/uk Esenbel's, IstanbuI'da Bir japon, YamadaTorajiro' Istanbul no. 9 April 1994, pp. 3642. It is interesting that Yamada used terms like 'Dowager Empress' for the Sultan's mother, and 'Empress' for his first wife, drawing parallels with Chinese court ritual which he would have been familiar with. See also, Self uk Esenbel, 'A fin de siecle Japanese Romantic in Istanbul: The Life of Yamado Torajiro and his Toruko Gakan.' SOAS Bulletin, LIX part 2 (1996) pp. 237-52. 2. See particularly, E. Hobsbawm and Terence Ranger (eds) The Invention of Tradition (Cambridge 1983); Sean Wilentz (ed.) Rites of Power. Symbolism, Ritual and Politics Since the Middle Ages (Philadelphia 1985) and David Cannadine and Simon Price (eds) Rituals of Royalty. Power and Ceremonial in Traditional Societies (Cambridge 1987). Also Gluck,Japan's Modern Myths. 3- See John Elliot, 'Power and Propaganda in the Spain of Phillip I V in Rites of Power, p. 151. 4. Cherniavsky, Tsar.and People, p. 151. 5. Ibid., p. 119. O n the 'Holy Russian Land' see Hobsbawm, Nations and Nationalism, p. 49. 6. Gluck, Japan's Modern Myths, pp. 49-60, particularly p. 60: 'By adding something to the political to make it patriotically unpalatable, politics was denatured even as the constitutional system was first beginning to function'. 7. Ibid., p. 249. 8. Istvan Deak, 'The Habsburg Monarchy: T h e Strengths and Weaknesses of a Complex Patrimony,' Monarchies Symposium, Columbia University October 26-7 1990. Unpublished.

184

Tke Well-Protected Domains

9. Walter Leif sch, 'East Europeans Studying History in Vienna' in Dennis Deletant and Harry Hanak (eds) Historians as Nation Builders (London 1988) pp. 139-56. In the event the young Poles, Czechs, Serbs etc. ended up writing their own 'National Histories'. 10. Giilru Necipoglu, Architecture Ceremonial and Power: The Topkapi Palace in the Fifteenth and Sixteenth Centuries (Cambridge Mass. 8c London 1991) p. 59. 11. Zeynep Celik, Displaying the Orient, Architecture of Islam at Nineteenth Century World Fairs (Berekley, Los Angeles 1992) p. 32-6; BBA Y.A H U S 303/87 5 July 1894, Sublime Porte Foreign Ministry, telegram from Vienna Embassy no. 307. 12. Matdin, Religion and Social Change, p. 129. O n the Ottoman effort to keep pace with the world-wide augmentation and competition in ceremonial, see Selim Deringil, 'The Invention of Tradition as Public Image in the Late Ottoman Empire, 1808 to 1908' Comparative Studies in Society and History, 35 (1993) pp. 3-24. 13. Gluck, Japan's Modern Myths, p. 249. 14. O n the conception of the average cultured Ottoman of nomadism as the antithesis of civilization see Mardin, 'Center-Periphery Relations'. 15. Mehmet Akif Ersoy, Safahat (Istanbul 1990). O n the conceptualization of these terms by Ottoman intellectuals see, Sukrii Hanioglu, 'Osmanli Aydmmda Degisme ve "Bilim"' Toplum ve Bilim vol. 27 (1984) pp. 1 8 3 92. 16. Mardin, Religion and Social Change, pp. 38,119,172. 17. There is, however, some debate on the actual extent of reading of the Mukaddimah undertaken by Ottoman/Turkish historians. See Cornell Fleischer, 'Royal Authority, Dynastic Cyciism, and 'IbivKhaldimism in Sixteenth-Century Ottoman Letters,' journal of Asian and African Studies X V I I I 3 - 4 (1983) pp. 198-219. Particularly footnote 2 where Fleischer points out that a few of the standard Turkish references on the subject are inaccurate. 18. Ahmed Cevdet Pasa, Tarih-i Cevdet (Istanbul 1309 [1891]) 12 vols. 19. Omit Meric, Cevdet Pasa'ntn Cemiyet ve Millet Gorusu. (The Views of Cevdet Pasa on Society and the State) (Istanbul 1979) p. 13.This is someting of an 'engaged-partisan view' of Cevdet Pasa, who is said 'not to have needed Montesquieu or Taine' because he was 'Muslim and Ottoman', p. 11. See also Cevdet Pa$a, Tezakir (Ankara 1986) p. 79. 20. Ibn Haldun, Mukaddime, (Istanbul 1990) pp. 103, 331, 364. T h e intricacies of Ibn Khaldun's asabiyya are far too complex to be discussed here. My aim is merely to mention the adaptation of his work to the ideological core of late Ottoman statecraft.

Notes

185

21. Brinkley Messick, The Calligraphic State: Textual Domination and History in a Muslim Society (Berkeley 1993) pp. 54-8. Messick's work deals with the heritage of Ottoman law in Yemen. 22. Serif Mardin, The Genesis of Young Ottoman Thought (Princeton 1962) p. 179. 23. Ibid. Serif Mardin quotes Sadik Rjfar Pasa as stating that his aim was to establish, 'une force legale de nature a contenir le peuple et en meme temps empecher tout acte d'injustice.' p. 187. 24. Ahmed Cevdet Pasa, Ma'ruzat. Transcribed and prepared for publication by Yusuf Hallacoglu (Istanbul 1980) p. 42. My translation. 25. Ibid. 26. BBA YEE 14/1188/126/9. Baghdad 9 Ramazan 1309/7, April 1892. 27. Clifford GeertZj Negara. The Theatre Sfate in Nineteenth Century Bali (Princeton 1980) p. 135: 'Ideas are not and have not been for some time, unobservable mental stuff. They are envehicled meanings, the vehicles being symbols ... a symbol being anything that denotes, describes, represeats, exemplifies, labels, indicates, evokes, depicts, expresses-anything that somehow or other signifies ... Arguments, melodies, formulas, maps, and pictures are not idealities to be stared at but texts to be read, so are rituals, palaces, technologies, and social formations.'. 28. Irfan Gundiiz, Qsmanldarda Dtvht-Tekke Munasebetleri (Istanbul 1989) pp. 150-1. 29. Ehud Toledano, State and Society in mid-Nineteenth Century Egypt (Cambridge, England 1990) pp. 5 0 - 3 . Toledano points out that the Khedive of Egypt, Abbas Pasa, derived his legitimation from his status as vassal of the Ottoman sultan. This was to contrast greatly with the behaviour of the khedives of Egypt after rhe British invasion of 1882, when, with British encouragement, they posed as independent rulers and even laid claim to the Caliphate. See below, Chapter 2. 30. BBA Y. Mtv 230/24, 6 Safer 1320/15 May 1902, Vilayet of Ankara to Imperial Secretariat no. 136. It is interesting that the portrait should actually have been paid for. N o doubt in a similar situation today, it would simply be confiscated. 31. Islam forbids the depiction of the human image but, if his priceless photograph collection is anything to go by, Abdulhamid seems to have been less concerned with this than is commonly thought. His hesitancy over displaying his personal likeness may also have been prompted by security considerations. 32. In Japan this form of vocal acclaim was used for the first time in the presence of the Emperor in 1889 after the promulgation of the constitution 'patterned as had been advocated, after the European 'hooray'.' See Giuck, Japans Modern Myths, p. 45. O n the augmentation of ceremonial in the British case see, David Cannadine, 'Splendour at Court: Royal

186

33. 34. 35. 36.

37. 38. 39.

40. 41.

42.

43. 44.

45.

The Well-Protected Domains Spectacle and Pageantry in Modern Britain, c.1820™ 1977' in Kites ofPower, pp. 206-43. Siikrii Hanioglu, Osmanli Ittibad ve Terakki Cemiyeti vejon Turkliik 1B&91902 (Istanbul 1985) p. 63. Aptullah Kuran, 'The Evolution of the Sultan's PaviEion in Ottoman Imperial Mosques,' Islamic Art I V (1990-1991) p. 284. Ibid., p. 283. Mehmet Ip§irli, 'Osmanhlarda Cuma Selamhgi: Halk Hiikumdar Miinasebetleri Acisindan Onemi' in Prof hekir Kiitukoglu'na Armagan (Istanbul 1991) pp. 4 5 9 - 7 1 . Louis Rambert,'Noteset impressions de Turauie. UEmpire Ottoman sous AbdulHamid. (Geneva and Paris n.d.) p. 76. Celal Esad Arseven, Sanat ve Siyaset Hatiralartm (Istanbul 1993) pp. 8 1 - 2 . Selcuk Esenbel, 'Istanbul'da bir japon: Yamada Torajiro', p. 40.1 am grateful to Selcuk Esenbel for the observation concerning the sultan's change of carriages. Another contemporary noted that in 1879, the sultan actually rode to ceremonial prayers. See Ali Said, Saray Hattralart (Istanbul 1992) p. 72. Arseven, Sanat ve Siyaset Hattratanm, pp. 8 2 - 4 . T h e author was a member of the Palace Guards and was present at the scene. Ibid., p. 40. There is an interesting similarity here with the practice in China, where the empress dowager, a contemporary of Abdulhamid, would travel to her summer palace only after the roads were strewn with 'yellow powder', yellow being the imperial colour. See, Sterling Seagrave, Dragon Lady: The Life and Legend of the Last Empress of China (London 1992). See also Rambert, Notes et impressions, p. 132: '[The] roads would be covered by a fine yellow sand.' O n the historic origins of the visit to the shrine of Ebu Eyyub el Ansari see, Nicolas Vatin, 'Aux origines du Pelerinage a Eyiip des Sultans Ottomans' T U R C I C A vol. 27 (1995) pp. 91-9. Hagop Mintzuri, Istanbul Amlan 1897-1940 (Istanbul 1993) pp. 14-16. By 'Arab soldiers who did not speak Turkish' Mintzuri means the Regiment of the Turbanned Zouaves (Sankh Zuhaf Alayi), fashioned after the N o r t h African French regiment by the same name. This regiment was created by Abdulhamid in 1892 to 'demonstrate the loyalty of the Sultan's Arab subjects'. O n the Zuhaf Alayi see, Necdet Sakaoglu, 'Gecmis. Z a m a n Olur ki' Skyline. March 1997, p. 112. For a very good recent study of these issues see Ronald Grigor Suny, Looking Toward Ararat (Bloommgton and Indianopolis 1993) pp. 15-30. Ibrahim Hakki Konyah, SoglXt'de Ertugrul Gazi Ttirbesi ve thtifah (Istanbul 1959). My thanks to Dr Cemal Kafadar for this reference. On the cult of Ertugrul Gazi see below. Mintzuri, Istanbul Amlan, pp. 13-14.

Notes

187

46. Ali Said, Sarny Hattralan: Sultan Abdiilbamid'in Hayati, (Istanbul 1993) pp. 49-50. 47. BBA Yikhz Resmi Maruzat (hereafter Y.A RES) 135/22 Official correspondence between Yikhz Palace and the Ministries.

48. Ibid. 49. Personal collection of Dr Edhem Eldem. I owe thanks to Dr Eldem for bringing this medallion to my attention. T h e medallion is in bronze and was struck in Brussels. 50. Salname-i VilayetA Hudavcndigar, 1303/1885, pp. 110-33. 51. For a discussion of the 'Mirrors for Princes' literature see: Christine Woodhead, '"The Present Terror of the World?" Contemporary Views of the Ottoman Empire c!600' History vol. 72 (1987) pp. 20-37, particularly p. 29. 52. T h e first comprehensive chronicles of the origins of the Ottoman state were written as late as the fifteenth century. They too were written with a view to legitimising contemporary regimes. See Cemal Kafadar, Between Two Worlds: The Construction of the Ottoman State (Berkeley 1995) p. 122: 'That they hailed from the Kayi branch of the Oguz confederacy seems to be a creative 'rediscovery' in the genealogical concoction of the fifteenth century.'. 53. Compare Salnamc-i Devlet 1271, Salnamt'i Devlet 1297, Salname-i Devlet 1298, Salname-i Devlet 1302. Also the sheer size of the volumes is the best indication of the growth of the state apparatus. Compare SalnameA Devlet 1270,110 pages; Salname-i Devlet 1298, 522 pages; Salname-i Devlet 1308, 892 pages; Salname-i Devlet-1324,1107 pages. 54. BBA Y.A H U S 261/91 Imperial Chancery to Yikhz Palace.ll Zilkade 1309/7June 1892. 55. BBA Yiidiz Miitenevvi Maruzat (Y.Mtv) 39/61,4 Zilkade 1306/2 July 1889, Sublime Porte, Ministry of Interior, no. 30. Minister of Interior Miinir Pa$a to Yiidiz Palace Imperial Secretariat. 56. See Richard Wortman, 'Moscow and Petersburg: T h e Problem of Political Center in Tsarist Russia 1881-1914' in Sean Wilentz (ed.) Rites of Vower (Philadelphia 1985) pp. 244-71. See also Richard Wortman, Sce~ narlos of Power, Myth and Ceremony in Russian Monarchy, vol. 1 (Princeton 1995). 57. See E. H . Kantarowicz, The Kings' Two Bodies (Princeton 1957) p. 82: 'The Byzantines had claimed that the so-called 'haloed' essence of ancient Rome on the Tiber, or her sempiternal genius, had been transferred to the New Rome on the Bosphorous.'. 58. Necipoglu, Architecture Ceremonial and Power, p. 13. 59. Joseph Fletcher, 'Turco Mongolian Monarchic Tradition in the Ottoman Empire' Harvard Ukrainian Studies vol. Ill (1979-80) p. 246.

188

The Well-Protected Domains

60. Gulru Necipoglu, 'The Ottoman Hagia Sophia,' paper delivered at 'The Structure of Hagia Sophia' symposium, 19 May 1990, Princeton University. Cited with permission of the author. 61. BBA Y.Mtv 26/3; 7 Receb 1304/1 April 1887. Ministry of Pious Foundations no. 2. Minister Mustafa Pasa to Yildiz Palace Imperial Secretariat. T h e document specifies that the dates scrawled on the wall are from 'some thirty years back', which leads one to suspect that Abdulhamid's predecessors were not as sensitive as he was. 62. Gulru Necipoglu, 'Dynastic Imprints on the Cityscape. T h e Collective Discourse of Ottoman Imperial Mausoleums in Istanbul'. Unpublished paper p. 14. Cited with permission of the author. 63. BBA YEE 11/1419/120/5 Circular from Abdiilhamid II to his Ministers, Cemaziyelevvel 1319/August 1901. N o precise day is given. J. Armstrong could not be more wrong when he states, 'The Ottomans, while retaining the Byzantine capital city rejected specific court mythic formulas' or,.' T h e Ottoman's lukewarm attitude towards Constantinople presents a striking contrast to the identification of the Byzantines with the city. See Armstrong, Nations Before Nationalism, pp. 144,174. 64. See Godfrey Goodwin, History of Ottoman Architecture (London 1971) p. 419 for photographs of clock towers in Ottoman towns. My thanks to Andrew Finkel for this reference. 65. BBA Y.Mtv 67/10, 15 Safer 1310/8 September 1892, Minister of the Privy Purse Mikail Pa§a. It is interesting that the chronograms seem to follow the historic sequence of reigns. Was the aim to build a mosque dedicated to each sultan of the dynasty?. 66. BBA Y.Mtv. 2 8 / 3 7 , 1 7 Mubarrem 1305/5 October 1887. Yildiz Palace Imperial Secretariat, no. 190. 67. BBA Y.Mtv 131/39,17 Cemaziyelevvel 1313/5 November 1895 Ministry of Pious Foundations, Secretariat no. 47. 68. BBA Y.A H U S 182/93,21 Zilkade 1302/1 September 1885. The Grand Vizier Said Pasa was instructed to inspect the construction in person, 'from door to pulpit' (min bob mm el mibrab). Also photographs were taken of the interior of the Ulu Cami. 69. BBA Y.A H U S 245/43; 10 Receb 1308/19 February 1891; Vali of Baghdad S i m Pa$a to Ministry of Public Works no. 141. 70. Mardin, 'Center Periphery Relations', pp. 169-90. Mardin is quoting from W. M. Ramsay, 'The Intermixture of Races in Asia Minor. Some of its Causes and Effects' Proceedings of the British Academy (1915-1916) p. 409. 71. Cherniavsky, Tsar and People, p, 151. 72. Paul Dumont and Francois Georgeon (eds), Villes Ottomanes a la Fin de I'Empire. (Paris 1992). T h i s is a remarkable collection of essays which give a very detailed appreciation of how this 'project of modernization' was put into application in cities as diverse as Bitola, (Manastir) and

Notes

73. 74.

75. 76. 77. 78.

79.

80.

189

Salonka in the Balkans to Ankara and Van in Anatolia. The overall picture that emerges from the very competent contributions is one of concerted efforts expended by the Ottoman modernising officials which were doomed by the spread of nationalism. Timothy Mitchell, Colonising Egypt (Berkeley 1991) p. 65. Zeynep Ceiik, The Remaking of Istanbul Portrait of an Ottoman City in the Nineteenth Century (Berkeley, Los Angeles, Boston 1993). See particularly chapter 3 pp. 49-63. Ibrahim Hakki Konyah, SUgtit ve Ertugrul Gazi thtifali. (Ankara 1957). BBA Y.Mtv 285/167; 25 Safer 1324/20 April 1906, Communication from Vilayet of Bursa to Imperial Secretariat, no. 27. See Professor Colin Imber's forthcoming work on the foundation myth of the empire, Konyah, Sogtid'de Ertugrul Gazi Turbesi, pp. 14-19. For a photograph of the fountain chronogram see p. 15, Osman's grave was later removed to Bursa when that city became the Ottoman capital. Ibid., p. 23. The term used is 'en matcher rivayetleregore'. Kafadar also points out that, 'Sultan Abdiilhamid instigated the discovery of the tomb of "Ertugrul's wife'". See Kafadar, Between Two Worlds, p. 185 n 9. BBA Y.Mtv 228/58; 21 Zilhicce 1319/31 March 1902. Administrative council of the Kaza of Sogiid to the Mutasarnf of the Sancak of Ertugrul, no. 27. It is interesting to note that the buildings in question are specified as 'Christian dwellings'. They were paid for in full, and 'because the inhabitants were poor' they were awarded the wreckage. Includes list of dwellings and names of owners.

81. Ibid., p. 34. 82. BBA Y.Mtv. 222/54; 7 Te ? rin-i Evvel 1317/20 October 1901. Telegram from the Headman of the Karakecili Tribe, Haci Bekir, to Yildiz Palace. 83. BBA Y.Mtv 47/92.Selh-iCemaziyelevvell308/12Febmaryl89l.Keeper of the Privy Purse, Agop Pasa no. 612. 84. BBA Y.Mtv 301/37; 6 Receb 1325/15 August 1907; Minister of the Privy Purse Agop Pasa. Giving details about the installation of a new chandelier in the shrine. 85. See, Enver Kartekin, Ramazanogullart Beyligi Tarihi (Istanbul 1979) pp. 149-50. 86. Kafadar, Between Two Worlds, p. 95. 87. BBA Y.Mtv 38/59; 24 Mart 1305/6 April 1889. From the Mutasarnf of Jerusalem to the Yildiz Palace Imperial Secretariat. 88. BBA Y.Mtv 132/156; 20 Cernaziyelahir 1313/8 December 1895 Emir of Mecca Avn El Refik to Yildiz Palace Imperial Secretariat, no. 5. 89. BBA Y.Mtv 66/87; 13 Safer 1310/6 September 1892. Minister of the Privy Purse Mikail Pasa. Even in cases where there was no obvious publicity value, such as the construction of a special building for the

190

90. 91.

92. 93. 94. 95. 96. 97.

98.

99. 100. 101. 102.

103.

104. 105.

106.

The Well-Protected Domains preservation of the religious court records (§eriyye Sicilleri), the Privy Purse (Cib-i Humayun) payed the building expenses. See BBA Y.Mtv 63/17. BBA Y.A H U S 184/65 Grand Vizier Kami! Pasa to Yildiz Palace. 14 Muharrem 1303/23 October 1885. Y.A H U S 306/46. Grand Vizier Cevad Pasa. Sublime Porte Receivers Office no. 589 12 Safer 1312/16 March 1894. It is interesting to note that, to this day, non-Muslim places of worship display the Turkish flag (very prominently) yet one never sees a mosque displaying the national colours. BBA Y.Mtv 68/27; 6 Rebiyiilevvel 1310/28 September 1892. Chief of Istanbul Municipality (§ehremini) Ridvan Pasa to Yddiz Palace. Annuaire Commercialk de Constantinople, 1893. BBA Y.Mtv 282/6; Gurre-i Zitkade 1323/28 December 1905. Minister of Interior Mahmud Memduh Pasa to Sublime Porte. BBA Irade Hususi 188/51; 14 Sevval 1310/12 May 1893. Yildiz Palace Imperial Secretariat 8836. Rambert, Notes et Impressions de Turquie, p. 159. O n the Ottoman adoption of the Western practice of giving decorations, see: Jacob Landau, 'Nishan,''Encyclopaediaof Islam. New Edition, (Leiden 1993) pp. 57-9. In this sense Landau is mistaken in his assertion that in Abdulhamid's reign the number of decorations, 'increased so much that their intrinsic value declined'. See 'Nishan', p. 58. BBA Y.A H U S 306/91,16 Safer 1312/19 August 1894. BBA Y.A H U S 261/141, Grand Vizier Cevad Pasa to Vilayet of Konya, 23 Zilkade 1309/19 June 1892. BBA Y.Mtv 296/46,10 Mart 1323/23 March 1893. Ottoman High Commissioner Ferik Sadik al-Muayyed to Imperial Secretariat, no. 131/16. BBA Y.Mtv 290/183 Letter from Prof. Adolphe Strauss to Imperial Secretariat, 18 March 1906. Enclosing article entitled: 'Le S u l t a n Abd-ul-Harnid II' in Revue d'Orient et de Hongrie. O n the Hamidian regime's attempt to assimilate the Kurdish tribes into politically reliable elements of the empire, see below Chapters 2 & 3. Also, Ali Karaca, Dogu Anadolu hlabatt vc Ahmet $akir Pasa (1838-1899), (Istanbul 1993) pp. 173-203. BBA Y.A H U S 251/34 27 Agustos 1307/11 September 1891. Governor of Ttabzon Ali Pasa to Ministry of the Interior. BBA Y M t v 66/84 13 Safer 1310/6 September 1892. Keeper of the Privy Purse Mikaii Efendi no. 420. Mikail Efendi reports that five such banners had now been completed in addition to seven others which had been presented before. Naci Kiciman, Medine Miidafaasii Hicaz bizden nasil ayntdt? (Istanbul 1971) pp. 135, 229. See also Chapters 1 and 2.

Notes

191

107. BBA YMtv 61/80 24 Ramazan 1309/22 April 1892;Imperial Privy Purse no. 98. Keeper of the Purse Mikail Efendi. The regiment was named after the legendary founder of the Ottoman Empire, Ermgrul Gazi, and it was recruited from the town of Sogiit which was purported to be the first camp of the tribe of Osman. 108. BBA Y.A H U S 182/36, 5 Ramazan 1302/18 June 1885, Grand Vizier Said Pasa to Yildiz Palace. 109. BBA YEE 1 4 / 2 9 2 / 1 2 6 / 8 , 7 Tesrin-i Ewel 1306/20 October 1890. 110. Butrus Abu Manneh, 'The Sultan and the Bureaucracy: the AntiTanzimat Concepts of Grand Vizier Mahmud Nedirn Pasa' International journal ojMiddle East Studies (1990) pp. 268-9. 111. BBA Y.A H U S 235/28,14 Ramazan 1307/4 May 1890; Prime Minister Kamil Pasa to Imperial Private Secretariat no. 19. Although the name in the document is spelled 'Perpinyani* it is more likely that the person in question was from the well known Istanbul Levantine family of Italian origin, the Perpigniani. My thanks to Prof. Cem Behar for bringing this to my attention. The Ghassanids were an Arab dynasty of the sixth century, who were an ally of the Byzantines. See Ana Britannka (Turkish edition of the Encyclopaedia'Britannka)vol. 9, p. 299. 112. BBA Y.A H U S 257/64 15 Receb 1308/24 February 1891. Minutes of the special session of the Council of Ministers. 113. BBA Y.A H U S 257/64,13 Saban 1309/13 March 1892; Grand Vizier Cevad Pasa to Yildiz Palace. 114. BBA Irade Hususi, 62 20 Muharrem 1311/3 August 1893. Yildiz Palace Imperial Secretariat no. 450. 115. BBA Y M t v 71/98 11 Cemaziyelevvel 1310/1 December 1892. Yildiz Palace Imperial Secretariat no. 4235. Imperial Private Secretary Sureyya Pasa. 116. Ibid. Report of Minister of Police Nazim Bey 15 Cemaziyelevvel 1310/ 5 December 1892. 117. BBA Y.Mtv 57/57 21 Cemaziyelevvel 1309/23 December 1891. Imperial Private Secretary Sureyya Pasa. 118. Mitchell, Colonising Egypt, pp. 138-46. In his excellent discussion and critique of how Orientalism and nineteenth-century linguistics established and maintained a 'hierarchy' between 'normal' Western languages and Eastern languages Mitchell warns against seeing words as fixed meanings which are in fact 'just the same only different'. I am aware that what I am proposing here runs the risk of establishing a fixity which may not have been there. Yet, I will proceed just the same because I feel that the constant repetition of certain formulae in Ottoman official documentation was not fortuitous, and indeed was an indication of how the Ottoman elite perceived their own 'hierarchy' as legitimate.

192

The Well-Protected Domains

119. James Finn, Stirring Times (London 1878) pp. 405-7. Emphasis in original. 120. Bernard Lewis, The Political Language ofhlam (Chicago 1988) p. 62. 121. BBA Irade Dahiliye 56 11 Eykll 1308/24 September 1892. Sublime Porte Grand Vizier's Office. See Chapter 3 below on the Ottoman campaign to convert the Yezidis. T h e number of all of these examples can be increased almost indefinitely, I have chosen a representative sampling. 122. BBA Irade Dahiliye 68120. 29 Muharrem 1299/21 December 1881. Vilayet of Yanya to Prime Minister, no. 24. 123. BBA Irade Dahiliye 99649.19 Cemaziyelahir 1309/21 December 1892. Vilayet of Syria to Grand Vizier, no. 32. 124. BBA Y.Mtv. 53/108 2 Zilkade 1308/9 June 1891. Mutasamf of Tokad Lutfi Bey to Ministry of Interior, no. 159. 125. BBAY.A RES 134/71 20 Zilkade 1323/16 January 1906.YudizHususi Maruzat Defteri, no. 13613. 126. BBA Y.Mtv 59/22 28 Kanun-u Sani 1308/10 February 1893. Yildiz Palace Imperial Secretariat. T h e Redhouse Dictionary also provides 'fickle, inconsistent and irresolute' as a second string of meanings for stbukmagz. 127. BBA Y.Mtv 43/117,4 Haziran 1306/17 June 1890. Vali of Basra Esseyid Mehmed bin Yunus to Imperial Secretariat, no. 52. 128. BBAY.A B U S 379/8 27 Cemaziyelahir 1315/24 October 1897,Ministry of Interior no. 3380. T h e document refers to an irade which ordered that such serious matters be published in the form of pamphlets (risale) rather than in newspapers. 129. Hasan Kayali, 'Arabs and Young Turks: Turkish-Arab Relations in the Second Constitutional Period 1908-1918'. Harvard P h D Thesis 1988, pp, 109-10. T h e quote in Kayak is probably the old Hamidian formula of'those who cannot tell good from evil'. 130. Naci Kiciman, Medine Mudafa&st p. 435. 131. BBA Bab-i Ali Evrak O d a s i (BEO) 24234 Mosul Giden 336. 21 Cemaziyelevvel 1311/30 November 1893; BBA Irade Dahiliye 4 Gurrei Ramazan 1313/15 February 1896. 132. Mardin, Religion and Social Change, p. 169. 133. Armenouhie Kevonian, Les Noces JSloires de Gulizar. (Paris 1993). p. 112. 134. BBA Y.ARES25/14,2Agustos 1300/15 August 1884. Vilayet of Bingazi to Ministry of Interior. 135. BBA Y.Mtv 37/73; 3 Agustos 1304/16 Agustos 1888. Vali of Hicaz, Osman Nuri Pa$a, to Sublime Porte. T h e word used in such instances, tevahhu}, is defined in Redhouse as 'a being or becoming timid like a wild beast'.' 136. BBA Y.Mtv 51/74, 23 Zilkade 1308/1 July 1891. Report by the Commission for Military Affairs relating to the constitution of irregular Hamidiye units in Bingazi; BBA Irade Dahiliye 56,11 Eyluf 1308/24 Sep-

Notes

137. 138. 139.

140. 141.

142.

193

tember 1892 see also above Chapter 1; BBA Y.Mtv 53/108, 2 Zilkade 1308/9 June 1891 Murssarn/ofTokad to Ministry of Interior, no. 159. BBA Y.Mtv 34/39 22 Kanun-u Evvel 1305/4 December 1889, BBA 53/108 26 Zilhicce 1308/2 August 1891. Governor of Sivas Vilayet Mehmed Memduh Pasa to Ministry of Interior. trade Hususi 878/123; 17 Muharrem 1310/11 August 1892;YildizPalace Imperial Secretariat no. 678. T h e matter at hand was the selection of photographs to be sent to the Chicago World's Fair of 1893. Mardin, Religion and Social Change, p. 129. BBA YEE 31/1950 Mukerrer/45/83 22 Zilkade 1299/5 October 1882. Report by Grand Vizier Said Pasa on the difficulties of the budget, the invasion of Egypt by the British, and the legal basis of rule of the Ottoman State in the provinces. Report by Osman Nuri Pasa on conditions in the Hicaz. BBA YEE 14/ 292/126/8, 5 Temmuz 1301/18 July 1885; Kantarowicz, Tfce King's Two

Bodies, p. 79. 143. Gellner, Nations and Nationalism, p. 77. 144. Edward Shils, Tradition (Chicago 1980) p. 198. 145. Cannadine and Price (eds), Rituals of Royalty, Vowtr and Ceremonial in Traditional Societies (Cambridge 1987) p. 19.

Chapter T w o 1. Naci Kasif Kiciman, Medine Mudafaast, p. 121,181, 215. 2. Elie Kedourie, 'The Surrender of Medina, January 1919,' Middle Eastern Studies vol. 13 (1977) p. 130. Kedourie shows that the story put about by Lawrence and the British that they chose not to attack the city for fear that the Ottomans would do wilful damage to the holy places, was simply not true. The Arab forces were never a match by themselves for the strong Turkish garrison, and Fahreddin Pasa had no. intention of doing damage to places he deeply respected. See p. 135: 'The picture which emerges is one of orderly and equitable rule in which property rights were respected, food supplies assured, and morale maintained.'. 3. 4. 5. 6.

Kiciman, Medine Mudafaast, p. 192. Hobsbawm, Nations and Nationalism, p. 84. Gellner, Nations and Nationalism, p. 16. Bernard Lewis, The Emergence of Modern Turkey, p. 105. Although Lewis uses the term 'Reorganization', I think the aims and implications of the movement justify the use of'Reordering'. 7. Butrus Abu Manneh, 'The Islamic Roots of the Giilhane Rescript,' Die Welt des Hams 34 (1994), pp. 173-203. 8. Lewis, Emergence, pp. 107,112,117,120.

194

The Well-Protected Domains

9. Hobsbawm, Nations and Nationalism, p. 73.

10. Ibid. 11. Carter Vaughn Findley, 'The Advent of Ideology in the Islamic Middle East/ Studia hlamica LV (1982). Although Findley credits Ottoman statesmen with 'some appreciation of the force of patriotism' he mistakenly concludes that they,' [had] little grasp of the forms it took in the nineteenth century world.' p. 165. 12. Joan Haslip, The Sultan. The Life of Abdul Hamid 11 (New York 1958), p. 124. 13. Thomas Arnold, The Caliphate (London 1965) p . 129-62. 14. Halil Inalcik, 'The Ottoman State and its Ideology' in Halil tnalcik and Cemal Kafadar (eds), Suleyman the Magnificent (Chicago and Princeton, Princeton Occasional Papers 1992) pp. 49-72. 15. Arnold, The Caliphate, pp. 146-7. 16. Lewis, The Political Language of Islam, p. 49. 17. Halil tnalcik, and Donald Quataert (eds), An Economic and Social History of the Ottoman Empire 1300-1914 (Cambridge 1994) pp. 2 0 - 1 . 18. Cornell Fleischer, Bureaucrat and Intellectual in the Ottoman Empire (Princeton 1986) p. 276. 19. O n the Arab Challenge to the legitimacy of the Ottoman Sultan's position as Caliph see, U. Haarmann, 'Ideology and Alterity; T h e Arab Image of the Turk,' International journal of Middle East Studies 20 (1988) pp. 1 7 5 96. 20. Anderson, Imagined Communities, pp. 83-111. 21. Stephen Duguid, 'The Politics of Unity: Hamidian Policy in Eastern Anatolia,' Middle Eastern Studies (9) 1973 p. 139. 22. India Office Library and Records. Political and Secret Home Correspondence (hereafter L / P & S / ) 7/27 p. 617. Lieut. Col. E. C. Ross, Political Resident in the Persian Gulf, to A. C. Lyall, Secretary of the Government of India Foreign Department, no. 240 Bushire 13 December 1880. 23. Ramsay, 'The Intermixture of Races', pp. 359-422: 'In 1880 to 1882 a hopeless despondency about the future of the country reigned everywhere in Turkish society. Prophecies were current that the end of Turkish power was at hand.' I owe thanks to Serif Mardin for this reference. 24. Albert Hourani, A History of the Arab Peoples (London 1991) p . 280. 25. H u r i Isiamoglu-Inan, 'Mukayeseli tarih y&zim icjn bir oneri: Hukuk, Mulkiyet, Mesruiyet/ Toplum ve Bilim 62 (1993) p. 30. 26. Mardin, The Genesis of Young Ottoman Thought, p. 18. 27. See Halil Inalcik, State and Ideology, pp. 49-72: 'Under Suleyman Hanafism was declared ,.. as the exclusive school of law ..,' See alsoTarifc Deyimleri ve Terimleri Sozliigii (Istanbul Milli Egitim Basimevi 1983) p. 728. It is interesting that even in this modern publication, which is a standard reference, there is a clear bias to this school of jurisprudence. It is clearly

Notes

195

(and in somewhat exaggerated language) stated that it was the official belief of the Ottoman state which was propagated 'from the Balkans to

Japan'. 28. It must be noted however that Abu Hanifa himself first stated that the legitimate ruler had to be from the Qureish. It would seem that the more pragmatic interpretation of the Caliphate occured through the force of historical circumstance and the need to accomodate such obviously powerful rulers such as the Mongol and the Turks. See Ann K. S. Lambton, State and Government in Medieval Islam (London 1985) pp. 5,9, 32,182. 29. Fleischer, bureaucrat and Intellectual, p. 280. 30. Snouck Hurgronje, Mekka In the Latter Part of the 19th Century. (London 1931) pp, 182-3. 31. BBA YEE 14/1188/126/9, Baghdad 9 Ramazan 1309/7 April 1892. Suleyman Hiisnii Pasa is a good example of the type of bureaucrat/soldier/intellectual who ran the empire. Something of an illustmous exile he was banished to Baghdad by a suspicious Abdulhamid for his central role in the deposition of Sultan Abdiilaziz in May 1876. H e was the author of a well known tract proposing reform, the Hts-i Inkilab, and was also the author of some twenty books and monographs. On him see Turk Meshurlan Amiklopedisi (Istanbul 1943) p. 360.

32. Ibid. 33. It is worth noting here that Suleyman Hiisnii Pasa had written a 'World History' (Tarih'i Alem) dealing with matters ranging from the 'big bang to the sequence of Chinese dynasties. See, Kemal Ziilfti Taneri, Suleyman Husnu Pasa'mn Hayatt ve Eserleri (Ankara 1963). 34. Ibid, It is also worth noting that the man whom Suleyman Hiisnii Pasa recommended to write the chapter on Chistianitv, Moulvi Rahmetullah, is referred to by Snouck Hurgronje as, 'The highly revered assailant of Christianity, Rahman Ullah an exile from British India (living in the Hijaz).' See Hurgronje, Mekka , p, 173. 35. Mardin, Hellion and Social Change, p. 76. 36. Serif Mardin, 'Some Notes on an Early Phase in the Modernisation of Communications in Turkey,' Comparative Studies in Society and History (3) 1960-61. pp. 257-71. 37. Duguid, 'The Politics of Unity', quoting G. P. Gooch and H , Temperley, British Documents on the Origins of the War, vol. v. p. 27 (London 1938). 38. Milash Durmuszade Hafi^ Mehmed Sadik, Milas'da Aga Camii muderds ve Cami-i Kebir Vaizi, (Preacher Mehmed Sadik from Milas, Preacher in the Great Mosque and teacher at Aga Mosque School.) Miivaiz-i Diniye (Izmir 1328/1910) p. 38. 39. Ahmed Cevdet Pasa, Mecdle-x Ahkam-t Adliyc. (Istanbul 1876). Also AH Himmet Berld, Actklamah Mecelk, (Istanbul 1990) p. 7. The Mecelle Com-

196

40. 41. 42. 43.

Tfce Well-Protected Domains mission began work in 1869 but the multi-volume work did not appear until 1876. An Arabic translation was made. Berki, Afiklamalt Mecelle, p. 9. Ibid., p. 8. Ibid., p. 10. Messick, The Calligraphic State, p. 56. Messick also mentions that the effect of this codification was to strengthen the hand of the Suitan. See pp. 57-9.

44. BBA IradeDahiliye 72560.18 Cemaziyelahir 1301/15 April 1884. Yddiz Palace Imperial Secretariat. 45. BBA Y.Mtv 183/16 5 Cemaziyelahir 1316/31 October 1898. Vali of Yemen Hiiseyin Hilmi Pasa to Yildiz Palace Imperial Secretariat, no. 192. 46. Messick, The Calligraphic State, p. 50. 47. BBA YEE 14/292/126/8. 5 Temmuz 1301/18 July 1885. 48. Ibid. T h e 'civilizing motif in Ottoman terminology relating to the indigenous population of the Hijaz is reflected in the official almanacs. In a salname for the vilayet of Hija2 ( the 'primitive condition' of the area and the people is expressed in the following terms: 'Before [the arrival of the Ottomans} because the people of Mecca were rather short of intelligence, the Haram-i Sharif was deserted to the point where deer would wander in the Bab al-Safad from the Cabal-i Qabis and walk around the Haram.' See: Salname-i Vilayet-i Hicaz, 1309. p. 185. 49. BBA Y.Mtv 37/73; Dispatch from Vali of Yemen Osman Nuri Pasa to Yildiz. Palace. Yildiz Palace Imperial Secretariat, no. 1562. T h e document is undated but bears a reference to an immediately previous communication dated 3 Agustos 1304/16 August 1888. 50. BBA YEE 14/292/126/8 7 Tesrin-Ewel 1306/20 October 1890. Report on tour of duty in Yemen. 51. BBA Y.A RES 16/21,10 Sabanl299, 27 June 1882 Grand Vizier's Office. 52. Messick, The Calligrahic State, p. 65, 53. BBA Y.A R E S 9 3 / 3 8 Memorandum by Minister of education Zuhdii Pasa to the office of the Seyhulislarn, 20 Receb 1315/15 December 1897.

54. Ibid. 55. Ibid. O f course this practice had only been forbiden since 1859. It is noteworthy that Snouck Hurgronje also referred to, ' T h e Government printing press (in Mecca) which had recently been opened there (188586), See Hurgronje, Mekka, p. 165. Hurgronje refers to works such as the 'Universal History of Mosltm Conquests ... from the time of Mohammed till the year 1885 ' printed at the press under goverment auspices. 56. Ibid. Also memorandum by Council of State dated 16 Saban 1314/20 January 1897. There was also considerable discussion as to what to do

Notes

57. 58. 59.

60. 61.

62. 63. 64. 65. 66.

197

with the 'faulty' copies of the Holy Book, as destroying even faulty copies was objectionable. It is contended that one of the reasons put forward in the fetwa deposing Abdiilhamid was that he had ordered the burning of Qur'ans who had failed the text of orthodoxy. Despite their condemnation of the Sultan the Young Turk regime continued the practice of 'vetting' Qur'ans and after 1909 they established the 'Superintendancy for the Printing of Q u r ' a n s a n d legal works'. See: MecltS'i Vukela Mazbatalan (Minutes of the Ottoman Cabinet) vol. 184 no. 730, and First Encyclopedia of Islam, 'Shaikh al-Islam' pp. 276-8. T h e practice still holds today. T h e State Religious Bureau (Diyanet isjeri) must give the seal of approval to any Qur'an before it is sold to the public. O n the history of the formation of the Diyanet, see istarTarhanh, Miisluman Toplum 'Laik' Dcvlet, Turkiye'de Diyanet isleri Baskanhgi (Istanbul 1993). BBA Y.Mtv 222/20,5 Receb 1319/18 October 1901. Minister for Education Z u h d i i Pa$a, no. 26. BBA Y.Mtv 227/139,19 March 1902. Ottoman Embassy Stockholm to Foreign Ministry, tel. no. 20. Riza Tevfik, Biraz da Ben Konusayitn. (Istanbul 1993), pp. 242-3. Tevfik also claimed that the authorities confiscated other classics such as the Buhari-i Serif, and Taftazani's AkaiM Nesefi and had them burned in the Istanbul bath-houses. H e alleged that the Buhari was censored and a new version, entitled Sabih'i Buhari (The True Buhari) was issued. j . W. Redhouse, A Vindication of the Ottoman Sultan's Title of'Caliph' (London 1877) pp. 6 - 7 . BBA Irade Meclis-i Mahsus 4625. 26 Rebiyiilevvel 1307/20 November 1889. This 'exclusivist' tone is apparent in the very 'Law on Ottoman Nationality' (19 January 1869), which states in Article 8: 'The children of one who has died or has abandoned Ottoman nationality, even if they are minors, are not considered to be the same nationality as their father and continue to be considered Ottoman subjects. (However) the children of a foreigner who has taken Ottoman nationality, even if they are minors,-will not be considered the same nationality as their father and will be considered as foreigners.' See Dustur 1. Tertip pp. 16-18. O n the matter of the 'French Algerians' see: Pierre Bardin, Algeriens et Tunisiens Dans UEmpire Ottoman de 1848 a 1914, (Paris 1979) pp. 54-78. Gervais Courtellemont, Mon voyage a la hiecaue (Paris 1896) pp. 46 and 58. My translation. BBA Y.A R E S 15/38 17 Cemaziyelevvel 1299/6 April 1882. Surayi Devletno.2317. BBA YEE 1 4 / 2 9 2 / 1 2 6 / 8 5 Temmuz 1301/18 July 1885 Report by O s m a n N u r i Pasa. Ibid. Courtellemont, Mon voyage, p. 151.

198

The Well-Protected Domains

67. Ibid., p. 95. 68. BBA Y.A RES 15/38 Memorandum from the Administrative Council of the Hicaz. 24 Safer 1299/15 January 1882. 69. Bernard Cohn 'Representing Authority in Victorian India' in Hobsbawm and Ranger (eds), The Invention of Tradition, p.174. Cohn indicates that the title was borrowed from Mughal representations of power. 70. India Office Political and Secret H o m e Correspondence. L / P & S / 3 / 2 3 9 vol. 52 p. 937. Acting Consul Moncrieff to Lord Granville.

71. Ibid. 72. BBA Irade Dahiliye 68044, 15 Rebiytilahir 1299 (6 March 1882) See also Selim Deringil, 'Legitimacy Structures in the Ottoman Empire; The Reign of Abdiilhamid II1876-1909,' International journal of Middle East Studies, 23 (1991) pp. 345-9. 73. Courtellemont,Mon voyage, p. 101. Abou Muttalib was the Emir of Mecca, technically an Ottoman official position although the Emirs enjoyed considerable autonomy. 74. Y.A RES 98/38 Memorandum of the Council of State, 16 Saban 1314/ 20 January 1897. 75. BBA Yildiz Arsivi Hususi Maruzat (Y.A H U S ) 295/108,10 Zilkade 1311/15 May 1894. Sublime Porte Receivers Office 2968. 76. BBA Y.A H U S 297/25,12 Zilkade 1311/17 May 1894 Sublime Porte Receivers Office 4002. 77. BBA Y.A H U S 245/45 10 Saban 1308/21 March 1891 Sublime Porte no. 7. O n late Ottoman relations with Morocco see Edmund Burke III, 'Le Pan-islamisme et les origines du nationalisme en Afrique du nord, 1890-1918/Cabiers de I'Unite de I'Antbropolog'te socials et culturelle. Universite d'Oran (Algeria) 1987, 21-40. 78. O n the competition in aims giving between the Ottomans, Moroccan Sultans and the Mughals in earlier centuries see: Suraiya Faroqhi, Pilgrims and Sultans, p. 143-8. 79. For a very competent study of mid-century Egypt see Toledano, Egypt in the Nineteenth Century. 80. British forces invaded Egypt in September 1882 as a response to the nationalist movement led by Ahmad Urabi Pasa. After this date, although the British nominally recognized Ottoman suzerainty in Egypt, the country was in fact turned into a 'Veiled Protectorate' and the ruling khedive, technically an Ottoman official, became a British puppet. See Selim Deringil, 'The Ottoman Response to the Egyptian Crisis of 1881-82,' Middle Eastern Studies, vol. 24 (1988) pp. 3-25. O n the relationship of the sultan and the khedive see: L. Hirsnowicz, 'The Sultan and the Khedive 1892-1908/ Middle Eastern Studies, vol. 8 (1977) pp. 287-311.

Notes

199

81. BBA Y.A H U S 159/80, 6 Zilkade 1295 / l November 1878 Memorandum by Foreign Minister Safvet Pasa. 82. BBA Y.A H U S 188/101, 17 Cemaziyelevvel 1303/21 February 1886. Grand Vizier Kamil Pasato the Omce of the Imperial Secretariat; Snouck Hurgronje also was to note that, 'The Turkish Sultanate would lose its prestige all over the Muslim world if the yearly gifts .,. of money and corn from Egypt were stopped.' See Hurgronje, Mekka,^. 173. Although the provisions were sent from Egypt, the official position was that the khedive was simply acting as the conduit for the sultan's bounty. 83. BBA YEE 14/292/126/8. Osman Nuri Pasa is very representative of the provincial administrator who is something of an 'unknown soldier' in late Ottoman history. Although much has been said and written on such notable figures as Re$id Pasa, A!i Pasa, Fuad Pasa, Midhad Pasa etc., men who dedicated their lives to the thankless task of patching up what was an increasingly more leaky boat have been forgotten. Osman Nuri Pasa is a good example of the diligent provincial administrator and his views are representative as such of the late-Ottoman world view. 84. Ibid. T h e Hijaz had a special status and considerable autonomy, one of its privileges was that its population was exempted from military service. 85. Ibid. T h e imagery of the tree is significant here. One of the symbolic representations of the Ottoman state was the plane tree. 86. Ibid. 87. Fleischer, 'Ibn Khaldunism'. 88. See Faroqhi, Pilgrims and Sultans, p. 141: 'The Ottoman Sultans could not legitimize themselves by their role in early Islamic history, nor could they put forth a claim... of being descended from Chingiz Khan or Tamerlane. Thus their legitimacy depended on practical service to the Islamic community at large, and ... service to the pilgrims was of special importance.'. 89. William Ochsemvald, Religion Society and the State in Arabia. The Hicaz under Ottoman Control 1840-1908 (Columbus, Ohiol984). 90. BBA Irade Dahiliye 27,9 Saban 1313/25 January 1896. Instructions for the yearly committee of envoys sent from Istanbul to the Hijaz to secure the safety and comfort of pilgrims. 91. Ibid. 92. BBA Irade Hususi 102 23 Cemaziyelahir 1311/1 January 1894. 93. BBA Y.Mt. V 67/51, 21 Safer 1310/14 September 1892. Vali of Hicaz Osman Nuri Pasa, to Yildiz Palace. 94. The most detailed works in English on the Hicaz Railway remain; Jacob Landau, The Hijaz Railway, (Detroit, Michigan 1971) and William Ochsenwald, The Hijaz .Railroad (Charlottsville Va. 1980). 95. German engineers were used but were kept somewhat in the background. See Ochsenwald, Hijaz Railroad, pp. 60,66,152.

200

The Well-Protected Domains

96. For a detailed report including detailed drawings and specifications of the projected fleet see, BBA Irade Meclis-i Mahsus 5368,18 Rebiyiilevvel 1309/22 October 1891, Minutes of Cabinet meeting where the issue was discussed; see also, Yildiz Miitenevvi Maruzat (Y. Mtv) 65/117 25 Muharrem 1310/19 August 1892. Report by the Ottoman Admiralty. 97. Feroze Yasamee, 'Abdulhamid II and the Ottoman Defence Problem,' Diplomacy and Statecraft vol. 4 (1993) p. 20. 98. India Office L / P & S / 7 / 2 7 p. 613. 27 November 1880 and 10 December 1880. From John Kirk H M Consul General Zanzibar to A. C. Lyall C. B. Secretary to Government of India, Foreign Department. 99. Ibid., Muscat 6 December 1880. From Lieut. Col. S. B Miles, Political Agent and Consul, Muscat; to Lieut. Col. E. C Ross, Political Resident in the Persian Gulf. 100. Ibid., p. 617. Bushire 13 December 1880. From Col. E. C Ross, Political Resident in the Persian Gulf to A. C Lyall, Secretary of the Government of India Foreign Department. 101. T h e Ottomans had never recognized the occupation of Aden and Ottoman atlases at the turn of the century continued to refer to it as Adin Sancagu 102. India Office L / P & S / 7 / 2 8 p. 715. Aden 4 October 1880. From Major G. R. Goodfellow, Acting Resident to C. Gonne, Chief Secretary to the Government of Bombay Political Dept. 103. Ibid. Aden, 14 February 1881. From Aden Residency, Major Gen. F. A. E. Loch to Government of India. 104. Ibid. T h e resident hastened to add chat Ah Mokhbil told him chat the petitions had been obtained under duress. 105. Ibid. L / P & S / 7 / 2 7 p. 628. Simla, 2 February 1881. Letter from the Government of India to Secretary of State for India. 106. BBA Y.A H U S 209/15,5 Rebiyukhir 1305/21 December 1887 Kamil Pasa to the Sultan. 'Arab Government' (Hukumet-i Arabiyyc) had become something of a catchword for all objectionable activity on the part of the Arabs -such as Arab nationalism. O n Ottoman reactions to early Arab nationalism, particularly the Urabi movement, see Deringil, 'The Ottoman Response', pp. 6-24. 107. BBA Y.A H U S 209/15. Kamil Pasa to Yildiz Palace Imperial Secretariat. 108. BBA Y.A H U S 214/46. Kami! Pasa to Imperial Secretariat. 21 Ramazan 1305/1 June 1888. 109. Mardin, Religion and Social Change, p. 112. It might be more a case of coopting a few of them and leaving the rest to their own devices. T h e numerous pleas relating to the piteous situation of the medreses in Iraq bear out this interpretation. See below. 110. As an example of this genre see, Mehmed Sadik, Muftti of Milas, Alem-i Islam'da Qkad-t Bkber. T h a t a small town religious functionary should so

Notes

201

actively publish religious tracts on so radical an issue is indicative of this new militancy among the lower ulema. 111. Irfan Giinduz, OsmanhlardaDevkt/Tekke Munasebetleri (Istanbul 1989) p. 221. This is a very comprehensive study which sheds light on this critical issue. 112. Ibid., p. 222-3. 113. BBA Y.A H U S 203/69 Sublime Porte 29 Ramadan 1304/21 June 1887. Grand Vezier Kamil Pasa to Imperial Secretariat. It was, however, decided that, as a precautionary measure, he would travel via Istanbul in order to avoid provoking the ire of his followers. 114. BBA YEE 14/2065/74/30, 18 Tesrin-i Sani 1306/1 December 1890. Report by Mehmed Bey, Imperial A D C , Major of Cavalry. I owe thanks to Butrus Abu Manneh for this reference. 115. Ibid. O n the various activities of Gumushanevi see, Butrus Abu Manneh 'Shaylch Ahmed Ziya'iiddin El-Gumushanevi and the Ziya'i-Khalidi Suborder' Frederick De Jong (ed.) Shi'a Islam, Sects and Sufistn (Utrecht 1992) pp. 104-17. This event is also discussed in Raymond Lifchez,Tfce Dervish Lodge (Berkeley 1992) p. 223. Gumushanevi Naksjbendis are still very active in Turkey today. 116. Abu Manneh, 'Shaylch Ahmed Ziyauddin' p. 113. 117. Ibid. 118. BBA Y.A H U S 263/6; 1 Muharrem 1310/26 July 1892. Sublime Porte to Yildiz Palace, Report by Imperial A D C Cevad Pa$a. 119. Ibid. 120. Previous sultans had also made an effort to harness the popular energy of the tarikat, or mystic orders. Sultan Mahmud II, who is reknowned for his crushing of the Janissaries in 1826, sought to centralize the appointment of tekkc shaikhs with the establishment of the Ministry for Pious Foundations (Evkaf-i Humayun Nezflretf).The Tanzimat had continued the process. See: Giindiiz, Ostnanlt tmparatorlugunda Devlet-Tekke Munasebetleri p. 197: 'Although all these measures were in fact contrary to our traditions, the state, which was leaning towards centralization in all things, had no alternative.'. 121. Butrus Abu Manneh, 'Sultan Abdiilhamid II and Shaikh Abulhuda AlSayyadi/ Middle Eastern Studies, vol. 15 (1979) p; 138. 122. Ibid., pp. 136-9. O n Ahmad Es'ad's role in that capacity see Deringil, 'The Ottoman Response'. 123. Abu Manneh, 'Sultan Abdiuhamid II and Shaikh Abulhuda', pp. 140-1. 124. BBA YEE Kamil Pasa Evrakina Ek. (KPE) 86-9/880. It must be noted, however, that Kamil Pasa was no admirer of Ebulhuda. 125. J. Spencer Trimingham, The Sufi Orders in Islam (Oxford 1971) p. 126. 126. Haslip, The Sultan, p. 246.

202

The Well-Protected Domains

127. H . Anthony Salmone, 'The Real Rulers of Turkey/ The Nineteenth Century, vol. 37 (May 1895) pp. 719-33. In view of the evidence emerging from the archives it is highly unlikely that Abdulhamid should have allowed himself to be 'terrorized' by men whom he kept on his staff. 128. Trirningham, Tfce Sufi Orders, p. 127. 129. Smith, The Ethnic Origins of Nations , p. 79. From earlier centuries the Ottoman elite had conceived a very defitinite self-image. See Fleischer, Bureaucrat and Intellectual, particularly pp. 410 and 261-72. 130. Mardin, Religion and Social Change, p. 80. 131.BBA YEE 14/1188/126/9. Also see below Chapter 3. 132. Mardin, Religion and Social Change, pp. 32 and 253, extract from Bitiis Gazette of 17 October 1889: I n respect of this caza, though it is hoped that under the Sultan's auspices ... order tranquillity, and civilization ... will be established-yet in its present condition is entirely wanting in such order and civilization.' For the filtering down and 'popularization' of science in the Ottoman provinces see Ibid., p. 76. 133. Ahmet Yasar Ocak, 'Abdulhamid Donemi islamciliginm Tarihi Arka Plam: Klasik Donem Osmanh Islami'na Genel Bir Bakis'. Sultan II Abdulhamid ve Devri Semineri. Istanbul Oniversitesi Edebiyat Vakultesi, Bildiriler (Istanbul 1994) p. 123. Emphasis in original, Ocak uses the term 'modernist' (yenilesmeci). 134. Gluck, japan's Modem Myths, p. 204. 135. Eugene Weber, Peasants into Frenchmen, (Stanford 1976).

Chapter Three

1. O n this dictum see: Lewis, The Political Language of Islam, p. 29: 'The basic rule for Muslim social and political life, commonly formulated as "to enjoin good and forbid evil" is thus the shared responsibility of the ruler and the subject.' Lewis is quoting from the Qur'an. (111,104,110,114; V I U 5 7 ; IX, 67, 71,112; XXII,41; X X X I , 17 as cited in Lewis fn 8 p. 129). O n the same principle see Lambton, State and Government, pp. 37, 42, 36,105,145. 2. John Guest, The Yezidis. A Study in Survival (London & New York 1987) pp. 132-3. This paragraph is my fictional re-creation based on Guest's book. 3. O n heresy in the Ottoman Empire see: Ahmet Yasar Ocak, 'Kanuni Sultan Siileyman devrinde bir O s m a n h Heretigi: Seyh M u h y i d d i n Karamani,' in Prof. Befeir Kutiikoglu'na Armagan, Istanbul Oniversitesi Edebiyat Fakiikesi, (Istanbul 1991) pp. 473-84. Also by the same author: "Les reactions socio-religieuses contre 1'ideoiogie officielle ottomane et la question de zendeaa ve ilhad. (Heresie et Atheisme au X V I siecle) Turcica 21-3, (1991) pp. 71-82.

Notes

203

4. Weber, Peasants into Frenchmen. 5. It is interesting to note that these tactics were very similar to methods used in other imperial systems at the time. For instance, the British in India used very similar methods, see Gyanendra Pandey,' 'Encounters and Calamities'; T h e History of the North Indian Qasba in the Nineteenth Century in Selected Subaltern Studies, Ranajit Guha and Gayatri Chakravorty Spivak (eds) (Oxford 1988) pp. 106-7: Pandey gives an account of how a local 'disturbance' in the eyes of the Raj officials was settled in 1893-94 by a 'punitive police [who] were quartered in the town (Mubarakpur) for several months'. My emphasis. 6. Guest, The Yezidis. This remains the best study to date of these people who still inhabit eastern Anatolia, northern Iraq and the territories of the ex-Soviet Union. For information on the Yezidi faith, see Isya Joseph, Devil Worship. The Sacred Books and Traditions of the Yezidis. (London 1919) reprinted by the Health Research Institute, California (1972), Although they strenously denied that they worshipped the Devil, the reputation seems to have stuck. It is interesting to note that present day Turkish historians seem to have remained loyal to the Ottoman line that the Yezidis were heretics from Islam. See Mehrnet Aydin, 'Yezidiler ve Inane Esaslan,' Belleten (Official Journal of the Turkish Historical Society) vol. 52 (1988) particularly p. 33: 'According to the latest research, Yezidism is a heretical (heretique) branch of Islam. In other words, it is a deviant sect' The French term occurs in the original. 7. Guest, The Yezidis, p. 126. 8. O n the Hamidiye regiments see Selim Deringil, 'Ottoman to Turk, Minority^Majority Relations in the late Ottoman Empire,' paper delivered in conference: ' M a j o r i t y - M i n o r i t y Discourse. P r o b l e m a t i z i n g MukiculturaHsm,' East-West Centre, Honolulu Hawaii, 11-13 August 1994. Publication of proceedings forthcoming. Stanford University Press. 9. BBA Yjldiz Miitenevvi Maruzat (Y. Mtv) 50/21,5 Mayis 1307/18 May 1891. Report signed by three local ulema, Seyyid Mehmed N u r i from Dahok, Seyyid Nureddin from Befirkan, Farukizade Abbas Buhari, and Leader of the Advisory Commission, Major Abdelkadir. Note that three local religious notables were present in the commission. 10. Ibid. The names of ten chiefs are given. 11. BBA Yildiz Miitenevvi Maruzat (Y.Mtv) 51/61,19 Mayis 1307/31 May 1891. Major Abdelkadir to Yildiz Palace Imperial Secretariat. 12. BBA Y.Mtv 53/6,11 Zilkade 1308/18 June 1891. Yildiz Palace Imperial Secretariat, Imperial Private Secretary Sureyya Pa$a. 13. BBA Y.Mtv 5 1 / 6 1 , 1 8 Zilkade 1308/25 June 1891. Memorandum by Imperial Aide de Camp Mujir Sakir Pasa. 14. Ibid.

204

The Well-Protected Domains

15. BBA Y.Mtv 6 1 / 1 8 , 1 9 Mart 1308/28 March 1892, Yildiz Palace Imperial Secretariat. Cipher telegram from the Commander of the Fourth Army, Mii;ir Mehmed Zeki Pasa to Yildiz Palace. ';_ 16. BBA Irade Dahiliye 97775,9 Rebiyulewel 1309/13 October 1891 Y.ldiz Palace Imperial Secretariat no. 818. 17. Guest, T V Yezidis, p. 129. 18. Ibid, p, 130. 19. BBA Irade Dahiliye 53,7 Agustos 1307/20 August 1892.Telegram from Omer Vehbi Pa$a to Sublime Porte. 20. BBA Irade Dahiliye 53,8 Agustos 1397/21 August 1892.Telegram from Omer Vehbi Pasa to Sublime Porte. T h e document includes a list of 11 villages by name and the amount of money needed for the schools. The total came to 229,930 kurus. It is unclear exactly who the 'Sebekli' were. There is a contention that they are Muslims of the Sha'afi school. Yet others claim they are a sort of Shi'ite. See Harry Charles Luke(somerime Assistant Governor of Jerusalem), Mosul and its Minorities {London 1925) p. 14-15: '[In Mosul] you find Shebeks, a dim agricultural tribe, who may be a sect of Kurdish Shiahs but have an odd dialect and no mosques, and are believed by their neighbors to be a survival of one of the great Mongol invasions, a living relic of Hulagu or Timurlenk.' 21. Ibid. T h e Yezidi sanctuary at Lalis was the venered shrine of Sheikh Adi b. Musafir, the prophet of the Yezidi religion who had lived in the l i t h century. O n him and the sanctuary see: Guest, The Yezidh, pp. 15-27. 22. BBA Irade Dahiliye 53, 27 Muharrem 1310/21August 1892. Administrative Council of Mosul Vilayet to Sublime Porte no. 74. T h e Council also referred to the Sebekli as 'a heretical sect who live in a region to the south of Mosul'. 23. BBA Meclis-i Vukela Mazbatalan (M.V) 71/10, 2 Safer 1310/26 August 1892. T h e Sheikhs in question were Mirza Bey, Ali Bey, Hamza Bey, and Huseyin Bey. 24. BBA Irade Dahiliye 53, 31 Agustos 1308/13 September 1892. Omer Vehbi Pasa to Sublime Porte. 25. BBA Irade Dahiliye 56,11 Eylul 1308/24 September 1892. Omer Vehbi Pasa and Governor Osman Pasa to Sublime Porte. T h e reference here is to the Yezidi kavals, or holy men who passed from village to village bearing the sacred relics, the bronze peacock or 'Melek Tatts, and collecting contributions. Omer Vehbi Pasa's allegations of the collection of vast sums which would then be misappropriated were not entirely unfounded. One Yezidi kbfek, or medium, had, according to French consular sources, ammassed the equivalent of £9000 on the basis of alms and had proceeded to build himself a castle. See Guest, The Yeztdis, pp. 28-41 and p. 134. 26. Guest, The Yezidh, p. 130.

Notes

205

27. BBA trade Dahiliye 56,12 Eyliil 1308/25 September 1892. Omer Vehbi Pa§a and Governor Osrnan Pasa to Sublime Porte. 28. BBA Irade Dahiliye 56 14 Rebiyiilevvel 1310/6 October 1892. Sublime Porte no 744. T h e Imperial irade was issued on 18 Rebiyiilevvel 1310/11 October 1892. 29. BBA MechVi Vukela Mazbatalan (M.V) 71/37 24 Safer 1310/17 September 1892. Minutes of Ottoman Council of Ministers discussing telegram from Vilayet of Mosul. 30. Ibid. 31. BBA Y.Mtv 68/90,28 Eyliil 1308/11 October 1892. Vali of Mosul Osman Pasa to Yildiz Palace Imperial Secretariat. 32. Guest, The Yezidis, pp. 130-1. Although Guest claims that the civil Governor was not notified the Ottoman documentation points in the opposite direction. 33. Ibid., p. 133. 34. BBA Y.A H U S 282/27,13 Tesrin-i Evvel 1309/26 October 1893. Governor of Mosul Aziz Pasa to Sublime Porte. 35. BBA Y.A H U S 283/55, 22 Rebiytilahir 1311/2 November 1893. Sublime Porte Receiver's Office no. 1394. Grand Vezir Cevad Pasa to Yildiz Palace. 36. BBA Y.A H U S 283/55,19 Tesdn-i Evvel 1309/1 December 1893. Governor of Mosul Aziz Pasa to Sublime Porte. 37. BBA BEO 25835 Mosul Giden 336,3 Receb 1311/10January 1894. Sublime Porte special sitting of the Council of Ministers. 38. BBA Y.Mtv 71/25,4 Cemaziyelevvel 1310/24 November 1892. Imperial ADC's Ibrahim Dervis bin Ibrahim and Sakir Pasa. 39. BBA Y.Mtv. 71/53, 8 Cemaziyelevvel 1310/28 November 1892. Office of the General Staff. 40. BBA Y.Mtv 74/33, 12 Kanun-u Sani 1308/25 January 1893. Imperial A D C Dervis. 41. BBA Y.Mtv. 74/91,24 Kanun-u Sani 1308/6 February 1893. Ferik Sakir Ferik Sadilc, Mir Liva Kamil, Yaveran-i Hazret-i Sehriyari Dervis. 42. BBA Irade Hususi 22, 6 Safer 1311/19 August 1893. Yjldiz Palace Imperial Secretariat no. 865. 43. BBA BEO 24234 Mosul Giden, 27 Cemaziyelevvel 1311/6 December 1893. 44. BBA M.V 184/733. 27 Safer 1332/25 January 1914. Also during the Armenian massacres of the Great War it is claimed that the Yezidis gave shelter to their erstwhile enemies. See Luke, The YeziMs, p. 25: 'They gave shelter to hundreds of Armenian refugees who crawled from Deir-ezZ o r (sic) to the jebel Sinjar in the course of the great Armenian massacres, and stoutly refused to surrender them despite the persuasion and threats of the Turks.'

206

The Well-Protected Domains

45. BBA YEE 11/1554/120/5 1 Ramazan 1300/6 July 1883. Yikhz Palace to Seyhulislarn's Office. 46. BBA Y.Mtv 65/33. 9 Muharrem 1310/3 August 1892. Seyhiilislam's Office. Seyhulislam Cemaleddin; T h e Redhouse Turkish and English Lexicon defines 'da'i' as: 'i) One who calls, summons or invites, a supplicant, ii) A one who calls men to religion; also a functionary who initiates novices, a missionary. Hi) Members of the corps of ulema use this term to others of the corps, to designate themselves and others from the same corps.' From the context of the word in the documents seen by this author the closest meaning seems to be the second, although it remains possible that the term may have had a generic significance meaning simply, ulema. See below chapter 4 on 'The Missionary Problem'. 47. BBA Y.Mtv 231/79, 14 Rebiyulevvel 1320/1 July 1902. Seyhulislam Mustafa Celaleddin to Imperial Secretariat. 48. BBA M V 9 6 / 1 , 9 6 / 2 , 1 2 Safer 1302/1 December 1884.

49. Ibid. 50. O n the Zaidi sect see Lambton, State and Government, pp. 28-34. Also Messick, Tfee Calligraphic State., pp. 27,114,129. • 51. BBA Y.Mtv 183/93,17 Cernaziyelahir 1316/2 November 1898. Vilayet of Yemen to Yildiz Palace Imperial Secretariat, no 205. Vali of Yemen Hiiseyin Hilmi Pasa. 52. BBA Y.Mtv 195/34,6 Cernaziyelahir 1317/12 October 1899. Governor of Yemen Hiiseyin Hilmi Pa$a to Imperial Secretariat. Enclosing syllabus for primary schools. 53. BBAY,Mtv92/127,17Temmu31315/20 July 1899. Yildiz Palace Imperial Secretariat. 54. Eugene Rogan, 'The al-Karak Revolt of 1910; Ottoman Order at Odds with Local Order.' Paper presented in the panel, 'The New Order and Local Order: Continuity and Crisis in Everyday Life'. MESA meeting Toronto November 16 1989. (cited with the permission of the author). 55. BBA Y.Mtv 189/165.27 Zilhicce 1316/8 May 1899. Bab-t Fetva Dairei Me^ihat. Seyhulislam Mehmed Cemaleddin. T h e hocas to be sent to Ankara were named as Ishak Efendi of the Suleymaniye mosque, and Abdelhalim Efendi from the Beyazid mosque, both Istanbul ulema, whose 'good conduct and talent is known'. It was also stated that similar preachers had been sent to the Vilayet of the Aegean Islands (Cezayir-i Bahr-i Sefid Vilayetleri). 56. Suny, Looking Towards Ararat pp. 103-6. 57. Sakir Pa$a was born in 1838 in Istanbul to a family with a long history of government service. Graduating from the military academy in 1856, Sakir Pasa served on the Refugees Commission for the Vilayet of the Danube (Bulgaria), where he worked with Mithad Pa§a. In the early 1870's he served as the Mutasamf of Bagdad. After the calamity of the war of 1878

Notes

207

he was one of the few military commanders to emerge with a clean record. In May 1878 he was appointed Ambassador to St-Petersburg. In 188990 he served as the vali of Crete. As a result of what was a remarkable career he was made chief aide de camp of Sultan Abdulhamid (Yaver-i Ekrem) in July 1890. His interest in the Armenian question and his knowledge of Armenian activities in Russia determined his appointment to the post of Inspector of Anatolian vilayets. Sakir Pasa spoke fluent French and Russian as well as some Arabic. See. Ali Karaca, Anadolu Islahatt ve Ahmet Sakir Pasa (Istanbul 1993) pp. 17-29. 58. BBA Y E E / A / 2 4 - X / 2 4 / 1 3 2 , 3 Kammusani 1314/16 January 1898, Vilayet of Trabzon to General Inspector of Anatolia in Amasya. 59. Ibid., Vilayet of Trabzon to Amasya. 23 Zilhicce 1316/4 May 1899. Secret. no.346. There are references in the document to repeated alarms and investigations. 60. Anthony A. M. Bryer, The Empire ojTrebizond and the Pantos, Variorum reprints, (London 1980) pp. 268-72. My thanks to Dr Gun Kut for this reference, 61. E W. Hasluck, Christianity and Islam under the Sultans. (Oxford 1929) vol.2 pp. 469-70. 62. BBA YEE A / 2 4 - X / 2 4 / 1 3 2 , Sakir Pasa to the Vilayet of Ankara. 11 Agustos 1313/23 August 1897. no. 337. 63. Bryer, The Empire of Trebizond, p. 48 ; Because they were exempt from taxation as Christians, and untill the 1890's avoided military service as Muslims: 'The Crypto-Ghristians were therefore no more than people who, from an Ottoman point of view, did not exist'. 64. BBA YEE A - 2 4 / X / 2 4 / 1 3 2 , Sakir Pasa to the vilayet of Ankara. 11 Agustos 1313/23 August 1897. no. 337. 65. Ibid. 66. Mutasarnf of Yozgad Ahmed Edib to Inspector General Sakir Pasa. Cipher no.9476. 67. BBA YEE 3 1 / / 7 6 - 4 5 / 7 6 / 8 1 ; 6 Saban 1316/20 December 1898. General Inspectorate of Anatolia to Yildiz Palace Imperial Secretariat, no.768. 68. Ibid. Yemen, Libya, and H k a z were infamous as sinkholes of military force. The Pomaks were a Slavic peoples that had converted to Islam and were reknowned for their zeal. 69. Ibid. 70. BBA Y.Mtv 230/96, 27 Nisan 1318/10 May 1902. Cipher telegram no. 80 from the Vilayet of Ankara to the Ministry of Interior. 71. Ibid. 18 Safer 1320/27 May 1902. Vali of Ankara Mehmed Cevad Bey to Imperial Secretariat. Vilayet of Ankara, tel no. 177. 72. R. Janin, 'Musulmans Malgre Eux, Les Stavriotes'. Echos D'Orient. vol. X V (1912) pp. 495-505. My thanks to Dr G u n Kut for this reference.

208

The Well-Protected Domains

73. BBA Ayniyat Defteri (AD) 1422 page 115, 23 Rebiyuievvel 1307/17 . November 1889. Order from the Sublime Porte to the office of the Chief of Staff (Serasker). 74. BBA M V 68/44., 26 Ramazan 1309/24 April 1892. T h e donme were jews who had converted to Islam under the leadership of Shabbatai Tzvi in 1676. The term Avdeti was used for them in official documentation. O n the donme of Salonica see, Stanford Shaw, The jews of the Ottoman Empire and the Turkish Republic. (London 1991) pp. 177-9. 75. Ibid. Shaw also notes that the donme prohibited marriage with 'real Muslims' and jews. See, The jews ojthe Ottoman Empire, p. 177. 76. Mardin, Religion and Social Change, p. 149. 77. BBA Y.Mtv 65/68,17 Muharrem 1310/11 August 1892. Imperial Privy Purse (Hazine-i Hassa-i Sahane) no. 356. 78. BBA Ayniyat Defteri no. 1423,19 Mayis 1307/1 June 1891, p. 172. 79. T h e best recent study on the Kizilbas is Altan Gokalp, Tetes Rouges et Bouches Nairn, Une confrerie tribale de I'ouest Anatolien (Paris 1980). 80. BBA Ayniyat Defteri 1422, p. 181. 23 Kanun-u Evvel 1305/4 January 1890. Sublime Porte to Ministry of Education. Central Anatolia is still an area with a considerable Kizilbas population, see Gokalp Tetes Rouges, p. 31, 81. BBA Y.Mtv 53/108,2 Zilkade 1308/9 June 1891- Mutasamfof Tokad to Ministry of Interior, no. 159. T h e tenor of these instructions are reminiscent of instructions sent to Christian missionaries in the field. 82. BBA Y.Mtv 53/108,26 Zilhicce 1308/2 August 1891. Governor of Sivas, Mehmed Mernduh Pasa. There seems to be an interesting confusion here. Memduh Pasa stated that the Kizilbas. worshipped Melek Taus, the sacred figure of the Yezidi, and one of the titles of Satan, hence earning the Yezidis the reputation as Devil Worshippers. It seems that Memduh Pasa was confusing the two. See Guest, The Yeztdis. 83. Ibid. T h e recruitment of the Ki&ilbas into the ranks of the Ottoman armed forces had been a problem since the seventeenth century, see Gokalp, Tetes Rouges pp. 229. 84. Hasluck, Christianity and Islam, vol. 2. p. 445. 85. BBA Ayniyat Defteri 1422 pg 35. 18 Muharrem 1307/14 September 1889. Order to the Ministry of Education. See also: Ana Britannica (Turkish edition of the Encyclopaedia Britannica) p. 624: T h e Nuseyri believed in the divinity of the Prophet Ali, and were therefore considered as Shi'a. Yet, their belief also differed considerably from the Shi'ite, including such elements as the belief that Ali created the Prophet Muhammed and Muhammed created Selman, who is responsible for the natural cosmos. They did not worship Sve times a day and their prayers were directed at Ali, Selman and Muhammed. T h e current strong man of Syria, Hafez Es'ad, is a Nuseyri.

Notes

209

86. BBA Ayniyat Defteri 1422 pg 57.16 Receb 1307/8 March 1890. 87. BBA Irade Meclis-i Mahsus 4867, 13 Haziran 1306/26 June 1891. Mutasarrif of La:zkiye Mohammed Hassa to Sublime Porte. 88. Ibid. 89. Ibid. 90. O n the struggle between the Ottoman state and the missionaries over the Nuseyris see; A. L. Tibawi, American Interests in Syria (Oxford 1966) particularly pp. 2 6 2 - 3 . 91. BBA YEE 3 1 / 7 6 - 4 5 / 7 6 / 8 1 ; 6 Saban 1316/20 December 1898. Genera! Inspector of Anatolian Vilayets to Yildiz Palace, no. 768. 92. Ibid. 93. BBA Irade Maarif 1.7 Safer 1314/18 July 1896. 94. Hasluck, Christianity and Islam, p . 469. 95. BBA Yildiz Resmi Maruzat (Y.A H U S ) 5/38,14 Rebiyfilewei 1297/25 February 1880. Grand Vizier Said Pasa to Sultan. It must be pointed out however, that Christians had always served in specialist units such as in the Arsenal and the Navy,

96. Ibid. 97. R. M. Ramsay, Impressions of Turkey During Twelve Years' Wanderings (London 1897)163. 98. BBA Dahiliye Nezareti, Hulcuk Miisavirligi. (Ministry of Interior, Legal Advisors Bureau) (hereafter D H - H M S ) 13/47; 23 Haziran 1320/6 July 1904. General no. 244. File no.62570. In the catalogue of the documents of the Ministry of Interior Legal Advisors Bureau there is a separate category entitled, 'Conversion' (thtida). 99. See,Seyit Solak, 1571'den Gunumuze Ktbrts Turk Yonetimleri (Nicosia 1989) p. 107. Solak is citing the memoir of a muftit, Ahmed Bin Mehrned Emin Hoca, who served as the muftu of Larnaca in 1658-1662. See, Ahmed Bin Mehrned Emin Hoca Hatiratt (Larnaca 1672) p. 63, as cited in Solak. The conflict between the Ottoman authorities and the Greek Patriarchate was particularly acute in Cyprus. Cypriot sources maintain that, 'as O t toman power increased on the island, many young Greeks sought to join Islam. T h e duty of the priest was to ascertain that there had been no coercion and rhat the convert was of age. However, subsequently they always contrived to assert that the convert was not of age, and had not wanted to convert'. See, Ali Nesim, Kibrtsh Turkler'tti Kimligi. (The Identity of Cypriot Turks) (Nicosia 1990) p. 40. My thanks to Basar Ozal for these references. 100. D H - H M S 13/47. 101. Ibid. 102. BBA D H - H M S 13/46; 8 Rebiyiilahir 1312/9 October 1903. Sublime Porte. Ministry of Interior, no.110, 103. Ibid.

210

The Well-Protected Domains

104. BBA DH-HMS 13/47, 30 Tesrin-i Sani 1327/12 November 1911.

105. 106.

107.

108. 109.

110. 111. 112.

113.

114. 115. 116.

117.

General Directives of the Ministry of Interior relating to the verification of the age of converts, no. 503. File 3. BBA D H - H M S 13/49, 25 Mayis 1329/7 June 1913. Ministry of Interior no.79107; D H - H M S 13/50, 3 Agustos 1329/16 August 1913. BBA Y.A H U S 352/1. 18 Subat 1311/3 March 1895. Subiime Porte. Office of the Grand Vizier. Report prepared by Accountant of the Evkaf of Aleppo Ali Riza, President of the Court of Aleppo Mustafa, Special Agent of Britain, Fkzmaurice. Ibid. 18 $ubat 1311, Baztstnm eskt dinlerinc donmekfikri evzalarm&an istidld olunuyor. The fact that the sentence is also incongruously dated again while the document was already dated also leads one to speculate that a Turkish member wanted to get on record as having guessed that the Armenians would re-convert. BBA Ibid.,Gurre-iZilhicce 1313/14 May 1896. Grand Vizier RifatPasa. BBA Y.A H U S 352/60,15 Mayis 1312/28 May 1896. Sublime Porte Office of the Grand Vizier. Cipher telegram from the Vilayet of Aleppo, ValiRaufPasa. Ibid. Ibid. Ibid. This does not mean, of course, that the Vali's suggestions were taken up in Istanbul, or that the centre necessarily agreed with him. But, he must have known the boundaries of what it was permissible to suggest. BBA Y.A H U S 355/38, 3 Temmuz 1312/13 Temmuz 1896. Sublime Porte. Office of the Grand Vizier. Cipher telegram from the Vilayet of Aleppo. Vali Rauf. BBA Y.A H U S 352/4, 29 Zitkade 1313/12 May 1896. Sublime Porte. Receiver's Office no. 893. Foreign Minister Tevfik Pasa. BBA Y.A H U S 352/133,9 May 1897. Subiime Porte. Cipher telegram from the London Embassy dated 9 May 1897. no. 383. BBA Y.Mtv 50/57,17 Mayis 1307/30 May 1891. Yildiz Palace Imperial Secretariat. Cipher from the Vilayet of Ankara. It is worth noting the hesitency on the part of the police official, as this was a time of tension when the slightest provocation could result in inter-community violence. From the Madenataran Depository of Manuscripts in Yerevan. This series of letters are titled 'Copies of letters sent to the Holy Patriarchate of Constantinople. This one is no. 159. It is a letter adressed to 'Most Honourable Holy Father, Patriarch of Constantinople and Greatly Honoured Members of the National Political Council'. It is signed Krikor Vartabet Aghvanyan. I owe this reference and its translation to Ara Sarrafyan who kindly shared this information with me. H e informs me that 'Dajig' is a generic term used by Armenians to denote Muslims.

Notes

211

118. BBA Y.Mtv. 33/88,15 Sevval 1305/25 June 1888. Number 2411-111. Imperial A D C Ibrahim Edhem Pasa co Imperial First Secretary Siireyya Pasa. 119. A r m e n o u h i e Kevonian, Les noccs notres de Gulizar (Paris 1993). Presentations historigues de Anahide Ter Minassiati et Keram Kevonian. I would like to thank Prof. Ter Minassian for bringing this work to my attention. 120. BBA Y.Mtv. 54/12,18 Muharrem 1308/3 September 1890. Vaii of Sivas, Mehmed Memduh Bey, to Imperial A D C Dervis Pasa. 121. BBA Bab-! Ali Evrak Oda S1 (BEO), 149343.6 Zilkade 1320/4 February 1903. Vilayet of Mosul to Ministry of Interior. 122. BBA BEO 149434,9 Zilkade 1320/7 February 1903. Vilayet of Mosul to Ministry of Inrerior. 123. BBA BEO 149900, 16 Zilkade 1320/14 February 1903. From the Vilayet of Mosul to Ministry of Interior. 124. BBA trade Dahiliye 97963. 30 Rebiyulevvef 1309/3 November 1891. Yildiz Palace Imperial Secretariat. 125. B B A Y E E / 3 1 / 7 6 - 4 5 / 7 6 / 8 1 . 126. BBA Irade Hususi 123, 3 Rebiyutahir 1315/1 August 1897. Yildiz Palace Imperial Secretariat no. 3659. 127. BBA YEE A - 2 4 / X / 2 4 / 1 3 2 ; 28 Kanun-u Evvel 1314/11 January 1898. Decoded cipher telegram from Palace Scribe Mehmed Kamil Bey to Ahmet Sakir Pasa. Chapter Four 1. Smith, The Ethnic Origin Of Nations, p. 142. This passage above is my fictional re-creation based on Serif Mardin's study of Said-i Narsi. 2. Hobsbawm. 'Mass Producing Traditions' in The Invention of Tradition, p. 282. Hobsbawm, Nations and Nationalism, p. 110. 3. The standard work on Hamidian education policy still remains: Bayram Kodaman, Abdiilhamid Donemi Egitim Sistemi (Istanbul 1980). This study has benefited greatly from conversations with Aksin Somel, currently finishing a doctorate on the Hamidian education policies. His thesis should provide some pioneering research. 4. 5. 6. 7.

Weber, Peasants into Frenchmen, pp. 7 2 - 3 . Anderson, Imagined Communities, p . 83. Karaca, Anadoh Islahatt ve Ahmet $akir Pasa, pp. 184-5. BBA Y.Mtv 25/52 6 Cemaziyelahir 1304/2 March 1887; Bab-i Fetva Enciimcn-i Maarif (Seyhulislam's Office Educational Commission), Seyhiiiislam Ahmed Es'ad El-Uryani.

8. Ibid. On the gradation of Ottoman Schools sec Kodaman, Abdiilhamid donemi Egitim Sistemi.

212

The Well-Protected Domains

9. Ibid. BBA Y.Mtv 25/52. 10. Ibid. 11. Ibid., enclosed memorandum by Ahmed Recai Efendi, Director of Gaktasaray lycee... It must be pointed out that both Ahmed Es'ad's and Recai's memoranda were suggestions, yet both men were probably putting forward views which they hoped would appeal to the conservatism of Abdiilharnid. 12. Ibid. Detailed breakdown of curriculum by year. 13. BBA Y.Mtv 175/234,15 ZUkade 1315/7 April 1898. Serasker Riza Pasa to Minister for Education. 14. BBA Y.Mtv 181/62,16 Rebiyukhir 1316/3 September 1898. Serasker Riza Pasa to Minister for Education. 15. BBA Ayniyat Defteri no. 1422. p. 297. Gurre-i Rarnazan 1307/21 April 1890. 16. BBA Y.Mtv 189/184 24, Kanun-u Bvvel 1307/6 January 1891. Circular from Ministry of Education. 17. Ibid. 18. BBA Y.Mtv 113/49, 22 Receb 1310/9 February 1893. Minister for Education Ziihdii Pasa. Ministry for Education no. 62, 19. BBA Y.Mtv 172/20, 29 Saban 1315/23 January 1898. Ministry of the Navy. Minister Hasan Pasa to Yildiz Palace. 20. BBA Y.Mtv 172/20,10 Kanun-u Sani 1313/23 January 1898. Imperial A D C and Minister for All Military Schools Mustafa Zeki Pasa to Yildiz Pakce. 21. BBA Y.Mtv 260/200, Selh-i Rebiyiilevvel 1322/15 June 1904. General Inspector for the Provinces of Rumelia Huseyin Hilmi Pasa to Yildiz Palace Imperial Secretariat no. 418. On the donme commercial elite called the Cavelleros in Judeo-Spanish, or Kapanakr in Turkish see; Shaw, The Jews of the Ottoman Empire, p. 178. 22. BBA Y.Mtv 199/74,18 Sevval 1317/19 February 1900; Imperial A D C and Minister for Military Schools, Mustafa Zeki Pa$a to Yildiz Pakce. 23. BBA Y.Mtv 180/16,2 Rebiyukhir 1316/20 August 1898. Yildiz Pakce Imperial Secretariat no. 320. 24. Findley, Ottoman Civil Officialdom, p. 243. 25. Ibid., pp. 243-5. Although at first he seems unable to make up his mind as to whether this attitude existed or not, he does seem in the end to come down on the positive side. 26. BBA Y.Mtv 189/184, 30 Zilhicce 1316/11 May 1899 Ministry of Education. Memoramdum signed by Minister for Education Ahmed Zuhdii Pasa, Director of the Mekteb-i Sultani Gaktasaray Abdurrahman Seref, Director of the Mekteb-i Mulkiye Mehmed Recai. Abdurrahman Seref later became a rather famous Turkish historian whose Tarih Musahabeleri became a standard work of nineteenth-century Ottoman history.

Notes

213

27. Ibid. 28. See Cory Blake, 'Arab Students in the Mekteb-i Miilkiye' (PhD dissertation, Princeton University 1991). 29. Ibid. 30. BBA Yildiz Esas Evrakj (YEE) 14/292/126/8. 'The views of the late Governor of Yemen and Hicaz, Osman Nuri Pasa, on nomadic administration, religious matters, and reform measures'. 5 Temmuz 1301/18 Temmuz 1885. O n Osman Nuri Pass's relations with the Sharifs of Mecca see: Butrus Abu Manners, 'Sultan Abdulhamid II and the Sharifs of Mecca/ Asian and African Studies, vol. 9 (1973) pp. 1-21. 31. Ibid. 32. Ibid. 33. Ibid. Osman Nuri Pasa even implied that the traditional tax-exempt status of the local inhabitants should be revised. 34. Ibid. O n the teaching of the language of the centre seeWeber, Peasanti into Frenchmen, pp. 7 2 - 3 : 'Teaching people French was an important facet of 'civilising' them, in their integration into the superior modern world ... There can be no clearer expression of imperialistic sentiment: a white man's burden of Francophony, whose first conquests were to be right at home.' There is a striking similarity here with the way in which education was perceived in late nineteenth century France, see Weber, pp. 329-30: 'The polite forms it inculcated softened the savagery and harshness natural to peasants'. 35. BBA YEE 3 1 / 7 6 - 4 5 / 7 6 / 8 1 ; Amasya, 6 Saban 1316/20 December 1898. General Inspector of Anatolia Sakir Pasa to Yildiz Palace Imperial Secretariat. no. 768. 36. BBA Irade Dahiliye 98525 28, Cemaziyelevvel 1309/30 December 1891. T h e irade includes the list of the names of the boys. The two Sunnis are marked with the letter sin which appears above their names. See also BBA Irade Dahiliye 98993,19 Cemaziyelahir 1309/20January 1891: The Sultan gave 200 liras out of his privy purse to pay for their bedding when they got to Istanbul. H e further allocated 5000 kurtis monthly for their food, servants, laundry, and private tutors. O n the whole issue of Ottoman counter-propaganda see: Selim Deringil, T h e Struggle Against Shiism in Hamidian Iraq/ Die Welt Des Islam 30 (1990) pp. 45-62. 37. BBA Y.Mtv 72/43, 20 ReMyulahir 1310/11 November 1892. Vali of Mosul Ismail Nuri Pasa to Yildiz Palace. O n the conversion campaign against the Yezidis see above Chapter 2. 38. Ibid. The exact terms used are: 'Kahailin ulum ve maarijle meydan-t medeniyet ve cemaat-i mumtaztA beseriyete is'ali.'. 39. BBA Irade Dahiliye 100672.24 Zilkade 1309/21 April 1892. Yildiz Palace Imperial Secretariat to vilayet of Mosul, no. 8037.

214

The Well-Protected

Domains

40. John Presland, Deedes Bey. A Study ofSir Wyndham Beedei 1883-1923. (London 1942) p. 87. Deedes was an officer in the Turkish gendarmerie. T h e Turkish attitude to education is slow to change, and the adage 'roses bloom where the teacher strikes' (ogretmenin vurduguyerdeguller acar ) is still taken literally by Turkish teachers. 4 1 . For a good account of the reasons for its establishment see Bayram Kodarnan, Sultan II Abdiilhamid'in Dogu Anadolu Politics! (Istanbul 1983). See also: Alisan Akpinar, Osmanh Devletinde Asiret Mektebi (Istanbul 1997). 42. 43. 44. 45.

Ibid., pp. 106-8. Ibid., p. 108-9. Ibid., p. 117. BBA Y.Mtv 67/90,27 Safer 1310/20 September 1892. Minister of Interior Halil Rifat Pasa to Yildiz Palace. Sublime Porte Ministry of Interior no. 42.

46. For a different view as to the degree of coersion and consent involving tribal sheikhs. See : Eugene Rogan, 'Asiret Mektebi : Abdulhamid II's School for Tribes 1892-1907', IJMES 28 (1996) pp. 83-107 47. BBA Y.Mtv. 68/41,20 Eylul 1308/3 October 1892. Vali of Mamuretulaziz Enis Pasa to Yildiz Palace Imperial Secretariat. 48. Ibid. 49. BBA D H - H M S 2 4 / 1 , 1 5 Zilkade 1321/2 February 1904; Ministry of the Interior, secret, no. 249. 50. Mardin, Religion and Social Change, pp. 47,61,62, 226. 51. BBA Y.Mtv 70/14, 4 Rebiyiilahir 1310/26 October 1892. Minister for Education Ziihdu Pasa to Yildiz Palace Imperial Secretariat. 52. Ibid. 53. BBA Y.Mtv. 73/99. T h e Directives for the Administration of the Tribal School. (Asiret Mektebi'n'm Talimatname'i Dab'tliyesid'tr.). 54. Ibid. 55. BBA Y.Mtv 7 6 / 8 8 , 2 1 Ramazan 1310/8 April 1893. Minister of Education Ziahdii Pasa to Yildiz Palace. 56. BBA Y.Mtv 114/80; 26 Kanun-u Sani 1310/8 February 1894. Minister of Education Ziihdii Pasa to Yildiz Palace. 57. Kodarnan, Sultan II Abdulhamid'in Dogu Anadolu Siyaseti, p. 116. 58. BBA Y.Mtv 167/189, 23 Rebiyiilahir 1315/21 September 1897. Minister of Military Schools MtJjir Mustafa Zeki Pasa to Yildiz. O n Mustafa Zeki Pasa see Salname-i Devlet 1315, p. 218. 59. BBA Y.A R E S 134/17,18 Muharrem 1323/25 March 1905. Ministry of Interior to Office of Grand Vizier, no. 161. 60. Serif Mardin also points out chat the food riot 'seems to have been a preliminary to more concrete demands.' See Religion and Social Change, p. 126 ff.

Notes

215

61. BBA YEE 11/1419/120/5 Circular from Yildiz Palace Imperial Secretariat to Sublime Porte. T h e date is given as 1319 Cemaziyelevvel (August 1901) no day is recorded. 62. For instance the 1869 law stipulated that in communes where there were more than 100 non-Muslim households the state had to establish a Christian rusdiyc which would teach in the local language but would also emphasise T u r k i s h language learning. See, Bertold Spuler, Die Minderbeitenschulen der europaischen Turkei von der Refortnze'tt bis zum Weltkrieg. (Breslau 1936) p. 75. (My thanks to Aksin Some! for this reference). 63. Ibid., pp. 56-7. 64. Voyn Bojinev, Bulgarskaya Prosveta v Makedonya i Odrinska Trakya 18781913. (Sofia 1982) pp. 33-4. I am indebted to Aksin Somel for this reference. 65. Richard Clogg, 'The Greeks and their Past' in, Dennis Deletant and Harry Hanak (eds) Historians as Nation Builders, p. 29. 66. BBA Ayniyat Defteri (Payments Register) For Christian schools in the vilayet of Syria, N o 1244, 7 Muharrem 1301/8 November 1883 p. 85. For Armenian schools in Van, no 1419,26 Saban 1297/3 August 1880 p. 46. For an Armenian school in Yalova N o 1423 14 Saban 1308/25 March 1891 p. 34. Inter alia. I owe thanks to Aksin Somel for these references. 67. Kodaman, Abdulhamid Danemi Egitim Sistem't, p. 94, 68. Atilla Cetin, 'II Abdiilhamid'e sunulan Beyrut vilayetindeki yabanci okullara dair bir rapor'. Turk Kulturu vol. 22 (1984) pp. 316-24. 69. BBA Ayniyat Defteri 1422 p. 267,28 Receb 1307/20 March 1890. This does not mean of course that they were. 70. BBA Ayniyat Defteri 1422 p. 577, 20 Kanun-u Evvel 1306/2 January 1891. 71. Ibid. 72. H . F. B. Lynch, Armenia, Travels and Studies, lb the Turkish Provinces (London 1901) pp. 213-15. 73. Andreas Tietze, 'Ethnicity and Change in Ottoman Intellectual History,' Turcica vol. 2 2 - 3 (1991) p. 393. Tietze refers to the literature in Turkish, but written in Greek or Armenian characters. 74. Ibid. Sufficient allowance must be made, however, for the difference between Istanbul and provincial non-Muslims. 75. BBA Ayniyat Defteri 1422,16 Ramazan 1307/6 May 1890. p. 317. Grand Vizier's office to Ministry of Education. 76. BBA Ayniyat Defteri 1422,10 Muharrem 1308/26 August 1890 p. 411. Grand Vizier's Office to Ministry of Education and Ministry of Religious Sects. 77. BBA Ayniyat Defteri 1422,19 Zilkade 1307/7 July 1892. p. 347. Grand Vizier's Office to Ministry of Education.

216

The Well-Protected Domains

78. Kozmas Politis, Yitik Kent'm Kirk Ydi trans. Osman Bleda (Istanbul 1992) pp. 6 1 - 2 . It must be pointed out however that Politis is no narrow chauvinist and he tells his story with humour and compassion. For instance the boys are overcome by the tragic tone of their teacher, but they have distinct difficulty recalling the ancient names of mountains, blurting out the first thing that comes naturally, i.e. the Turkish names. 79. Eric Hobsbawm, 'Mass Producing Traditions' in The Invention of Tradition, p. 265. 80. BBA Ayntyat Defteri 1422,14 Receb 1307/6 March 1890 p. 255. Grand Vizier's Office to the Ministry of Education. 81. BBA Ayniyat Defteri 1422,1 Cemasiyelahir 1307/23 January 1890 p. 215. Grand Vizier's Office to Ministry of Education. 82. BBA Y.A RES 107/50, 1 Saban 1317/5 December 1899. Vilayet of Kastamonu to Ministry of Interior, no. 386 Governor Mehmed Enis Pasa. 83. Findley, Ottoman Civil Officialdom, pp. 134-5,139. 84. A. P. Comte de Cholet, Voyage en TuroM-ie d'Asie, (Paris 1892) pp. 2 - 3 . T h e writer was visiting a village primary school in the region of Haci Bektas in central Anatolia, near Cappadocia. My translation. 85. Princess Annie de Lusignan, The Twelve Years' Reign of His Imperial May esty Abdul Hamid II, Sultan of Turkey (London 1889) pp. 176-7. 86. Sinasi Tekin, 'Imperial Self Portrait'. See also below. 87. Anderson, Imagined Communities, p. 82. 88. Cherniavsky, Tsar and People, p. 150; BBA YEE 11/1419/120/5 Circukr from Palace to all Ministries. 89. Jeffrey Brooks, When Russia Learned to Read (Princeton 1985) p. 216. 90. Ibid., p. 218. 91. Ibid., p. 47: 'The state and Church officials who designed the programs considered religion the foundation of the Russian state system.' Also p. 53. 'Religion and nationality were interwoven as in Bukanov's popular primer: "Our fatherland is Russia. We are Russian people ... There are churches in the cities and villages. The Church is the house of God.'" . 92. Ibid., p. 48 and p. 49. See also above. It must however be pointed out that the official curricula in Russian schools were no more than a suggested outline which the teachers filled as they saw fit. This is very unlike the Ottoman system where teachers were strictly ordered not to stray from the set curriculum. Another obvious dissimilarity is that the problems of an empire which is expanding were not to be same as an empire which was contracting and fighting for its very survival. The aim of the comparison here is merely to bring out the idelogtcal content of the curricula in both systems. 93. Ibid., p. 53. 94. O n Namik Kema! see: Turk ve Diinya Meshurlart Ansiklopedisi. 95. Anderson, Imagined Communities, p. 83.

Notes

111

96. Lesley Blanche, The Sabres of Paradise (London 1960) pp. 162-75. 97. Kodaman, Sultan II. Abdulhamid'in Dogu Anadolu Politikast. 98. See Brooks, When Russia Learned to Read, p. 42; 'By the 8-11 [years old] standard ... Ministry of Education officials calculated that as of January 1915, only 58 per cent of all 8 to 11 year-olds were m school in European Russia and only 70 per cent of the necessary teachers were available. T h e figures for the empire as a whole were 51 per cent and 61 per cent, respectively.' O n the Ottoman case see Findley, Ottoman Civil Officialdom, p. 52. 99. Anderson, Imagined Communities, p. 86. 100. Gluck, Japan's Modern Myths, particularly pp. 102-56, Chapter V 'Civil Morality'. 101. Ibid., p. 105. 102. Ibid., pp. 108-9103. It must be emphasised, however, that Japanese education always remained much more open to western ideas and the emphasis on learning of the world's nations' never entirely disappeared. Also the level of pedagogic debate seems to have been of a much more sophisticated level than in the Ottoman case. See Gluck, Japan's Modern Myths, pp. 104,108. 104. BBAYEE 14/292/126/8. 105. iSiikru Hanioglu, Osmanh tttihad ve Terakki Cemiyeti vejon Turkluk 18891902. (Istanbul 1985) p. 176. Hanioglu's work is remarkable for its great wealth of primary sources and other documentery evidence drawn from numerous archives. There is considerable literature on the Committee of Union and Progress. T h e two seminal studies are by Serif Mardin, Jon Turklerin Siyasi Fikirleri W5-19QS (Istanbul 1983) (First published in 1964) and Feroz Ahmad, The Young Turks: The Committee of Union and Progress in Turkish Politics 1908-1914 (Oxford 1969). These have been followed by other important works. To name just a few see, Sina Aksin, Jon Tiirktcr ve htihat ve Terakki (Istanbul 1980) Zafer Toprak, Turkiye'de Milli tktisat (Istanbul 1983), Tank Zafer Ttmaya Turkiye'de Siyasi Partiler (Istanbul 1984). There is also a veritable library of published memoires of prominent C U P members. 106. Hanioglu, Osmanh pp. 178,181,182. Hanioglu notes however that their advances were not taken that seriously by the Embassy who found their revolutionary jargon somewhat far fetched. 107. Ibid., pp. 4 0 1 - 2 . C h a p t e r Five 1. See below. T h e passage at the beginning of this chapter is my fictional recreation. 2. Gluck, Japan's Modern Myths, p. 57.

218

The Well-Protected Domains

3. Ibid., pp. 133-5. It must, however, be pointed out that the most severe criticism of Christians came from Buddhist circles who were themselves feeling the pressure of the increasingly secular policies of the Meiji regime. 4. Ibid., p. 138. It must, however, be pointed out that the Japanese and Ottoman attitude to religious teaching in schools was diametrically opposed. Whereas the Japanese prohibited religious instruction in schools in 1899, the Ottomans, as an act of deliberate policy, gave it more weight. 5. Seagrave, Dragon Lady, p. 296. 6. Edward Said Orientalism (Basingstoke 1985) p. 294. Emphasis in origi' nal. 7. Jean Haythorne Braden, 'The Eagle and the Crescent: American Interests in the Ottoman Empire, 1860-1870/ Ohio State University P h D Dissertation {1973} p. 22. 8. Tibawi, American Interests, pp. 256-7. 9. Ibid., p. 260. 10. Ibid., p. 269. 11. Jeremy Salt, 'A Precarious Symbiosis: Ottoman Christians and Foreign Missionaries in the Nineteenth Century,' International Journal of Turkish Studies, vol. 3 (Winter 1985-86) no. 2. p. 56. 12. Ibid., p. 65. 13. BBA Irade Dahiliye 100258. YMiz Palace Imperial Secretariat, no. 6975, 27 Sewal 1309/25 May 1892. 14. Hobsbawm, The Age of Empire, p. 71. See also his evaluation of the acceleration of missionary activity on the same page: 'Between 1876 and 1902 there were 119 translations of the Bible, compared to 74 in the previous thirty years, and 40 in the years 1816-54. T h e number of new Protestant missions in Africa during the period 1886-95 was twenty three or about three times as many as in any previous decade.'. 15. Salt, 'A Precarious Symbiosis', p. 56. 16. Jeremy Salt, Imperialism Evangelism and the Ottoman Armenians, 1878-1896 (London 1993) p. 37, 17. A n E a s t e r n S t a t e s m a n , ' C o n t e m p o r a r y Life and T h o u g h t in Turkey,'Contemporary Rev tew, vol. 37 (1880) p. 343. 18. Ibid., p. 344. 19. Samuel M. Zwemer, The Law ojApostasy in Islam. Answering the Question why there are so Few Moslem Converts, and giving Examples of their Moral Cowri e and Martyrdom. (London, Edinburgh and New York) Although the book bears no formal publication date, the author's preface is dated 1924. See particularly p. 24: 'President C. F. Gates, of Robert College, Constantinople, states: 'The fear of deatli is certainly one cause for the fewness of converts from Islam to Christianity. Every Moslem knows that his life is in danger if he becomes a Christian. I have known a good many in-

Notes

20. 21. 22. 23.

24. 25.

219

stances of Moslems who would secretly assert themselves as Christians, but would make no open statement because of the danger attending it.'. Salt, Imperialism,Evangcli$m, p. 35. BBA Y.A RES 78/54, 7 Sevval 1313/22 March 1896. Grand Vizier's Chancery no. 2360. Salt 'Precarious Symbiosis', p. 55. See also above Chapter 3. BBA Irade Dahiliye. 99649, 19 Ceraaziyelahir 1309/20 January 1892. Vilayet of Syria Receivers Office no. 32 Vali Osman Nuri Pasa to Sublime Porte. Ibid. BBA (YEE) Kamil Pasa Evrakma Ek (KPE) 86-8/798,30 Kanun-i Ewel 1314/12 January 1898. T h e Vilayet of Aydm was of particular significance because it contained a very lively commercial community centred around the port city of Izmir.

26. BBA YEE KPE 86-11/1098,18 Haziran 1316/1 July 1900. Vali Kamil Pasa to Sublime Porte. 27. BBA Irade Dahiliye 99649,29 Recebl309/28 February 1892 Grand Vizier and Imperial A D C Cevad Pasa. 28. BBA Irade Dahiliye 100687 29, ZiIkadel309/25 June 1892. Yildiz Palace Imperial Secretariat no.8185. Imperial Secretary Siireyya Pas> 29. BBA Ayniyat Defteri no. 1313. p. 57. 28 Saban 1309/28 March 1892. Sublime Porte to the Ministry ofJustice and Religious Sects, the Ministry of the Interior, and the Foreign Ministry. 30. BBA Y.A RES 96/14 11 Cemaziyelahir 1316/27 October 1898. Report of the Special Commission on Education. (Enciimen-i Mahsus-u MaariF). 31. Ibid. 32. BBA Y.A RES 137/45,4 Cemaziyelevvel 1324/26 June 1906. Grand Vizier Ferid Pasa. Sublime Porte Grand Vizier s Chancery. Register number 1056 document number 4694, Including memo by Council of Ministers. T h e memo specified that repeated orders had been received from the Palace that measures be taken to upgrade the teaching in Islamic schools. 33. BBA Y.A H U S 257/153,29 Saban 1309/29 March 1892. Sublime Porte Receiver's Office. Grand Vizier Cevad Pasa. The Porte did however allow the Bible Society to carry out printing and distribution of the Old Testament in Albanian using Greek characters. See Y.A H U S 261/19, 3 Zilkade 1309/30 May 1892. 34. BBA Y.A H U S 257/153,9 Mart 1898/9 March 1898. Sublime Porte. Bureau of Translation for the Foreign Press, no. 308. 35. BBAY. Mtv. 57/65,25 Cemaziyelevve! 1309/27 December 1891. Maiyeti Askeri Komisyonu. 36. BBA Ayniyat Defteri no. 1420 p. 299,11 Rebiyiilahir 1299/2 March 1882. Sublime Porte to Ministry of Education.

220

The Well-Protected Domains

37. BBA Ayniyat Defteri no. 1422 p. 57. 5 Safer 1307/1 October 1889. Vilayet of Iskodra to Ministry of Education. 38. Guest, The Yeztdis, pp. 49,50. 39. BBA Irade Dahiliye 100258.10 Mayis 1308/23 May 1892. Cipher from Vali of Mosul, Kemali Pasa. 40. Salt, 'A Precarious Symbiosis', p. 55. 41. Guest, The Yezidis, p. 130. 42. BBA Irade Dahiliye 56.12 Eylul 1308/25 September 1892. Cipher Tel. from Commander of Reformatory Force Omer Vehbi Pasa to Sublime Porte. As it turned out on this particular occasion the French consul was instructed by his superiors in Istanbul not to become involved. See Guest, The Yezidis, p. 130. 43. For the French presence in Lebanon see Akarli, The Long Peace. 44. BBA Y E E K P E 86-1 /SO unsigned and undated letter in Arabic. 45. Ibid. Mention of the Franco-Prussian War (1870) and the Russo-Ottoman war of 1877-78 makes it possible to date this document as having been written around 1880. 46. BBA Y.A H U S 208/77,25 Rebiyulahir 1305/10 January 1888. Office of the Grand Vizier. Signed by Kami! Pasa. no. 565. 47. Ibid. 48. BBA Y.A H U S 208/79. Military Doctor Corporal Yusuf Zeki bin Mihail Hamoye. no date. 49. Ibid. 50. BBA Y.A RES 24/47., 10 Sevval 1301/3 August 1884, Minutes of the Council of State, no. 598.

51. Ibid. 52. BBA Irade Hususi 50. 21 Muharrem 1315/22 June 1897. Yildiz Palace Imperial Secretariat no. 735. Imperial Private Secretary Siireyya Pasa. 53. BBA Y.A RES 80/5, 4 Muharrem 1314/15 June 1896. Sublime Porte Receiver's Office, no. 55. 54. BBA K P E 86-15/1500,17 Zilhlcce 1319/11 March 1902. It is interesting that Karnil Pasa should refer to Christian children in these terms (evlad-t vatan), 55. BBA K P E 86-14/1336,18 Tesrin-i Evvel 1318/31 October 1902. Minister of Interior Memduh Pa$a to Vilayet of Aydin. The Porte had received information that although France was expelling Jesuits from its own soil, it was moving to arrange for their reception and protection in third countries. 56. BBA Y.A H U S 207/79,3 Safer 1305/1 October 1887. Report by Grand Vizier K a m i ! Pasa on the visit of the Papal legate in Istanbul to Thessaloniki where he used complimentary language regarding the Sultan; BBA 209/59 6 Kanun-u Sani 1888/19 January 1888 Sublime Porte, Foreign Ministry.

Notes

221

57. BBA Y.Mtv 112/12; 1 Kanun-u Ewel 1310/14 December 1894. Catholic Patriarchate to Yildiz Palace Imperial Secretariat. 58. BBA Irade Hususi 96, 6 Sevval 1315/28 Februarv 1898. Yildtz Palace Imperial Secretariat 12249; Irade Hususi 16 6 Zilkade 1315/29 March 1898 Yildiz Palace Imperial Secretariat 13509. 59. BBA Y.Mtv 54/65 29, Agustos 1307/11 September 1891. Armenian Patriarchate to the Ministry of Justice and Religious Sects. 60. O n the British occupation and the so-called 'temporary administration' see Nesirn Zia, Ktbrts'm Ingiltere'ye Gecisi ve Ada'da kurulan Ingiliz Idarcst. (Ankara 1975). O n foreign schools in general see, liknur Polat Haydaroglu, Osmanlt Imparatorlugu'nda Yabanct Ohullar (Ankara 1990) for a somewhat sketchy survey of the topic. 61. BBA Y.A R E S 2 5 / 9 , 8 Zilkade 1301/30 August 1884. Minister of Pious Foundations Kamil Pasa. The document has a certain poignant tone as Kamil Pasa was a native of Nicosia. 62. Ibid. It seems however, that his fears were unfounded, the British Administration did continue to finance and support Muslim schooling on the island. See Nasim Zia, Ktbrts'tn Ingiltere'ye Gegisi p. 93-4: 'The British Administration established at least one school in every village... [in 1881] There were 122 Muslim primary schools on the island.'. 63. BBA Y.A RES 24/21,28 Saban 1301/23 June 1884. Sublime Porte. Foreign Ministry. Translation of note received from the British Embassy. 64. BBA BEO 245757,6 Receb 1325/16 August 1907. Vilayet of Basra no. 44. Vali Abdurrahman Hasan Pa$a to Ministry of Interior. 65. BBA Irade Dahiliye 99013. 24 Cemaziyelahir 1309/26 January 1892. 66. BBA BEO Yemen Gelen. no. 366, entry no. 1307.22 Saban 1308/3 April 1891. 67. BBA Y.A H U S 292/51, 12 Ramazan 1311/19 March 1894. Sublime Porte. Grand Vizier's Office. Grand Vizier Cevad Pasa. 68. BBA Y.A H U S 307/7,16 Temmuz 1894/16 July 1894. Foreign Ministry. Communication no. 188 from Ottoman legation in Washington. For further information on Anglo-American co-operation see Salt, A Precarious Symbiosis. 69. Salt, Imperialism, Evangelism, p. 31: '[The American minister at Istanbul in rhe 1880s, S. S. Cox, gave this list of missionary organizations established in the Ottoman state: T h e American Board of Missions, T h e Presbyterian Board of Foreign Missions (Syria), The United Presbyterian Missionary Board (Egypt), T h e Board of Missions of the Methodist Episcopal Church of N e w York (Bulgaria), T h e Foreign Mission Board of the Reformed Presbyterian Church of New York (active among the Nusayris of northern Syria), the American Bible Society, the Trustees of Bible House Istanbul, the Trustees of the Syrian Protestant College in

222

The Well-Protected Domains

Beirut and of Robert College in Istanbul... Cities and towns 'occupied' by the missionaries amounted to 394.' 70. For the best survey to date of American Missionary activity in Turkey see, Frank Andrews Stone, Academies for Anatolia. ( New York, London 1984), For the only Turkish work on the subject which makes use of the American Board Archives, yet does not go very far beyond the official nationalist perspectives and does not use Ottoman archival sources, see: Uygur Kocabasoglu, Anadolu'daki Amerika (Istanbul 1989). The standard work on the topic to date has been Joseph L GrabiH's Protestant Diplomacy and the Near East. Missionary Influence on American Policy. (Minneapolis 1971) which does not use any Turkish sources but remains a fair discussion of the problem from the American perspective. 71. BBA Y.A RES 1/2 no date. Memorandum prepared by Foreign Mintstry at the order of the Yildtz Palace. According to the millet system each non-Muslim community was administered on a daily basis by their own religious and secular leaders, who were responsible to the Porte for their affairs. 72. Ibid. It is interesting that the Reform Edict should be brought up in this context, as it was on the basis of the same document that foreign powers often took the Ottoman government to task for not fulfilling its obligat i o n s . T h i s was p a r t of the tendency of the O t t o m a n s to use interpretations of international law to defend their interests. 73. BBA Y.A H U S 352/89,5 May 1896. Communication no. 240 from the Ottoman legation in Washington. Enclosing translation of Missionary Herald article. It is interesting to compare Ottoman archival data with Stone's Academies because often it is possible to find details on names and information that come up in Ottoman despatches. For instance, on the Bartlett family which is mentioned in the above article we learn that they were a missionary family based in Izmir who ran a kindergarten for Armenian children, and the evangelical centre in Kayseri developed into a 'network of churches And elementary schools. By 1909 forty-four schools with almost two thousand children in them were operating.' See Stone Academies, pp. 89-90. 74. BBA Y.A H U S 291/63, 26 Saban 1311/4 March 1894. Sublime Porte Receiver's Office, no. 3008. Grand Vizier Cevad Pasa. Stone also points out on several occasions that 'In most cases American Board personnel in Turkey were opposed to the Armenian revolutionaries,' and found their views 'irreligious impractical and dangerous'. See Academies pp. 122,191. 75. BBA Y.A H U S 282/120,15 Rebiyulewel 1311/26 September 1893. Foreign Ministry. Dispatch no. 182. 76. Ibid. 77. BBA Y.A H U S 282/120,15 Rebiyulahir 1311/27 October 1893. Sublime Porte Receiver's Office no. 1298.

Notes

223

78. BBA Y.A H U S 376/68. Consul General Munci Bey to Foreign Ministry. 8 Agustos 1313/21 August 1897. O n this see also: Selim Deringil, 'An Ottoman View of Missionary Activity in Hawaii/ Hawaiian journal of History (1993) pp. 119-25. 79. Ibid. 80. Ibid. 81. BBA Y.A H U S 352/18,1 Zilkade 1313/14 April 1896. Ottoman Embassy in Washington to Foreign Ministry. 82. O n the affair of Dr Knapp see Stone, Academies p. 122. 83. BBA Y.A H U S 352/18, Gurre-i Zilhicce 1313/14 May 1896. Foreign Minister Tevfik Pasa to Grand Vizier. 84. BBA Y.Mtv 197/16, 2 Saban 1317/6 December 1899. Foreign Ministry no. 332. 85. BBA Y.Mtv 19/77, 30 Tesrin-i Sani 1315/13 December 1899. Telegram from Vilayet of Erzurum to Foreign Ministry. 86. BBA Irade Husui 86. 25 Rebiyiikhir 1315/23 September 1897. Yildiz Palace Imperial Secretariat no. 4724. 87. BBAY.ARES.125/103,29Zilkadel312/24Mayl895.CouncilofState no.3294.

88. Ibid. 89. BBA Y.Mtv 53/108, 26 Zilhicce 1308/3 August 1891. Vali of Sivas Mehmed Memduh Pa§a to Sublime Porte. 90. John Joseph, Tfce Nestor'tans and their Muslim Neighbors (Princeton 1961} p. 123. 91. BBA Y.A H U S 365/39, 16 Kanun-u Ewe! 1312/29 December 1896. Despatch from Consulate of H o y and Selmas to Foreign Ministry. Compare Stone, Academies, p. 130: 'Evidence indicates that the American Board Missions in Turkey were always viewed by the Atmenian revolutionaries as palliatives that inhibited their objectives . . . ' . 92. 93. 94. 95. 96.

Stone, Academies, p. 127. Ibid., p. 140. Grabill, Protestant Diplomacy, p. 47. Ibid. p. 4 1 . BBA Y.Mtv 188/118,22 Zilkade 1316/4 April 1899. Although the document does not make any direct reference to the massacres, the vilayets listed are those where massacres did occur i.e. Bitlis, A l e p p o , Mamurettilaziz, Trabzon, Erzurum, Diyarbekir, and Sivas. T h e total number of orphans listed is 6386.

97. Stone, Academies, pp. 12-13. 98. Joseph, The Nestor'tans, p. 123. 99. BBA Y.A H U S 374/103, 27 Safer 1315/28 July 1897 Sublime Porte. Grand Vizier's Office 267. Grand Vizier Rifat Pasa. 100. Ibid. Enclosed cutting from Worcester Mass. Daily Spy, 6 December 1895.

224

The Well-Protected Domains

101. Ibid. Cutting from Daily Spy, 1 January 1896. 102. Ibid. Cutting from Boston Ideas no. date on clipping. Knowles also gave list of the decorations he already held. 103. Ibid. Detailed discussion of the issue in Cabinet minutes. 104. BBA Irade Hususi 86. 2 Rebiyiilahir 1323/6 June 1905. Yildiz Palace Imperial Secretariat.no. 1258. 105. BBA Y.A RES 122/145,26 Receb 1321/18 October 1903. Enclosure of detailed table of American educational establishments with licenses, giving location, level of education and date of establishment. 106. BBA Y.A RES 127/49,9 Cemaziyelahir 1322/21 August 1904. Minutes of meeting of Council of Ministers. 107. BBA Y.Mtv 294/22, 21 Zilhicce 1323/16 February 1906. Foreign Minister Tevfik Pasa to Yildiz Palace Imperial Secretariat. 108. Ramsay, Impressions of Turkey, p. 221. 109. Salt, Imperialism,'Evangelism,p. 32. 110. Ibid., p. 227. O f course the rest of the statement illustrates just how far off the mark Ramsay was on things Ottoman: 'An educated middle class is almost entirely absent from Oriental countries and Oriental society in which there is only a monarch and his slaves.'. 111. Karl K. Barbir, 'Memory, Heritage, History: Ottomans and Arabs,'. In Carl L. Brown, Impend! Legacy, p. 106. 112. Mardin, The Genesis of Young Ottoman Thought., p. 206. 113. See Pashalis Kitromilides, '"Imagined Communities" and the Origins of the National Question in the Balkans' European History Quarterly vol. 19 (1989) pp. 149-94. Particularly p. 183: 'By the turn of the twentieth century, a whole new mentality shaped by the values of nationalism crept gradually into the politics of the Ecumenical Patriarchate.'. 114. Mardin, Religion and Social Change on what Mardin calls the 'demonstration effect' see pp. 103-22. 115. Ibid., p. 51. 116. Stone, Academies, p. 142. 117. Hafiz Mehmed Sadik, Alem-i Islam'da Cihad-i Ekber (Istanbul 1342-39) p. 25. T h e writer is introduced as the Mutftu of Milas, a small town in south-western Turkey. Throughout the last quarter of the 19th century and well into the 20th century such pamphlets continued to appear in Turkish literary circles. 118. Rev. Dr J. Muhleisen-Arnold, The Society for Propagating the Gospel among the Moslems, In Connection with the Church of England; Its First Appeal on BehalfoflBO millions of Mohammedans. (London 1860) p. 4. 119. Ibid. 120. Samuel M. Zwemer, The Law of Apostasy, p. 16. 121. Tibawi, American Interests, p . 269. 122. Said, Culture and Imperialism, pp. 4 5 - 6 .

Notes

225

C h a p t e r Six 1. R. G. Latham M A M D (Late Fellow of King's College Cambridge) Russian and Turk, from a Geographical, Ethnological, and Historic Point of View (London 1878) p. 160. 2. Ahmed Cevdet Pasa, Maruzat, p. 4. 3. Mardin, The Genesis of Young Ottoman Thought, p. 388. 4. Said, Culture and Imperialism, p. 6. 5. BBA Yildiz Peralcen.de (Gazeteler). 6. BBA Y.A H U S 181/63 7 Nisan 1885/7, April 1885. Musurus Pasa to Sublime Porte Translation Bureau t e l no. 149, 7. BBA Y.A H U S 186/75,17 Rebiyulevvel 1303/24 December 1885. Sublime Porte Foreign Ministry. 8. Musurus Ghikis Bey, %'avenir de l'lslam' Questions Diplomatises et Coloniales. vol. X I (1901) pp. 595-7. 9. Henry Elliot, 'The Death of Abdul Aziz and of Turkish Reform' The nineteenth Century February 1888 pp. 276-96. 10. Ibid., p. 296. 11. BBA Y.A H U S 210/53; 21 Cemaziyelevvel 1305/4 February 1888 Sublime Porte Foreign Ministry n o . 1027; Y.A H U S 2 1 1 / 6 5 - A , 21 Cemaziyelahir 1305/5 March 1888. 12. BBA Y.A H U S 212/19,10 Mart 1888/10 March 1888 Ottoman Embassy London to Sublime Porte tel. no. 80. 13. Lusignan, The Twelve Years' Reign p. 4. 14. Ibid., p. 33. Of course it is impossible to describe the 'special tribunal' that tried Midhat Paja and his colleagues as an impartial court of'ordinary law', nor did the public get anywhere near it. O n the Yildiz Trials see: Ismail Hakla Uzun?arsili, Midhat Pasa vc Ytldtz Mahkemesi (Ankara 1967). 15. BBA Y.A H U S 222/71 2 Mart 1889/2 March 1889 Ottoman Embassy in London to Sublime Porte Foreign Ministry communication no. 50. 16. BBA YEE 14/1337/126/10 nd. (But judging from ehe context the text seems to be written in the early 1890s). Draft article by Selim Melhame and covering letter presenting the article to the Sultan. Selim Melhame was one of Abdulhamid's 'official Arabs', There is no proof that the article was actually published and Melhame was obviously out to flatter his master. Nonetheless, the document is interesting precisely because it gives a good idea of the image Abdulhamid wanted to project. 17. Lusignan, Tfoc Twelve Years' Reign, p. 198. 18. Feroze Yasamee, 'The Ottoman Empire and the European Great Powers/ S O A S P h D Dissertation (1984) pp. 56-60. 19. BBA Y.A H U S 219/75 2 Kanun-u Evvel 1888/15 October 1888. Cipher telegram from the Ottoman Embassy in Vienna to the Sublime

226

i'be Welt-Protected Domains Porte. O n Arminius Vambery see M . Kemal Oke, lngiliz Casusu Prof. Arminius Vambery'nin gizli Raporlarmda II Abdulbamid ve Donemi (Istanbul 1983). There is also a section in the Public Record Office, London, called the 'Vambery Letters' dealing with the material he sent to the Foreign Office from Istanbul.

20. Ibid. Letter from Vambery in Turkish to an unknown recipient, probably Munir Bey of the Vienna Embassy dated 3 Kanun-u Evvel 1888 / 1 6 October 1888. 21. BBA Y.A H U S 220/32,11 Rebiyulahir 1306/15 December 1888 Foreign Ministry to Grand Vezier. Foreign Minister Said Pasa. 22. BBA Y.A H U S 223/2,16 subat 1889/18 February 1889. Ottoman Embassy Vienna to Foreign Ministry communication no. 6 1 . 23. BBA Y.A H U S 226/42, 14 Mayis 1889/14 May 1889 Ottoman Embassy at St Petersburg to Sublime Porte dispatch no. 70. 24. BBA Y.Mtv 51/45.15 Zilkade 1308/22 June 1891 Ministry of Customs no. 52. 25. BBA Y.A H U S 303/87; 5 July 1894 Sublime Porte Foreign Ministry Telegram from Vienna Embassy no. 337; Sublime Porte Receiver's Office no. 254 Grand Vizier Cevad Pasa. 26. BBA Irade Hususi 34,5 Safer 1311/18 August 1893 Yildiz Palace Imperial Secretariat no. 836. 27. BBA Irade Hususi, 3 Gurre-i Sevval 1311/7 April 1894. Yildiz Palace Imperial Secretariat no. 6880. 28. BBA Y.Mtv 132/76,15 Tesrin-i Sam 1311/28 November 1895. Ottoman Ambassador Alexander Mavroyeni to Yildis; Palace Imperial Secretariat. 29. Said, Orientalism, pp. 49-73. 30. BBA YEE Kamil Pasa Evrakina Ek (KPE) 8 6 - 3 / 2 6 4 6 Ramazan 1307/ 26 April 1890. Yildiz Palace Imperial Secretariat. 31. BBA Y.A H U S 283/54,21 Tesrin-i Evvel 1896/3 November 1896. Sublime Porte. Foreign Ministry. 32. BBA Y.A H U S 284/74,7 Te ? rin-i Sani 1896/20 November 1896. Ottoman Embassy London to Sublime Porte. Dispatch no. 500. 33. BBA Y.A H U S 287/49, 14 Kanun-u Evvel 1893/27 December 1893. Ottoman Embassy Rome to Sublime Porte Foreign Ministry no. 413. 34. BBA Y.A H U S 237/50,19 July 1890 Ottoman Embassy London to Sublime Porte Foreign Ministry, confidential no. 182. 35. Ibid.The communication went on to state that Osman Bey was born to Dr Millingen's first wife, 'a Greek from Chios later divorced by the doctor because of evil behaviour', the 'Greek' lady went on to marry Kibrish Mehmed Pasa, a leading figure in mid century Istanbul. Osman was then adopted by Mehmed Pasa and brought up as a Muslim. T h e information of the London Embassy was only partly accurate. T h e 'Greek woman'

Notes

36. 37. 38. 39. 40.

41.

42. 43. 44. 45. 46.

227

whose name was M a d e Dejean and who later took the Muslim name of Melek H a n u m had a Greek grandmother, Armenian grandfather and French father. After the break up of her marriage to Dr Millingen, and Mehmed Pasa being forced to divorce her also for her involvement in various petty mid century intrigues in Istanbul, in 1866 she was forced to flee to Paris, where Osman became a Christian. She wrote her memoirs called, Thirty Years in the Harem or the Autobiography of Melek Hanum, Wife ofH. H. Kihrtzh Mebemet Pasha (New York 1872). This was evidently the sort of literature that the ambassador in London was referring to as 'vulgar and scandalous trash' that the lady was wont to publish. BBA Y.Mtv 100/26, 17 July 1894 Ottoman Legation at the Hague to Sublime Porte Foreign Ministry, no. 1780. Ibid., T h e outline of the sketch is in French. It is an interesting sample of particularly low grade Orientalism. Rambert, Notes et Impressions, pp. 175-6. BBA Y.A H U S 309/31,18 Rebiyulewel 1312/19 September 1894. Grand Vizier's Office no. 1077. Grand Vizier Ekrem Cevad to Yildiz Palace. BBA Y.A RES 124/80,17 Zilhicce 1321/5 March 1904 Grand Vizier Mehmed Ferid Pa ? a; Y.A RES 129/54,5 Zilkade 1322/11 January 1905. Sublime Porte. It appears that Abdiilhamid was aware of the implications of the rise ofJapan as a world power. Well before the Russo-Japanese war, in 1892, a Japanese Muslim by the name of Abdul Halim Noda Efendi was engaged to teach Japanese at the Military Academy see: BBA Y.Mtv 66/61,3 Safer 1310/27 August 1892. General Staff Receivers Office no. 146. BBA Y.A H U S 203/70, 19 June 1887. Ottoman Embassy London to Sublime Porte Foreign Ministry, O n the occupation of Egypt and preceding events see: Selim Deringil, ' T h e O t t o m a n Response to the Egyptian Crisis of 1881-82/ Middle Eastern Studies vol. 24 (1988) pp. 3~ 24. BBA Y.A H U S 203/68,27 Ramazan 1304/19 June 1887 Sublime Porte Receivers Office no. 4. BBA Y.A H U S 193/28, 30 June 1896 Ottoman Embassy London to Sublime Porte. L. Hirszowicz, T h e Sultan and the Khedive', pp. 287-311. BBA Y.A RES 86/104; 29 Zilhicce 1314/31 May 1897. Yildiz Palace Imperial Secretariat. BBA Y.A RES 57/52 Letter by Queen Victoria to Abdiilhamid II dated 25 January 1892. It is a very simple document edged with black, penned by a secretary and signed in the Royal hand, 'Your Imperial Majesty's good Sister, Victoria Regina.' It was quite common practice among royal houses to inform each other about familial changes.

228

The Well-Protected Domains

47. Ibid. Letter from Abdiilhamid to Queen Victoria addressed, 'To Her Imperial H i g h n e s s O u r Exalted Friend' (Dost-w Biilend-i Itibartmiz Hazretleri) evidently a draft, as it is only dated Receb 1309 (January 1892). 48. Ramberr, Notes et Impressions p. 100. Diary entry dated 5 September 1900. 49. BBA Y.A H U S 284/85 7 Tesrin-i Sani 1893/20 November 1893 Ottoman Embassy London to Sublime Porte Foreign Ministry. T h e paragraph s.t the beginning of this chapter is my fictional re-creation of this event. 50. Charles Williams, The Armenian Campaign, A Diary of the Campaign of 1877 in Armenia and Koordistan. (London 1878) p. ix, x. 51. Latham, Russian and Turk, p. 160. T h e reference to'flower of rhetoric'by 'influential authors' may well have been a reference to Gladstone's contemporary pamphlet, 'The Bulgarian Massacres', published in 1876. 52. Roger Adelson, Mark Sykes, Portrait of an Amateur. (London 1975) p. 110. O'Connor had been the British Ambassador to the Porte since 1898. This passage refers to 1905-1906. 53. Ibid., p. 98. T h e school referred to is obviously the Mekteb-i Asiret see above. 54. Ibid., pp. 6 4 - 5 . 55. Karl Blind, 'Young Turkey' Forthnighdy Review (London 1896) LXVI p. 840. 56. BBA Y.A H U S 191/123 23 May 1883, Ottoman Embassy London to Sublime Porte, Foreign Ministry no. 358. 57. BBA Y.A H U S 189/25 10 February 1886, Ottoman Embassy London to Foreign Ministry. 58. Lusignan, The Twelve Years' Reign, p. 147. 59. Constance Sutcliffe, 'Turkish Guilds,' Fortnightly Review vol. LXVI (1896) p. 828. 60. Fatma AHye, Nisi/an-i Islam (Istanbul 1891). 61. Mehmed tzzct, Yeni Afrika. (Istanbul 1308} p. 3. Izzex Bey takes pains to point out that he is writing this book on commission from the Palace. 62. Y.Mtv. Numerous entries in Catalogue no. 5. (Dreyfus meseksine dair). 63. India Office Library and Records. L / P & S / 3 / 2 2 6 , vol. 39, p. 787, Layard to Salisbury, Constantinople 15 April 1880. 64. L / P & S / 3 / 2 2 6 vol. 39 pp. 1315-20. Sir Louis Malet to Foreign Office and the Government of India. 65. L / P & S / 7 / v o l . 26, part 6 p. 1252. Government of India to H M Secretary of State for India. Simla, 28 September 1880. 66. Ibid., p. 1256.15 July 1880. Memorandum by Major P. D. Henderson on intrigues between Constantinople and Mahometans in India. 67. Side Emre, 'Political Imagery in the journal Servet-i Funun' unpublished Masters Thesis, Bogazici University, 1996.

Notes

229

Chapter S e v e n 1. Said, Orientalism, p. 204. 2. BBA Y.Mtv 66/66,10 Safer 1310.5 September 1892 Ottoman minister to Washington Alexander Mavroyeni to Yiidiz Palace Imperial Secretariat. The paragraph at the beginning of this chapter is my fictional recreation based on this document. 3. Ibid. T h e fact that Mavroyeni was a Greek defending the honour of Islam is an additional irony. 4. BBA Y.A H U S 285/66,21 Cemaziyeiewel 1311/1 December 1893Subtime Porte Receiver's Office no. 1821. Memorandum by Grand Vizier and Imperial A D C Cevad Pasa. 5. Ibid. 6. Carney E. S. Gavin, Sinasi Tekin and Gonul Alpay Tekin (eds) 'Imperial Self Portrait. The Ottoman Empire as Revealed in Sultan Abdul Hamid's Photographic Albums/ journal of Turkish Studies, vol. 12 1988. 7. Ibid., p. 47. 8. Ibid., p. 48. 9. Ibid., pp. 112,114,115. 10. BBA Y.Mtv 70/67,12 Rebiyiilevvel 1310/4 November 1892 Minister of War Riza Pasa to Yiidiz Palace. Given that the poses to be taken and the very size of the pictures is defined, it is almost certain that the photographs were designed for the albums. O n the Asiret School see Ch. III. 11. 12. 1314. 15. 16.

Gavin, Tekin and Tekin (eds). Imperial Self Portrait, pp. 162-4. Hobsbawm, Age of Empire, p. 123. Gavin, Tekin and Tekin (eds), Imperial SelfPortrai, pp. 193,194, 203, 205. Anderson, Imagined Communities, p. 27. Mitchell, Colonising Egypt, pp. 13 - 1 5 . BBA Y.A R E S 54/30,18 February 1891. Ottoman Embassy London m Foreign Ministry; Y.A RES 55/1 Sublime Porte Council of Ministers. The proposed Ottoman delegation was to consist of the Director of the Imperial Galatasaray Lycee, Ismail Bey, and the famous popular writer Ahmed M k h a d Efendi.

17. BBA Irade Hariciye 6,7 Cemaziyeiewel 1320/13 August 1902. 18. BBA Y.A R E S 62/27, Gurre-i Cemaziyelahir 1310/21 December 1892. Sublime Porte Special Council no. 1517. 19. BBA Y.A RES 106/71,22 Zilhicce 1317/24 April 1900. Sublime Porte Council of Ministers, no. 3541. T h e only Ottoman possession in Africa at this time was the vilayet of Trablusgarp in North Africa comprising the sancaks of Tripoli and Benghazi. Ostritch feathers and tortoise shell were major exports from this province, as well as esparto grass, a plant which grew on the Tripolitanian plains and was used to make high quality paper. O n this see, Michel Le Gall, 'Pashas Bedouins and Notables: T h e Ottoman Administration in Tripoli and Benghazi 1881-1902.' Princeton University P h D Dissertation (1986) pp. 9 5 - 7 , 1 5 3 - 4 .

230

The Well-Protected Domains

20. BBA Irade Hariciye 5,9 Rebiytilevvel 1320/17 June 1902. Sublime Porte Receiver's Office no. 500. O n e of the portraits of the albums in Imperial Self Portrait was the photograph of two blind boys from the Imperial School of the Blind. See Imperial Self Portrait, p. 177. 21. The idea that the 'whole world is watching' is derived from Carol Gluck's study of very similar phenomena in a Japanese context. See Giuck, Japan's Modern Myths, 22. There is a great deal of recent literature on the World's Fairs. See for example: Burton Benedict, The Anthropology of Worlds Fairs. San Francisco's Panama Pacific Exposition of 1915 (London & Berkeley 1983); Mitchell, Colonising Egypt; Paul Greenhalgh, Ephemeral Vistas. The Expositions Unwrselles, Great Exhibitions and World's Fairs 1851-1939 (Manchester 1988); Robert W. RydetL The Book of the Fairs (Chicago and London 1992). O n the Ottoman Empire's participation in the fairs the only work for a long time was Rifat Onsoy, 'Osmanh imparatorlugu'nun katildigi ilk Uluslararasi sergiler ve Sergi-i Umumi-i Osmani,' Belleten vol. 47 (1983) pp. 195-235. This situation has now been remedied by the excellent study of the subject by Celik, Displaying the. Orient. 23. Celik, Displaying the Orient; Rifat, Onsoy Osmanh tmparatorlugu ve Diinya Sergilen. 24. Benedict, Anthropology of World's Fairs, pp. 6-7. 25. BBA Y.Mtv 33/72,2 Haziran 1304/15 June 1888, Ottoman Embassy in Madrid to Foreign Ministry, no. 2379. 26. Y.A H U S 224/96, 27 Saban 1306/27 April 1889. Sublime Porte Foreign Ministry. In the event, the Ottomans participated quite successfully with the Ministry of Navy Bulletin winning a prize. See Y.Mtv. 63/33; 7 Zilkade 1309/4 June 1892. It was remarkable that the monarchies still felt this strongly about the revolution one century after the event, on this see: Eric Hobsbawm, Echoes of the Marseillaise (New Brunswick 1990) pp. 69-70. 27. BBA Y.A RES 58/33,25 Sevval 1309/24 May 1892Sublime Porte Council of Ministers. 28. Ibid. O n the matter of prestige going with location see Mitchell, et al. 29. Ibid. Contract between T h e Ottoman Ministry of Trade and Public Works and Ilyas Suharni Saadullah and Co. 30. Celik, Displaying the Orient, pp. 24-5 on the Egyptian exhibition at the 1889 Paris fair. 31. BBA Y.Mtv 77/114, 28 Nisan 1309/11 May 1893. Ottoman Ambassador Mavroyeni Bey to Yildiz Palace Imperial Secretariat. 32. Said, Orientalism, p. 21, emphasis in original. 33. BBA Irade Hususi 878/123 17 Muharrem 1310/12 August 1892 YildiK Palace Imperial Secretariat no. 678.

Notes

231

34. BBA Y.Mtv 75/202, 28 Subat 1308/13 March 1893. Ottoman Representative at the Chicago Pair, Hakla Bey, to Yildiz Palace. 35. BBA Y.Mtv 76/35, 13 Mart 1309/26 March 1893; Y.Mtv 79/163, 11 Muharrem 1311/26 July 1893 Minister of Trade and Public Works Hasan Tevfik Pasa to Sublime Porte. Fifty specially made ceremonial flags were also to be dispatched to Chicago. 36. BBA Irade Hususi 1746/59, 21 Receb 1310/11 February 1893 Sureyya Pasa, Imperial Private Secretary, Yildiz Palace Imperial Secretariat, no. 5746. 37. BBA Irade Hususi 1310/141, 20 Rebiyulevvel 1310/3 October 1982. Letter from Raci [Bey] subject of His Imperial Majesty from Acre. 38. Ibid. Programme for Ottoman Hippodrome. 39. Ibid. (Osmanli At MeydamTalimatnamesi). 40. Irade Hususi 1310/141, 25 Rebiyulevvel 1310/8 October 1892, Yildiz Palace Imperial Secretariat no. 3027. 4 1 . Celik, Displaying the Orient, p. 23; See also Gavin, Tekin and Tekin (eds) Imperial Self Portrait, pp. 198,199. 42. BBA Y.Mtv. 75/167,26 Subat 1308/11 March 1893 Hakki Bey, the Ottoman Commissioner at the Chicago Fair to Yildiz Palace Imperial Secretariat; Y.Mtv 76/36,12 Mart 1309/25 March 1893 Letter from Ali Bey Z a d e Nuri notable merchant of Gernlik. 43. BBA Y.Mtv 76/36,13 Mart 1309/26 March 1893 Hakki Bey, Ottoman Commissioner at the Chicago Fair to Yildiz Palace. 44. BBA Irade Nafta 9/5.S.1310; Ministry of Trade and Public Works, Receiver's Office no. 169. Letter from Hakki Bey dated 25 Receb 1310/13 February 1893. 45. Ibid., 3 $aban 1310/21 February 1893; Sublime Porte Council of Ministers. no. 2 1 1 5 . O n 17 O c t o b e r the Porte decided to send two English-speaking guides, Seraphim Efendi and Ekerin Efendi, see Irade Nafta 1/8.R.1311. 46. BBA Irade Nafia 3/23.ZA.1310. Letter from Hakki Bey to the Ministry of Trade and Public Works dated 15 May 1893. Letter no. 3. 47. Ibid., 4 Tesrin-i Evvet 1309/17 October 1893. Letter from Hakki Bey to Ministry of Trade and Public Works. 48. BBA Irade Nafia 3 23.ZA.1310/15 May 1893. Letter from Hakki Bey to Ministry of Trade and Public Works. Letter no. 2. 49. Ibid.The additional funds were approved on the 9 June 1893, see Irade Nafia 3 23.ZA.1310. Sublime Porte Receiver's Office no. 2877. 50. Rydell, All the World's a Fair, pp. 6 4 - 5 . 51. See: The Chicago Fair Illustrated. Proprietor S. K. Bistany. (Official gazette of Ottoman delegation) 1 June 1893, plan of the fair on p. 7. My thanks to Gultekm Yildiz for this reference.

232

The Well-Protected Domains

52. BBA Y.Mtv 103/15, 21 Safer 1312/25 August 1894 Report by the Imperial jeweller Istepan Mihran Dikran, Cubukcuyan Efendi. T h e details of the design of the display cases are in: Irade Nafia 3/23.ZA.1310. T h e style of jewellery and porcelain which can be seen in the Yikhz Palace Museum is actually quite garish, totally in keeping with Victorian taste. 53. BBA Y.Mtv 100/38,10 Temmuz 1894/10 July 1894. Embassy at Brussels to Sublime Porte Foreign Ministry. 54. BBA Y.Mtv 99/55, 7 Muharrem 1312/12 July 1894. Foreign Minister Said Pa§a to Ottoman Embassy in Brussels no. 21. Sulukule is a district well known even today for its gypsy dancers. 55. BBA Y.Mtv 100/38,15 Muharrem 1312/20 July 1894. Foreign Minister Mehmed Said Pasa, Sublime Porte Foreign Ministry no. 21. 56. BBA Y.Mtv 154/53, 8 Zilkade 1314/11 April 1897 Minister Of Trade M a h m u d Celaleddin Pasa, enclosing letter from C. b\. Sudre, the Ottoman honorary consul in Brussels, and Paul Kupelyan, representative of the firm in charge of the Brussels fair. 57. BBA Y.A H U S 354/41,23 Muharrem 1314/5 July 1896. Sublime Porte Foreign Ministry no. 1610. 58. BBA Y.A RES 72/47, 21 Rebiyulahir 1312/23 October 1894; Sublime Porte Council of Ministers no. 1117. 59. In the 1890s photography had become a gentleman amateur's hobby and some focused particularly on hirsute dervishes begging in the streets. This was something which was frowned upon by the Ottomans. O n photography of dervishes see Nancy Micklewright, 'Dervish Images in Photographs and Paintings' in R. Lifchez (ed.) The Dervish Lodge, pp 2 6 9 84. 60. BBA Y E E 14/1163/74/14; n.d. Letter from Ebuziyya Tevfik to Imperial secretariat. 61. Ibid. 62. BBA Y.A R E S 82/43, 18 Safer 1314/30 July 1896. Ministry of Trade and Public Works. Minister Mahmud Celaleddin Pasa to Grand Vizier no. 39. 63. BBA YEE Kamil Pasa Evrabna Ek 1087/1, 8 Haziran 1312/21 June 1896. Minister of Interior Memduh Pa§a. 64. BBA Y.A RES 83/99,11 Cemaziyelevvel 1314/18 November 1896. Ministry of Trade and Public Works no. 63. 65. BBA Y A R E S 85/21, 2 Ramazan 1314/4 February 1897. Minister of Trade and Public Works Mahmud Celaleddin Pa$a to Grand Vezier. no. 82. 66. BBA Y.A R E S 91/30,24 December 1897 Ottoman Ambassador Miinir Bey to His Excellency the Foreign Minister Tevfik Pasa. 67. BBA Y.Mtv 178/194; Sel'h-i Muharrem 1316/21 June 1898. Director of Istanbul Municipality to Sublime Porte.

Notes

233

68. BBA Y.A RES 94/68, Y.A RES 95/24,20 Rebiyulahir 1316/8 September 1898. Sublime Porte Council of Ministers no. 1198. 69. BBA Y.A R E S 95/24; 8 Cernaziyiilevvel 1316/25 September 1898. Ministry of Trade and Public Works no. 58. Usak and Gordes are regions in central Anatolia still famous for their prestige carpets. 70. BBA Y.Mtv 206/143; 29 Rebiyulahir 1318/25 October 1900 Imperial General Staff, Serasker Rtza Pasa to Grand Vmer. 71. <^.t\ik, Displaying the Orient, p. 89. 72. BBA Y.A RES 124/59; 17 Kanun-u Evvel 1903/30 December 1903. Sublime Porte Foreign Ministry. Letter from banker Konra to Ottoman Ambassador in Washington. I owe thanks to Mr Benjamin Fortna for this reference. 73. See above Chapter 1. 74. BBA Y.A RES 124/59; 15 Zilkade 1321/3 February 1904. Foreign Minister Tevfik Pasa to Grand Vizier. 75. BBA Y.A RES 134/10; 25 Saban 1323/15 October 1905. Council of State Financial Department no. 2813. 76. BBA Y.Mtv 2 5 8 / 5 1 ; 7 Muharrem 1322/25 March 1904. Ottoman Coraissioner in Bulgaria Ali Ferruh Bey to Sublime Porte, no. 5791/33. Ottoman Bulgaria had been split into two by the Berlin Congress of 1878. T h e two halves were called Westen and Eastern Rumelia. Western Rumelia became in effect independent in 1885, with nominal Ottoman suzereinty. 77. BBA Y.A RES 59/21; 1 Zilhicce 1309/27 June 1892. Sublime Porte Council of Ministers. Includes translation from the Bulgarian newspaper Plovdiv bemoaning the fact that 'although much of our agricultural produce goes to Istanbul the Turks are not present in our fair'. 78. Celik, Displaying the Orient, pp. 10-11. 79. Benedict, The Anthropology of World's Fairs, pp. 45-6. 'People as trophies' are one of Benedict's categories for the display of conquered peoples. 80. Ibid., p. 7. 81. Qelik, Displaying the Orient, pp. 3, 39. 82. Ibid., p. 48.

Chapter Eight 1. In Habermas's terms this was caused by: 'The structural dissimilarity between areas of administrative action and areas of cultural tradition, [which] constitutes, then, a systematic limit to attempts to compensate for legitimation deficits through conscious manipulation'. See Habermas, Legitimation Crisis, (Boston 1983) p. 71. 2. Shils, Tradition, p. 22.

234

The Well-Protected Domains

3. Mardin, The Genesis of Young Ottoman Thought, p. 171. Emphasis in original. 4. Ibid., p. 70. 5. Carter Findley, 'Factional Rivalry in Ottoman Istanbul: The Fall of Pertev Pasa, 1837,'. Raiyyet Rusumu. Essays Presented to Halit Inalak. Journal of Turkish Studies., vol. 10 (1986) pp. 130,131; See also Findley, Ottoman Civil Officialdom, pp. 70-80. 6. I. H . Uztmcarsih, Midhat Pasa ve Yildiz Mabkemesi. (Ankara 1967). 7. I. H . Uzuncarsih, Midhat Pasa ve Taif Mahkumlan (Ankara 1985) pp. 88, 95,106,112. 8. Uzuncarsih speaks of the 'martyrdom of Mithat Pasa' (Mithat Pasa'ntn sehadetf). 9. For a classic view of Mahmud II as the dynamic reformer comp; Niyazi Berkes, The Development of Secularism in Turkey (Montreal 1964). Also Stanford Shaw, History of the Ottoman Empire and Modern Turkey ( C a m ' bridgel986) vol. II p. 20: 'Strong measures followed to hunt out the remaining Janissaries.'. 10. BBA YEE 18/1858/93/39. N o date. 11. BBA YEE 31/1950 miikerrer/45/83,22 Zilkade 1299/6 October 1882. Memorandum by Said Paja. 12. Kantorowicz, The King's Two Bodies p. 19. 13. Ortayli, Imparatorlugun En Vzun Yuzydi (Istanbul 1983) p. 171. 14. BBA YEE 18/1858/93/39, 15. Ibid. 16. Ibid. 17. Ibid. 18. Ibid. It is interesting that Cevdet seems to have been ahead of later republican historians of the early Ottoman state, like Fuad Kopriilu, who sought somehow to 'prove' that Mihal was actually Turkish in the first place. See Kafadar, Between Two Worlds, pp. 1-30. 19. Ahmet Cevdet Pasa, Tarth-t Cevdet (2nd Edn) vol. Ill p. 160. My thanks to Dr Stephanos Yerasimos for kindly bringing this passage to my attention. 20. Said, Imperialism and Culture, p. 23. 21. The burning and sacking of the Chinese Emperor's Summer Palace by British troops in 1862 had immediate and widespread repercussions in the Ottoman world. See, Munif (Reis-i Sani-i Ticaret), 'Mukayese-i Ilm ve Cehl,' Mecmua-i Eiinun, no.l (1279) pp. 21, 22, 25,29-30: 'If the Chinese had not insisted in maintaining their old ways and trieir imperfect civilization would they have suffered such an insult at the hand of a few thousand foreigners?' As quoted in Hanioglu, Osmanlt Ittihad ve Terakki Cemiyeti, p. 19.

Notes

235

22. Ahmed Cevdet Pasa, Maruzat p. 4 1 . (transliterated and annotated by Yusuf HaUcoglu). 23. Roderic Davison, Reform In the Ottoman Empire. 1856-1876 (Princeton 1963). 24. Selim Dermgil, 'Les Ottomans et le partage de l'Afrique'. Studies on Ottoman Diplomatic History, V (1992) pp. 121-33. 25. Kantorowicz, The King's Two Bodies, pp. 36-9, 26. Yavuz Selim Karakisla, 'Exile Days of Abdulhamid II (1909-12) and the Confiscation of His Wealth'. M A Diss. Bogazici University (1991). Karakisla is quoting from Ali Cevad, Ikinci Mesrutiyet'in Ham ve 31 Mart Hadisesi (Ankara 1985) pp. 269-70. 27. O n the matter of the deposition and trial of Louis XVI, see Mona Ozouf, Xe Proces du roi' in Dictionnaire Critique de la Revolution Francaise, (Paris 1988) pp. 134-45. Particularly p. 136: 'II a repris son titre originel, il est homme'. 28. Ali Cevad Ikinci Mesrutiyet, ('Biz Meclis-i Mehusan tarafmdan geldik. Fetva-i Serifvar. Millet sens kai'etti. Ama hayatimz emindir). My thanks to Dr Edhem Eldem for drawing my attention to this wording. 29. Dustur, Tertib-i Sani, Numero 57-p. 166; "Sultan Abdulhamid Han-t Sanintn Hilafet ve Saltanat-i Osmaniyeden hkatiyla Sultan Mehmed Han-i Hamis Hazretterimn Asad ve Iclasi hakkinda Fetvayt Serife ve Meclis-i Umumi-i Milli Kararnamesi'. (Thefetwa and the Act of the National Parliament relating to the removal of Sultan Abdulhamid Khan from the caliphate and sultanate and the enthronement of Sultan Mehmed Khan the Fifth.) Signed by Seyhulislam Mehmed Ziyaadin Efendi. 30. See above Chapter 2. It is however not at all certain that the confiscated copies of the Qur'an were actually burnt, as stated in thtfetwa. 31. Dustur. 32. Tiirkiye Buyiik Millet Meclisi. Gizli Celsc Zabittart (Secret Minutes of the Closed Sessions of the Grand National Assembly) (Ankara 1985) "Bunlar kbyte edecck ve bu berifi ahp gtdecekler,' p. 1046. See also, Selim Deringil, 'Ottoman Origins of KemaHst Nationalism: Namik Kemal to Mustafa Kemal/. European History Quarterly, vol. 23 (1993) pp. 165-93. 33. Deringil, 'Ottoman Origins', pp. 165-93. 34. Halil tnalcik, 'The Caliphate and Ataturk's Inkilab,' Belleten CXLVI (1982)353-65. 35. Unsigned Declaration Entitled, 'Califat et Souverainete, Nationale' in Revue du Monde Musulman no. 59 (1925). This issue of the journal is very interesting as it is a special issue dealing with the 'Caliphate Question'. T h e document also appeared in Turkish as 'Hilafet ve Milii Hakimiyet', and in Arabic. T h e first article is the document cited above. Subsequent articles deal with how the matter was being received in Egypt. Others are

236

36.

37. 38. 39. 40. 41. 42. 43. 44. 45.

The Well-Protected Domains concerned with the implications for Central Asia. T h e section on India is written by the well known Indian radical, Barakatallah. Ibid., p. 7. Unlike the French version which was unsigned, the Turkish text is made up of articles signed by major figures in the early Kemalist movement. See, Hilafet ve Milti Hakimiyet. Hilafet ve Milli Hakimiyet mesaiti hakktnda mubtclif zevattn makalat ve tnutalaatindan murekkeb bir risakdir. (Pamphlet consisting of articles and views of various personages on the questions of the Caliphate and National Sovereignty) (Ankara tstihbarat Matbaasi 1339). Ibid., pp. 55,56,57; pp. 12,16. Hamid Enayat, Modern Islamic Political Thought, (London 1982) pp. 55-6. Youssef M. Choueiri, Arab History and the 'Nation State. (London 1989) p. 197. • Toledano, State and Society p. 22. Philip Khourv, Syria and the French Mandate. The Poll tics of Arab Nationalism. (Princeton 1987) p. 4. David Fromldn, A Peace to End All Peace. Creating the Modern Middle East, 1914-1922 (London 1990) p. 43-54. Mardin, Genesis of Young Ottoman Thought, p. 120. B B A Y E E 18/1858/93/39. Ilber Ortayh, Imparatorlugun En Vzun Yuzyih (Istanbul 1983) p. 10.

Bibliography

Official Records Bajibakanlik Arsivi (Prime Ministry Archives, Istanbul) Yildiz Esas Evraki Yildiz Sadaret Hususi Maruzat Yildiz Sadaret Resmi Maruzat Yildiz Miitenevvi Maruzat Yildiz Perakende Gazeteler Bab-i Ali Evrak Odusi Gelen/Giden Ayniyat Defterleri tradeler/Hariciye/Meclis-i Mahsus/Dahiliye Salname-i Devlet 1270,1271,1297,1298,1299,1302,1308,1324 Salname-i Vilayet-i Hudavendigar 1303/1885. Salname-i Vilayet-i Hicaz 1309. Diistur l.Tevtip. Tiirkiye Buy&k Millet Meclisi. Gizli Celse Zabitlan (Secret Minutes of the Closed Sessions of the Grand National Assembly) (Ankara 1985). India Office Library and Records. Political and Secret Home Correspondence (LP&S).

Books and Articles Abou-El-Haj, Rifa'at, Formation of the Modern State. The Ottoman Bmpre-Sixttenth to Eighteenth Centuries (Albany 1991). Abu Manneh, Butrus, 'Shaykh Ahmed Ziya'iiddin El-Gtimushanevi and the Ziya'i-Khalidi Suborder,' in Frederick De Jong (ed.) Shi'a Islam, Sects and Sufism (Utrecht 1992) pp. 104-17. —'Sultan Abdiilhamid II and Shaikh Abulhuda Ai-Sayyadi.' Middle Eastern Stud' its vol. 15 (1979).

237

238

The Well-Protected Domains

— ' T h e Sultan and the Bureaucracy; T h e and-Tanzimat Concepts of Grand Vizier Mahmud Nedim Paja,' International Journal of Middle East Studies 22 (1990) pp. 268-9. —'The Islamic Roots of the Gulhane Rescript/ Die Welt des Mams 34 (1994) pp. 173-203. Adelson, Roger, Mark Sykes, Portrait of an Amateur (London 1975). Ahmad, Feroz, The Young Turks: The Committee of Union and Progress in Turkish Politics 1908-1914 (Oxford 1969). Ahmed Cevdet Pasa, Tezakir (Ankara 1986). —Tarih-i Cevdet (Istanbul 1309 [1891] ) 2 vols. —Ma'zurat Transcribed and prepared for publication by Yusuf Hallacoglu (Istanbul 1980). —Meceife-i Ahkam-\ Adliye (Istanbul 1876). Akarh, Engin, The Problems ofExternal Pressures, Power Struggles, and Budgetary Deficits in Ottoman Politics Under Abdulhamid II (1876-1909): Origins and Solutions, Princeton University P h D Dissertation (1976). —The LongPeace. Ottoman Lebanon 1861-1920 (Berkeley, Los Angeles, London 1993). •—'Abdulhamid II Between East and West'. Paper presented at the colloquium: 'Education, Nation Building and Identity in the Period of Abdulhamid IF. Bad Homburg, Germany, July 12-15 1993. Akpmar, Alisan, Osmanli Devletinde Asiret Mektebi ( T h e Tribal School in the Ottoman State) (Istanbul 1997). Aksin, Sina, Jon TtirMer velttihat ve Terakki (The Young Turks and the Committee of Union and Progress) (Istanbul 1980). Ali, Cevad, Ikinei Me$rutiyet (The Second Constitutional Period), n.d. n.p. An Eastern Statesman, 'Contemporary Life and Thought in Turkey,' ContempO' rary Review 37 (1880) p. 343. Ana Britann'tca (Turkish edition of the Encyclopaedia Britannka) vol. 9. Anderson, Benedict, Imagined Communities (London 1991). Anderson, Perry, Lineages of the Absolutist State (London 1974). Anderson, Mathew, 'The Eastern Question 1774-1923 A Study in International Relations (London 1966). — The Ascendancy of Europe 1815-1914. Aspects of European History (London 1972). Annuaire Commercial de Constantinople 1893. Arai, Masami, 'An Imagined Nation: T h e Idea of the Ottoman Nation as a Key to Modern Ottoman History,' Orient vol. 27 (1991) pp. 1-20. Armstrong, j . . Nations Before Nationalism (Charlottsville 1982). Arnold, Thomas, The Caliphate (London 1965). Arseven, Cela! Esad, Sanat ve Siyaset Hattralanm (My Political and Artistic Memoirs) (Istanbul 1993). Aydm, Mehmet, Tezidiler ve Inane Esaslan' (The Yexidis and their Beliefs) Belleten (Official Journal of the Turkish Historical Society) vol. 52 (1988) pp. 33-45.

Bibliography

239

Barbir, Karl, 'Memory, Heritage, History: Ottomans and Arabs,' in Carl Brown (ed.), Imperial Legacy. The Ottoman Imprint on the Balkans and the Middle East (New York 1996). Bardin, Pierre, Algtriens et Tunisians dans I'Empire Ottoman de 1848 a 1914 (Paris 1979). Benedict, Burton, The Anthropology of World's Fairs. San Francisco's Panama Pacific Exposition of 1915 (London & Berkeley 1983). Berki, Ali Himmet, Aciklamah Mecelle (Istanbul 1990). Blanche, Lesley, The Sabres of Paradise (London 1960). Blind, Karl, 'Young Turkey,' Fortnightly Review L X V I (London 1896) pp. 829-40. Bojinev, Voyn, Hulgarskaya Prosveta v Makedonya i Odrinska Trakya 1878-1913 (Sofia 1982). Braden, jean Haythorne, 'The Eagle and the Crescent: American Interests in the Ottoman Empire, 1860-1870,' Ohio State University P h D Dissertation (1973). Brooks, Jeffrey, W h e n Russia Learned to Read (Princeton 1985). Brown, L. Carl, International Politics and the Middle Bast. Old Rules Dangerous Game (Princeton & Londonl984). Bryer, Anthony A. M., The Empire ofTrebizond and the Pontos. Variorum Reprints (London 1980), Burke, Edmund HI, Xe pan-islamisme et les origines du nationalisme en Afrique du nord, 1890-1918,' Cahiers de 1'Unite de d'Anthropologie Sociale et Culturelle. Universite d'Oran (Algeria) 1987,21-40. Cahun, Leon, Excursions sur les lords de I'Euphrate (Paris 1885). 'Califat et souverainete nationale,' in Revue du Monde Musulman no. 59 (1925). Unsigned. Cannadine, David, 'Splendour at Court: Royal Spectacle and Pageantry in Modern Britain, c.1820-1977,' in Rituals ofRoyalty, Power and Ceremonial in Traditional Societies (Cambridge 1987) pp. 206-43. Cannadine, David and Price, Simon (eds), Rituals of Royalty. Power and Ceremonial in Traditional Societies (Cambridge 1987). Carney, Gavin E. S-, SJnasi Tekin and Goniil Alpay Tekin (eds), '"Imperial Self Portrait", The Ottoman Empire as Revealed in Sultan Abdul Hamid's Photographic Albums,' Journal of Turkish Studies 12 (1988). Carmichael, Joel, The Shaping of the Arabs (New York 1967). Celik, Zeynep, Displaying the Orient. Architecture of Islam at Nineteenth Century World's Fairs (Berkeley, Los Angeles, Oxford 1992). —The Remaking of Istanbul. Portrait of an Ottoman City in the Nineteenth Century (Berkeley, Los Angeles, Boston 1993). Cherniavsky, Michael, Tsar and People. Studies in Russian Myths (New Haven and London 1961). Choueiri, Youssef M., Arab History and the Nation State (London 1989).

240

The Well-Protected Domains

Clogg, Richard, 'The Greeks and their Past/ in Dennis Deletant and Harry Hanak (eds). Historians as Nation Builders (London 1988) pp. 22-36. Cohn, Bernard, 'Representing Authority in Victorian India,' in E. Hobsbawm and T. Ranger (eds), The Invention of Tradition (Cambridge 1988) pp. 1 6 5 209. Couitellemont, Gervais, Man voyage a la Mecaue (Paris 1896). Davis, Owen, 'Those Dear Turks. A Lecture at the Congregational Church, Lee, on 1st November 1876,' Papas on the Eastern Question (London 1876). Davison, Roderic, Reform In the Ottoman Empire 1856-1876 (Princeton 1963). Deak, Istvan, 'The Habsburg Monarchy: The Strengths and Weaknesses of a Complex Patrimony,' Columbia University, Monarchies Symposium October 26-27 1990, unpublished paper. Deringih Selim, 'Legitimacy Structures in the Ottoman Empire: Abdulhamid II 1876-1909/ International journal of Middle Bast Studies, 23 (1991) pp. 345-59. —'The Struggle Against Shi'ism in Hamidian Iraq/ Die Welt des Islctms, 30 (1990) pp. 45-62. —'The Invention of Tradition as Public Image in the Late Ottoman Empire, 1808 to 1908/ Comparative Studies in Society and History, 35 (1993) pp. 1-27. —'The Ottoman Response to the Egyptian Crisis of 1881-82/ Middle Eastern Studies, 24 (1988) pp. 3-24. —'Les Ottomans et le Partage de 1'Afrique/ Studies on Ottoman Diplomatic History, V (1992) pp. 121-33. —'The Ottoman Origins of Kemalist Nationalism: Narmk Kemal to Mustafa Kemal/ European History Quarterly, 23 (1993) pp. 165-93. —'Ottoman to Turk, Minority-Majority Relations in the late Ottoman Empire', in D m C. Gladney (ed.) Making Majorities: Constituting the Nation in Japan, China, Korea, Malaysia, Fiji, Turkey and the US (forthcoming, Stanford University Press). —'An Ottoman View of Missionary Activity in Hawaii/ Hawaian journal of His* tory, 27 (1993) pp. 119-25. Duguid, Stephen, 'The Politics of Unity: Hamidian Policy in Eastern Anatolia,' Middle Eastern Studies 9 (1973) pp. 130-55. Dumont, Paul and Georgeon Francois (eds), Vilks Ottomanes a la fin de {'empire (Paris 1992). Elliot, Charles, Turkey in Europe (London 1965). Elliot, Henry, 'The Death of Abdul Aziz and of Turkish Reform' in The Nineteenth Century (1888), pp. 276-96. Elliot, John 'Power and Propaganda in the Spain of Phillip IV,' in Sean Wilentz (ed.), Rites ofPower. Symbolism, Ritual and Politics Since the Middle Ages (Philadelphia 1985)p. 151 Emre, Side, 'Political Imagery in the Journal Servet-i Funun/ unpublished Masters Thesis, Bogazici University (Istanbul 1996). Enayat, Hamid, Modern Islamic Political Thought (London, Austin 1982). Ersoy, Mehmet Akif, Safahat (Istanbul 1990).

Bibliography

241

Esenbel, Selcuk, 'A fin de siecle Japanese Romantic in Istanbul: The Life of Yamado Torajiro and bis Toruko Gakan,' SOAS Bulletin, 59, part 2 (1996) pp. 237-52. 'Istanbul 'de bir japon,' (A Japanese in Istanbul) Istanbul Dergisi, 9, pp. 36-42. Faroqhi, Suraiya, Pilgrims and Sultans. The Hajj under the Ottomans (London 1994). Fatma AHye, Nisvan-i Islam (Muslim Women) (Istanbul 1891). Fawaz, Leila, An Occasion jor War. Civil Conjlict in Lebanon and Damascus in 1860 (London 1994). Findley, Carter, Ottoman Civil Officialdom. A Social History (Princeton 1989). —'Factional Rivalry in Ottoman Istanbul: T h e Fall of Pertev Pasa, 1837/ in Raiyyet Rusumu. Essays Presented to Halil Inalcik. Journal oj Turkish Studies 10(1986)pp.l20~35. —'The Advent of Ideology in the Islamic Middle East,' Studia Islamica LV (1982). -—Bureaucratic Reform in the Ottoman Empire. The Sublime Porte 1789-1922 (Princeton 1980). Finkel, Caroline, The Administration ojWarjare: The Ottoman Military Campaigns in Hungary 1593-1606 (Vienna 1988). —'French Merceneries in the Habsburg-Ottoman war of 1593-1606,' SOAS Bulletin, LV, part 3 (1992) pp. 452-71. Finn, James, Stirring Times (London 1878). Fleischer, Cornell, Bureaucrat and Intellectual in the Ottoman Empire; The Historian Mustafa 'li (1541-1600) (Princeton 1986). —'Royal Authority, Dynastic Cyclism, and "Ibn-Khaldunism" in Sixteenth-Century Ottoman Letters,' Journal oj Asian and African Studies X V I I I 3-4 (1983) pp. 198-219. Fletcher, Joseph, 'Turco Mongolian Monarchic Tradition in the Ottoman Empire,' Harvard Ukrainian Studies III (1979-80) Euchansterion: Essays presented to Omeljan Pritsak on his Sixtieth Birthday by his Colleagues and Students pp. 246-52. Fromkin, David, A Peace to End all Peace. Creating the Modern Middle East 19141922 (London 1990). Geeitz, Clifford, Negara. The Theatre State in Nineteenth Century Bali (Princeton 1980). Gellner, Ernest, Nations and Nationalism (Oxfordl983). Gluck, Carol, Japan's Modern Myths (Princeton 1985). Gooch, G. P. and Temperley, Harold (eds), British Documents on the Origins ojthc War 1898-1914 (London 1938). Goodwin, Godfrey, History of Ottoman Architecture (London 1971). Gokalp, Altan, Tetes Rouges et Bouches Noires. Vne confrerie tribale de I'ouest Anatolien (Paris 1980). Giinduz, Irfan, Osmanltlarda Devlet-Tekke Muhasebetleri (The Relationship of the Stare and Sufi Lodges in the Ottoman Empire) {Istanbul 1989) pp. 150-1.

242

TJ?e Well-Protected Domains

Grabill, Joseph, L., Protestant Diplomacy and the Near East. Missionary Influence on American Policy (Minneapolis 1971). Greenhalgh, Paul, Ephemeral Vistas. The Expositions Univenellts, Great Exhibitions and World's Fairs 1851-1939 (Manchester 1988). Guest, John, The Yezidis. A Study in Survival (London SC New York 1987). Haarman, Ulrich, 'Ideology and Alterky: The Arab Image of the Turk from the Abbasids to Modern Egypt,' International Journal ofMiddle East Studies 20 (1988) pp. 175-96. Habermas, Jurgen, Legitimation Crisis (Boston 1983). Hafiz, Mehmed Sadik, AlemA hlam'da Cihad-t Ekber (The Great Holy War in the Islamic World) (Istanbul 1342-1339) p. 25. Hanioglu, Siikrii, The Young Turks in Opposition (Oxford 1995). —Osmanli tttihad ve Terakki Cemiyeti vejon Turkluk 1889-1902 (The Young Turks and the Committee of Union and Progress) (Istanbul 1985). —'Osmanli Aydmmda Degisme ve "Bilim",' (Change, "Science", and the Ottoman Intellectual), Toplum ve Bilim 27 (1984) pp. 183-92. —Osmanli ittihad ve Terakki Cemiyeti vejon Turkluk (The Young Turks and the Committee of Union and Progress)(Istanbul 1983). Haskell, Thomas L, 'Objectivity is not Neutrality,' History and Theory vol. 29 (1990} pp. 129-57. Haslip, Joan, The Sultan. The Life of Abdul Hamid 11 (New York 1958). Hasluck, F. W., Christianity and Islam under the Sultans (Oxford 1929) 2 volumes. Haydaroglu, Ilknur Polat, Osmanh Imparatorlugu'nda Yabana Okullar (Foreign Schools in the Ottoman Empire) (Ankara 1990). Herbert, William von, The Defence of Plevna 1877; written by one who took part in it (Ankara 1990). Hilafet ve Milli Hakimiyet, Hilafet ve Milli Hakimiyet mesaili hakhnda tnuhtelif zevatin makalat ve miitalaattndan miirekkeb bir risaledir (Pamphlet consisting of articles and views of various personages on the questions of the Caliphate and National Sovereignty) (Ankara Istihbarat Matbaast 1339). Hirszowicz, L. 'The Sultan and the Khedive 1892-1908/ Middle Eastern Studies vol. 8 (1977) pp. 287-311. Hobsbawm, Eric, Nations and Nationalism (Cambridge 1990). —Echoes of the Marseillaise (New Brunswick 1990). —The Age of Empire (London 1987). Hobsbawm, Eric and Ranger, Terence (eds), The Invention of Tradition (Cambridge 1983). Hourani, Albert, A History of the Arab Peoples (London 1991). — ' H o w Should We Write the History of the Middle East? International Journal of Middle East Studies 23 (1991) pp. 125-36. Hurewitz, j . C , 'Ottoman Diplomacy and the European State System' Middle East Journal vol. 15 (1961) pp. 141-52. —The Middle East and Africa in World Politics (New Haven 1975).

Bibliography

243

Hurgronje, Snouck, Mecca in the Latter Part of the 19th Century (London 1931). Inalcik, Halil, and Quataert, Donald (eds). An Economic and Social History of the Ottoman Empire 1300-1914 (Cambridge 1994). Inalcik, Halil, T h e Ottoman Scate and its Ideology' in Halil Inalcik and Cemal Kafadar (eds), Conference on Siileyman the Magnificent, Princeton Occasional Papers (1992) pp. 49-72. — T h e Caliphate and Ataturk's Inkifab' Belleten, CXLVI (1982) pp. 353-65. —The Ottoman Empire in the Classical Age: 1300-1600 (London 1973). Ipsirli, Mehmec, 'Cuma Seiamhgi. Halk Hiikiimdar Miinasebetleri Acismdan Onemi' (The Institution of Friday Prayer in the Ottoman State. Its Importance for the Relationship of the Ruler and the People) in Prof. Bekir Kutiikoglu'na Armagan (Istanbul 1991) pp. 459-71. Islamogki-inan, Huri, 'Mukayeseli Tarih Yazim Icin bir Oneri: Hukuk, Miilkiyet, Mesruiyet' (An Agenda for Comparative History: Law, Property and Legitimacy) Toplum ve Bilim 62 (1993) pp. 19-34. —'Koyliiler, Ticarilesme Hareketi ve Devlet Guciiniin Mesrulasmasi' (Peasants, Commercialization and the Legitimation of State Power) Toplum ve Bilim 4 3 / 44 (1988-89) pp. 7 - 3 1 . Janin, R. 'Musulmans malgre eux, Les Stavriotes'. Echos D'Orient XV (1912) pp. 495-505. Joseph, Isya, Devil Worship. The Sacred Books and Traditions of the Yezid'ts (London 1919) reprinted by the Health Research Institute (California 1972). Joseph, John, The Nestorians and their Muslim Neighbors (Princeton 1961). Kafadar, Cemal, Between Two Worlds. T V Construction of the Ottoman State (Berkeley 1995). Kantarowicz, E. H., The King's Two Bodies (Princeton 1957). Karaca, Ali, Dogu Anadolu hlahatt ve Ahmet Sakir Pasa 1838-1899 (Ahmet $akir Pasa and Reform in Eastern Anatolia) (Istanbul 1993) pp. 173-203. Karakisla, Yavuz Selim, 'Exile Days of Abdiilhamid II (1909-1912) and the Confiscation of His Wealth'. MA Diss. Bogazici University (1991). Kartekin, Enver, Kamazanogullan Tarihi (History of the Ramazanoglu Dynasty) (Istanbul 1979). Kayah, Hasan, 'Arabs and Young Turks: Turkish Arab Relations in the Second Constitutional Period 1908-1918,' Harvard University P h D thesis (1988). Kedourie, Elie, 'The Surrender of Medina, January 1919,' Middle Eastern Studies 13 (1977) pp. 124-43Kent, Marian (ed.), The Great Powers and the End of the Ottoman Empire (London 1984). Kevonian, Armenouhie, Les Noces Noires de Gulizar (Paris 1993 ). Khoury, Philip S., Syria and the French Mandate: The Politics of Arab Nationalism (London 1987). Kiaman, Naci Kajif, Medine Mudafaast. Hicaz Uzden nastl aynldi? (The Defence of Medina or How we Lost the Hijaz) (Istanbul 1971).

244

The Well-Protected Domains

Kisakiirek, Necip Fazil, Ulu Hakan II. Abdulhamid Han (Istanbul 1981). Kitromilides, Pashalis, "Imagined Communities" and the Origins of the National Question in the Balkans/ European History Quarterly 19 (1989) pp. 149-94. Kocabasoglu, Uygur, Anadolu'daki Atnerika (America in Anatolia) (Istanbul 1989). Kodaman, Bayram, Abdulhamid Donemi Egitim Sistemi (Educational Policies during the Reign of Abdulhamid) (Istanbul 1990). —Sultan Abdulhamid in Dogu Anadolu Politikast ( T h e Policies of Sultan Abdulhamid in Eastern Anatolia) (Istanbul 1983). Konyali, Ibrahim Hakki, Sogui'de Ertugrul Gazi Tiirbesi ve tbtifali (The Commemoration Ceremonies at Ertugrul Gazi's Shrine in Sogiit) (Istanbul 1959). Kuran, Aptullah, 'The Evolution of the Sultan's Pavillion in Ottoman Imperial Mosques,' Islamic Art I V (1990-1991). Lambton, Ann K. S., State and Government in Medieval Islam (London 1985). Landau, Jacob, 'Nishan', Encyclopaedia of Islam. New Edition (Leiden 1993) pp. 57-9. —The Hijaz Railroad. Latham, R. G., MA, M D , Russian and Turk, from a Geographical, Ethnological and Historic Point of View (London 1878). Le Gall, Michel, 'Pashas Bedouins and Notables: T h e Ottoman Administration in Tripoli and Benghazi 1881-1902/ Princeton University P h D Dissertation (1986). Leitsch, Walter 'East Europeans Studying History in Vienna' in Dennis Deletant and Harry Hanak (eds), Historians as Nation Builders (London 1988). Lewis, Bernard, The Political Language of Islam (Chicago 1988). —The Emergence of Modern Turkey (London 1973). Lifchez, Raymond The Dervish Lodge (Berkeley 1992). Luke, Harry Charles (Sometime Assistant Governor of Jerusalem) The Yezidis or Devil Worshippers of Mosul (Bombay 1925). —Mosul and its Minorities (London 1925). Lusignan, Princess Annie de, Tfce Twelve Years' Reign ofH.I.M. Abdul Hamid II, Sultan of Turkey (London 1889). Lynch, H . E B., Armenia, Travels and Studies II: The Turkish Provinces (London 1901). Mardin, Serif, Religion and Social Change in Modern Turkey. The Case ofBediiizzaman SaidNursi (New York 1989). —'Center-Periphery Relations, A Key to Turkish Politics?' in Daedalus (1973) pp. 169-90. —'Some Notes on an Early Phase in the Modernization of Communications in Turkey/ Comparative Studies in Society and History 3 (1960-61) pp. 257-71. —Tfce Genesis of Young Ottoman Thought (Princeton 1962). —Jon Tiirklerin SiyasiFikirleri 189S-1908 (The Political Thought of Young Turks) (Istanbul 1983).

Bibliography

245

Mehmed, Izzet, Yeni Afrika (The New Africa) (Istanbul 1308). Mehmed Sadik, Alem-i Islam'da Cihad-i Ekber (The Great Holy War in the Islamic World) (Istanbul 1904). —Milash Durmuszade Haftz Mehmed Sttdtk, Milas'da Aga Camii Miidcrris vt Cami-i Kebir Vaizi (Preacher Mehmed Sadik from Milas, Preacher in the Great Mosque and Teacher at Aga Mosque School) —Muvaiz-i Diniye (Religious Sermons) Izmir 1328 (1910). Melek Hanum, Thirty Years in the Harem or tht Autobiography of Mekk Hanum, WifcafHBKtbrtzU Mehemet Pasha (New York 1872). Meric, Omit, Cevdet Pasa'mn Cemiyet ve Millet Gorusu (The Views of Cevdet Pasa on Society and the State) (Istanbul 1979). Messick, Brinkley, The Calligraphic State, Textual Domination and History in a Muslim Society (Berkley, Los Angeles, Oxford 1993). Micklewright, Nancy, 'Dervish Images in Photographs and Paintings' in Raymond Lifchez (ed.) Tfje Dervish Lodge, pp. 269-84. Mintzuri, Hagob, Istanbul Amlan 1897-1940 (Istanbul Memoirs) (Istanbul 1993). Mitchell, Timothy, Colonising Egypt (Berkeley 1991). Mouradian, Claire, Armenie, que saisje (Paris 1995). Muftiioglu, Mustafa, Abdiilhamid. Ktztl Sultanmi? (Abdiilhamid, The Red Sultan?) (Istanbul 1985). Muhleisen-Arnold, Dr J., The Society for Propagating the Gospel Among the Moslems, In Connection with the Church ofEngland; Its First Appeal on Behalfof180 millions of Mohammedans (London 1860). Musurus, Ghikis Bey, 'L'avenir de I'lslam,' Questions Diplomatiaues et Coloniaks XI (1901) pp. 595-7. Necipoglu, Giilru, Architecture Ceremonial and Power: Tfce Topkapt Palace in the Fifteenth and Sixteenth Centuries (Cambridge Mass. & London 1991). — T h e Ottoman Hagia Sophia,' paper delivered at Princeton University 'The Structure of Hagia Sophia' symposium, 19 May 1990. Cited with permission of the author. —'Dynastic Imprints on the Cityscape. T h e Collective Discourse of Ottoman Imperial Mausoleums in Istanbul' unpublished paper p. 14. Cited with permission of the author. Nesim, Ali, Kibrish Turklerin Kimligt ( T h e Identity of Cypriot Turks) (Nicosia 1990). Niyazi, Berkes, The Development of Secularism in Turkey (Montreal 1964), Ocak, Ahmet Yasar, 'Abdiiihamid Donemi Islamcihgmm Tarihi Arka Plani: Kiasik Donem Osmanh Islami'na Genel Bir Bakis' (A General View on the Background to Islamism in the Hamidian Period) Sultan II Abdiilhamid vt Devri Semineri. Istanbul tJniverskesi Edebiyat Fakultest, Bildiriler (Istanbul 1994) pp. 10725. —'Kanuni Sultan Siileyman devrinde bir Osrnanh Heretigi: seyh Muhyiddin Karamani' (An Ottoman Heretic at the Time of Siileyman the Magnificent:

246

The Well-Protected Domains

Shaikh Muhyiddin Karamani), Prof, Bekir Ktitukoglu' na Armagan (Istanbul 1991) pp. 473-84. —'Les reactions socio-religieuses contre I'ideologie officielle ottomane et la question de zendeqa ve ilhad (Heresie et Atheisme au XVTeme siecle) Turcica 2 1 - 2 (1991) pp. 71-82. Ochsenwald, William, Religion Society and the State in Arabia. The Hijaz under Ottoman Control 1840-1908 (Columbus 1984). — B e Hijaz Railroad (Charlottesville Va. 1980). Onsoy, Rifat, 'Osmanli Irnparatorlugun'un Katildigi ilk Uluslararasi sergiler ve Sergi'i Umumi-i Osmani,' (The First World Fairs: Ottoman Participation and the Ottoman General Fair) Belleten 47 (1983) pp.195-235. Ortayh, liber, I I . Abdulhamid Devrinde Tasra Burokrasisinde Gayrimuslimler' in Sultan II Abdulhamid ve Devri Semineri, Istanbul Univemtesi Edebiyat Fakultesi (Istanbul 1994). —Impamtorlugun En Uzun Yuzyili (The Longest Century of the Empire) (Istanbul 1983). Tanzimattan Sonra Mahalli tdareler (1840-1878) (Ankara 1974). Oke, M. Kemat, Ingiliz Casusu Prof Arminius Vambery'nin gizli Raporlarinda II Abdulhamid ve Donetni (Abdulhamid II and his Time in the Reports of the British Spy Arminius Vambery) (Istanbul 1983). O'Rourke, P. j . , 'Give War a Chance,' Atlantic Monthly (New York 1992). Owen, Roger, Tfoe Middle East in the World Economy 1800-1914 (London 1987). Ozouf, Mona, 'Le Proces dur Roi,' in Dktionnaire Critique de la Revolution Francaise (Paris 1988) pp. 134-45. Pamuk, Sevket, The Ottoman Empire and World Capitalism (Cambridge 1987). Pandey, Gyanendra, 'Encounters and Calamities: The History of the North Indian Qasba in the Nineteenth Century' in Ranajit Guha and Gayatri Chakravorty Spivak (eds) Selected Subaltern Studies (Oxford 1988) pp. 106-7. Presland, John, Deeds Bey: A Study of Sir Wyndham Deedes 1883-1923 (London 1942). Politis, Kozmas, Yitik Kentin Kirk Ytli (Forty Years of the Lost City), trans. Osman Bleda (Istanbul 1992). Rambert, Louis, Notes et impressions de Turquie. VEmpire Ottoman sous Abdul-Hamid (Geneva and Paris n.d). Ramsay, R. M., Impressions of Turkey During Twelve Years' Wanderings (London 1897). Ramsay, William Mitchell, 'The Intermixture of Races in Asia Minor' in Proceedings of the British Academy 1915-1916, pp. 359-422. Redhouse, James W , A Vindication of the Ottoman Sultan's Title of'Caliph' (London 1877). Rogan, Eugene L., 'Asiret Mektebi: Abdulhamid II's School for Tribes (18921907),' International journal of Middle East Studies vol. 28 (1996) pp. 83-107. —'The al-Karak Revolt of 1910: Ottoman Order at Odds with Local Order.' Paper presented in the panel, 'The New Order and Local Order: Continuity

Bibliography

247

and Crisis in Everyday Life'. M E S A meeting Toronto November 16 1989 (eked with the permission of the author). Rydell, Robert W, The Book of the Fairs (Chicago and London 1992). Suleyman Hiisnii Pasa, Tarih-i Alem (World History) (Istanbul 1872). Said, AIL Saray Hatiralaru Sultan Abdulhamid'in Hayatt (A Memoir of the Palace. The Life of Abdulhamid II) (Istanbul 1993), Said, Edward, Culture and Imperialism (London, Melbourne, Sydney 1994). —Orientalism (Basingstoke 1985). Sakaoglu, Necdet, 'Gecmis Zaman Olurki,' Skyline (March 1997) p. 112. Salmone, Anthony H., 'The Real Rulers of Turkey,' The Nineteenth Century 37 (May 1895) pp. 719-33. Salt, Jeremy, Imperialism, Evangelism and the Ottoman Armenians 1878-1896 (London 1993). —'A Precarious Symbiosis: Ottoman Christians and Foreign Missionaries in the Nineteenth Century,' International Journal of Turkish Studies, 3 (Winter 198586) p. 56. Salzmann, Ariel, 'An Ancien Regime Revisited: "Privatization" and Political Economy in the Eighteenth-Century Ottoman Empire,' Politics and Society vol. 21 (1993) pp. 393-423. Scholch, Alexander, Egypt for the Egyptians: The Sociopolitical Crisis in Egypt 18781882 (London 1981). Seagrave, Sterling, Dragon Lady. The Life and Legend of the Last Empress of China (London 1992). Seton Watson, Hugh, 'On Trying to be a Historian of Eastern Europe,' in Harry Hanak (ed.). Historians as Nation Builders (London 1988). Shaw, Stanford, The Jews of the Ottoman Empire and the Turkish Republic (London 1991). —History of the Ottoman Empire and Modern Turkey (Cambridge 1986). Shils, Edward, Tradition (Chicago 1980). Sirma, Ihsan Sureyya, 'Abdulhamid' in Islam Birligi Siyaseti (Abdulhamid and the Policy of Pan-Islamism) (Istanbul 1983). ' Smith, Anthony D„ The Ethnic Origins of Nations (Oxford 1986). Solak, Seyit, lS71'den Guniimiize Ktbrts Turk Yonetimlefi (Turkish Administrations m Cyprus from 1571 to the Present) (Nicosia 1989). Spuler, Bertold, Die Minderheitenschulen der europaischen Turkei von der Reformzeit bis zum Weltkrieg (Breslau 1936). Stone, Andrews Frank, Academies for Anatolia (Lanham, New York, London 1984). Suny, Ronald Grigor, Looking Toward Ararat. Armenia in Modern History (Bloomington and Indianapolis 1993). SutcIifFe, Constance, 'Turkish Guilds' Forthnightly Review LXVI (1896), 828-35. Taneri, Kemal Zuhu., Suleyman Hiisnii Pasa'mn Hayati ve Eserleri (The Biography and Works of Suleyman Htisnu Pasa) (Ankara 1963).

248

The Well-Protected Domains

Tarhanli, Istar, Musluman Toplum 'Laik' Devkt. Tiirkiyede Diyanet if fen Baskanltgt {Islamic Society and 'Secular' State. T h e Directorate of Religious Affairs in Turkey) (Istanbul 1993). Tarih Deyimlert ve Terimleri Sozlugii (The Dictionary of Historical Sayings and Terms) (Istanbul 1983). Tevfik, Riza, Biraz da Ben Konusaytm (My Turn to Speak) (Istanbul 1993). Tekin Sinasi and Tekin, Goniil Alpay, 'Imperial Self Portrait. The Ottoman Empire as Revealed in Sultan Abdul Hamid's Photographic Albums,' journal of Turkish Studies, 12 (1988). Tietze, Andreas, 'Ethnicity and Change in Ottoman Intellectual History,' Turcica, 22-31 (1991) Timur, Taner, Osmanli Cahsmalari (Ottoman Studies) (Ankara 1989). —Osmanh Kimligi (Ottoman Identity) (Istanbul 1986). Tadotova, Maria, 'The Ottoman Legacy in the Balkans' in Carl Brown (ed.), Imperial Legacy. The Ottoman Imprint on the Balkans and the Middle East (New York 1996) pp. 4 6 - 7 7 . Toledano, Ehud, State and Society in mid-Nineteenth Century Egypt (Cambridge 1990). Toprak, Zafer, Tiirkiyede Milli tktisat (The National Economy in Turkey) (Istanbul 1983). Trimingham, j . Spencer, The Sufi Orders in Islam (Oxford 1971). Trumpener, Ulrich, Germany and the Ottoman Empire 1914-1918 (Princeton 1968), Tunaya, Turik Zafer, Turkiye de Siyasi Parttler (Political Patties in Turkey) (Istanbul 1984). Turk Meshurlan Ansiklopedisi (Encyclopaedia of Famous Turks) (Istanbul 1943). Turkgeldi, Ali Fuad, Mesail-i Muhimme-i Siyasiye (Important Political Matters) (Ankara 1966). Uzuncar^ili 1 H-, Midhat Pasa ve Taif Mahkumlart (Midhat Pasa and the Prisoners ofTaif) (Ankara 1985). —Midhat Pasa ve Ytldiz Mahkemesi (Midhat Pasa and the Yildiz Trials (Ankara 1967). Vatin, Nicolas, 'Aux Origines du Pelerinage A Eyiip des Sultans Ottomans,' Turcica 27 (1995). • Voltaire, Francois-Marie, Essaie sur les moeurs. Bt I'esprit des nations (Paris 1858). Weber, Eugene, Peasants into Frenchmen. The Modernization of Rural France 18701914 (Stanford 1976). Wilentz, Sean (ed.), Rites of Power. Symbolism, Ritual and Politics Since the Middle Ages (Philadelphia 1985). Williams, Charles, The Armenian Campaign, A Diary of the Campaign of 1877 in Armenia and Koordistan (London 1878) pp. ix,x. Wilson, Keith M (ed.), British Foreign Secretaries and Foreign Policy (London 1987). Woodhead, Christine, 'The Present Terror of the World'? Contemporary Views of the Ottoman Empire cl600,' History 72 (1987) pp. 20-37.

.1

Bibliography

249

Whitman, Richard, Scenarios of'Power: Myth and Ceremony in Russian Monarchy vol. 1 (Princeton 199S). —'Moscow and Petersburg: The Problem of Political Center in Tsarist Russia 1881-1914/ in Sean Wilentz (ed.), Rites ofPower (Philadelphia 1985) pp..244~ 71. Yasamee, E A. K., Ottoman Diplomacy, AbdiXlhamid U the Great Powers 1878-1888 (Istanbul 1996). —'Abdulhamid II and the Ottoman Defence Problem' in'Diplomacyand Statecraft 4 (1993). —'The Ottoman Empire and the European Great Powers' PhD Dissertation, School of Oriental and African Studies (1984) pp. 56-60. Zia, Nesim, Ktbris'm Ingiltere'ye Gecisi vt Ada'da kurulan Ingiliz Idaresi (The British Take-over in Cyprus and the British Administration on the Island) (Ankara 1975). Zwemer, Samuel M,, Tbe Law of Apostasy in Islam. Answering the Question why there are so few Moslem Converts, and Giving Examples of their Moral Courage and Martyrdom (London, Edinburgh and New York 1924).

Index

Abbas Hilmi Pasa, Khedive of Egypt (1892-1914) 58,144 Abbasids, Abbasid dynasty 46,170, 174 Abdeikadir, Major 70 Abd El Montaleb 57 Abdulaziz, Sultan (1861-1876) 18, 22,33,60,63,138,172 Abdulhamid II (Sultan) 18761909,5,11,17,58,144,145,163; deposition of 81,174; and education 97,101,103; exile and execution of Midhat Pasa 167-8; Hijaz railway 6 0 - 1 passim; interpretations of reign 4-18; isolation of 18-19; on foreign missionaries 115; and Friday prayer 22-6 passim; titles 48, 56; and ulema 6 3 - 6 passim; and conversion policy 70,83 see also caliphate

Aden 61 Africa, Africans 3,148,153,154 ahkk'i milliye ('national morals') 11 Ahmed I, Sultan 30 Ahmed Cevdet Pasa, (historian, statesman and jurist) 2 0 - 1 , 5 0 1,136,168,169-70,172,176 Ahmed Efendi 63 Ahmed Muhtar Pasa, Gazi, (Ottoman diplomat) 146 Ahmed Sakir Pasa 7 0 , 7 8 - 8 0 , 8 3 , 91,92,94,99,101 Asci Dede 7 Akarli, Engin Deniz 3 Akdag Madeni (Anatolia) 7 8 - 9 AkifPasa 167 al-Azhar 53 Alaeddin Keykiibad I (Selcuk

Abdullah Freres, Imperial photographers 152 Abdtilmecid, Sultan (1839-1861) 18, 22,27,45,168 Abhaz 170 Abu El Haj, Rifa'at 9,10 Abu Hanifa ( Hanefi imam) 95, 173 Abu Manneh, Butrus 38,45,64 adab-t milliye ('national traditions') 110 Adana 30,88

119 Albanian House Guards 23, 24 Albert Victor, Prince Consort 145 Aleppo 65,83,87,123; vilayet 101, 127 Alevis 40,42 Alexander III, Tsar (1881-1894) 17,28 Algeria, Algerians 55 Ali ibn Abi Talib, Caliph (656-661)

Sultan 1219-1237) 2 Albania, Albanians 81,101,109,

5

250

Index AH Ni^ami Pasa, Ottoman diplomat 144 Ali Pasa Mehmed Emin, grand vizier 64,115,139 Ali Resid Efendi 83 Aikus (Basra) 90 Alliance Evangelique 124 America, Americans 4 1 , 1 2 5 - 3 2 fwai«135,143,149,150 American Bible Society 127-8 American Mission in Syria 114 Anatolia 8,33,69,77~S2 passim, 86,89,92,93,99,121,141,147 Anderson, Benedict 152 Angora 78 Ankara 22, 63,77,80, 81, 89, 91, 176 Antakya (Antioch) 83 Anvers Fair (1894) 160 Arai, Masami 7 architecture see monuments Armenians 2, 34,42,49, 86-92 passim, 98,105,106,123,127-8, 131-2,137; massacres (1890s) 77,92,129,137,146 Arnold, Dr Muhleisen 46,124,134 As'ad, Ahmad 65 AsimBey 68,73,74 Austria, Austrians 3, 20 Avdm (vilayet) 117 Aziz Pasa, Vaii of Mosul 7 Biichner 54 Baghdad 8, 66, 76,100,101; vilayet 107, 99 Balkans 1 Barbir.Karl 132 Barcelona Fair (1888) 154 Basmaciyan, Karabet 34 Basra 66,99,100,101,123; vilayet 41 Batum 63 Besiktas 102

251

Bebek Seminary 130 bedeUi askeri (military exemption tax) 85 Bedouins 41,42,77,91,157,159, 164 Beirut 105; vilayet 131; French schools in 120-1 Benedict, Burton 154 Berlin Africa Conference (1884) 172 Beyazidll, Sultan 30 Bible Society 118 Bingazi, sancak 59,101,104 Birecik (Anatolia) 86,88 Bismarck 99,138 Bitlis 89,93; Boys Academy 127; viiayet 66 Blind, Karl 146 Britain, British Empire 22, 56,114, 140,142,143-4,146,154 British Museum 151 Brittany 8 Brussels 162 Budapest 140; Fair (1896) 161 Buddhism, Buddhists 109,126 Bukhara, Emir of 57 Bulgaria, Bulgarians 36,137,148, 154,164; 'Turkish atrocities' 46 Bursa 31; vilayet 106 al-Bustam, Siileyman 156 Byzantine Empire 29 Caeser, Julius 10 Cahun, Leon 7 - 8 Cairo 142-3 caliphate 169,170,173,174-5; Hamidian Hanefi 4 6 - 9 Caratzas, Kyrillos 81 Carnot, Sadi, President of France 142 Catalca 34 Catherine II, Empress of Russia 136

252

The Well-Protected Domains

Catholic orders: Capucin 119,121; Carmelite 119; Jesuit 116,11922passim,13Q Caucasus 63,94 al-Cawaih (newspaper) 61 Cebele 84 Celik, Zeynep 165 Cemaleddin, Sheikh 109 Central Turkey College (Antep) 133 ceremony 16-26, 3 1 - 2 see also selamhk Cevad Pa§a, grand vizier 124 Chicago Columbian Exhibition (1893) 126,155-61,164-5; Ottoman Hippodrome 157, 159,164 China, Chinese 111,112,125 Choueiri, Youssef 174 Christianity, Christians 8,36,68, 72,77,78,113,134,148,169, 170; Catholics 115-22 passim, (Armenian 33,122); Chaldeans 49,119; conversion to Islam 8 4 91; crypto see Istavn; Greek Orthodox 28,92,134; Maronites 120; Nestorians 49,119,122, 124,156; Protestants 92,113-33 passim see also Armenians, Catholic orders, missionaries Circassians 170 Comedie Franchise 142 Committee of Public Instruction (India) 109 Committee of Union and Progress 109-11 see also Young Turks. Concert of Europe 136 Confucianism 109 conversion see religious policy (Hamidian) Corfu 80 courts, religious (ser'i) 5 1 - 2 courts, secular (nizami) 45, 5 0 - 2

Courtellemont, Gervais 55, 56 Crete 28 Crimean War (1854-1856) 10, 135,136,137 Cromer, Lord 58 Crown Prince of Morocco 58 Currie, Sir Philip 143 Cyprus, 85; British occupation 123 Cyprus Convention (1879) 113 Da'i al-rasbad of Sheikh Ebulhuda al-Sayvadi 65 Dagistan, Dagistanis 63,109 Dajigs 89 Damascus 8,63,76,120 Darwin, Charles 54 Dashnaktsutiun 128 de Lusignan, Princess Anne Marie 108,138,139,147 Debats 137 decorations (nisan) 21,24, 27, 35-7, 142; Wat (robe of honour) 37 dervishes 150 see also Sufism devlet salrtameleri (state almanacs) 27-8 Diderot 137 Diplomatic Flashes 138 Dix'Neuvietne Siecle 141 Diyarbekir 102 donme (Jewish converts) 81,97 Dreyfus affair 148 Duguid, Stephen 47 Dumont, Paul 31 Dutch East Indies 171 Eastern Question 6 Ebuziyya Tevfik 161 Edirne 34 education (mass) curriculum 9 4 - 5 , 97, 98, 110; elite perception of 98-9; extension of 94,107-8; Law of Education (1869) 105, 116; and national identity 108-9,

Index 110; of non-Muslim population 104-7; policy and aims 96-101 passim, 107,110; religious instruction 9 4 - 5 , 9 7 - 8 ; of Shi'a 99-100 see also schools and colleges Egypt, Egyptians 3,22,31,41,46, 53,58,59,65,150,171; British occupation 113,120; Napoleonic invasion 120 Elbasan (Albania) 81 elite, Ottoman 2 - 3 , 1 5 , 4 5 , 6 6 ; intellectual conditioningl9-22; mission civilhatricc 41,94,168; language of 43; as a service elite 3,166 ('tacit knowledge' of 16670) sec also education Elliot, M r 128 Elliot, John 16 Elliot, Sir Henry 138 Emetullah H a t u n 33 Emin Efendi 167 Emir of Bukhara 57 Emirgan 152 Enayat, Hamid 175 ErtugrulGazi 3 0 , 3 1 - 2 Ertugrul Regiment 25, 32, 37 Erzincan 63 Erzumm 36,106,127; vilayet 121 Esad Pasa, Ottoman diplomat 155 Eyiip, shrine of 24 Fahreddin Pasa 41, 44 Fatma Aliye, daughter of Ahmet Cevdet Pasa 147 Fatma Sultan Mosque 63 Fehim Efendi, Haci 63 FerahAUPasa 169 Feyztulah Efendi, Haci 81 Findley, Carter 7,97,107,167 'fine tuning' 8-11 Finkel, Caroline 4, 30 First World War 171

253

Fitzmaurice 86,87 Fleischer, Cornell 46,48 Franco-Prussian war 120 France, French 68,72,114,118-9, 120-3,130-2,153,170 French Revolution 171 Fuad Pasa 5,64,172 Gumiishane (Trabzon vilayet) 78, 89 Gumiishanevi Ahmed Ziyauddin Efendi (Naksibendi sheikh) 6 3 4 Galatasaray Lycee (Mekteb Soltani) 6,96,98,104 Garber, Maria 131 Gazi Osman 32,170 Geertz, Clifford 21 Gellner, Ernest 10,43,45 Georgeon Francois 31 Germans, Germany 3, 20,154 Ghassanids 38 Gladstone 147,148 Gluck, Carol 3,17,19 Grabill, Joseph 129 Grand National Assembly 174 Greek-Ottoman war (1897) 144 Greece, Greeks 18,28,40,85,86, 105,136 Gregory XV, Pope 119 Giilmez Brothers (imperial photographers) 35 Gumiishanevi' 64 Hiiseyin Gagman, Kadi 50 Habermas,Jurgen 9 Habsburg, Habsburgs 17,47 Hagia Sophia see also Ayasofya 29, 31 haj 57-60 passim Hakla Bey 158-60 passim, 163 Halckari 70 Hamburg 153

i

254

The Well-Protected

Domains

Hamtdiye regiments 36,69,70,75, 81,102 Hamlin, Cyrus 130 Harh(Yezidisect) 73 Hasan Tarhan Pasa, Ottoman diplomat 154 Hasluck, William 83 HaymeAna 32-3 Henry jessup 134 Hepworth, George H . 128 heretics, heretical sects (ftrak't dalle) 69 Hija:z (vilayet) 4 4 , 4 6 , 5 1 - 6 2 />oMim,65,80,101,171 Hijaz railway 44,60,164 Hidaiye (sect) 84 historiography 175 Hobsbawm, Eric 6,45,93 Hourani, Albert 47,175 Howard, M r 128 H o y (Khoi, Iran) 128 Hungary, Hungarians 139-40,141 Hutgronje, Snouck 48 IbnKhaldun 20 iconography 26-35; on chronograms (ebced hesabt) 30; Ottoman coat of arms 26-7; medallions 27; tugra (sultan's monogram) 29-30, 34,35; mythical genealogies as 2 7 - 8 see also monuments ideology, Hamidian 18-19,29,46, 59,66,68,91,96,107, i « also eduction, legitimation, nationalism, religious policy Ilya Suhami Saadullah 8c Co. 155 Imperial Tobacco Regie 35,143 imperial systems/states 3; compared with Ottoman Empire 17-18; 9 4 , 1 0 8 - 1 0 , 1 1 2 , 1 7 1 - 2 Inalcik, Halil 46,175

India, Indians 3,55,56,57,59,62, 149,171 India Office 148 International congresses, Ottoman participation 153-4 Iran 52,100 Iraq 3,48,50 Irish Question 147 Iskodra (vilayet) 119 Islam, Muslims (Shi'i) 19,41,82, 99-100; Zaidis 51,76 Islam, Muslims (Sunni) Hanefi mezhcb (school) 3,14,40.48,66, 68,83,85,100,174-5;Sha'afi mezbefe (school) 70,77,100; Wahhabism 61 see also religious

policy Islamoglu lnan, Huri 47 Ismail Hakki Pasa 62 Ismail Nuri Pasa, Vali of Mosul 100 Istanbul 2,29, 33,57,58,60,69,89, 99,133 Istavri (Stavriotae) 78-81, 91 Italy, Italians 20 Izmir (Smyrna) 107,121 Izzed, Mehmed 148 Janissaries 4,162; massacre of (1826)168 Japan, Japanese 3,16,17,22,93, 108,109,110,113,126,171 Java, Javanese 56, 59 Jeddah 55,57,62 Jerusalem 8,101 Jessup, Henry 113 Jews 49 Johore, Kingdom of 160 Ka'ba 33,57 Kami! Pasa, grand vizier 33, 62,65, 120,122,123 Kantorowicz, E. H . 168,173

Index Kara Hisar 125 Karaca Pasa, Ottoman diplomat 143 Karak (Jordan) 77 Karakyan, Leon Bey 162 Kars 135,145 Kastamonu (vilayet) 107 Kayali, Hasan 41 Kayseri, missionary schools in 125 Kemal,Namik 109 Kemalists 174,175,176 Kerbela 99 Kethtidazade Haci Gogiis Efendi 133 Kevar (Anatolia) 124 Khoury, Philip 175 Kizilbas 82, 91 Knapp, George Perkins 127 Knowles, Revd Edward Randall 130 Konta, Monsieur 163-4 Kenya 41; (vilayet) 130 Koprtilu, Ahmed 27 Koprulii, Mehmed 27 Korea 113 Ku$iik Said Pasa 168 Kuran, Aptullah 23 Kurds 19,40,41,49, 81 see also Yezidis L'lllustration 149 Lalis, Yezidi sanctuary 73 Latakia (Syria) S3 Latham, R . G . 146 Layard, Sir Henry 46,116,149 Lazkiye (Syria) 83,84 Lebanon, French influence in 120 legitimation 43,45; crisis of in Ottoman state 8,10,11,166; definition 10; 'delegitimation' 173-6; policies in Hamidian era 2,47-8, 52-3,93-4,107; Weber on 2

255

Lekner, Dr G. W 153 Lewis, Bernard 40,46 Library of Congress 151 Libya 58 London 140,143-4 148,154 London Chronicle 149 London Daily News 118 London World Exhibition (1855) Louis XVI, King of France 173 Lourdes 8 Lynch, H . B . 106 Ma'an valley (Jordan) 77,91 Macauley, Thomas Babington 109 Madrid 58 Maharajah of Jahore 152-3 Mahmud II, Sultan (1808-1839) 9,11,18,22,47,167,168 Mahmud Efendi 79 Mahmud Memduh Pasa 80 Maksimoff 58 Malet, Sir Louis 149 Malluh 150 Mamuret-itl-aziz (Elazig), vilayet 102,121 Manastir vilayet (northern Greece) 128 Manolaki 34 Mardin, Serif 31,42, 49,63,81, 167 Markab 84 Mavroyeni Bey, Alexander (Ottoman diplomat) 150-1,156 Mazzini 99 Mecca 21,25,29, 33, 37,108 Mecelk~i Abkam-t Adliye (codification of ser'i rulings) 50 Medina 25,29,33,37,41,44 Mehmed Emin Ali Pasa 27 Mehmed Emir Ali Khan 149 Mehmed II, Sultan 27,142 Mehmed IV, Sultan 30 Mehmed Sakir Efendi 87

256

The Well-Protected

Domains

Mehmed Sadik, Hafrz Mehmed Vahdettm, Sultan 174 Meiji dynasty 16,112,171 Mekteb-i Asiret (Tribal School) 101-4,109,152 Mekteb-i Bahriye (Imperial Naval Academy) 96 Mekteb-i Harbiye (Imperial Military Academy) 94,104,110 Mekteb-i Mitlkiye-i Sahane (Imperial Civil Service School) 94,96,98,102,104,110 Mekteb-i Sultani see Galatasaray

Lycee Mekteb-i Tibbiye (Imperial School of Medicine) 94,110 Melhame, Selim 139 Merzifon (Anatolia) 124,129 Messkk, Brinkley 50, 51 Midhat Pasa, Ottoman statesman 137,138,167,168,173 MihalBey 170 Mihahccik (Anatolia) 77 Mihran Boyaayan Efendi 105 Milan 142 Mill, John Stuart 54 Millingen, Osman 142 Mintzuri, Hagop 2 4 - 5 Mir Ali Bey 71, 72 Mir Mirza Bey, Yezidi chieftan 70, 74 Mizrakli Ilmihal 54 Missionary Herald 125 missions, missionaries 4 0 - 1 , 1 1 2 34 passim', American 41,125-32 (and Armenian question 12731); British 123-4; French 119-23; Hanefi (da'iyan) 75-6; Shi'i 41 see also schools Mitchell, Timothy 31 Mohammed, the Prophet 142 Mokhbii, Sheikh Ali, Emir of Zhafi 61,62

Molokai Island, Hawaii 126-7 Mondros Armistice 44 monuments, symbolic use of 2 9 32; clock towers 29-30; mausoleums 31-2 Morocco, Moroccans 58 Moscow 152 Mosul 68, 71,72, 73,74; vilayet 41, 66,75,90,99,100,101,104,119 Mughal dynasty 2 Mukaddimab of Ibn Khaldun 20 Munci Bey, (Ottoman diplomat) 126-7 Munir Pasa, Minister of Protocol 36,145 M u r a d H I 30 Mus 89 Musa Bey (Kurdish sheikh) 89 Muscat, Emir of 61, 62 Mustafa I, Sultan 30 Mustafa Kemal (Ataturk) 174,176 Mustafa Resid Pasa 27 Mustafa Zeki Pasa 96 Musurus Ghikis Bey 137 Musurus Pasa (Ottoman diplomat) 137,144 mysticism see Sufism Nablus 118 Naci Kiciman 44 Napoleon Bonaparte (1804-1814) 120 Napoleon III, Emperor (18521870) 20 nationalism 45; 'official' 47; Arab 104; Ottoman 133; proto 59 Nawabof Rampur 57 NazimBey (Minister of Police) 39 Necipoglu, Giilru 29 al-Necri, Sheikh Ali 102 New York 150-1 Nicholas I, Tsar (1894-1917) 17, 94

Index Nicosia 123 Nigde (Anatolia) 30 The Nineteenth Century 138 Normandy 8 Norway 54 Noya Vremya (newspaper) 140 Nubar Pasa, Egyptian statesman 144 al-Nur al'sati of Sheikh Ahmad Es'ad 6 Nuri Bey 158 Nursi, Sheikh Said 102,133 Nuseyri, Nuseyris 8 3 - 4 , 9 1 Ocak, Ahmed Yasar 66 Odemis 125 Oguz tribe 2 7 - 8 Oman 61 Omdurman, battle of 1 Omer Vehbi Pasa, General 7 1 - 5 passim Ommayads, Ommayad dynasty 174 Orhan, Sultan (1326) 31 Orientalism 150 Ortayh, liber 167,176 Osman, Sultan (1299) 27, 29, 31 Osman II, Sultan 30 Osman Dikna (Sudanese Mahdi) 62 Osman Pasa, Gazi 154 Osman N u r i Pasa {Ottoman bureaucrat) 37, 51-2, 56, 59, 99, 101,110 Osman Pasa, Vali of Mosul 73 Otranto 1 Ottoman Empire foundation myth 27-8; administrative divisions 12-13; as a Great power 134, 168; as an imperial power 145-7; image of abroad 134-5; use of protocol 172; historiography 4 8; see also legitimation

257

Ottoman government (Sublime Porte) 58,74,77,86, Council of Ministers (Meclis-i Viikela) 7~ 3,76, 81, 84; intelligence services 128 Ottoman-Greek war (1897) 28 Ottoman-Russian war (1877-78) 154 Paik-i Islam (newspaper) 149 Pall Mall Gazette 135,138 Papacy, Vatican 122 Paris 67,141,14 Paris World Expositions (1867)18; (1899) 155-6,162; (1900) 1 6 1 2,172 Patrak 71 Pesderli (Kurdish tribe) 100 Persian Gulf 60 Pertev P a p , grand vizier 167 Peter the Great (1682-1725) 8 Petitjournal 139 Phiiiki Etheria 80 Philip IV, King of Spain 16 Philippines 130 photography, photographic albums 108,151-2 Picard, Monsieur 162 Plevna, battle of 1,154 Poland 94 Politis, Kozmas 106 Pomaks 80 press (domestic) as vehicle of religious dissemination 49 press (foreign) Ottoman image in 137-43; and missionaries 125-6, 131 power of popular press 147 Questions Diplomatiques et Coloniales 137 Qur'an 21,36, 37,52-4,57,83, 109,173 Qureisb (clan) 49

258

The Well-Protected Domains

RaciBey 157,159 Ramazanojnillan d a n 32 Ramberr, Louis 35,143,145 ResidPasa 29,45,108 Recai Bey 98,102 Red Crescent 144 Red Sea 60, 62 Redhouse, Sir James 55 relics, acquisition by Ottoman state 38-9 religious policy (Hamidian) 44-67; aims 47-8; codification of Seriat rulings 5 0 - 1 ; concept of religious freedom (serbesth idyan)HS-17; control of sacrality 52-6; 'correction of beliefs' (tashih-i akaid) 49,70; conversion (of Christians 8 4 - 9 1 ; of 'heretical sects' 68-92); education of missionaries 76; Hanefi mezbeb as official ideology 4 8 - 9 , 66-7; and Sufis 63-6; and utema 63-6 passim, 77; Resid Pasa 5 Richard II 173 . Rifat Pa$a, grand vizier 116 Risorgimento 20 Robert College (Bebek Seminar) 130 Romanov dynasty 47 Rome 142 Roseberry, Lord 142 Rosenstein, Mr 34 Rumania 144 Rumelia 97,136 Russell, Earl 114 Russia, Russians 3,17,28,31,53, 114,136,138,139,140,144,145, 146,149 Russo-Japanese War (1905) 7,135, 149 Russo-Ottoman war (1887-88) 140

Saadullah Efendi 163 SabitBey 159 Sadik Rifat Pasa, Ottoman statesman 20 Sadik, Hafiz Sadik 134

Sahyun 83 Said Nursi, Beditizzaman 66,93 Said Pasa, grand vizier 2, 37, 38, 84, 140 Said, Edward 113,136,142,150, 156,172 Salisbury, Lord 89,144,149 Salonica, Selanik 91,97 Salt, Jeremy 114 Salzmann 9 San Francisco 149 S a n a 50,51,76,107 Sanasarean College, School of Law 106,110 al-Sayyadi, Sheikh Abulhuda 6 5 - 6 schools ibtidatye (primary) 94,100; idadi (middle) 94, 98,100,105; medreses (religious) 71,98-9, 105; missionary 116-17,121-4, 130-4 passim; rusdiye (secondary) 82,84,94,95,100,107,118; 119 Sebekli (Yezidi clan) 71, 73 Selaheddin Bey, Colonel 135 selamhk (Friday prayer) 1 6 - 1 7 , 2 3 4,26 Selim 1,'Yavuz', Sultan (15121520) 30,44,46,91,169 Selim II, Sultan 4, 30 Selim III, Sultan 4 Senusi Brotherhood 42 Serbia, Serbs 18,36 seriat (sacred law of Islam) 20, 21, 26,77,80,85,116; Ottomanization 50-52 passim ServetA Funun (magazine) 137,143, 144 Seton-Watson, Hugh 47

Index Seyhan valley 70,71, 73,75,100 seyhuiislam (chief religious official of the Ottoman Empire) 52,64, 92,94,95 Sheikh Samil 109 Sheikh Adi (Yezidi shdne) 74,100 Sheikh Alt Efendi 83 Sheikh Fehmi 7 Sheikh Ibn Resid 37 Sheikh Irbili Mehmed Es'ad 64 Sheikh Muhammed N u r 90 Shils, Edward 43,167 Sincar.Mt 4 1 , 6 8 , 7 3 - 5 Sinart Pasa mosque 24 Sis 89 Sivas 90, 111; vilayet 78,82,128 Smith, Anthony 66 Snider,). 159 Socrates 136 Sofraaoglu Omer 80 Sogiid (Anatolia) 25,32 Spain, Spaniards 58 Spanish-American War (1898) Spencer, Herbert 54 St Petersburg 140 St Louis Fair 163 Stockholm 54 Strauss, Professor Adolphe 36 Suakin 62 Sudan, Sudanese 62 S u d r e , C H . 160 Sufis, Sufism 6 3 - 6 , 8 1 Sufi orders Khalidi Naksibendi 64; Melami 81; Mevlevi 161; Naksibendi 63,81; Rifai 65; Shadhili-Madani 65 Siileyman Hiisnii Pasa, Ottoman statesman 21, 48, 66,133 Siileyman the Magnificent 30,27, 46 Siileymaniye Mosque 64 Stirre Alayi Sultan of Zanzibar 61

259

Sultanahmet Square 155 Sutcliffe, Constance 147 Sweden 54 Sykes, Mark 146 Syria, vilayet 55,58,65,76,83,84, 101,116,117,149; French influence in 120-1 Tahir Efendi 39 Tahsin Efendi, Hoca 132 Taif (Yemen) 166 Taiwan 112 Tanzimat 9-10, 20, 31,45,48, 94, 108,110,115,133,167,168,169, 176; Imperial Rescript of the Rose Chamber (Hat-y Serif-i Gulhane 1839) 9,167,168; Reform Edict (Islahat Fermam.l856)78,115 Tarih-i Gevdet of Ahmet Cevdet Pasa 170 Tataristan 94 Teasdale, General Christopher 135, 145 Tel el-Kebir 1 TevfikBey 159 Tevfik Pasa, Khedive 144 Tevfik Pasa, foreign minister 132 Tevfik, Riza 55 The Hague 143 theatre, Ottoman image in 142-3 Thirteenth World Congress of Orientalists 153 Three Emperor's League 139 Tibawi,A.L. 113 Tietze, Andreas 106 The Times 137 Timur, Taner 5 Tir-i Miijgan H a m m (mother of Abdiilhamid) 30 Todorova, Maria 9 Tokad 4 2 , 8 2 Toledano, Ehud 175

260

The Well-Protected Domains

Topkapi Palace 24,155,160 Topcani, Esat 173 Torajiro, Yamada, 23 Trablus (Lebanon) 83 Trabiiis (Libya) 80 Trabzon (Trebizond) 36,77,78,79, 91; vilayet 78-9,121,125 Treaty of Berlin (1878) 115 Treaty of Paris (1856) 10,134 Trimingham,j. Spencer 65,66 Tripoli 73,101 Tzu Hsi, Empress Dowager of China 113 Ubeydullah Efendi 63 United States see America, Americans Oskiidar School of Trades and Crafts 152 Uskiib (Scopje) 34 Uvarov, Count 94,108 Uzuncarsih 16 Vajda, Sigismonde 36 Vakit 149 ValiofAydm 117 Vambery, Arminius 139-40,160 Van 49,75,93,96,99,102,127, 128 Vassaf Efendi 167 Vatican 122 Victoria, Queen of England (18371901) 56-7,144,145,147;

jubilee 3 Vienna 141; siege of (1683) 9; University of 17 Voltaire, Francois-Marie 4,136 Wailachians 105 Washington D C 150 Washington Post 126 Weber, Eugene 94 Weber, Max 2 West Sussex, Massachusetts 130 Westerly, Dr Michael 127

Williams, Charles 146 Woods Pasa, Admiral 145 Worcester, Massachussetts 125 World Archaeology Congress 153 World Congress to Improve the Conditions of the Blind 154 World fairs, Ottoman participation in 150-65 Yemen (vilayet) 5 0 , 5 1 , 5 8 - 9 , 6 1 , 76,77,80,101,104,124,135, 167 Yezidis 3,4041,91,68,69-75,78, 100; campaign to convert 120 Yildmm Beyazid 30 Yikhz archives 11-13,137 Yikhz Palace 16,18,23,26,33, 36, 77,94,140,151,156 Yildiz Tribunal 168 Young Turk'"Revolution' (1908) 104 Young Turks 173,174,175 Yozgad (Anatolia) 30,78,79,80,89 YusefRazi 162 YusefZeki Efendi 120 Yzmit 34 Zafir, Sheikh Hamza 65 Zanzibar 61 Zeld Pasa, Commander of military schools 104 zikr (ceremony of'remembrance') 150 Ziya Pasa, Ottoman statesman 136 ZoIa,Emile 147 Z o r (sancak) 73,101,102 Zuhdii Pa$a (Minister of Education) 52,103 Zulus 2

The papers in this volume, a multidisciplinary collaboration of anthropologists, linguists, and psychologists, explore the way in which cultural knowledge is organized and used in everyday language and understanding. Employing a variety of methods, which rely heavily on linguistic data, the authors offer analyses of domains of knowledge ranging across the physical, social, and psychological worlds, and reveal the crucial importance of tacit, presupposed knowledge in the conduct of everyday life. Many of the papers included examine American cultural knowledge; others, by anthropologists, provide accounts from very different cultures. Collectively, the authors argue that cultural knowledge is organized in "cultural models" - storylike chains of prototypical events that unfold in simplified worlds - and they explore the nature and role of these models. They demonstrate that cultural knowledge may take either proposition-schematic or image-schematic form, each enabling the performance of different kinds of cognitive tasks. Metaphor and metonymy are shown to have special roles in the construction of cultural models: the former allowing for knowledge to be mapped from known domains of the physical world onto conceptualizations in the social and psychological domains as well as in unknown physical-world domains; the latter providing different types of prototypical events out of which cultural models are constructed. The authors also reveal that some widely applicable cultural models recur nested within other, more specialpurpose models, thereby lending cultures their thematicity. Finally, they show that shared models play a critical role in thinking, one that has gone largely unappreciated in recent cognitive science - that is, that of allowing humans to master, remember, and use the vast amount of knowledge' required in everyday life. This innovative collection will appeal widely to anthropologists, linguists, psychologists, philosophers, students of artificial intelligence, and other readers interested in the processes of everyday human understanding.

Cultural models in language and thought

\

EDITED BY

University of North Carolina, Chapel Hill

Naomi Quinn Duke University

Tbt right 0/ ilu Unttmily of Csmb'tdgt to prim and self all mmnrr a/ book i was granted by Httiy V)ll in 1531, Tht Vnirinliy Aoi prigtid and publishtd eenil/tuousty Jl'SM /JW.

Cambridge University Press CAMBRIDGE NEW YORK NEWROCHELLE MELBOURNE SYDNEY

?fh

Published by the Press Syndicate of the University of Cambridge The Pitt Building, Trumpinglon Street, Cambridge CB2 IRP 32 East 57th Street, New York, NY 10022, USA 10 Stamford Road, Oakleigh', Melbourne 3166, Australia © Cambridge University Press 1987 First pubBshed 1987 Reprinted 1987,1989 Printed in the United States of America Library of Congress Cataloging-in-Publication Data Cultural models in language and thought. Papers originally presented at a conference. 1. Language and culture. 2. Cognition and culture. I. Holland, Dorothy C, II. Quinn, Naomi. P35.C8 1987 401'.9 86-17524 British Library Cataloguing in Publication Data Holland, Dorothy Cultural models in language and thought, 1. Ethnopsychotogy 2. Cognition I. Title II. Quinn, Naomi 390 GN502 ISBN 0 521 32346 0 hard covers ISBN 0 521 31168 3 paperback

Is

73732

Preface List of Contributors ' Introduction >0 1. Culture and cognition ,

page vii xi 3

NAOMI QUINN & DOROTHY HOLLAND

Part I. Presupposed worlds, language, and discourse 2. The definition of lie: an examination of the folk models underlying a semantic prototype

43

EVE E. SWEETSER

3. Linguistic competence and folk theories of language: two English hedges

67

PAUL KAY

4. Prestige and intimacy: the cultural models behind Americans' talk about gender types

78

DOROTHY HOLLAND & DEBRA SKINNER

5. A folk model of the mind

112

ROY D'ANDRADE

Part II. Reasoning and problem solving from presupposed worlds >0 6. Proverbs and cultural models: an American psychology of problem solving 151 GEOFFREY M. WHITE

7. Convergent evidence for a cultural model of American marriage

173

NAOMI QUINN

Part III. The role of metaphor and analogy in representing knowledge of presupposed worlds 'yj 8. The cognitive model of anger inherent in American f English GEORGE LAKOFF & ZOLTAN KOVECSES

195

CONTENTS

9. Two theories of home heat control

•'" '222'

WILLETT KEMPTON,

10. How people construct mental models

243

ALLAN COLLINS & DEDRE GENTNER

Part IV. Negotiating social and psychological realities 11. Myth and experience in the Trobriand Islands

269

EDWIN HUTCHINS

12. Goals, events, and understanding in Ifaluk emotion theory -

290

, CATHERINE LUTZ

13. Ecuadorian illness stories: cultural knowledge in natural discourse

313

LAURIE PRICE

14. Explanatory systems in oral life stories

343

CHARLOTTE LINDE

Part V. An appraisal 15. Models, "folk" and "cultural": paradigms regained?

369

ROGER M. KEESING

Index

395

This volume represents an interdisciplinary effort that has brought together.-:-/ anthropologists, linguists, and psychologists who study human cognition. In recent years, cognitive scientists from these three fields and others have converged in the study of knowledge, its organization, and its role in language understanding and the performance of other cognitive tasks. Here, we present a cultural view. We argue that cultural knowledge - shared presuppositions about the world - plays an enormous role in human understanding, a role that must be recognized and incorporated into any successful theory of the organization of human knowledge. As we summarize in the introductory chapter, cultural knowledge appears to be organized in sequences of prototypical events - schemas that we call cultural models and that are themselves hierarchically related to other cultural knowledge. This volume, then, is an interdisciplinary investigation of cultural models and the part they play in human language and thought. Earlier versions of most of the chapters in this volume were assembled and presented at a conference held in May 1983 at the Institute for Advanced Study in Princeton, New Jersey. However, to think of the book as a conference volume would be to fail to appreciate its history, which goes back some time before the Princeton conference. As histories should, this one has a lesson. It tells how, under felicitous circumstances, institutional support can enable scientific collaboration even across disciplinary boundaries. The developments described in this volume were underway in the late 1970s. One of us, Naomi Quinn, then a member of the Social Science Research Council Committee on Cognitive Research, organized an interdisciplinary workshop under the auspices of that committee to draw together some of the new ideas about culture and cognition. Held in August 1979 in La Jolla, California, under the rubric "The Representation of . Cultural Knowledge," that workshop numbered among its participants four of the contributors to the present volume - Roy D'Andrade, Edwin Hutchins, Dorothy Holland, and Naomi Quinn. As a substantive statement about the role of cultural knowledge in the understanding process, the workshop could be fairly characterized as premature. Many of the talks and much of the discussion had a tentative quality. Several of the formal

VU1

PREFACE

discussants, deliberately recruited from fields of cognitive science outside of anthropology, made clear their skepticism about that discipline's contribution to cognitive studies. The perspective represented in this volume was incipient at La Jolla, but undeveloped. Yet the workshop was a necessary first step toward defining a common enterprise and setting a theoretical agenda. Naomi Quinn's involvement in the activities of the SSRC committee enabled her to identify other people outside of her own field who were working toward similar ideas about cultural knowledge. She became better acquainted with the thinking of committee members Eleanor Rosch, a psychologist, and Charles Fillmore, a linguist, whose ideas and observations were to figure importantly in the approach developed in this book:: At La Jolla, she met for the first time psychologists Allan Collins and Dedre Centner and heard a paper on folk models they were presenting at an overlapping conference. At another committee activity that summer in Boulder, Colorado, she met linguist George Lakoff (though not for the first time, he reminded her) and obtained from him a copy of the book in manuscript, Metaphors We Live By, which he and Mark Johnson had just completed. Lakoff later invited Quinn to be an observer at his Conference on Cognitive Science, Language, and Imagery funded by the Alfred P. Sloan Foundation and held in Berkeley in the spring of 1981; there, she met Charlotte Linde and other linguists with similar interests. At neighboring universities, the two of us talked on about our common view of "folk knowledge," which was still crystallizing out of work in cognitive anthropology and related fields of cognitive science. We decided to organize a multidiscipiinary symposium for the 80th Annual Meeting of the American Anthropological Association in Washington, D.C., in December 1981. We called it "Folk Theories in Everyday Cognition." The resulting group of participants, and the papers they presented, encouraged our vision. Contributions by Lutz, Price, Sweetser, and White in this volume began as meeting papers delivered at that symposium; Holland, Hutchins, and Lakoff also participated, giving different papers than those they ultimately presented at the Folk Models conference that culminated in this book. The earlier La Jolla workshop had served as a beginning; the AAA symposium has a somewhat different but equally important role as a dry run for the conference to follow. Among members of the American Anthropological Association, it is popular to question the intellectual defensibility of meetings sessions, with the limited time constraints they place on paper and discussion length and the peripatetic audiences they attract. These critics overlook the important role of sessions like the one we organized as preliminaries to more ambitious professional activities. Relatively untaxing of organizational and fund-raising effort, the AAA symposium was an opportunity to gauge whether the new ideas about "cultural knowledge," "folk theories," and

PREFACE

k

"folk models" (which eventually became "cultural models") were sufficiently developed to merit a larger conference. It was also an occasion to experiment with the composition of the group, so that in the end we might identify and include individuals, whatever their disciplines; whose perspectives and enthusiasms matched our own in substantial w,aysi Finally, it served to orient individual efforts toward production of conference papers. It was shortly after the well-attended AAA symposium, with its high-quality papers, that we decided the time was ripe to seek funding for a full-scale conference. -.v"> :vBy then, unable to raise new operating funds, and having already sponsored a series of valuable conferences and workshops, the SSRC Committee on Cognitive Research was soon to be disbanded. The conference proposal we submitted to the Anthropology Program of the National Science Foundation was adapted from one Quinn had earlier drafted as a section of the final, unsuccessful umbrella proposal intended to fund the continuing activities of the SSRC Committee. NSF funded our proposal. Concerned that the grant might not cover all the expenses for this large conference, we applied to the Wenner-Gren Foundation for Anthropological Research for supplementary support. Working in consultation with NSF, Wenner-Gren contributed funds to fly our most distant participant, Roger Keesing, from Australia. Quinn was a member of the Institute for Advanced Study during the academic year 1982-83, as part of a group of researchers in cognition. Learning that the Institute sometimes hosted conferences, she explored the possibility of holding the conference there. The advantages, in terms of facilities, supporting staff, and location, quickly became evident. We formally proposed to Institute Director Harry Woolf and to Clifford Geertz, the anthropologist on the faculty of the School of Social Science, that they host the conference, and they graciously agreed. Subsequently, the project was granted an additional small amount by the Institute out of Exxon Educational Fund monies at its disposal; these funds allowed us to invite interested "observers" from the Institute and from surrounding universities to conference meals, to interact further with conference participants. It was clear to us by its close that a promising framework for the investigation of cultural knowledge was emerging at this conference, and that the research that had been reported in the delivered papers was sufficiently developed and interrelated to warrant publication. Scientists working independently along similar lines had been brought together to exchange ideas and to articulate a common approach. We are hopeful that publication of their chapters, with the integrating volume introduction we have provided, will convince other cognitive scientists of the heretofore largely neglected role of cultural presuppositions in human cognition and also demonstrate to other anthropologists the usefulness and promise of a cognitive approach to culture.

X

PREFACE

We have detailed the history of the efforts that led to this publication to make the point that institutional support of scientific projects such as this one has a cumulative effect not easy to assess in the short term. The book is the product of a lengthy, tentative process of regrouping and exchange, a process realized in several formal gatherings organized according to several different professional formats and made possible by the funding and facilities of an array of different institutions operating with different institutional mandates and designs. They were all indispensable. We hope Cultural Models in Language and Thought will testify'to the value of such repeated institutional support for organized meetings, large and small. We are indebted to all these supporting organizations, and to all their individual staff members with whom we worked. We came to appreciate keenly the special competencies that some of these individuals have for making the scientific process work. Lonnie Sherrod, staff associate at the Social Science Research Council, shepherded the Committee on Cognitive Research during most of Quinn's tenure on it and did so with an acute sense of what was happening in that quarter of the social sciences and what could be helped along. Stephen Brush, then the staff associate in the Anthropology Program at NSF who was responsible for oversight of our grant, shared much good advice about how to make an intellectually satisfying conference happen. Mary Wisnovsky, assistant to the director, and Grace Rapp, her assistant in the Office of the Director at the Institute for Advanced Study, are two unforgettable people with a special talent for making a conference happen smoothly and painlessly, even making it fun to give one. The postconference editing task has been lightened enormously by the skilled assistance of Carole Cain and Anne Larme, two anthropology graduate students at the University of North Carolina, Chapel Hill. We thank them all.

Contributors

Allan Collins Bolt, Beranek and Newman, Inc. 10 Moulton Street Cambridge, MA 02238 Roy D'Andrade Department of Anthropology University of California, San Diego La Jolla, CA 92093 Dedre Gentner Department of Psychology University of Illinois, Urbana-Champaign Champaign, IL 61801 Dorothy Holland Department of Anthropology University of North Carolina, Chapel Hill Chapel Hill, NC 27514

Roger M. Keesing : Department of Anthropology Research School of Pacific Studies Australian National University P.O.B. 4 ' Canberra, A.C.T. 2601, Australia Willett Kempton Center for Energy and Environmental Studies School of Engineering/Applied Science Princeton University Princeton, NJ 08544 Zoltan Kovecses Department of English ELTE, Pesti B.u. 1 Budapest, Hungary George Lakoff Department of Linguistics University of California, Berkeley Berkeley, CA 94720

Edwin Hutchins Institute for Cognitive Science University of California, San Diego La Jolla, CA 92093

Charlotte Linde Structural Semantics P.O. Box 707 Palo Alto, CA 94302

Paul Kay Department of Linguistics University of California, Berkeley Berkeley, CA 94720

Catherine Lutz Department of Anthropology State University of New York, Binghamton Binghamton, NY 13901

xii Laurie Price Multipurpose Arthritis Center Trailer 16, Building 272H Medical School ,. University of North Carolina, Chapel Hill Chapel HH1, NC 27514 : Naomi Quinn Department of Anthropology Duke University Durham, NC 27706 Debra Skinner Department of Anthropology University of North Carolina, Chapel HiU Chapel Hill, NC 27514

CONTRIBUTORS • Eve E. .Sweetser Department of. Linguistics University of California, Berkeley Berkeley, CA 94720 ;:

Geoffrey M. "White "" Institute of Culture and Communication East-West Center 1777 East-West Road Honolulu, HI 96848

Introduction

- -

1

Culture and cognition1

Naomi Quinn & Dorothy Holland

Undeniably, a great deal of order exists in the natural world we experience. However, much of the order we perceive in the world is there only because we put it there. That we impose such order is even more apparent when we consider the social world, in which institutions such as marriage, deeds such as lying, and customs such as dating happen at all because the members of a society presume them to be. D'Andrade (1984a:91) contrasts such culturally constructed things with cultural categories for objects such as stone, tree, and hand, which exist whether or not we invent labels for them. An entity such as marriage, on the other hand, is created by "the social agreement that something counts as that condition" (ibid.) and exists only by virtue of adherence to the rules that constitute it. •-•• -±. Such culturally constituted understandings of the social world point up not only the degree to which people impose order on their world but also the degree to which such orderings are shared by the joint participants in this world, all of whom behave as though marriage, lying, and dating exist. A very large proportion of what we know and believe we derive from these shared models that specify what is in the world and how it works. The cognitive view of cultural meaning The enigma of cultural meaning, seemingly both social and psychological in nature, has challenged generations of anthropologists and stimulated the development of several distinctive perspectives (see Keesing 1974 for an early review). Each of these ideational traditions in anthropology has had to address the same question: How are these meaning systems organized? Any convincing answer to this question should be able to account for at least the following properties of culture. It must be able to explain the apparent systematicity of cultural knowledge - the observation, old to anthropology, that each culture is characterized, and distinguished from others, by thoroughgoing, seemingly fundamental themes. Such a theory of culture also ought to explain how we come to master the enormous amount of cultural knowledge that the people of any culture have about the world and demonstrate in their daily negotiations with it (D'Andrade

4

NAOMI QUINN & DOROTHY HOLLAND

1981). Moreover, the large base of cultural knowledge we control is not static; somehow, we extend it to our comprehension of particular experiences as we encounter them. Given the uniqueness, sometimes radical and sometimes small, of these myriad daily experiences, cultural meaning systems must be adapted to the contingencies and complexities of everyday life. A theory of the organization of cultural knowledge must explain the generative capacity of culture. The approach in this volume makes progress and offers promise in accounting for all these properties of culture. The papers in the volume represent a cognitive approach to the question of how cultural knowledge is organized. For nearly three decades, cognitive anthropologists have been pursuing the question of what one needs to know in order to behave as a functioning member of one's society (Goodenough 1957:167). This school of anthropology came to stand for a new view of culture as shared knowledge - not a people's customs and artifacts and oral traditions, but what they must know in order to act as they do, make the things they make, and interpret their experience in the distinctive way they do. It is this sense of culture that is intended in the title of the present volume: Cultural models are presupposed, taken-for-granted models of the world that are widely shared (although not necessarily to the exclusion of other, alternative models) by the members of a society and that play an enormous role in their understanding of that world and their behavior in it. Certainly, anthropologists of other persuasions have arrived at the idea of "folk models" as a way of characterizing the radically different belief systems of nonwestern peoples (e.g., Bohannan 1957; Holy & Stuchlik 1981a). What is new in the present effort is ari attempt to specify the cognitive organization of such ideational complexes and to link this organization to what is known about the way human beings think. Cultural models, talk, and other behavior In practice, Goodenough's original mandate to investigate the knowledge people need in order to behave in culturally appropriate ways has been translated into a narrower concern for what one needs to know in order to say culturally acceptable things about the world. The relation between what people say and what they do has not gone entirely unconsidered by cognitive anthropologists. For example, this concern surfaces in an ongoing tradition of natural decision-making studies of which Geoghegan (1969), Gladwin and Gladwin (1971), Johnson (1974), and Fjellman (1976) are early representatives. In this line of research, behavioral decision models constructed with the help of informants' accounts of how they make decisions are then used to predict their actual choices. (See Nardi 1983 and Mathews in press for recent critiques of this approach from a perspective that would insist on the role of cultural knowledge in framing, not just

CULTURE AND COGNITION

5

making, decisions.) For the most part, however, cognitive anthropologists ; have specialized in talk/ ' ^: -^y-r:-'^ This definition of the research task r- explaining what people need to know in order to say the things they do - is simply taken for granted by the linguists with whom cognitive anthropologists exchange ideas,-andit is a conventional research strategy in other branches of cognitive science as well. In artificial intelligence, for example, apart from art Occasional robotic tour deforce, the major methodological arid theoretical challenge has been to build computer programs capable of story compreherision and other kinds of linguistic processing. This definition of the task is.alegacy of earlier attempts to solve the machine translation problem.* Artificial intelligence workers attempting machine translation discovered that language cannot be understood, much less translated, without reference to a great deal of knowledge about the world. The preoccupation of subsequent artificial intelligence research with this problem has captured the interest of cognitive anthropologists similarly concerned with what people have to know in order to use language. It has been colleagues from the more materialist traditions in anthropology, and indeed from some of the ideationalist traditions within the discipline as well, who have been at pains to point out the limitation of a research program for validating cultural models solely on the basis of linguistic behavior. These anthropologists observe that people do not always do what would seem to be entailed by the cultural beliefs they enunciate (for cognitive anthropologists' own critique of this issue, see Lave et al. 1977; Frake 1977; Clement 1982). Do cultural models, they want to know, influence more than talk, and if so how? Harris (1968) has proposed that cultural beliefs are epiphenomena altogether, reflecting the political economic circumstances that they arise, post Hoc, to rationalize. From a wholly different perspective, Levi-Strauss (1953) had earlier characterized native models as "home-made" ones, to be treated as repositories of false knowledge. The influence of his view can be gauged by the stance adopted in the work of anthropologist Barbara Ward (1965; 1966). Citing Levi-Strauss, she felt obliged to apologize for her interest in Hong Kong fishermen's native models of society, about which she wrote. A third, related strain in anthropological thought reflects this same tendency to discount the role, in people's behavior, of the cultural beliefs reflected in their talk. In this formulation, models for talking are separated, analytically, from models for doing. Paralleling Ryle's (1949) distinction between "knowledge how" and "knowledge that," and Geertz's (1966) distinction between "models for" and "models of," Caws (1974) presents an oft-cited argument for a tripartite typology of models (see also Holy & Stuchlik 1981b: 19-21). In addition to the scientist's "explanatory model," Caws proposes two types of native models: "representational" and "operational." The former are indigenous models of their world that people can more or less articulate; the latter are indigenous models that guide behavior

6

NAOMI QUINN & DOROTHY HpLLAND

in given situations and that tend to be out of awareness. Representational models, from this view, are not necessarily operational nor are tiie latter necessarily representational; thus, inconsistencies between what people say and what they do need not be cause for puzzlement. Holy (1979). applies this distinction in his attempt to resolve a long-standing debate in.social anthropology over the reported disparities between Nuer descriptions Of their kinship system and Nuer kinship behavior "on the ground.",^.^ Our vision of the role and importance of cultural models is at odds with the views of Harris and Levi-Strauss and that articulated in social anthropology by Caws. We do not assume that cultural models always translate simply and. directly into behavior. Indeed, the papers in this collection by Hutchins, Linde, and Price move toward a more precise understanding of the situations in which cultural models are invoked to rationalize and sometimes disguise behavior for other people and for ourselves. Nor do we expect cultural conceptualizations of the world to be the sole determinants of behavior. The work in this volume does suggest, however, that cultural models - which we infer from what people say - do relate to their behavior in complex, powerful ways. We are only beginning to specify the nature of these relations. Keesing is right, in his paper in this volume, to urge that cognitive anthropologists like ourselves take an active role in the emerging interdisciplinary study of "humans-insocieties." By linking meaning to action, cognitive anthropologists could substantiate Keesing's argument that "how humans cognize their worlds constrains and shapes how humans-in-societies reproduce them."We think it is a crucial first step to show, as these studies do, how cultural models frame experience, supplying interpretations of that experience and inferences about it, and goals for action. When interpretation and inference call for action, as discussed by Lutz with regard to the goals embodied in Ifaluk emotion words, and by White with regard to the dual conceptual and pragmatic functions of proverbs, then cultural understandings also define the actor's goals. (See also Jenkins 1981; Nardi 1983; Quinn 1981; Salzman 1981; and White 1985 for complementary views.) THE RELATION OF TALK TO ACTION AND AWARENESS

i

Seen as simultaneously interpretative and goal-embodying, cultural kno^edg£isjio£rjro^uctiveiy analyzedmto "jggdejs of ^and "models for," into^representational" and^bjp^rajionar' knowledge. Racier, in our view, underlying cultural models of the same order - and in some cases the same underlying cultural model - are used to performji variety of different cognitive tasks. Sometimes these cultural models serve to set goals for action, sometimes to plan the attainment of said goals, sometimes to direct the actualization of these goals, sometimes to make sense of the actions and fathom the goals of others, and sometimes to produce verbalizations that may play various parts in all these projects as well as in the subse-

CULTURE AND COGNITION

"f

quent interpretation of what has happened. Complexity in the relationship between what people verbalize about what they do and the execution of other, nonverbal activities is inherent in part because speakers so frequently undertake complex tasks with many goals that may or may not include producing a veridical verbal description1 of what they are about. Just to pose some possibilities in which verbal accounts are decidedly not veridical to the behavior they purport to describe, people may sometimes be concerned, simultaneously, to manage their affairs in a way advantageous to themselves and to present their goals in a favorable light; or to carry out their plans while hiding their true objectives from onlookers. In producing verbalizations, it is not so much that speakers invoke a different order of conceptualization of the activity about which they speak; it is rather that they invoke those cultural understandings pertinent to performing the linguistic part of the overall task at hand - say, in the task of presenting one's actions in a favorable light, a shaded model of the good person for whom one wishes to be taken; or, in the task of concealing one's plans, a shared model of plausible intentions with which to detract attention from one's real motives. Even when people are not wholly concealing or misrepresenting their behavior in what they say about it, they are characteristically called on to construct post hoc accounts of that behavior that are comprehensible, plausible, justifiable, and socially acceptable to themselves and other audiences, and that require a certain amount of smoothing, patching, and creative amendment to these ends. . Moreover, the multiple cognitive tasks and subtasks required to meet one's varied goals must often be executed simultaneously; the task demands of nonverbal behavior and those of concurrent verbal behavior may diverge, creating a further complexity in the relationship between the two. A waiter bent on getting a good tip, for instance, might be attempting to provide customers with swift, faultless service, silently anticipating their requests before these can be voiced, while at the same time keeping up a line of niceties and flattery. Even such ordinary daily activities as are. involved in doing one's job are multifaceted in nature, often requiring verbal expression and other action at once - sometimes in coordination, other1 times for independent purposes. Again, this is not to agree to the assumption that there exist, in the mind of the individual performing those different cognitive tasks simultaneously, two orders of cultural model. It is' simply to acknowledge that these differing tasks draw on a variety of cultural knowledge available for different purposes at different times. Indeed, talk itself involves such complex skills and understandings. As Sweetser (this volume) points out, even a single utterance may have multiple purposes. Her paper on lies and Kay's on hedges in this volume point up this complexity especially well; talk, as they demonstrate, may use much specialized cultural knowledge about linguistic utterances as well as other cultural knowledge about the nonlinguistic world being talked about.

8

NAOMI QUINN & DOROTHY HOLLAND

It is also a misleading simplification to imply, as Caws has, that one set of models (those guiding behavior, in his formulation) are out of awareness, whereas another set (those said to guide description) are not. It is no doubt true that some knowledge is more habitually, hence more readily, put into words than other knowledge; that some knowledge but not other knowledge is tidily "packaged" in memory, hence easily retrieved for the telling; and that some knowledge is under conscious and voluntary control whereas other pieces are less available for introspection and articulation. Hutchins, in this volume, provides an instance of .the latter case: a case in which inferences attributed by the analyst to the speaker in order to account for her interpretation of a Trobriand myth, appear to be out of the awareness of the speaker herself, Hutchins presumes, because they are so painful as to be repressed. At another extreme, some linguistic outputs, but by no means most, have the "canned" quality of well-worked and well-rehearsed rationalizations or idealizations. Perhaps ethnographers are especially likely to be proffered such accounts. Much of people's cultural knowledge, however, is likely to be somewhere in between these two extremes of accessibility and inaccessibility - as D'Andrade (this volume) found for the American college students he interviewed about the way the mind works. These interviewees could not provide a comprehensive, well-organized view of the entire cultural model of the mind but could certainly describe how it operates when they were asked questions about specific examples. Models such as this one of the mind, which people use in a variety of tasks such as making inferences and solving problems (for a different example, see Jorion 1978), will be brought into awareness and made available to introspection and articulation to varying degrees depending on the precise demands of those tasks for such introspection or articulation.; Equally, knowledge embodied in cultural formulations that Caws might want to call "representational," cannot easily be distinguished from "operational" models with regard to the function he assigns the latter, of guiding behavior. Well-articulated cultural models of the world may also carry "directive force" (a term borrowed from D'Andrade 1984a). An obvious example, provided by White in this book, is that of proverbs. Proverbs promote enactment of the dictums they contain, White argues, precisely because their formulaic and linguistically economical construction signals cultural wisdom. This claim on wisdom is enhanced by present tense verb forms, which give them a timeless, enduring quality, and by_„their disallowance of exceptions or hedges, which grants them a seeming universal validity. Cultural models, then, are not to be understood in either-or terms. That various anthropologists have proposed to sort cultural understanding into a kind for thinking and a kind for doing and to associate talking with the former may reflect more about the mind-body duality in our own

CULTURE AND COGNITION

9.

western cultural model of the person than it does about how.cultural knowledge is actually organized. .. .•;--h.<:• '-•:.•• TALKASACTION

.-.:.;.•. :> v : ;,,;., : ;-/i^.^

Were its only claim t o be able t o account for what people say^the presententerprise would still be an important one. The dismissive materialist stance that cultural models influence little more than talk neglects t h e pivotal social function of talk itself. A s m o d e m sociolinguistics teaches,us, talk is one of the most important ways in which people negotiate understand-; ing and accomplish social ends. Of course, discourse can be crucial to. the.; efforts of individuals t o create inner meaning for themselves, as illustrated vividly by Hutchins's analysis, in this book, of a Trobriand woman's a t - . tempt to comprehend her own experience in the terms of a familiar myth. However, these shared cultural understandings also figure large in the creation of social meaning. In Trobriand litigation, which is the subject of Hutchins's (1980) recent book, spoken claims and counterclaims are consequential acts. For the college-age women whom Holland and Skinner describe in this volume, labeling another woman's fiance" a nerd is not just inconsequential chatter. The illness stories Price collected from poor Ecuadorian city> dwellers (this volume) reveal the efforts to which people will go in order to establish public, legitimated accounts of their behavior (see also Early 1982). Lutz (this volume) details a case in which the future course of kin relations depends o n the accepted interpretation of an incident, an interpretation that emerges as the kinspeople involved talk t o one another, pro' posing and negotiating different possible emotional definitions of the event (see also Frake 1977; Young 1981). Other papers in the collection suggest how cultural models undergird such varied kinds of talk as negotiations about the justification for anger, marital disagreements, proverbial and other advice about the solution t o everyday problems, and inquiries into suspected lies. Such talk, in. turn, influences social relations among people and the subsequent actions they take toward one another. Talk is itself a kind of act, and speech acts can have powerful social consequences. THE DIRECTIVE FORCE OF CULTURAL MODELS

How do cultural models, whether invoked to persuade another or to order one's own inner experience, motivate behavior? The papers in this collection reveal differential sources of motivational force: One basis is in the authority and expertise with which cultural models may be invested, another in the intrinsic persuasiveness these models themselves have for us. White's analysis of proverbs, as mentioned, suggests that linguistic forms can grant a certain amount of persuasiveness to knowledge by packaging it as "cultural wisdom." Relatedly, Linde shows how explanatory systems for human behavior that are devised by one group of culturally

10

NAOMI QUINN & DOROTHY HOLLAND

designated experts - academic psychologists - have come to provide us with models for making our own life choices. This is so even though neither; ordinary people nor the "expert" psychologists themselves agree oh a single explanatory system. Cultural understandings would seem to gain force from their identification with expert knowledge and cultural .wisdom, in spite of the availability of alternative, equally expert explanatory systems . . and contradictory, equally wise-sounding admonitions. •.; -••; :• >..•.. Even though expert validation and cultural authority play a role in the persuasiveness of cultural models, explanatory adequacy in the face of; our experience can also be compelling. This effect is perhaps best illustrated in the present collection by Kempton's example of an informant who switches from a "valve" theory to a "feedback" theory of home heating in mid-interview, after realizing that the first of these analogies was contradicted by her memory of how an observable heating device actually. worked. Kempton shows elsewhere in his paper how acceptance of one or another of these alternative theories has consequences for thermostat settings. Collins and Gentner's paper, on the other hand, cautions against any conclusion that evidence drawn from real-world analogies is automatically compelling, showing as it does that a thinker such as their Subject PC, who relies heavily on analogies to phenomena he has observed or heard about, may shift among these local analogies without checking their consistency - failing to develop a coherent view of evaporation and often giving inaccurate answers. This tendency of individuals to check their understandings against expert opinion and test them against experience highlights the co-existence of alternative, often conflicting cultural renditions of that world. In the pages of this book, it appears that individuals find it relatively easy to entertain different theories of how the thermostat works and even abandon one theory for another; to combine components of different analogies in their attempted explanations of evaporation; to invoke conflicting proverbial advice for the solution of different problems; and to adopt one. " or another contradictory folk theory of language depending on which one best fits the linguistic case at hand. The latter example, of two contradictory folk theories of language, \ prompts Kay (this volume) to observe that cultural models are not to be thought of as presenting a coherent ontology, a globally consistent whole, in the way that the expert's theory is designed to be. Cultural models are ' better thought of, in Kay's view, as resources or tools, to be used when suitable and set aside when not. That there is no coherent cultural system of knowledge, only an array of different culturally shared schematizations formulated for the performance of particular cognitive tasks, accounts for the co-existence of the conflicting cultural models encountered in many domains of experience. What is not accounted for, in this view, is the • degree of apparent systematicity, best characterized as a thematicity, that does seem to pervade cultural knowledge as a body. In the final section v

CULTURE AND COGNITION


of this introduction, we argue that this thematic effect arises from the ; availability of a small number of very general-purpose cultural models: that are repeatedly incorporated into other cultural models developed, for; special purposes. This account of cultural thematicity does not rule but the kind of contradiction arising among variant cultural models that Kay and other volume authors describe. t Some cultural understandings people have, such as the models of men- J , tal processes, emotional states, marital commitment, career choice, gender relations, and kinship obligations described in this book, have a different feel from our models of heating devices. The metaphor of conceptual '^••••.. models as tools to be taken up and put down at will does not fit these other cultural models very well. They are compelling in a way that does not depend on what the experts say and often seems highly resistant to ; revision in the face of apparent contradiction. Largely tacit and unexam- '•' ined, the models embed a view of "what is" and "what it means" that seems wholly natural - a matter of course. Alternative views are not even recognized, let alone considered. But more than naturalness, these cultural models grant a seeming necessity to how we ourselves live our lives. . How do ideas gain such force? Partially, the answer lies in what we accept as the typical and normal way of life, judging from the lives of our fellows. When we look around us, we find confirmation for our own lives in the beliefs and actions of other people; cultural models that have force for us as individuals are often the historically dominant models of the time. This is so even though such cultural understandings have certainly undergone historical change, often radical, and certainly have contemporary competitors in any given historical moment. But the force cultural understandings can have is not simply a matter of people's conformity to the dictums popular in their time. In considering the directive force of cultural meaning systems, D'Andrade <1984a:97) returns to the ideas of Melford Spiro (1961), who argued persuasively that much socially required behavior comes to be inherently motivating for individuals, most often because it directly satisfies some culturally defined need (what Spiro called "intrinsic cultural motivation") or sometimes also because it realizes some strongly held cultural norm or value ("internalized cultural motivation," in Spiro's term). As D'Andrade (ibid.:98) summarizes, "through the process of socialization individuals come to find achieving culturally prescribed goals and following cultural directives to be motivationally satisfying, and to find not achieving culturally prescribed goals and not following cultural directives to be anxiety producing." D'Andrade adapts this argument to a cognitivist view of cultural meaning. He suggests that culturally acquired knowledge need not be purely representational, as the term cultural knowledge connotes, but may draw on socialized-in motivation as well. This directive force is "experienced by the person as needs or obligations to do something" (ibid.). Thus, in D'Andrade's (ibid.:98) example, the cultural meaning of sue-

12 '

:

NAOMI QUINN & DOROTHY HOLLAND

cess for Americans, accomplishment may be rewarding because it is.both instrumental in the satisfaction of culturally shaped needs for personal recognition and achievement and an objective that has come;tobe valued in its own right/Both sources of directive are learned as part of the : understanding of success and what it entails. This inner motivation to be successful, along with external sanctions for making a living and providing • for one's dependents, and social pressure toward conformity with the image and the life-style that mark success, together and in interaictipn: overdetermine the motivational component of tthjstcultutal: meaning system. As D'Andrade (ibid.) muses, ^'perhaps what is surprising is that anyone can resist the directive force of such a system." This complex of meaning and motivation is an American preoccupation even though, for most Americans, what constitutes success in our society is actually unattainable. In the course of human socialization, directive force seems to become attached to those understandings, such as the meaning of success for Americans, that are most closely bound up with the sense individuals have of themselves and the sense they make out of their lives. Perhaps such understandings, including culturally provided understandings about oneself and one's place in life, organize our knowledge of what D'Andrade (n.d.:23) has described as "highly general conditions which people want to bring about or avoid." Cultural models of self and life organize what are, literally, vital understandings. These understandings - however differently they may be delineated in different cultures - become, again in D'Andrade's (ibid.) words, "the most general source of'guidance,' 'orientation,' and 'direction' in the system." Socialization experiences may differ sharply in the degree to which they endow a given cultural model with directive force for an individual. Thus, "where there's a will there's a way," to the degree that this common proverb frames a model of the self as the agent of one's fate, may have special force for individuals whose socialization has led them to think of themselves as the sole or primary agents of their own fate. Other individuals, who learn from a quite different socialization experience that they are relatively powerless and blameless with regard to their own fate, > may, like the interviewee quoted in Linde's paper in this volume, find . behavioral psychology a particularly persuasive interpretation of their lives. Consider another example. Just as Americans learn to think of • themselves and their lives in terms of success, many American women grow up with the teaching that marriage is the measure of a woman's success " in life. If this lesson is amplified, as it was for one of Quinn's interviewees whose mother conveyed to her a personal sense of failure for having been unable to hold on to a husband, then the idea of marital success becomes conceptually powerful in the extreme. Thus, as D'Andrade (n.d.:23) points , out, what he calls "lower-level schemas," such as the model of marriage in this example, act as goals only when "recruited" by some more general ;

CULTURE AND COGNITION

* 13

"upper-level schema which is currently functioning to 'guide,':'orient' and 'direct'theflowof action," such as the model of the successful life in this example. Cultural models of all kinds gain directive force when theyare. recruited, whether in the course of uniquely individual experiences or those' more widely shared, by understandings of oneself arid one's ;HfeyX^ Consideration of the potential directive force of cultural models brings! us to Keesing's concern with the ideological force of some of these models and their use as instruments of ideological hegemony. Social life, as Spiro (1961) saw, depends on the fit between what is socially required and what is individually desired. So, too, the designs of those who would rule society, and those who would benefit from this control over others, depend upon the willingness of the populace to fill its role in these plans. Therefore, states and other agencies promulgate ideology persuading people to do what they otherwise might question or resist doing. In spite of the resources and power that may be brought to such attempts at persuasion, it is not always effective. To be successful, ideologies must appeal to and activate preexisting cultural understandings, which are themselves compelling. Even though ideologues may mold and adapt cultural models to their own devices, and often show a great deal of genius for doing so, they do not create these cultural ideas de novo, nor are they able to guarantee the power of any given cultural model to grip us. Specifically, Lewontin et al. (1984:64) observe that to be convincing, an ideology must pose as either legitimate or inevitable. For "if what exists is right, then one ought not to oppose it; if it exists inevitably, one can never oppose it successfully." These ideas about what is right and what is inevitable are largely given by cultural models of the world. The point made by Lewontin et al. leads to a further observation: Among alternative versions of what is legitimate and what is inevitable, a given ideology is most compelling if its Tightness engages the sense one has of one's own personal uprightness and worthiness, or if its inevitability engages the view one has of one's own inherent needs and capacities. These matters lie at the heart of our understanding of ourselves and our place in life. They also are largely cultural matters. Perhaps the contribution cognitive anthropology is poised to make (and poised may be slightly too optimistic a word) toward the study of "humansin-society" is this: insight into those conditions under which cultural models are endowed with directive force and hence with ideological potential.

A short history of methodological strategy The point from which the previous section departed, cognitive anthropology's focus on linguistic phenomena as the behavior to be accounted for, has proved to be a richly productive strategy, as the papers in the present volume illustrate. In the course of the enterprise, it will be seen, the original view of the relationship of language to culture, with which cognitive anthropology set out, has undergone significant modification.

14

NAOMI QUINN & DOROTHY HOLLAND

Our cultural understanding of the world is founded on many tacit assumptions. This underlying cultural knowledge is, to use Hutchins's (1980:12) words, "often transparent to those who use it. Once learned, it becomes what one sees with, but seldom what one sees" This "referential transparency" (ibid.), we note in the previous section, causes cultural knowledge to go unquestioned by its bearer. At'the same time, this transparency has posed an absorbing methodological problem for the analyst: how, and from what manner of evidence, to reconstruct the * cultural models people use but do not often reflect on or explicitly articulate. The problem has remained central to cognitive anthropology, but approaches to it have changed. Early efforts sought to describe the semantic structure of lexical domains. If analysts could recover or reconstruct what one needed to know / in order to label pieces and portions of the world correctly in the native's [ own language, it was reasoned, then the resulting model would capture"] an important part of those people's culturally constructed reality. Such analyses produced the formal taxonomic and paradigmatic descriptions} for which the emerging enterprise variously called "ethnoscience," "ethnographic semantics," and "the new ethnography" became known and with! which cognitive anthropology, evolved out of these earlier efforts, has been persistently associated long after its practitioners began exploring networks of semantic relations, schemas for decision making, and other alternatives to taxonomic and paradigmatic models (D'Andrade n.d.:19). The semantic structures recovered in these earliest analyses did provide insight into the organization of some domains of thelexicon. However, the organization of lexicon was soon recognized to offer only limited insight into the organization of cultural knowledge (D'Andrade 1976; D'Andrade et al. 1972; Good & Good 1982; Howe & Scherzer 1975; Lave et al. 1977; Randall 1976; White 1982). Notwithstanding the primacy attributed to referential meaning in the western positivist/empiricist tradition, what one needs to know to label things in the world correctly did not prove to be the most salient part of cultural meaning. Formal semantic analysis did not uncover the cultural models that individuals invoked for the performance of such naturally occurring cognitive tasks as categorizing, reasoning, remembering, problem solving, decision making, and ongoing understanding, but gave only such partial and selective glimpses of those models as had come to be embedded in the lexical structure. In the tradition of formal semantic analysis, special tasks were devised that induced subjects to rely on lexical structure for their performance; as Randall (1976) first pointed out, however, naming and discrimination tasks such as these are infrequently encountered in the ordinary course of life. Even the "psychologically real" analyses of people's judgments of semantic similarity, which followed on the heels of formal semantic analysis, proved to be of limited insight into the organization of cultural models.

CULTURE AND COGNITION

• .:\,fe;;f3.;v;,-::.

Such analyses revealed that people brought som^^ngmoT^jn^^^^^i'.^.'1'. of cultural understanding than word knowledge or even ('encyclopae|icJ?./h :'£ knowledge, to use Sperber's (1975:91-94) term, to their improvisations">;:/•"';'.?• of these unfamiliar sorting tasks. But what this something more was still ~ : • had to be,filled in. ....... •!^^^%ov((:••••. Several papers in this volume represent the culmination of this methodp-. logical tradition. Both the paper by White and that by Holland andSkinner : show how additional analysis of natural discourse must be introduced to make sense of the results of multidimensional scaling (and by implication : other multivariate analyses) of semantic similarity judgments. Holland and Skinner's paper is a particularly telling critique of the tradition in cognitive anthropology that has relied exclusively on the interpretation of such semantic similarity results. Their analysis argues that the items composing the lexical domain of gender labels used by college students are related in an interesting but oblique way to these students' presupposed knowledge of gender relationships. The terms label individuals who violate cultural expectations about the course of normal relationships between males and females. To understand these labels, one must understand the presumed relationship. An interpretation based on labels of gender types alone, then, would be missing the central assumptions of the cultural model. Lutz makes a similar point in her paper about Ifaluk emotion words: The meaning of these words cannot be fully grasped from an analysis of the words alone; one must have an understanding of the Ifaluk ethnothepry of emotion that underlies them. This is not to argue that semantic similarity-based multidimensional scaling analyses and other such techniques should be discarded. People do sometimes use semantic similarity of terms to accomplish such natural tasks as inferring information about acquaintances (D'Andrade 1965; 1974). Moreover, both White and Holland and Skinner demonstrate the utility of the method of analysis as a preliminary step in recovering cultural models. Elsewhere than this book, D'Andrade (1984b; 1985) has made the same use of these scaling techniques; he shows the considerable advantage of such analysis for sketching in the broad outlines of a large domain of American culture, that of person perception. Such an approach is highly efficient but relatively crude. It necessarily sacrifices depth for scope; description of how particular parts of the model work for rapid identification of key components and orientation of these components relative to one another. Such scope is important because, as is discussed further in the final section of this introduction, cultural models appear to interpenetrate one another, some of general purpose playing a role in many other more special-purpose models. Thus, for example, in this volume, assumptions about relations among thought, intention, and action, which figure in a folk model of the mind argued by D'Andrade to be widely shared by Americans, are shown by White to underlie our model of problem solving. Assumptions about difficulty, effort, and success, which

16

NAOMI Q0INN & DOROTHY HOLLAND

D'Andrade (1984b) has shown to be part of a shared American model of task performance, play a role in Americans' model of marriage, as de• scribed by Quinn (1985, this volume). In Samoan thought, Clement (1982) has found notions of valued social identities to underlie understanding of mental disorders. A sweeping view identifies these interconnections. Thus, a methodological division of labor seems to be emerging. Multivariate analyses of semantic similarity judgments, techniques by which a relatively large quantity of data can be efficiently collected and effectively reduced, are used to sketch in a map of the territory and orient it relative to other domains, while different methods provide the higher resolution needed to explore a given terrain closely. New approaches to the investigation of cultural models, then, reflect a recognition that the relationship between a model and any regularities in the terminology of semantic domains referenced by this model is likely to be complex and indirect at best. Consequently, there are no mechanical procedures by which the former can be derived from the latter. Now, word meaning and, indeed, all of language are viewed as holding possible clues to the underlying cultural knowledge that enters into linguistic and other behavior. Reconstructing the organization of this cultural knowledge, however, requires kinds of linguistic data richer in such clues than the data provided by naming and sorting tasks, and it requires eclectic exploitation of all possible sources of such data. The major new data sources that cognitive anthropologists have adapted to the task of reconstructing cultural models, represented in this volume, are two: systematic use of native-speaker's intuitions, and analysis of natural discourse. For many nonanthropologist practitioners of cognitive science, of course, neither method is new. The former is exemplified in the volume papers by Lakoff and Kfivecses, by Sweetser, and by Kay, all linguists for whom the method of developing one's analytic model out of one's native-speaker's intuitions, and verifying this model against fur- ,. ther intuitions, is a matter of disciplinary canon. Using his own native speaker's intuitions represents a methodological departure, however, for anthropologist D'Andrade, who draws not only on his intuitions about the language of mental processes, but also on a long tradition of introspec- ; tion about such matters by philosophers. What all these papers suggest is that the intuitions of native speakers , about their language are heavily dependent on the intuitions of these natives as culture-bearers. Sweetser, for example, demonstrates elegantly how our •> judgment that some speech act is or is not a lie, depends on cultural assumptions about the simplified worlds of communication and mutual assistance in which such acts occur. Kay shows that the co-existing, alternative folk theories of language that lie behind the two hedges, loosely speaking and technically, depend on cultural assumptions about the nature of truth and the authority of experts, respectively. It is of interest that neither Sweetser, on the one hand, nor D'Andrade "

CULTURE AND COGNITION

47

on the other are comfortable with a cultural analysis validated Solely against their native-speaker's intuitions. Sweetser goes on to show fthat the model she constructs on the basis of her own introspection and the accounts of linguistic philosophers can parsimoniously'account for experimental findings of Coleman and Kay (1981), who Elicited subject's judgments as to whether a lie had been told in each of a series iof systematically varied hypothetical cases. D'Andrade demonstrates that interview responses to questions about mental events are explicable in terms of his model of the mind. These efforts at independent verification bf analyses derived from native speaker's intuition, against the linguistic responses of other speakers, can be interpreted as attempts to satisfy alternative standards of evidence that co-exist in multidisciplinary enterprises such as the one in which this group of cognitive scientists is joined. This strategy of building accounts from native speaker's intuitions and then testing them against other, independent observations can be expected to become a methodological hallmark of future investigations into cultural knowledge. The models developed in other papers rely heavily on another method likely to become a mainstay of the new enterprise. This is an eclectic kind of discourse analysis fashioned, as necessary, out of borrowed parts. An important source of inspiration for this methodological approach has been Linde, a linguist whose earlier work on discourse types (1978; n.d.) has influenced most of the anthropologists in the group. Many of these papers - most explicitly, those of Hutchins, Kempton, Quinn, Collins and Gentner, and Linde herself - show how the type of discourse Linde calls explanation can be exploited to reveal the cultural models that underlie speakers' reasoning. Kempton, for example, infers the underlying folk theories of home heating devices that informants hold from the metaphors they use, as Quinn infers from interviewees' metaphors the underlying propositions they are asserting about marriage. Collins and Gentner are able to identify a limited number of schemas or "component models" that recur in their subject's explanations of evaporation. Likewise, Linde uncovers a small number of recurrent "explanatory systems" identifiable by characteristic themes in American interviewees' explanations of their occupational choices, such as the "split self" theme, which is part of the Freudian explanatory system, and the "non-agency" theme, which is characteristic of the behaviorist explanatory system. Price's paper mines another discourse type discussed by Linde, narrative, in Ecuadorian stories about illness episodes, to reconstruct cultural understandings about familial roles from their "traces" in these narratives: what narrators highlight, elaborate, leave unsaid, mark with counter-examples, and comment on in affective propositions. Another powerful influence on several of the anthropologists in this group has been Hutchins's (1980) book on Trobriand land litigation. Hutchins demonstrated how explanation in natural discourse could be

18

NAOMI QUINN & DOROTHY HOLLAND

decoded to reveal cultural schemas for the propositions on which the argument of the discourse was based. Schemas,' in Hutchins's usage, state prppositional relations in terms of variable ranges, so that a given schema serves as a "template" from which any number of propositions can be constructed (ibid.:51). Trpbriand litigation over land transactions, however, is a specialpurpose discourse that uses a limited set of such schemas composed of highly technical information about the specific rights in landthatmay be transferred as a result of particular prestations. Could his approach be used to discover the schematic structure underlying more generalpurpose explanation (Quirm 1982a)? Three papers in this volume, by Lutz, Quinn, and Hutchins himself, would seem to answer "Yes." : •:<>;.;-. Lutz analyzes word definitions, natural instances of word use, and more general propositional .statements elicited from interviewees to reveal the "basic level schemas" that enter into the Ifaluk cultural model of emotion and how these schemas concatenate to form statements and inferences about common situations and their associated emotions. Quinn's analysis identifies stable proposition-schemas and schemas of chained propositions used in reasoning about marriage. As she shows, it is necessary first to decipher the metaphorical speech in which propositions are cast, the referencing of earlier propositions by later ones, and the causal constructions linking one proposition with another, in order to reveal the common underlying schemas in this talk. Hutchins cracks an even less obvious code, showing that mythic schemas, as disguised representations of their repressed thoughts and fears, enable Trobrianders to reason about their relations to deceased relatives. Key to his interpretation is the identification of the propositional structure of the myth with an analogous structure outside the myth, in a "relevant bit of life." ,, Thus, although it is fair to say that much of the original ethnoscientific enterprise was driven by a seemingly powerful method - semantic analysis - and constrained by the unforeseen limitations of that method, the same is not true of modern cognitive anthropology. Current efforts are more intent on theory building than on the pursuit of any particular methodology. The theoretical question is: How is cultural knowledge organized? The methodological strategy is to reconstruct the organization of this shared knowledge from what people say about their experience. To this strategy, cognitive anthropology has adapted some of the time-honored methodological approaches of linguistics.

An account of cultural knowledge from artificial intelligence This volume presents some initial answers to the question: How is cultural knowledge organized? In doing so, it makes a contribution, not to the field of cognitive anthropology alone, but to the multidisciplinary enter/-

CULTURE AND COGNITION

19

prise of cognitive science. Cognitive science asks: How is knowledge organized? However, the central role of culture in the organization of this everyday understanding has only recently begun to be appreciated by cognitive scientists. Efforts within artificial intelligence to model understanding by computer confront culture when, as is often the case, the task sojutibii to be modeled depends on preexisting knowledge of the sort human beings draw on so readily. . .-.\6^:-A,:.4;..-.^JA ..o;^i:..^ Robert Abelson (1975:276) has referred to the difficulty of incorporating this knowledge into computer simulations as the -size problem,'':cdnciuding that "there is too much common sense knowledge of the world in even the humblest normal human head for present computer systems to; begin, to cope with." Recognizing that most artificial intelligence has avoided the problem either by dealing with very restricted domains, or by modeling very general cognitive mechanisms that work in principle but never operate in actual situations, Abelson himself has attempted, with Roger Schank (1977), to design a more knowledgeable understander. Because theirs is arguably the most thoughtful attempt, from this quarter of cognitive science, to build cultural knowledge into understanding, and because their formulation is widely known to cognitive science audiences, a brief discussion of their work will be useful in order to say why anthropologists find it lacking and to compare it with the approach represented in this volume. Schank and Abelson (ibid.) begin with the notion of scripts as basic building blocks of our everyday understanding. Scripts, derived from daily routine, are standardized sequences of events that fill in our understanding of frequently recurring experiences. The "restaurant script,": now famous in cognitive science circles, guides the customer through the series of interchanges required to get a meal at a restaurant - getting seated, ordering, paying, and even sending unacceptable food back to the kitchen or adjusting the size of the tip to reflect the quality of service received. All this is mundane, but undeniably cultural, knowledge. (A strikingly similar approach to cultural knowledge of routinized events has been offered by anthropologist Charles Frake, who provides an analysis of such a routine in another culture; see Frake 1975; 1977.) t h e cultural models to be described here bear an intriguing resemblance to Schank and Abelson's scripts. Their enactment is not tied to a concrete physical setting, as is that of the restaurant script. They do, however, have two features that Abelson (1981:3) has singled out to characterize scripts: The casual definition of a script is a "stereotyped sequence of events familiar to the individual." Implicit in this definition are two powerful sources of constraint. One is the notion of an event sequence, which implies the causal chaining of enablements and results for physical events and of initiations and reasons for mental events. . . . The other constraint generator comes from ideas of stereotypy and familiarity. That an event

20

NAOMI QUINN & DOROTHY HOLLAND

sequence is stereotyped implies the absence of fortuitous events. Also, for events to be often repeated implies that there is some set of standard individual and institutional goals which gives rise to the repetition. The papers in this volume illustrate how our knowledge is organized in culturally standardized and hence familiar event sequences that tell, for example, how marriage goes (Qutnn); or how anger is engendered, experienced, and expressed (Lakoff & Kovecses); or under what circumstances a lie has been told (Sweetser); or what to expect in a relationship between two young adults of opposite gender (Holland & Skinner); or that wishes give rise to intentions and intentions to actions (D'Andrade). These "stories" include prototypical events, prototypical roles for actors, prototypical entities, and more. They invoke, in effect, whole worlds in which things work, actors perform, and events unfold in a simplified and wholly expectable manner. These events are chained together by shared assumptions about causality, both physical and psychological, as Abelson's characterization of scripts suggests. Abelson's casual definition of a script has much in common with what we here call a "cultural model" (or sometimes, a "folk model") to capture both its dynamic role in guiding expectations and actions and its shared possession by the bearers of a culture. To this point, the account of shared knowledge rendered by Schank and Abelson is not dissimilar to our own. Beyond scripts, however, the two accounts begin to diverge. The first difference is one that would strike any anthropologist. Schank and Abelson are not explicit about the cultural nature of the knowledge they invoke. They write of "well-developed belief systems about the world" (ibid.:132); however, they tend to attribute such belief systems to pan-human experience of how the world works (ibid.:119) rather than questioning whether some of these belief systems might be unique to our own culture. Without trying here to settle the big question of cross-cultural universals in human thought (a question D'Andrade and Sweetser address in this volume), we assert that many of Schank and Abelson's examples invoke knowledge peculiar to Americans. Cultural knowledge is key to the higher-order structures that embody goals in Schank and Abelson's formulation. As their inventors were the first to point out, all is not scripts. There is more to understanding than knowing how get a meal at a restaurant and how to execute the numerous other scripts and plans for carrying out all our daily objectives. As Schank and Abelson are led to ask, how do these goals themselves arise? How ,, are story understanders and other observers of the everyday world able to assess actors' goals and predict their future goals? Schank and Abelson's answer is that related goals are bundled together in "themes." These themes are said to generate actors' goals as well as other people's inferences about these likely goals. It is possible to make such inferences about the goals' of other people, presumably, because knowledge of themes, no less than knowledge of scripts, is shared.

CULTURE AND COGNITION

21

Anthropologists found this account at once provocative and unsatisfactory. Themes generate related goals. But how are they related? Schank and Abelson (ibid.) propose three types of theme: "role themes;" like WAITRESS or SHERIFF; "interpersonal themes/Mike MARRIED or LOVER; and "life themes," like SUCCESS or LUXURY LIVING; Each of these labels conjures up to the anthropologist a vast store of cultural knowledge. However, merely naming themes MARRIED or SUCCESS begs the question of how this shared knowledge of being married or achieving success organizes the goals we associate with these respective themes. Perhaps because Schank and Abelson supply examples drawn from their own cultural knowledge, which has a seeming naturalness for them; they take for granted in their theoretical formulation the same complex knowledge that ordinarily goes unquestioned in their everyday lives. Schank's (1982) more recent reformulation of the theory of scripts is much more sophisticated about how knowledge must be hierarchically organized and continually modified in memory in order to account for such processes as reminding and the generalization of learning. At the same time, however, Schank's newer account more glaringly exposes the inadequacy of a theory of the organization of knowledge that gives an insufficient role to how human beings acquire most of their knowledge, especially their most general understandings. Failing to make a place in his account for knowledge that is culturally shared and transmitted, Schank is left with the awkward supposition that an individual's understanding of the world is accumulated through the painstaking generalization of knowledge from one firsthand experience to another. It is difficult to imagine how people could learn as much as they know, even by the time they reach adulthood, from personal experience alone. Many of Schank's favorite examples, such as that of learning a different routine for ordering, paying, and eating in fast-food restaurants than that followed in regular restaurants, may represent the kind of detailed knowledge of setting-specific conventions that is, in fact, normally picked up in personal encounter with each new setting. However, others of his examples are less readily assimilated to this model of learning from actual experience. Knowledge of the "societal conventions" (ibid.:98) surrounding the idea of CONTRACT, for instance, is said to be generalized from successive experiences in which services are procured - meals at restaurants, visits to doctors' offices, home visits from plumbers, and the like. It is implausible to suggest that people learn all they need to know about such complex cultural matters as are embedded in CONTRACT or MARRIED or SUCCESS (e.g., that MARRIED has something to do with the other two) solely from successive experiences with actual contractual relations, marriages, and personal successes. Indeed, we know that individuals have sizable expectations about such things before ever experiencing them personally. Moreover, as Schank stresses, what happens to different people and how they respond to these experiences differs; if direct experience were

22

NAOMI QUINN & DOROTHY HOLLAND .

the only source of knowledge, then each individual's understanding of the world would diverge from that of every other individual. Indeed, this is what Schank is led to conclude: "There is no reason," he notes (ibid.:224), "why structures that are based on experience should bear .a relationship to any other person's structures." But of course, even allowing for'their unique individual perceptions of the world, people somehow,do end up with considerable shared knowledge. To pursue one of Schank's. examples,, the homeowner who calls a plumber about a leaking pipe and the plumber who comes to replace it can negotiate a contract between them, even though they have never met one another before, grew up in different parts of the country, and have entirely different class and- ethnic backgrounds. How this comes to be is left unexplained in Schank's formulation. The research presented here assumes that individuals are heir to a great deal of knowledge about the world that they do not necessarily draw from firsthand experience. Cultural knowledge is typically acquired to the accompaniment of intermittent advice and occasional correction rather than explicit, detailed instruction; but it is learned from others, in large part from their talk, nonetheless (D'Andrade 1981). This is perhaps most clearly illustrated by highly abstract ideas, such as the theory of relativity, philosophical arguments about the meaning of existence, or cultural conceptualizations of self and group identity, which are transmitted and perpetuated through language and could hardly be learned without it (Holland 1985:406-407). There is perhaps no experience, however concrete or however novel, that is not informed in some way by the culturally transmitted understandings an adult individual brings to that experience. The work in this book goes on to address the question: How does this received knowledge organize our understanding? Cultural models, as conceived by this volume's authors, play the conceptual role blocked out by "themes" in Schank and Abelson's original formulation. Cultural models frame our understanding of how the world works, including our inferences about what other animate beings are up to, and, importantly, our decisions about what we ourselves will do. With Lutz (this volume), we want to claim that many of our most common and paramount goals are incorporated into cultural understandings and learned as part of this heritage. An account of cultural models from prototype theory Our view of cultural models has obvious connection to ideas about proto- , types, which have figured importantly in cognitive scientists' recent discussions of knowledge representation. Event sequences played out in "simpli{, fied worlds" (Sweetser's term, this volume) appear to serve as prototypes 7? for understanding real-world experience. The notion that schematic struc- tures, or schemas of some kind, systematically organize how experience is understood, has wide acceptance in cognitive science, including cognitive anthropology (Casson 1983). That prototypes (e.g., the most representa- "

CULTURE AND COGNITION

23

tive members of a category) might serve as schemas for categories of t!hiiigs, is an appealing implication of Rosch's experimental research on categorization (fully reviewed in Rosen 1977; as Rosch stressed in a 1978 assessment of this work, however, the identification of prototype effects is not to be mistaken for a theory of mental representation accounting for these effects). Anthropologists are perhaps more familiar with the prototype notion from its original application to color categories in the work of Paul Kay and his associates (Berlin & Kay 1969; Kay & McDanie! 1978). The extension of the notion from its earliest application to color and such physical objects as birds and furniture, to prototypical event sequences, has reached anthropology through linguistics. Linguists first came to see the necessity of incorporating cultural knowledge into their accounts of word use. Perhaps the single most-cited example of a folk model, at the conference from which this volume grew, was the analysis of bachelor provided by linguist Charles Fillmore (1975; 1982). Fillmore had argued that traditional "checklist" definitions of words such as bachelor were inadequate. In the checklist view, a bachelor is a man who has never been married (Katz & Fodor 1963:189-190). However, as Fillmore pointed out, this definition utterly fails to explain why we do not consider, for example, the Pope to be a bachelor, Or a wolf-boy grown to maturity in the jungle. (Or, we might be tempted to add if we were linguists, a male victim of brain damage who has been in a coma since childhood.) This critique of the traditional linguistic approach to word definition parallels anthropologists' dissatisfaction with componential analysis of lexical sets discussed in an earlier section: Both accounts appear to leave out a crucial part of what speakers have to know in order to use a word or a system of terminology. The alternative Fillmore proposed is that the word bachelor "frames," in his term, a simplified world, in which prototypical events unfold: Men marry at a certain age; marriages last for life; and in such a world, a bachelor is a man who stays unmarried beyond the usual age, thereby becoming eminently marriageable. (Fillmore might have noted that the bachelor's female counterpart, the spinster, suffers a different fate.) Here is an example of a folk model presumed by a single word. A similar analysis, of the word orphan, has been offered by the linguist Ronald Langacre (1979). Quinn (1982b) has argued that a cultural model of difficult enterprises underlies the polysemous meanings of the word commitment, as this word is used in reference to marriage. Fillmore has elsewhere (1977) suggested that a set of related verbs from the domain of commerce can be understood as elements in "the scene of the commercial event," which is activated by use of one of these words, such as buy or pay. Several of the volume papers emerge directly from this linguistic tradition, an approach Fillmore has labeled "frame semantics." Sweetser's analysis of lie is perhaps the most sustained linguistic analysis of the

24

NAOMI QUINN & DOROTHY HOLLAND

simplified world required to explain use of a single word. Kay's reconstruction of the folk theories about language and speech that inform use of loosely speaking and technically extends this tradition to the analysis of hedges. The paper by Lakoff and Kovecses can be viewed as extending the same linguistic approach to another feature of language, metaphor. They show that American English metaphors for anger are structured in terms of an implicit cultural model of human physiology and emotion, which they delineate. Understandably, linguists are most concerned with the important implications of underlying cultural models for their theories of word definition, metaphor, polysemy, hedging, and other linguistic phenomena (but see Lakoff 1984). Anthropologists tend to orient their analyses in the opposite direction, treating linguistic usages as clues to the underlying cultural model and working toward a more satisfactory theory of culture and. its role in such nonlinguistic tasks as reasoning (Hutchins 1980; and the papers by Hutchins, by Lutz, and by Quinn in this volume), problem solving (Kempton this volume; White this volume), and evaluating the behavior of others (D'Andrade 1985; Holland & Skinner this volume; Price, this volume). However, the different questions that draw linguists and anthropologists should not obscure the common insight that brought together this particular group of linguists and anthropologists in the first place: that culturally shared knowledge is organized into prototypical event sequences enacted in simplified worlds. That much of such cultural knowledge is presumed by language use is as significant a realization to anthropologists as to linguists. For the latter, these cultural models promise the key to linguistic usage; for the former, linguistic usage provides the best available data for reconstruction of cultural models. „• The forms cultural knowledge can take How is the knowledge embodied in cultural models brought to the various cognitive tasks that require this knowledge? Lakoff (1984) offers some extremely helpful starting suggestions about types of cognitive models, observations that are as applicable to the culturally shared cognitive models described in this volume as to the more idiosyncratic cognitive models individuals devise. PROPOSITION-SCHEMAS AND IMAGE-SCHEMAS

Lakoff (ibid.: 10) makes a useful distinction between what he calls propositional models and image-schematic models. Consonant with the work of Hutchins (1980), which demonstrates the utility of a notion of culturally shared schemas for propositions and sets of linked propositions, we adopt y the term proposition-schema to refer to Lakoff's "proposition^ model," • and for parallel syntax, image-schema to refer to LakofPs "imageschematic model." Image-schemas and proposition-schemas, then, are two r;

25:

CULTURE AND COGNITION

alternative forms in which knowledge may be cast. (Since a cultural model may be recast in one or the other type of schema, or may use the two; in combination, it seems clearest to reserve the term model, in the present) discussion, for the entirety of a prototypical event sequence embedded in a simplified world and to talk about "schemas"as reconceptualizatipjis; of given cultural models, or components of such models,for .particular; cognitive purposes. The reader should be/ware, however,qf thediffering::. and conflicting uses of model in related literature, including some papers in-; this volume; as is typical of new theoretical endeavors, this one has not yet < gotten its terminology under control.2). Indeed, we argue, proposition^ schemas and image-schemas seem suited to different kinds of cognitive tasks.

I

•;•.,*"•.•,'•*:'.*/>

In the present volume, various papers illustrate proposition-schemas, with D'Andrade's describing perhaps the most complex set of related proposition-schemas, those of Lutz and Quinn showing how fixed schemas of related propositions may be used in reasoning, and that of White showing how the proposition-schemas underlying proverbs may be invoked for problem-solving. Proposition-schemas specify concepts and the relations, which hold among them (Hutchins 1980:51; Lakoff 1984:10). As Quinn points out for reasoning about marriage, in the discourse type Linde calls explanation, the causal assumptions connecting proposition to proposition are often dropped out, making these connections seem "empty." In fact, the reasoner, and any listener who shares the same knowledge, can fill in the missing information as necessary for clarification. The capability, afforded by proposition-schemas, of dropping out this detailed knowledge allows speakers to present relatively lengthy arguments and arrive at their. conclusions with reasonable economy. Much more generally, the stable, culturally shared proposition-schemas available for instantiating such causal chains not only facilitate the task of communicating familiar inferences about the world but also allow these inferences to be made swiftly and accurately in the first place. r This is an implication of the ability shown by Quinn's American interviewee, Lutz's Ifaluk informants, and the Trobriand litigants in Hutchins's (1980) study alike to work readily through relatively complex reasoning sequences. It is brought home in a study by D'Andrade (1982), who demonstrates the dramatic improvement in American university students' performance of a reasoning task requiring a complex contrapositive inference, when abstract logical values are replaced with familiar concrete relationships such as that between rain and wet roofs. Presumably, that causa! relation, like the relation Americans recognize between marital difficulties and impending divorce, and that the Ifaluk recognize between the emotion of ker and subsequent misbehavior, is inferred from a readily available proposition-schema. Further, Lutz suggests, the structure of propositionschemas may enable children to learn the content of cultural models in stages, first mastering abbreviated versions of proposition-schemas - or

26

NAOMIQUINN & DOROTHY HOLLAND•

"chunks" in Hutchins's (1980:115-116) terminology - in which intervene ing links in & complex chain of causality are omitted and only laterunderstood. '.<• ••:•••.:.•:., -.(•'.••$;:yr--tti Image-schemas lend themselves to quite different uses. Lakoff (1981) regards image-schemas as gestalts just as Visual images are. However,'they: are much more schematic than what we ordinarily think of as visual imagery, and they may contain not just visual components but also kin* aesthetic information of all kinds. The examples he provides -."bur: knowledge afedUt baseball pitches includes a trajectory schema.Our* knowledge about candles includes a long, thin object schema" (Lakoff: 1984:10) - make clear that image-schemas convey knowledge of physical phenomena, such as shape and motion. :-.;,.; In the present volume, Lakoff and Kovecses provide an example ~ of anger conceptualized image-schematically in terms of hot liquid in a container. Rempton's informants provide another: For some of these people, the "valve"" theorists, home thermostats are imagined as faucet-like devices; for others, the "feedback" theorists, as on-off switches. The labels Collins and Gentner give to the various "component models" (our "imageschemas") on which their subjects draw to imagine how evaporation works - the "sand-grain" model, the "random-speed" model, the "heatthreshold" model, the "rocketship" model, the "container" model, the "crowded room" model, and so forth - graphically convey the imageschematic nature of these components that subjects combine into a runable model of the process by which molecules might be conceived to behave in the water at the outset of evaporation, escape from the water into the air, behave in the air, return to the water, change from liquid to vapor, and vice versa. Two of these studies suggest strongly, if they do not demonstrate conclusively, that image-schemas are actually being used to perform the cognitive task that is verbally described for the investigator, rather than just being used to construct the verbal account of that task (a cognitive task in its own right, but a different one). The reports Kempton's informants give of their thermostat adjustment habits agree with his predictions, based on which of the two image-schemas household residents are using. Collins and Gentner compare the content of subjects' verbal protocols to the adequacy of their explanations for mundane observations, such as why you can see your breath on a cold day. The answers given by the two subjects illustrate two divergent tendencies, which seem to reflect the greater success of the first subject in reasoning from imageschemas. The second subject was able to give fewer correct answers and often fell into inconsistencies. From his protocols this appears to be because, unlike the first subject, he had not established a stable set of , image-schemas with which to work through the hypothesized evaporation ' process and against which to check his reasoning for inconsistencies. Rather, he invoked a different component model for every answer and ,

CULTURE AND COGNITION

27

relied frequently instead on isolated analogies to different phenomena he happened to have seen or heard about. More extensive evidence that image-schemas indeed enter into the performance of a task comes from Gentner and Centner (1983). They argue that subjects use one of two different analogies for explaining electricity. This study shows convincingly that choice of analogy has consequences for reasoning: Subjects using each characteristically make different kinds of mistakes. The nature of these mistakes would seem tofavor the interpretation that subjects are reasoning from image-schemasofthe physical-: world analogy they use. For example, subjects who adopt the "teeming' crowd" model of electricity are predictably better able to understand the difference between parallel and serial resistors, which they view"as gates;" These people correctly respond that parallel resistors (viewed as two side- ' by-side gates) give more current than a single resistor; serial resistors (viewed as two consecutive gates) less. An image-schematic interpretation would argue that such subjects gain an advantage by being able to manipulate these "gates" mentally and visualize how electricity, like a racing crowd, might make its way through them. Subjects who use the "flowing fluid" model, on the other hand, typically err in their explanations of parallel and serial resistors, viewing resistors as impediments to the passage of a fluid. These latter people conclude that both combinations of resistors constitute double obstacles and thus that both result in less current. Image-schemas seem well-adapted to thinking about not only physical relations but logical ones as well, when that logic is amenable to reconceptualization in spatial terms. Johnson-Laird and Steedman (1978), for instance, have argued that subjects solve difficult syllogisms by conjuring up Venn-diagram-like schematic relationships among groups of imaginary entities and then consulting these mental diagrams to read off the overlap between groups. Image-schemas would seem to permit the scanning and manipulation required by certain kinds of complex reasoning. THE ROLE OF METAPHOR IN SUPPLYING IMAGE-SCHEMAS

Lakoff (1984:10) goes on to argue that metaphor plays an important role in cognitive modeling, mapping proposition-schemas and image-schemas in given domains onto corresponding structures in other domains. Such mappings have a characteristic direction, as Lakoff and Johnson (1980:56-68) observed: Metaphors appear to introduce information from physical-world source domains into target domains in the nonphysical world. Why this should be so has not been made entirely clear. Lakoff and Johnson (ibid.:57, 61-62) sometimes seem to be suggesting that the concepts metaphors introduce are more readily understandable because they are grounded in our bodily interaction with the physical environment. However, Holland (1982:292-293; see also Butters 1981) points out that this is demonstrably not the case for Lakoff and Johnson's prime example, the metaphor ARGUMENT IS WAR. Lakoff and Johnson assert that

28

NAOMI QUINN & DOROTHY HOLLAND

we understand war more readily because of its basis in our evolutionary history as human animals, equipped for physical conflict. Holland responds that our understanding of modern war, far from resting on a conception of primal physical combat, is just as culturally given as our notions about argument; and argument is the more directly apprehended experience for most Americans. We suggest, rather, that the advantage of metaphors from the physical world rests on the nature of physical experience itself, and the manner in which physical properties and relations are apprehensible to human beings: - a view that Lakoff and Johnson (1980:58, 60-61) at other points seem to be developing. The present discussion about the role of image, schemas in understanding allows us to be more precise. Image-schemas ; are constructed out of physical properties and relations, and the advantage of metaphors drawn from domains of the physical world is that these source domains provide the material for image-schemas. Metaphor is important to understanding, then, because it enables image-schematic thought. Thus, it does not really matter whether WAR is grounded in actual experience or genetic memory of physical combat, or known indirectly from depictions of such combat. What makes it a useful metaphor for ARGUMENT is that, unlike the latter, war is largely culturally defined for us in terms of physical space - battlegrounds, battle lines, routes of retreat, demilitarized zones, and so forth - occupied by physical occurrences - troop advances, cross-fire, body counts, and so forth. The metaphor allows the largely intangible social dynamics of argument to be reconceptualized in the image-schematic terms provided by the tangible events of war. The result of any such mapping, from physical experience in the source domain to social or psychological experience in the target domain, is that elements, properties, and relations that could not be conceptualized in image-schematic form without the metaphor can now be so expressed in the terms provided by the metaphor. Such a result is achieved, for example, by the metaphor of anger as a hot fluid in a container, which can be envisioned as boiling, producing steam, rising, arid exerting pressure on its container, which, as a consequence, can be imagined to explode. Similarly, the image-schema of marriage as an entity allows it to be conceptualized as a manufactured object more or less well constructed and hence more or less likely to fall apart; and the image-schema of a problem as a protrusion on the landscape allows it to be reduced, conceptually, from the size of a mountain to the more realistic, and hence surmountable, size of a molehill. Like other image-schemas, metaphorically derived image-schemas are gestalts that make multiple relations more immediately apprehensible. These gestalts can then be scanned to arrive at entailments among related elements and manipulated to simulate what would be entailed under different conditions. In their paper, Collins and Gentner decompose the process by which

CULTURE AND COGNITION

29

novices - struggling to explain a physical phenomenon, evaporation, about which they have probably never before been asked - generate.their ,exr. planations by analogy and the image-schemas it provides. The subjects manipulate image-schemas, or as these authors put it, "run mental models" to simulate what might happen under various conditions. Collins and Gentner argue convincingly that subjects perform mental simulations in new domains by partitioning the system they are trying to understand (here, evaporation) into a set of component processes. Then models (in our terms, image-schemas) can be:mappe
30

NAOMI QUINN & DOROTHY HOLLAND

Many of the papers in this volume illustrate Lakoff and Johnson's (1980) claim that ongoing understanding often relies on the rich mapping potential of metaphor. However, Quinn's paper and that of Lakoff and Kdvecses suggest that metaphors are extended, not willy-nilly from any domain to any other, but in closely structured ways (see also Holland 1982:293^294). A multiplicity of metaphors for marriage or for anger or"for' types oTnien (Holland & Skinner 1985) fail into a handful of classes. What appears to constrain these metaphors to these classes is the underlying cultural model in the domain to which they are mapped: The classes from which speakers select metaphors they consider to be appropriate are those that capture aspects of the simplified world and the prototypical events unfolding in this world, constituted by the cultural model. Chosen metaphors not only highlight particular features of the cultural model; as we discuss, they also point to entailments among these elements. Thus, one husband's metaphor of marriage as a "do-it-yourself project" at once suggests for him the durable quality of something made in this manner - "it was very strong because it was made as we went along" - and implies, additionally, the craft and care and effort that must go into such a thing to make it well. Speakers often favor just such metaphors, which allow two or more related elements of the source domain to be mapped onto a corresponding set of related elements in the cultural model (Quinn 1985) and a comment on that relation to be made. At the same time, other metaphors that fail to reflect, or even contradict, aspects of the cultural model in the target domain to which they are mapped are likely to be rejected. Quinn (ibid.) gives an anecdotal example in which marriage was likened to an ice-cream cone that could be eaten up fast or licked slowly to make it last loflger -'a metaphor in such clear violation of our understanding of marriage as an enduring relationship that it bothered and offended members of the wedding at which it was voiced, THE ROLE OF METONYMY IN STRUCTURING CULTURAL MODELS

In Lakoff's (1984) formulation, metaphor does not exhaust the possible devices for structuring our understanding; metonymy has a central role to play. A metaphoric model maps structures from one domain to another; what Lakoff terms a metonymic model structures a domain in terms of one of its elements. Something is gained by this substitution of part of a category for the category as a whole: the former "is either easier to understand, easier to process, easier to recognize, or more immediately useful for the given purpose in the given context" (ibid.:12). Thus, for example, the social world in which some men are bachelors is structured, not by our full knowledge of the many possible courses men's lives may take, but by what Lakoff calls a typical example of a male life course. This life course, treated as canonical for men, provides the presupposed world within which bachelor is an applicable term.

CULTURE AND COGNITION

31

Typicality is not the only metonymic relation that may \hbld between a domain and some element in this domain. -Again, the papers'm this volume provide examples of several types of metonymic.model treated by.Lakoff (ibid.:12~15). Just as there exists in bur minds, against:the backdrop of a typical male life course, a stereotype of :the sprtfbf man who would deviate from this course to remain a bachelor (ibid;:21)v so Holland and Skinner's data argue that their interviewees, conceptualizing interactions between college men and women in terms of hbw such a rela-: tionship typically proceeds, understand individuals who violate, the expecr tations engendered by this canonical relationship in terms of (largely) negative social stereotypes. These social stereotypes of various kinds of inept and exploitative men are quite different from interviewees'notions of the (proto)typical man. Price discusses how knowledge about social roles is embedded in salient examples, remembered illness episodes that are used to characterize appropriate social role behavior both by exhibiting instances of that behavior and, more dramatically, by elaborating counterexamples. The proposition-schemas about marriage that Quinn enumerates exemplify another type of metonymy - ideals. Even though Americans might agree that most marriages are difficult, they would probably not agree that most marriages are enduring. This proposition-schema, that MARRIAGE IS ENDURING, derives not from any notion of the statistically dominant pattern, but from an ideal of the successful marital enterprise. Just as a successful marriage is enduring, a happy marriage is mutually beneficial, and a real marriage is lived jointly (Quinn 1985). Finally - in an example that constitutes an addition to Lakoff's list of metonymic types - Hutchins shows that a myth can be understood as a symbolic reformulation of events in life, a culturally given yet disguised representation that serves as a defense mechanism against realization of painful and unacceptable sentiments. Lakoff's discussion gives us a better sense of why cultural models have the prototypical nature they do: They are constructed out of various types of metonymy. In his words, (1984:11), Prototype effects are superficial phenomena. They arise when some subcategory or member or submodel is used (often for some limited and immediate purpose) to comprehend the category as a whole. In either proposition-schematic or image-schematic form, by way of metaphor or not, cultural models draw on a variety of types of idealized events, actors and other physical entities in these events, and relations among these, all of which are available to our understanding of ordinary experience: the typical, the stereotypical, the salient in memory, the mythic, the idea! successful, the ideal happy, and so on. Just as Fillmore has pointed out that simplified worlds provide the context of our understanding, Lakoff has drawn our attention to the fact that these presupposed

32

NAOMI QUINN & DOROTHY HOLLAND

worlds are simplified in different ways and that the different types of simplification put our understanding in different perspective. ' f::-/.< * Any given cultural model may be constructed out of several: types of metonymy. We have seen, for example, that the American cultural model of marriage, depending on the metonym in focus, allows "propositions about the ideal successful marriage, the ideal happy marriage; and what can be considered a "real'* marriage, as well as what the typical marriage is like. All of these ideas are part of the cultural understanding of marriage. Moreover, as is discussed in the final section, the several metbnymic types can stand in causal relation to one another, such prototypical causal links yielding the relatively complex chains of event sequences that characterize cultural models. These complexities are constructed out of simple metonymies. Our discussion of the prototypical nature of cultural models applies equally to our models of these models. Linde (this volume) discusses the interaction between the models of culturally designated "experts," or scientists, and the models of the "folk." Folk models of the world incorporate expert knowledge, as Linde's analysis of life stories shows. Conversely, as suggested by Kay's (this volume) observation that each of the two folk theories of language he identifies has its counterpart in an academic linguistic theory, the former penetrate the latter. As analysts, we cannot expect our "explanatory models" of cultural models, to adopt Caws's (1974) term for them, to be of a wholly different order than the cultural models we seek to explain. Simplification, by means of metonymy, is a feature of both. Nonetheless, by constantly questioning how cultural knowledge is organized, we aspire to a kind of analysis that can be successively improved to capture the native model and the tasks, explanatory and otherwise, to which it is brought by the native user. Not only do we hope to recognize and make explicit the cultural assumptions in our own analytic models, we also hope to minimize the kind of distortion of other people's cultural models that Keesing (this volume) cautions against, that arises from a too-facile reading of metaphysical theories of the world out of formulaic ways of talking. Cultural models and human cognitive requirements Given the observation that cultural models are composed of prototypical event sequence set in simplified worlds, we can begin to say something more about the organization of such models and the properties that make them readily learned and shared. In the simplified worlds of cultural models, complicating factors and possible variations are suppressed. In the world of Fillmore's bachelor, males are either old enough to marry 'I or not; and if of marriageable age, they are either already married or yet unmarried - there are no problematic thrice-married divorces, Sweetser , (this volume) points out. As we have seen, the papers in the present volume "

CULTURE AND COGNITION

;33

' provide many additional examples of presupposed worlds defined by such simplifying assumptions. In Americans' folk model of ithe institution, MARRIAGE IS ENDURING; in the folk model of communication that informs our understanding of lying, INFORMATIONS HELPFUL; in our cultural model of the mind, WISHES GIVE RISE TO INTENTIONS; in Americans' model of the emotion of anger, AN OFFENSE TO A PERSON CAUSES ANGER IN THAT PERSON; in the different world of Ifaluk emotion theory, JUSTIFIABLE ANGER-IS-CAUSE-FOR SUICIDE. •'\--:_.^i:;:^-:X-:L:ir^iiu-^ Even further, these worlds are ordered and simplified by implicit presuppositions about how such propositions may be linked one to, another.; In the Ifaluk model of emotion (described in more elegant notation by Lutz, this volume) because MISBEHAVIOR IN ONE PERSON LEADS TO JUSTIFIABLE ANGER IN ANOTHER, and because JUSTIFIABLE ANGER CAN LEAD THE ANGRY PERSON TO REPRIMAND THE PERSON WHO MISBEHAVED, and because A REPRIMAND CAUSES FEAR AND ANXIETY IN ITS RECIPIENT, then it follows that JUSTIFIABLE ANGER IN ONE PERSON OVER ANOTHER'S MISBEHAVIOR PRODUCES FEAR AND ANXIETY IN THE OTHER. In our own culture (Lakoff & Kovecses, this volume), because AN OFFENSE TO A PERSON PRODUCES ANGER IN THAT PERSON, and because RETRIBUTION CANCELS AN OFFENSE, then predictably, ANGER DISAPPEARS WHEN RETRIBUTION IS EXACTED. The predictable sequence of events, played out in the simplified world of the cultural model, allows that world to be characterized not only by proposition-schemas but also in terms of a smaller number of more complex schemas that specify sets of such propositions and the causal relations in which they stand to one another. It is these "causal chainings," to use Abelson's phrase quoted in an earlier section, that give the events occurring in cultural models their quality of unfolding stories. What we need to learn and remember and communicate about the world is vastly reduced by being packaged in such units. Further, these models articulate with one another in a modular fashion. As D'Andrade (this volume) makes explicit, a given schema may serve as a piece of another schema. D'Andrade uses Fillmore's example of the commercial transaction to make this clear. To know whether BUYING is taking place, one must invoke the other terms of the relationship to judge whether PURCHASER, SELLER, MERCHANDISE, PRICE, OFFER, ACCEPTANCE, and TRANSFER are involved. Each of these components, in turn, is constituted by a complex schema; but one need not know details of each event such as how the price was actually set or whether it was fair, to know that a sale has taken place. The significance of this latter point, D'Andrade argues, is that this hierarchical organization of cultural knowledge is adapted to the requirements of human short-term memory. To perform any particular cognitive task, such as judging whether

34

NAOMI QUINN & DOROTHY HOLLAND

something has been sold by one person to another, an individual heed invoke and hold in mind only a small set of criteria - a number not exceeding the limits of short-term memory storage. i The nestedness of cultural models one within another lends a further, far-reaching economy to cultural knowledge. This hierarchical structure in which models of wide applicability recur as elements of models in many domains of experience has implications for long-term memory as well. These general-purpose models considerably reduce the total amount of cultural knowledge to be mastered. A component model such as BARGAINING - a possible way in which price can be set ~ presupposes and draws on the BUYING schema within which it is nested. In the same way, nested within the cultural model of anger that Lakoff and JCovecses describe is a more widely applicable cultural model of exchange and balance in human affairs; this more general model includes a schema that yields the proposition RETRIBUTION CANCELS AN OFFENSE; In Quinn's model of American marriage, the proposition-schema, MARRIAGE IS MUTUALLY BENEFICIAL, makes sense in terms of the more widely applicable schema for social relationships, VOLUNTARY RELATIONSHIPS ARE MUTUALLY BENEFICIAL; and, since marriage is distinctive among voluntary relationships in that THE BENEFITS OF MARRIAGE ARE FULFILLMENT OF NEEDS, the knowledge that MARRIAGE IS MUTUALLY BENEFICIAL can be filled in by a further model, of need fulfillment, general to our understanding of the self, Reliance on general-purpose models for filling in knowledge is perhaps even more striking in the case described by Collins and Centner; in which subjects were called on to answer questions on a subject - evaporation - , about which they were untutored in the scientific model and had available " no ready-made common-sense theory. In this situation, the interviewees fell back on their understanding of other physical phenomena and attempted to apply very general principles such as that of a heat threshold or that of molecular attraction drawn from their models of these other phenomena. Clearly, complex proposition-schemas such as those for bargaining, retributive justice, mutual benefit, need fulfillment, and molecular attraction have application across multiple domains of our experience. The capability of such general-purpose cultural models for filling in the details of other cultural models creates a further simplification. This was demonstrated by Quinn's (this volume) interviewee, who was able to reason about the benefits, difficulty, and enduringness of marriage without having to explain the implicit theory of need fulfillment she knows she shares with her addressee. A great deal can be taken for granted. Parenthetically, it is just these cultural models of wider applicability, -[, serving as modular components of many other models, that give a culture its distinctiveness. As D'Andrade points out in his volume paper, understanding a culture depends on knowledge of at least these widely

35

CULTURE AND COGNITION

incorporated models. Anthropologists have long attempted to capture the distinctiveness of one culture from another in concepts such as that of "cultural themes," or "cultural belief systems," or differing "world views." However, such accounts have typically failed to,specify,tlie range ptdomains in which a given theme, alleged to be central to a culture, applies. Nor have they been able to explicate how such central premises articulate with particular domains of knowledge in which they figure (Clement $976),. The theory of cultural models under construction here promises to identify pervasive cultural premises and to reveal the structural linkages between these premises and the more circumscribed models specific to emotion, problem solving, the mind, gender relations, and the myriad other topics of cultural knowledge. -:'":"'/~'-iThe account emerging from this volume, then, is one in which cultural understanding is organized into units smaller and simpler in construction and fewer in number than might have been supposed. It is an account that offers a beginning solution to Abelson's "size problem," the problem of how we can learn and use as much knowledge as human beings do. The prototypical scenarios unfolded in the simplified worlds Of cultural models, the nestedness of these presupposed models one within another, and the applicability of certain of these models to multiple domains all go far to explain how individuals can learn culture and communicate it to others, so that many come to share the same understandings.

Notes

•--.'.•.

1. This introduction has benefited immensely from the long, careful readings and comments given an earlier draft by Roy D'Andrade, Edwin Hutchins, Paul Kay, Richard Shweder, and Geoffrey White, as well as from the briefer but telling reactions of Ronald Casson, Susan Hirsch, Alice Ingerson, John Ogbu, and Laurie Price. There are points on which each of these people would still disagree with us. On other points, years of talk with Roy and Ed have sometimes made it difficult to know where our ideas end and theirs begin. Both of them have contributed to our thinking about numerous matters. 2. Also variant in these papers is the plural form of schema. The editors recommended to the authors that all adopt a regularized plural, "schemas," in place of the Latin plural, "schemata," which is grammatically correct but awkward for many English speakers. There is precedent for both variants in the cognitive science literature. However, one author, Paul Kay, argued that technically, "schemas" was improper usage. We have honored his wish to use the longer form in his paper.

References Abelson, R. 1975. Concepts for representing mundane reality in plans. In Representation and Understanding: Studies in Cognitive Science, D. G. Bobrow and A. Collins, eds. New York: Academic Press. Pp. 273-309.

36

NAOMI QUINN & DOROTHY HOLLAND

1981. Constraint, construal, and cognitive science. In Proceedings of the Third Annual Conference of the Cognitive Science Society. Berkeley: .University of California. Pp. 1-9. Berlin, B. and P. Kay 1969. Basic Color Terms: Their Universality and Evolution. Berkeley: University of California Press. ". •• . •.^;v-;;. Bohannan, P. 19S7. Justice and Judgment among the Tiv. London: Oxford University Press. . Butters, R. 1981. Do "conceptual metaphors" really exist? Southeastern Conference on Linguistics Bulletin 5(3):108-117. Casson, R. 1983. Schemata in cognitive anthropology. Annual Review of Anthropology 12:429-462. Caws, P. 1974. Operational, representational, and explanatory models. American Anthropologist 76(1):1-11. Clement, D. H. 1976. Cognitive anthropology and applied problems in education. In Do Applied Anthropologists Apply Anthropology? M. Angrosino, ed. Athens, Ga: University of Georgia Press. Pp. 53-71. 1982. Samoan folk knowledge of mental disorders. In Cultural Conceptions of Mental Health and Therapy, A. J. Marsella and G. White, eds. Dordrecht, Holland: D. Reidel PubEshing Company. Pp. 193-215. Coleman, L. and P. Kay 1981. Prototype semantics: The English verb lie. Language 57(l):26-44. D'Andrade, R. 1965. Trait psychology and componential analysis. American Anthropologist 67(5,2):215~228. 1974. Memory and the assessment of behavior. In Measurement in the Social Sciences, H. Blalock, ed. Chicago: Aldine-Atnerton. Pp. 159-186. 1976. A propositional analysis of U.S. American beliefs about illness. In Meaning in Anthropology, K. H. Basso and H. A. Selby, eds. Albuquerque: University of New Mexico Press. Pp. 155-180. 1981. The cultural part of cognition. Cognitive Science 5(3):179-195. 1982. Reason versus logic. Paper presented at symposium, The Ecology of Cognition: Biological, Cultural, and Historical Perspectives, April, Greensboro, N.C. 1984a. Cultural meaning systems. In Culture Theory: Essays on Mind, Self, and Emotion, R. Shweder and R, LeVine, eds. Cambridge, England: Cambridge University Press. Pp. 88-119. 1984b. Cultural models and person perception. Washington, D.C: Progress Report to the Anthropology Program, National Science Foundation. 1985. Character terms and cultural models. In New Directions in Cognitive Anthropology, J. Dougherty, ed. Urbana: University of Illinois Press. Pp. 321-343. n.d. Culture and personality: A false dichotomy. Unpublished manuscript, Department of Anthropology, University of California, San Diego. D'Andrade, R., N. Quinn, S. B. Nerlove, and A. K. Romney 1972. Categories of disease in American-English and Mexican-Spanish. In Multidimensional Scaling, Theory and Applications in the Behavioral Sciences, Vol. 2: Applications, A. K. Romney, R. N. Shepard, and S. B. Nerlove, eds. New York: Seminar Press. Pp. 9-54.

CULTURE AND COGNITION

37

Early, E. A. ;--;:?!;.';i-; 1982. The logic of well being: Therapeutic narratives of Cairo. Social Science and Medicine 16(4):1491-1498. •••*..;.,,*•Fillmore, C. • ". V;-. .^-jyv^.-R : ' 1975. An alternative to checklist theories of meaning. In Proceedings iof the • F i r s t Annual Meeting of the Berkeley Linguistics Society, "C. Cogeri, H. Thompson, G. Thurgodd.K. Whistler, and J. Wright, eds. Berkeley: University of California. Pp. 123-131. ::.^u:, .-U .;,- Jl.,::.„:.,u-,i;; 1977. Topics in lexical semantics. In /Current Issues in Linguistic Theory, R. W. Cole, ed.Bloomington: Indiana University Press. Pp.76-138; 1982. Towards a descriptive framework for spatial debus. In SpeedvPIace, and Action, R. J. Jarvella and W. Klein, eds. New York: John Wiley and Sons^ Pp. 31-59. • : • • . • . • • . • . ' • ' • :.- .,- •: -?•'<',:•• ,: :\i .'-Wl Fjellman, S. .-••:'.-•• ..? 1976. Talking about talking about residence: An Akamba case. American Ethnologist 3(4):671-682. . ' • Frake, C O . ; 1975. How to enter a Yakan house. In Sociocultural Dimensions of Language Use, M. Sanches and B. Blount, eds. New York: Academic Press. Pp. 2540. 1977. Plying frames can be dangerous: Some reflections on methodology in cognitive anthropology. Quarterly Newsletter of the Institute for Comparative Human Development l(3):l-7. Geertz, C. 1966. Religion as a cultural system. In Anthropological Approaches to the Study of Religion, M. Banton, ed. A.S.A. Monographs 3. London: Tavistock Publications. Pp. 1-46. Gentner, D. and D. R. Gentner >: 1983. Flowing waters or teeming crowds: Mental models of electricity.//! Mental Models, D. Gentner, and A. L. Stevens, eds. Hillsdale, N.J.: Lawrence Erlbaum Associates. Pp. 99-129. Geoghegan, W. 1969. Decision-making and residence on Tagtabon Island. Language-Behavior Research Laboratory Working Paper No. 17. Berkeley: University of California. Gladwin, H. and C. Gladwin 1971. Estimating market conditions and profit expectations of fish sellers in Cape Coast, Ghana. In Studies in Economic Anthropology, G. Dalton, ed. Anthropological Studies No. 7. Washington, D.C.: American Anthropological Association. Pp. 122-142. Good, B. J. and M.-J. D. Good 1982. Toward a meaning-centered analysis of popular illness categories: Trightillness' and 'heart distress' in Iran. In Cultural Conceptions of Mental Health and Therapy, A. Marsella and G. White, eds. Dordrecht, Holland: D. Reidel Publishing Company. Pp. 141-166. Goodenough, W. H. 1957. Cultural anthropology and linguistics. In Report of the Seventh Annual Round Table Meeting in Linguistics and Language Study, P. Garvin, ed. Monograph Series on Language and Linguistics, No. 9. Washington, D.C.: Georgetown University. Pp. 167-173. Harris, M. 1968. The Rise of Anthropological Theory. New York: Thomas Y. Crowell Company.

38

NAOMI QUINN & DOROTHY HOLLAND

Holland, D. 1982. Conventional metaphors in human thought and language. Review article, Metaphors We Live By, by G. Lakoff and M. Johnson. Reviews in Anthropology 9(3):287-297. .i 1985, From situation to impression: How Americans get to know themselves and one another. In Directions in Cognitive Anthropology, J. Dougherty, ed. Urbana: University of Illinois Press. Pp.: 389-411. Holland, D. and D. Skinner ,•'>••-•:.•; 1985. The meaning of metaphors in gender stereotyping. North Carolina Working Papers in Culture and Cognition No. 3. Durham, N - C : Duke University Department of Anthropology.; - ;>•. H o l y ,

L .

• • • ' ' , . • •

.:•-.-,^

1979. Segmentary lineage structure and its existential status. In Segmentary Lineage Systems Reconsidered, L. Holy, ed. Queen's University Papers in Social Anthropology 4. Belfast: Queen's University. Pp. 1-22... Holy, L. and M. Stuchlik ,; 1981a. The Structure of Folk Models. A.S.A. Monographs 20. New York: Academic Press. 1981b. The structure of folk models. In The Structure of Folk Models, L. Holy and M. Stuchlik, eds. A.S.A. Monographs 20. New York: Academic Press. Pp. 1-35. Howe, J. and J. Scherzer 1975. Take and teE: A practical classification from the San Bias Cuna. American Ethnologist 2(3):435-460. Hutchins, E. 1980. Culture and Inference: A Trobriand Case Study. Cambridge, Mass.: Harvard University Press. Jenkins, R. 1981. Thinking and doing: Towards a model of cognitive practice. In The Structure of Folk Models, L. Holy and M. Stuchlik, eds. A.S.A. Monographs 20. New York: Academic Press. Pp. 93-119. Johnson, A. r 1974. Ethnoecology and planting practices in a swidden agricultural system. American Ethnologist 1(1):87-101. Johnson-Laird, P. and M. Steedman 1978. The psychology of syllogisms. Cognitive Psychology 10(l):64-99. Jorion, P. 1978. Marks and rabbit furs: Location and sharing of grounds in coastal fishing. Peasant Studies 7(2):86-100. Katz, J. and J. Fodor 1963. The structure of a semantic theory. Language 39(2):170-210. Kay, P. and C. McDanie! 1978. The linguistic significance of the meanings of basic color terms. Language 54(3):610-646. Keesing, R. M. 1974. Theories of culture. Annual Review of Anthropology 3:73-97. Lakoff, G. 1981. Introductory remarks. Paper presented at conference, Cognitive Science, Language, and Imagery, May, Berkeley, California. 1984. Classifiers as a reflection of mind: A cognitive model approach to proto?. type theory. Berkeley Cognitive Science Report No. 19. Berkeley: University ' of California Institute of Human Learning. Lakoff, G. and M. Johnson 1980. Metaphors We Live By. Chicago: University of Chicago Press.

39'-

CULTURE AND COGNITION

Langacker, R. W. - ; ' ;'-••?.; .;* .:k-?••':.• ' ;-..";'-•; • ,;/. UL:.-^ „

1953. Social structure. In Anthropology Today, A. L. Kroeber, ed. Chicago: University of Chicago Press. Lewontin, R. C , S. Rose, and L. J. Kamin t , :. *; 1984. N o t i n O u r G e n e s : B i o l o g y , I d e o l o g y , a n d H u m a n N a t u r e . : N e w Y o r k : • • P a n t h e o n B o o k s . - •-...'.••.-••;.•.••."•:"' •••:'••• '•:-' ",• v< •..•''><; Linde, C.

,.. •

-

. • • • • • • . > • .

v<;^--

;

1978. The organization of discourse. Ch. 4 in Style and Variables in English, T. Shopen and J. M. Williams, eds. Cambridge, Mass.: Winthrop Publishers. Pp. 84-114. n.d. The creation of coherence in life stories. Unpublished manuscript in preparation, Structural Semantics, Palo Alto. Mathews, H. (in press). Predicting decision outcomes: Have we put the cart before the horse . in anthropological studies of decision making? Human Organization. Nardi, B. 1983. Goals in reproductive decision making. American Ethnologist 10(4): 697-715. Quinn, N. 1981. Marriage is a do-it-yourself project: The organization of marital goals. In Proceedings of the Third Annual Conference of the Cognitive Science Society, Berkeley: University of California. Pp. 31-40. 1982ai Cognitive anthropology comes of age in the Trobriands. Review article, Culture and Inference: A Trobriand Case Study, by E. Hutchins. Reviews in Anthropology 9(3):299-311. 1982b. "Commitment" in American marriage: A cultural analysis. American Ethnologist 9(4):775-798. 1985. American marriage through metaphors: A cultural analysis. North Carolina Working Papers in Culture and Cognition No. 1. Durham, N.C.: Duke University Department of Anthropology. Randall, R. 1976. How tall is a taxonomic tree? Some evidence for dwarfism. American Ethnologist 3(3):543-553. Rosch, E. 1977. Human categorization. In Studies in Cross-cultural Psychology, Vol. 1, N. Warren, ed. London: Academic Press. Pp. 3-49. 1978. Principles of categorization. In Cognition and Categorization, E. Rosch and B. Lloyd, eds. Hillsdale, N.J.: Lawrence Erlbaum Associates. Pp. 27-48. Ryle, G. 1949. The Concept of Mind. London: Hutchinson House. Salzman, P. C. 1981. Culture as enhabilmentis. In The Structure of Folk Models, L. Holy and M. Stuchlik, eds. A.S.A. Monographs 20. New York: Academic Press. Pp. 233-257. Schank, R. 1982. Dynamic Memory: A Theory of Reminding and Learning in Computers and People. Cambridge, England: Cambridge University Press.

40

NAOMI QUINN & DOROTHY HOLLAND

Schank, R. and R. Abelson 1977, Scripts, Plans, Goals and Understanding: An Inquiry into Human Knowl. edge Structures. Hillsdale, N.J.: Lawrence Erlbaum Associates. Sperber, D. 1975. Rethinking Symbolism. Cambridge, England: Cambridge University Press. (First published in 1974 as Le Symbolisme en General.) Spiro, M. 1961. Social systems, personality, and functional analysis. In Studying Personality Cross-Culturally, B. Kaplan, ed. New York: Harper and Row Publishers. Pp. 93-127. '-•'- v..-,:;.-,., Ward, B. E. 1965. Varieties of the conscious model: Thefishermenof South China. In The Relevance of Models for Social Anthropology, M. Banton, ed, A.S.A. Monographs 1. London: Tavistock Publications. Pp. 113-137. . -.:,-.•>. 1966. Sociological self-awareness: Some uses of the conscious models. Man 1(2);201-215. White, G. 1982. The ethnographic study of cultural knowledge of 'mental disorder.' In Cultural Conceptions of Mental Health and Therapy, A. h Marsella and G. White, eds. Dordrecht, Holland: D, Reidel Publishing Company. Pp. 69-97. 1985. Premises and purposes in a Solomon Island ethnopsychology. In Person, Self and Experience: Exploring Pacific Ethnopsychologies, G. White and J. Kirkpatrick, eds. Berkeley: University of California Press. Pp. 328-366. Young, A. 1981, Rational m&n and the explanatory model approach. Culture, Medicine and Psychiatry 6(i):57-73.

PARTI Presupposed worlds,-language, and discourse

2 The definition of Me AN EXAMINATION OF THE FOLK MODELS UNDERLYING A SEMANTIC PROTOTYPE 1

Eve.E, Sweetser •-,...,•

.

This paper investigates how the semantic structure of one English word depends on, and reflects, our models of relevant areas of experience. As a linguist, my original concern was with the problems posed by the word lie for traditional semantic theories; but these problems led inexorably to the cultural models of informational exchange that motivate the existence of a semantic entity meaning lie.21 begin by posing the semantic problem and go on to the cultural solution. George Lakoff (1972), Fillmore (1977), and Coleman and Kay (1981) have argued against traditional generative and structuralist "checklists" of semantic features that constitute necessary and sufficient conditions for set-membership in the category denoted by a word. Lexical categories can have better or worse members, or partial members.3,Kay and McDaniel (1978) have shown that color categories lack necessary and suf-. ficient conditions; red is a gradient quality whose category-boundaries are best described by fuzzy set theory rather than by traditional set theory. Checklist feature-definitions, which do not allow for color's being "sorta red," must be replaced by a theory capable of dealing with fuzzy setmembership. Prototype semantics views word-meaning as determined by a central or prototypical application, rather than a category-boundaries. Clear definitions can thus be given for words with fuzzy boundaries of application. We define the best instance of a word's use, and expect realworld cases to fit this best example more or less, rather than perfectly or not at all. Coleman and Kay (1981) show that prototype theory is needed to explain the usage of the verb lie.4 As is natural in prototype semantics (but not in traditional set-membership semantics), lying is a matter of more or less. Clear central cases of lies occur when all of Coleman and Kay's proposed conditions are fulfilled; namely, (a) speaker believes statement is false; (b) speaker intends to deceive hearer by making the statement; and (c) the statement is false in fact. Conversely, a statement fulfilling none of a-c is a clear nonlie. But when only one or two of a-c hold, speakers are frequently confused and find it difficult to categorize an action as lie or nonlie. Further, these conditions (unlike checklist-features)

44

EVE E. SWEETSER

differ in weight, (a) being strongest and (c) weakest in influencing'speakers' categorization of acts as lies. ••:•;•: :;-v .: ,;A Prototype semantics has been attentive to the grounding of language . in the speaker's world. Kay and McDaniel found physical perceptual reasons for color-term universals; Rosch (1978) and Mervis and Rpsch (1981) demonstrate that linguistic categories depend on general human category~formation abilities. Fillmore (1977) discusses some ways in which the social world shapes word-meaning. Bachelor is a classic difficult case: Why is it difficult to say whether the Pope, or a thrice-married divorce, can be called a bachelor? The answer, Fillmore says, is that bachelor depends on a simplified world-view in which people are marriageable at a certain age, mostly marry at that age, and stay married to the same spouse. In this simplified world, a bachelor is simply any unmarried male past marriageable age; outside the simplified world, the word bachelor just does not apply. Bachelor necessarily evokes a prototypical schema of marriage within our cultural model of a life-history. I argue that like bachelor, lie is inherently grounded in a simplified or prototypical schema of certain areas of human experience. This, I suggest, is why Coleman and Kay found that lie needs a prototype definition. Basing my analysis on their experimental findings, I motivate those findings by relating them to work on discourse pragmatics and conversational postulates. It is necessary to examine folk understandings of knowledge, evidence, and proof; our cultural model of language (or at least of lying) cannot be analyzed independently of beliefs about information. I hope to show that He has a simpler definition than thas been thought, in a more complex context; since the cultural-model context for a definition of lie is independently necessary, our analysis is simplified ;.. overall. A cultural model of language Is there a simplified "prototypical" speech-act world, as there is a simplified marriage history? Although such a world has not been examined in detail, Kay (1983) suggests that the word technically evokes a "folk theory" of language use that assumes that experts are the arbiters of correct worduse. Grice's (1975) conversational maxims, and Searle's (1969) felicityconditions, are constraints on the appropriateness of utterances - speakers are assumed to follow these rules in the default situation. Kay's folk theories, Grice's maxims, and Searle's felicity-conditions all describe parts of our cultural understanding of discourse-interaction. Grice's "Be as informative as necessary," for example, is a maxim of which speakers are conscious; one can criticize an interlocutor for informational .. insufficiency. But informational content is irrelevant to a speech activity * such as joke-telling. Robin Lakoff *s (1973) work on politeness rules and Goffman's (1974) work on frame semantics show that conversation often ; has its primary purposes at the level of social interaction; making someone

45^

THE DEFINITION OF LIE

happy, or negotiating the interaction-frame, may be a more important! goal than informativeness. The maxim of informationality is thus binding V, precisely to the degree that we consider ourselves to be operating in'iaf simplified world in which discourse is informational, so that the default, purpose of an utterance is not joking, politeness, or frame-bargainings Our covert discourse-purposes are only made possible by a cultural model • that establishes our overt purpose as informational; frame-bargaining, and most indirect speech, depend on having "direct" speech say something else.

.

.A.::\

...LrX:S.^-..\L::yd:C^hf-.;:

I sketch some relevant aspects of our folk understanding of informational language-use and then use this cultural model to explain the meaning of lie as presented by Coleman and Kay. First, let us posit two basic principles as parts of our model of general social interaction rather than of our specific model of speech acts. These principles, which are assumed to operate in the default case (like Gricean maxims), are (1) Try to help, not harm and (2) Knowledge is beneficial. Together, the two principles yield the result that giving knowledge (since it is beneficial) is part of a general goal of helping others. Thus, in cases in which (2) is true, (1) translates at least partly as (3) Try to inform others. The rules just proposed constitute the cultural motivation for a folk understanding of language as informational. Before going on to a folk theory of knowledge and information, one issue needs clarification: the status of these cultural models, or folk theories. What does it mean to say that language is assumed to be informational in the "default" case? I do not mean that purely informational discourse is statistically more common than, or acquisitionally prior to, other kinds of discourse; indeed, it would be hard to separate discourse modes cleanly, since one utterance may have multiple purposes. However, the informational mode is the "direct" mode on which indirect speech is parasitic; and it may be viewed as more basic in the sense that all discourse involves the conveyance of information (if only about a speaker's intentional state), whereas not all discourse participates in all of the other purposes of language use. Our cultural model presents this "basic" discourse-mode, which is a vehicle for other modes, as being in its pure form the unmarked mode, the norm. Unlike maxims and conditions, this cultural model does not constitute rules of language use, but rather beliefs about what we do when we use language. These beliefs in turn make general social rules applicable to the domain of discourse: Grice's maxim of informationality is the manifestation of a general "Help not harm" maxim, in a simplified (folk-model) world in which information is always helpful. Now, on to our cultural model of information. A folk theory of information and evidence Any truth-conditional semantics assumes that we can "know" the propositional content of "true" statements; this begs the vexed question of what

46

EVE E. SWEETSER

knowledge is. I intend to pass over the philosophers' view of knowledge.: and instead examine our cultural idea of what counts as knowledge, since, this is what underlies our understanding of lies and truths in discourse, , Clearly, we do not imagine that all our beliefs can be proyen logically.; Nonetheless, we consider our beliefs sufficiently justified, and we are not; really worried that their truth is not known from logical proof (few iof: "us" speakers know formal logic) or personal experience. Evaluation of evidence is thus frequently an important issue: "knowledge" is not so much, a relationship between a "fact" ( = true proposition) and a knower^as a socially agreed-on evidential status given by a knower to a proposition. Rappaport (1976) demonstrates just how "social" the difference between statement and truth, between belief and knowledge, really is. He observes that a normative standard of truthfulness in informational exchange is essential to ensure that our belief-system (and our social existence) is not constantly undermined by distrust of new input. (Actual statistical likelihood of a random statement's truth is irrelevant to this norm.) He argues that a central function of liturgy and ritual is to transform a statement or belief into accepted, universal truth - that is, into something that can be unconditionally believed and treated as reliable. Rappaport is mainly concerned with social "facts," not with such falsiiiable information as "Ed is in Ohio." But let's remember that knowledge has many socially acceptable ("valid") sources - and that we do not in fact tidily separate messy socially based knowledge from clean falsifiable facts. We know promises can get broken - yet certain ritual aspects of oaths and promises still make us treat them as extratrustworthy, maintaining our social norm of truthfulness. Or, take a modern scholar who "knows" Marx's or Adam Smith's economic teachings - this "knowl- •; edge" may seem to a cynic as faith-based as religious belief, but that does not prevent a whole community of social scientists from acting on it as fact. Hard scientific knowledge and evidence often turn out to be as paradigm-dependent as social-science argumentation. What is crucial is; not whether scientists always have objectively true hypotheses, but that any society agrees on a range of socially acceptable methods of justifying belief; without such agreement, intellectual cooperation would be impossible. What counts as evidence or authority is thus a cultural question. In reply to a college student's scoffs at a medieval philosopher who appealed to classical authority, I once heard a professor ask how the student "knew" what Walter Cronkite had told him. Many natural languages formally mark with evidential markers the difference between direct and indirect (linguistically or logically mediated) experience, and/or between various sensory modalities as sources of a statement's information. Some priority ..: or preference seems to be given universally to both direct experience v (especially visual) and culturally accepted ("universal") truths. But failing these best sources of universal truth or personal experience, we trust some ;

THE DEFINITION OF LIE

'47.;.

input more than others; and we constantly make (nonlogical) deductions based on our observations of correlations in the world. We doriotbother to distinguish these generally trustworthy deductions from "fact? except when observed correlations break down and deductions fail. ; , ; , ; ; Whatever our rules of practical everyday jjrt|erericV"are: like/ ^ t n i s t . : them, in the default case. Thus, belief isribrriiallytaken'as haVirig; adequate justification, and hence as equivalent to knowledge,-which would entail truth. Gordon (1974) demonstrates the close, complex relatioriship of belief and knowledge in bur cultural understanding; he shoVs that, in adult as well as child use, factivity of verbs such as kribw is hot fixed, especially if the person said to "know" is not the speaker.1 A theory of knowledge as a cultural status given to certain beliefs isriiorecompatible with this flexibility than is a theory of knowledge as a link between an objective fact and a person's mind. In our cultural model of knowledge, the default case is thus for belief to entail justification and hence truth. Conversely, untruth will entail lack of evidence and impossibility of belief. Let us combine these entailments with the informational model of language. I start with a norm-establishing "meta-maxim": (0) People normally obey rules (this is the default case). Our general cooperative rule is: (1) Rule: Try to help, not harm. Combined with a belief such as (2), we can instantiate (1) as a Gricean conversational rule of informativeness, as in (3): (2) Knowledge is beneficial, helpful. {Corollary: Misinformation is harmful.) (3) Rule: Give knowledge (inform others); do not misinform. Our model of knowledge and information gives us the following proof of (6) from (4) and <5): (4) Beliefs have adequate justification. (5) Adequately justified beliefs are knowledge ( = are true). (6) .'. Beliefs are true (are knowledge). (6) allows us to reinterpret our helpfulness-rule (3) yet again: (7) Rule: Say what you believe (since belief = knowledge); do not say what you do not believe (this - misinformation). The hearer, in this cultural model, is presumed ready to believe the speaker; why refuse help from a speaker who is assumed to be not only helpful but also well-informed (having well-justified beliefs)? Putting together the whole chain of entailments, we reach the startling conclusion that (in the simplified world of our cultural model) the speaker's saying something entails the truth of the thing said:

48

EVE E. SWEETSER

(a) S said X. (b) S believes X. (a) plus (7) and the meta-maxim) (c) ;*. X is true, (b) plus (6)) Logically (outside our model), or statistically, this conclusion is rubbish. But as a folk model of language by which we all operate from day to, day, : it makes good sense - in fact, it seems doubtful that we could ever live our lives questioning the truth of every statement presented to us. We "question truth if we fear that our simplified discourse-world is too far from reality: when our source might be ill-informed (a broken link between belief and justification), naive (breaking the entailment between justification/ evidence and truth), or might want to deceive us (invalidating our assumption that folks are out to help, and so wish to inform correctly). Note that even in these cases, the usual cultural model is in effect: We know our interlocutor expects us to take what is said as an instance of information-giving. But in general, we take people's word. The next section examines cases in which we should not take someone's word; we now look at lying in the simplified discourse-setting established by our cultural understanding of linguistic exchange as informational. Prevarication in a simplified world Coleman and Kay proposed three components of a prototype-definition of lie: 1. Speaker believes statement to be false. 2. Speaker said it with intent to deceive. 3. The statement is false in fact. Now, in the simplified world we have outlined, any one of these conditions would entail the others. In particular, if we assume both a folk model of evidence in which a speaker's belief constitutes evidence of truth anda model of discourse as informational (intending to be believed), then we find that a factually false statement must be known to be false by the speaker, and (if made) must be intended to induce (false) belief and thus to deceive. The reasoning runs as follows: Premise: X is false. So S did not believe X, since beliefs are true. Therefore S intended to misinform, since we know that in order to inform one says only what one believes. Further, assuming that even uninformative speakers do not randomly discuss areas in which they have no beliefs (people act purposefully), we ; can go beyond "S did not believe X" to assert S believed Xto be false." We do not premise the meta-maxim that S is obeying the rules, since S's obedience to the Cooperative Principle is precisely what we are trying to prove or disprove.

49

THE DEFINITION OF LIE

• Figure 2.1 gives a taxonomy of speech settings; the box on the right encloses the idealized informational-discourse world. Lie must be defined within this restricted World; outside of this world, the word lacks application. Only within this world can the'hearer properly link utterance with infofmativeness, sincerity/and factual truth. The feature [+ Truth Value Relevant] on the tree indicates that the informational-exchange view of language is in effect; when truth value is relevant, knowledge is beneficial ACTIONS

("Help, don'I harm")

(''Information = he!pful";^tell the truth"}

Figure 2.1. A taxonomy of speech settings

50 ;,and

EVE E . SWEETSER

informing helpful. [+ Know] indicates that our folk theory of .knowledge and evidence is in effect; when belief is justified and hence true, the speaker can be assumed to have knowledge about what is said. Thus, we can define lie as a false statement, if we assume the statement occurs in a prototypical (informational) speech setting. This definition is elegant and would also help explain why native speakers tend to ; define lie as a false statement. Not only is this the first definition given "out of the blue" by many speakers, but it is (according to Piaget (1932)), also common for children to pass through a stage in which lie is used to denote any false statement. Wimmer and Peraer's (unpublished data) more recent experimental work shows that children up to age nine class "good faith" false statements and lies as alike, even when they themselves aire tricked into being the "good faith" false informer. Four-year-olds understand sabotage (physical manipulation to obstruct a precondition of an opponent's goal) well; butfive-year-oldsare only starting to understand manipulation of an opponent's belief-system. The social motivations of such manipulation entail an understanding of the speech setting as social interaction. Children only come to differentiate lies from other falsehoods as they learn the sociocultural background of speaking and acquire the folk theories that are a backdrop to the more restricted adult use of lie as a false statement made in a certain world.5 A fascinating parallel to child usage is found in Gulliver's explanation of lying to the Houynhnms. His definition, "saying the thing which is not," is perfectly comprehensible to him, but proves incomprehensible to the Houynhnms, precisely because (as Gulliver says) they have little experience of deception in any area; they lack the sociocultural background that makes a falsehood a lie. Adult English speakers (like Gulliver) have a complex set of possible discourse-worlds (cf. Figure 2.1); it is not strange that in one setting (4- Truth Value Relevant, - Know) a false statement should be called a mistake, whereas in another setting (+ Truth Value Relevant, + Know) a false statement is a lie. * Thus, the simple definition of lie as a false statement is natural given an understanding of our cultural model of knowledge and discourse. The taxonomy of speech settings in Figure 2.1 also motivates the order of Coleman and Kay's three features. First, it is clear why factual falsity is the least important feature. Outside of the prototypical (informational) speech environment, falsehood is not particularly connected with lying (we shall see that lie's moral status also depends on this setting; for now, suffice it that we experience a false statement differently when factors like truthrelevance vary). In a sense, lie is closer to tell the truth than to joke, although jokes are often factually false. Coleman and Kay's most important feature, the speaker's belief that the statement is false, corresponds to my 4 - / - Know branching: Given that a statement is false (another Coleman/Kay feature), the speaker's correct belief in its falsity merely constitutes full and correct information (the:

THB DEFINITION OF LIE

-51

informational part of our simplified cultural model of discourse).;rBeing the first tree-branching above the box enclosing the simplified world jthis feature is most important in speakers' judgments as to whether: ;we:are in that world (and hence whether the term lie applies).-.The.next,;tree^ branching, + / - Truth Value Relevant, corresponds-to Coleman:and Kay's "intent to deceive"; a falsehood can only intend to deceive if truth value is assumed to be relevant (information == beneficial) r not if-we are joking or story-telling. This branching is above the •+• / 'r- Know branch- < ing and farther from the break between the simplified world and ;other worlds - so it is a less important feature in a definition that crucially depends on that break. . >• ^.:J;^oa Coleman and Kay's least important feature is the definitional one: factual falsity. In the environment of their experiment, which actively stretched speakers' consideration beyond the prototypical informational setting, falsehood does not distinguish lies as a unified class. Within the simplified world, however, truth value criterially distinguishes between the two possible kinds of speech act - hence falsehood becomes the defining characteristic of lie, and native speakers reasonably cite it as such. Thomason (1983) (who also tries to ground Coleman and Kay's analysis in the speech setting) adds two more features to the semantic prototype of lie: "unjustifiability of belief" and "reprehensibleness of motive." However, he himself remarks that unjustified belief in the truth of X directly conflicts with "speaker believes X is false," which he retains; how could both be part of the meaning of lie? Under my analysis, the general maxims enjoining us to inform will also condemn misinformation, even if. not deliberate. Thus, unjustified statements will automatically be judged as like lies in some ways (without changing our definition of lie - false statement in prototypical informative setting). Mere unjustified (sincere) belief does not, however, greatly contribute to my actual classification of even a false statement as a lie. Furthermore, if "unjustified belief" were part of a definition of lie, then even true, sincere, unjustified statements would have to be considered lies to some degree: not a promising result of an admittedly self-contradictory definition of lie. The informationaiity maxims give a more general, coherent explanation of any perceived likeness between lies and unjustified statements. We shall see that Thomason's proposed feature of reprehensibility also follows from a more general understanding of informational exchange and is superfluous to a definition of lie. Notice how rules and maxims change form as they change setting: The general "Help don't harm" is manifested as "Inform others" in the setting in which information/truth is the most relevant beneficial factor. In the domain of politeness, the same general supermaxim is manifested as R. Lakoff's (1973) politeness rules. This model agrees, I think, with our experience: Both information and politeness are considered good and helpful (in their contexts), although in fact the two may conflict when we are unsure which setting takes priority.

52

EVE E. SWBETSER

A lie, then, is a false statement made in a simplified informationalexchange setting. All rules enjoining veracity are in effect, and the speaker is a fully knowledgeable imparter of information to a credulous hearer, Lie has a simple definition within a matrix of cultural models that are independently necessary. The prototype seems to be in the context/rather than in the definition itself. Speakers have difficulty judging whether an action is a lie when they are not sure the action's setting sufficiently matches the prototypical setting specified by the cultural model of informational exchange.6 The next section fits a larger sector of English vocabulary into the cultural model we have outlined; I then go on to motivate our moral condemnation of lying in terms of our cultural models as well. Less simplified worlds, less simple words English has words for false nonlies, or palliated/justified lies. These words mark deviations from the simplified world of the cultural model; thus, examining the deviations may elucidate the model. Common terms include white lie, social lie, exaggeration, oversimplification, tall tale, fiction, fib, and {honest or careless) mistake, some of which appear in Figure 2.1. First, as stressed in the previous section, a lie is not committed if truth is irrelevant. Thus jokes, kidding, and leg-puffings, which exist in a world where humor rather than information is the basic goal, are outside the informational model and cannot be considered lies. Of course, every culture also has a model for humor, and humorous discourse (like all speech) uses some aspects of the informational model. When we cannot decide which model predominates in a given situation, we ask the com; mon (and intelligible) question, "How serious was that remark?" Serious*" ness characterizes contexts, not statements; the same remark may be serious or not, depending on context. Since interlocutors constantly negotiate context (including the predominance of informational or humorous goals), one may ask about a statement's seriousness, meaning the speaker's perception of its micro-discourse context. Tall tales,fiction,and fantasy, when not referring to literature, palliate falsehoods by looking at them as literary, rather than as prototypically informational. The discourse in question is looked at more as a story (with a goal of artistic entertainment) than as facts with relevant truth values, Grandpa's tall tales of fifty-foot snowfalls in his childhood are fun and harmless. Similar claims in a history book, however, would be mistakes, to say the least. Tall tales of huge fish I caught are lies if we are still on the fishing trip and I convince you there is fish for dinner when there is not. I personally only use fantasy and fiction to refer to literature (or to internal, unspoken fantasizing). When fantasy refers to a false statement; however, it seems not only to mean a more artistic story than the truth, but also to include an element of 5e//-deception that further palliates the offense of deceiving others. Any departure from the prototypical infor--

THE DEFINITION OF LIE

53

mational setting, such as weakened truth-value relevance (literary, not informative goals) of less complete control of facts by the speaker, can make the difference between our judging a falsehood as a lie (within the simplified informational world) or as something else (in some other world), such as a tall tale. " - '><".'vjlv^r. v^hu' ' Mistakes are cases in which, without speakers' knowledge, the normal chain of entailment from belief to truth breaks down. Both speaker and hearer think they are in the simplified world delineated by cultural models of knowledge and evidence, but there is an uhknowftdeviation;1 Fofan honest mistake, in particular, the entailment between belief and evidence does hold: The speaker has normally sufficient reason to believe what was said. Carelessness is charged if the broken entailment is between belief and evidence - the speaker should have realized the evidence was insufficient, but failed to. Speakers are responsible for evaluating evidence, so we blame irresponsibility where we would not blame an honest mistake. In either case, however, we assume that the rules ought to hold: Mistake marks a disruption of our simplified informational world's assumptions, rather than an agreed-on suspension (in favor of other goals), as in the case of joke. Lie, on the other hand, denotes a wrong moral choice, with no disruption or suspension of the informational model. As further indication that speech acts are subcases of actions (rather than some separate, parallel category), note that the same word mistake denotes both an unintentional falsehood and a wrong turn taken, or a typo. Ideally, we should be able to justify any act, speech or otherwise; the graver the consequences, the higher the standards for justification. But blameless wrong choices do occur; and if we did our best with available information and resources, unintentional harm can be forgiven. The category mistake is a recognition of human frailty as an allowable out. In exaggerations, oversimplifications, understatements, and other distortions, the informational-exchange rules are more or less consciously bent, rather than suspended or disrupted. Such cases do not strictly follow the dictates of our cultural model; we feel we are being less informational (less truthful) than we might be, hence less helpful. But distortions are not necessarily in direct opposition to truth; they may indicate a subjective personal reaction better than the strict truth could, and hence be truthful at another level. Or, it may be more informational for an expert to oversimplify than to fail totally to communicate with a nonexpert. Many such distortions are indisputably literally false. Whether we judge them as lies depends on (1) whether the setting is prototypically informational and (2) if so, whether they advance or obstruct the informational goals of interaction. White lies and social lies are generally like lies, but they occur in settings in which information might harm rather than help. They are still called lies: even nonreprehensible, deliberate misinformation counts as a lie. In these cases, the entailments of speaker's knowledge, evidence, and

54

EVE E. SWEETSER

intent to be believed (seriousness) still hold; likewise the supermaxim "Help don't harm" holds; but the usual helpfulness of truth cannot be assumed. For a social Ue, the politeness maxims have superseded theJnjunction < to truthfulness. Truth is seen as more harmful to the social situation than minor misinformation would be. In the case of white lies, truth might harm in some other, sometimes more direct, way: Some people would call it a white lie to tell a dying person whatever he or she needs to hear tojdie in peace. Some speakers would also call a (less altruistic) lie told in selfdefense a white lie if it helped them and hurt nobody else.: As with politeness, self-defense is clearly only supposed to be allowed to supersede the informational mode if the consequences of the resulting ^deception are small. The compounds white lie and social lie show in their two elements the conflicting worlds in which the actions take place (it is a lie as an informational utterance, but it is also a social utterance). Figure 2.1 puts them under more than one heading to show this dual categorization.7 There are lies which most people would think justified by some higher good achieved but which would not be called white lies, since their informational consequences are too major (however moral) for us to diminish their status as lies. I would think it moral to lie to the Gestapo about the location of a Jew, but I would call that an unqualified lie. The informational paradigm is fully, even saliently, in effect in this instance - it is only that we feel our uncooperativeness to be justified. Last and least, a Jib is a small or inconsequential Ue, and thus a palliated offense, since the seriousness of an offense of lying is a function of its harmful consequences. However, a fib is nonetheless an offense (though minor). in that it is considered to have at most only a selfish and unimportant reason ' for overriding the usual motivations for veracity. This brings us to the question of the importance of a falsehood or a deception. As Coleman and Kay observe, we can only judge major versus minor deviations from the truth in terms of human consequences. They" contrast an error in the millions column of a city's population (a deception) with an error in the ones column (no deception, because it has no serious consequences). It is clearly only felt allowable to override the truthis-beneficial maxim when the truth-violation could have no negative conr sequences as serious as the negative results of truthfulness. A social Ue cannot be justified as poUte (hence helpful) if it gravely and harmfully misinforms. When truth is more important than politeness, the informational mode cannot be overridden. This merely repeats that our judgement of a lie depends on the extent to which the relevant cultural models are in effect. Knowledge as power: the morality of lying The cultural models relevant to lying also help explain the generally accepted reprehensibility of lies. Coleman and Kay, noting that a lie is no

THE DEFINITION OF LIE

;:;\S§.M^y'^::

more or less a lie because of reprehensible motives on the speakefS'-^i^i^-^f.^/" (consider my Gestapo example as a case of a real lie with good motives^ ^: decide that such motives are typical rather than prototypical of lying.' That ;;-'••. is, lies tend in the real world to be selfishly motivated, just as real surgeons ; ; currently tend to be male; but one cannot claim that maleriess'is iri ;an^^ y;,: way part of the meaning of surgeon. ° i vt-xi:;^ • Placed in the framework of cultural models of discourse and informa-; tion, the variable reprehensibility of lies follows naturally. To the extent that information really is beneficial at a higher level, and false informa:. tion harmful, a lie will harm. General social judgements will condemn deliberate harmful actions. ••'^.•."" Thomason (1983) disagrees that lies are typically reprehensibly motivated; he suggests that social lies are the most common sort of He and are nonreprehensible. I differ with him; social lies are rarely altruistic, though their element of selfishness may not be deeply harmful; and their statistical predominance is unprovable, as a valid survey is surely impossible in this domain. Coleman and Kay correctly reflect a folk understanding that deceit usually profits the deceiver, to the listener's detriment. Thomason's wish to include reprehensibility in the prototype of lie shows that he shares this folk belief in a deep connection between deceit and harmfulness. This deep judgment of falsehoods as inherently harmful goes beyond what we can so far predict from cultural models examined; our informational-exchange model would ask us to condemn falsehood only when, in fact, truth is beneficial and misinformation harmful, so that the simplified world is in effect. I now turn to an examination of the cultural links between information and power, in order to explain why a stigma of immorality attaches to even well-intentioned prevarication. Let us first examine what we do in making an "ordinary" informational statement, true or false, R. Lakoff's (1973) Rules of Politeness, now recognized as a necessary part of our understanding of speech acts, are: 1. Don't impose. (Formality) 2. Give options. (Hesitancy) 3. Make interlocutor feel good; be friendly. (Equality/Camaraderie) Lakoff says (2) explains why a direct command is less polite than an indirect command with the surface form of a request or of a query about the hearer's willingness or ability to do the task. Indirect forms give the hearer options besides obedience or disobedience; the hearer can negatively answer a query about ability without having to refuse compliance directly. Alternatively, indirectness allows compliance without implicit acceptance of the felicity-conditions of a command and recognition of the speaker's authority. Hedged commands avoid assuming ungranted authority over an addressee. Without details of the motivation, Lakoff also says that

56

EVE B. SWEETSER

the same factors make it more polite to qualify assertions with "I guess" or "sorta." This seems a puzzle at first: Why should it be, more polite to guess than to assert, or to make a hedged assertion rather than an unhedged one? Statements have so many purposes that the issue is messier than for * "commands* but the answer (as Lakoff at least implicitly .noticed) is that a statement does something to the hearer, just like other speech acts. It pushes at the hearer's belief-system. An informative speaker requires a hearer ready and willing to believe, or information cannot be imparted. This cooperative hearer grants the speaker a good .deal of power to push around certain aspects of his or her belief system.8 English reflects the equation of knowledge with power, in the uses of a group of hedges that mark the evidential status of statements. Some examples of evidentiality-hedges are: to the best of my knowledge; so far as I know; if I'm not mistaken; as far as I can tell; for all I know; as I understand it; my best guess is; speaking conservatively; at a conservative estimate; to put it mildly; beyond question. The literal use of these hedges is to limit the speaker's normal responsibility for the truth of assertions. An assertion has the precondition (Searie 1969) that the speaker be able to provide evidence for its truth. Or, in terms of our cultural models of information and evidence, in an. informational setting a hearer knows that a cooperative speaker will only state justified beliefs. However, even reliable-looking evidence can turn out to be insufficient. Evidentiality-hedges allow the hearer access to the evidence-evaluation and thus transfer some of the speaker's evaluative responsibility to the hearer. They avoid potential charges of carelessness or irresponsibility by not allowing the hearer to over- or undervalue the evidence supporting the hedged assertion. (Cf. Baker 1975 on some related ,',. hedges that signal and excuse potential discourse violations.) G. Lakoff points out (personal communication) that responsibilitytransfer goes even further. Not only can we qualify a statement's evidential status, but we can also evade personal responsibility for the original ,-<•• (prequalification) statement. For example: to the best of our current knowledge to the extent to which this phenomenon is understood at all so far as can be judged from work to date according to the current consensus in the field This last set of hedges makes criticism or disagreement difficult; whereas . if the speaker had simply evidentially qualified his or her personal evaluation, the hearer could easily disagree (though not accuse the speaker of irresponsibility or prevarication). At the opposite end of the spectrum, .; hedges such as speaking conservatively commit a speaker to an assertion's high evidential status (another example is all the evidence points to the conclusion that). Evidentiality-hedges, then, allow the speaker to modify

THE DEFINITION OF LIE

57.

the normal degree of responsibility for a statement's truth by qualifying its evidential status. Unqualified statements presumably take on a default level of responsibility, varying with context. ,• However, evidentiality-hedges have another function besides the metalinguistic evaluation usage just described; they also function aspragmatic deference-markers. However sure a student may be of one of the foIldw-V ing assertions, he or she might have social motivation to mark uncertainty with an evidentiality-hedge: .-•••.,• i'>r.v.. - C ^ ^ K ^ ^ : - - ^ But, Professor Murray, as far as I can tell, this parallels Andrews'. example, which suggests another interpretation.: ^:; :.i Professor Jones, if I'm not mistaken, haven't Smith's recent results made the Atomic Charm hypothesis look dubious? When social authority is low, the right to push people's belief systems is correspondingly low. Especially if our hearer may be unwilling to listen and change opinions, we have to be socially careful; we have no more authority to command belief changes than any other action against the will of our interlocutor. Evidentiality-hedges thus hedge both kinds of authority that underlie an assertion: informational authority (evidence) and social authority (we cannot as readily command belief-systems of people higher on the social scale). This is a natural pairing, considering our understanding of assertion as manipulation of belief systems. In a prototypical informational exchange, the hearer is as ignorant and credulous as the speaker is knowledgeable and ready to inform. Who has the upper hand in such an exchange - the knowing and manipulative speaker, or the ignorant and passive learner? Teaching (a relatively one-way exchange, at least in early stages) has aspects of authority even without a surrounding institutional power-structure. To a lesser degree, any assertion has the same inherent power structure. In further support of this analysis, note that a person with both kinds of authority can lay aside either kind with an appropriate evidentialityhedge. A professor who wants to get a point out of a student rather than giving the answer may thus lay aside both aspects of authority, in a statement like: But as I understand it, semantics is the study of meaning - so how does it strongly depend on spelling, Mr. Smith? Too many such hedges from the professor would sound sarcastic, since it is insincere to deny the existence of one's power position while leaving its broader social presence unchanged. As further evidence that speakers link assertion with (a) request for belief and (b) assumption of an authority position, consider the following hedges:

58

EVE E. SWEETSER

(Please) believe me:. . . / don't ask anyone to believe this, but . . . i" can't expect you to believe me, but . . . " These hedges mark unreasonable belief-requests, tacitly assuming that an ordinary belief-request is just a matter of course. / can't expect you to believe me needs to be stated, even though our normal right to such an expectation passes unnoticed and unstated. Phrases like the strength of an assertion, or the authority for a statement, are not random. Both social and informational authority structure our discourse world, and the strength of an assertion depends on both. If either kind of authority is extremely strong, it may overcome opposition from the other: An undergraduate who is very sure of a fact may correct a department chair, and a dean may feel freer than a student to speculate, having more social protection from contradiction. Thus, our cultural model of information as power motivates evidentiality's relationships with politeness and authority. Incidentally, Grice's (1975) maxims are often cited as barring assertions that are obvious or well known to the hearer because they are useless and uninformative. However, I have not seen it overtly said that obvious statements are also often insulting. Their rudeness cannot be deduced from their uninformativeness but follows directly from viewing them as unwarranted assumptions of informational authority ("I know better than you").9 This view may help explain the Coleman example (P. Kay, personal communication) "Crete is sort of an island," where sort of appears to hedge neither the choice of the word island nor the precision of the truth-value, but the act of asserting is weakened to avoid rudeness. V Conversely, Jef Verschueren (personal communication) points out to me that the idea of informational authority gives added motivation (besides Lakoff '$ rules) for seeing questions about ability or willingness as politer than direct commands. Question form has the inherent courtesy of giving ; the addressee a presumed informational authority. It is no huge politeness to assume an individual is the best authority on his or her own wishes and abilities. The contrary assumption, however, is ipso facto particularly counter to the rules of politeness, unless either camaraderie or unusual social authority overrides politeness. A direct command thus indicates presumed unconcern for whether the addressee has opinions, let alone what they are - and in a domain in which that person is the evident authority (i.e., his or her own internal state). Verschueren also drew my attention to the contrast between an indirect but less polite "The window's open" (in a rude tone, to hearer who sees the window) and a direct but more polite request or command "(Please) close the window." Here I feel, the chosen mode of indirectness is more insulting than a direct command ~ the statement implies either (I) that the hearer is so unaware of the obvious that the assumption of informa- -.

THE DEFINITION OF LIE

';-'::W^S'--''.

tional authority is warranted OR (2) even greater social authi^ty^hariF^^ a command; the hearer is expected not only to obey, but also to iducgivtand meet the speaker's wishes before they are stated (the hearer-jkt^^^^f^ seem to mind the open window).10 For me, the politeness-contrast-Me|;;^v;-''••/:•• verses (as expected) if 'The window's open" is said courteously, to:f:;|>^|5:& son who somehow (mental absorption? a physical barrier?) just:hasftnW^7' noticed but might reasonably share the speaker's concern. These exainples:. .• demonstrate the complex interplay between informational and sociaI;au^ •: thority in determining politeness. .••';..:-.-. ., •-;:• :}^i^y^;::y From the preceding discussion, lying emerges as serious authority^ abuse. Authority relations structure the prototypical informational ex~ change, the setting in which lie is defined. As we get further from the simplified world in which the credulous hearer depends on the speaker for some crucial information, truth becomes less relevant and falsehood less reprehensible. In the simplified world, however, (barring major reversal of social authority and morality judgments, as in the Gestapo example), falsehood constitutes a deliberate use of authority to harm someone in a weaker, dependent informational position. We thus naturally judge it as immoral, barring exceptional extenuating circumstances. As salient examples of our view of lying as authority-abuse, let me cite the anger of patients lied to by doctors, or of children systematically lied to by adults (e.g., about sex). Doctors in particular derive much of their authority from large amounts of knowledge that is not otherwise accessible to patients. By refusing information or misinforming, they can control important decisions for patients. To a lesser degree, any possessor of information can influence or control less knowledgeable hearers. To the extent that we feel people should control themselves, lying is immoral because it undermines the potential for self-determination." This deep identification of lying with power abuse may explain why for some people all lies retain some reprehensibility, however good the motive. Deception and lying Lies are only a subclass of deception. Any deception, in that it induces false beliefs in a credulous hearer, is a culpable abuse of informational authority and naturally liable to the same moral charges leveled at a lie. But oddly enough, speakers often feel less immoral if they manage to deceive rather than to lie straight out. Victims conversely feel that such a deception is a dirtier trick; they cannot complain of being lied to and resent the deceiver's legal loophole. There thus seems to be a further folk belief that literal truth and real truth (honest information-transmission) are prototypically connected. A literally true statement thus retains vestigial legality (if not morality), even if it misleads, whereas a deliberate factually false statement retains some stigma of reprehensibility, even with strong moral justification. Folklore

60

EVE E. SWEETSER

gives magical power to literal truth, and a common folk theory is/that law also emphasizes literal truth rather than informativeness (I do not know, about modern perjury laws). Some people would find lying to the Oestapo immoral; yet most of them would think it laudable to mislead villains, ^ saving an innocent victim. In any case, complete dissociation between literal and "real" truth, or between the latter and morality, is regarded as highly atypical. ... .; . , .;^ A common way to mislead is to imply, but not overtly state, the false proposition to be communicated. The overt statement and the false proposition are often linked by Gricean conversational implicature; the utterance is irrelevant or insufficient in context, unless the hearer also assumes the unspoken falsehood. In such cases, the speaker could without selfcontradiction go on to cancel the deceitful implicature. Taking a case from Coleman and Kay: "Mary, have you seen Valentino lately? ".. Mary: "Valentino's been sick with mononucleosis all week." Mary could go on, "But I've visited him twice." Part of people's disagreement about the morality of misleading (and about whether it constitutes lying) may be genuine disagreement about the degree to which a conversational implicature constitutes a "statement" and hence makes the speaker responsible for having said it. As Thomason says, some speakers are so sure the implicature was present that they include it in a restatement: "Mary said No, Valentino had been sick." The plot thickens as the implicatures become more closely bound to the linguistic form. Such implicatures seem to me to be closer to statements than Mary's implicature about Valentino. Thus, I would predict that an utterance such as "Some of my students cut class," (used when not one showed up) would impress speakers as closer to a prototypical lie than ). Mary's statement. An even more difficult case is that of presupposed falsehoods. How close to lies are statements such as "He's only a sophomore, but he got into that course," used of a student at a two-year college where sopho-; mores are the most privileged students, and said to deceive the hearer about the nature of the college or the course? I personally rate these examples high. I hope in the future to investigate what constitutes "stating," as well as what constitutes lying. Our cultural model of representation is essential to our understanding of misrepresentation. Cross-cultural parallels Anthropologists interested in cultural models, or linguists interested in culturally framed semantics, now ask "How universal or culture-bound are the cultural models we have just examined? " I have used English data (like Coleman and Kay); studies of French (Piaget) and German (Wimmer and Perner, above) child language agree with each other and are highly compatible with my proposed analysis of the English verb lie. These

THE DEFINITION OF LIE

61

linguistic communities also share the accompanying moral judgements of lying, probably due to shared understanding of power structures and informational exchange. However, a first glance at more distant cultures shows a startling degree of surface variance as to the morality of misleading or lying. Ochs Keenari (1976) discusses the frequency (and acceptability) of vague or misleading answers to questions in a small Malagasy-speaking community. Gilsenan (1976) states that successful lying is a major positive status-source for males in a Lebanese Arabic-speaking community. In what respects do these groups differ from English speakers? My answer is that, on examination, these cultures differ from ours much less than the isolated statements above might indicate. At least, the differences are not in their understanding of informational exchange, evidence, or abuse of informational power. Ochs Keenan's Malagasy community, while agreeing with English speakers that information-giving is cooperative and useful, has a different idea of when a hearer has a right to such cooperation. Europeans or Americans might think of their own contrast between "free goods" (any stranger gets a reply to "What time is it?") and other facts (e.g., one's age, or middle name) that need a reason to be told. Malagasy speakers place an even higher power-value on information than do English speakers (news is rare in small communities) and naturally hoard precious and powerful knowledge; questioners cannot expect as broad a spectrum of free goods in such a society, and day-to-day informational demands have less right to expect compliance. Malagasy speakers are not uncooperative when refusing information could seriously harm (e.g., if asked "Where's the doctor?" by an injured person). Our classic informational-exchange setting is just not in place as often as in an English-speaking community; since Malagasy speakers all know this, their equivocations do not manipulate unsuspecting addressees. The Malagasy community shares basic cultural models of information and truth with English speakers, but evokes them under different circumstances. We might note here that lying to enemies is often culturally accepted. Many English speakers think such lies less immoral than lies to trusting friends, who are "owed" more sincerity (Coleman and Kay cite speakers who, extending this scale, said Mary did not "owe" John the truth about Valentino, as they were not engaged). In some cultures, lying may be forbidden primarily within the group; but such a culture does not lack our judgment of lies as harmful. Rather, their rule about who should not be harmed is different. Gilsenan's Lebanese village is an even more complex case. He states that this community thinks lying immoral, probably for the same reasons we do. Community members caught lying lose status and honor. However, certain restricted kinds of undetected lies told by adult males can be extremely status-productive. First, verbal self-presentation is highly competitive for Lebanese men,

62

EVE E.SWEETSER

so false (or unfalsifiable) boasts are profitable, though detection causes corresponding status-loss. Conventional verbal competition gives nonm:, formational aspects to Lebanese boasts (though not as formalized as, 'e.g^ Turkish, or urban black American,,boys*.boasting or insults)..English. "speakers might lie competitively in other areas, and less conventionally; but the Lebanese view of lying is not in serious conflict with .our„.o.wn ,5 The second way a Lebanese man can gain status by lying, is to Ie.ad another man "up the garden path" and subsequently reveal the deception.; He must avoid detection, or it may be difficult to prove he did not mean to deceive permanently. A "garden path" is crucially not real lying, since it achieves its goal only by eventual truth-revelation. Thus, such deceptions do not show a different idea of lying from ours; but why do these. play-lies give status? ••:-a<:-^ Gilsenan explains that discernment is a major source of prestige for Lebanese men: A reputation for telling truth from falsehood is valued especially in religious leaders, but also in any adult male. He tells of a visiting religious leader who upstaged the village religious leader (a man with a long-built reputation for discernment, even omniscience). A village man, resenting the intruder, perpetrated and then publicly revealed a successful minor hoax on him; he left, discredited. Lebanese "garden-path" lies are usually less important, but do cause real status - gain or loss unlike American April-fools or leg-pulling. Lebanese society evidently has conventionalized competitive uses of informational power; men overtly gain.power by forcing false beliefs on others or by seeing through false claims (exposing the author as nohauthoritative, dishonorable, or simply unsuccessful at one-upping). Serious use of this power by lying would be immoral, but one can conventionally . display power without using it - as a martial arts victor does not kill but shows that he has overcome his opponent and could kill. A martial arts victor's status need not indicate corresponding cultural approval of actual killing or assault; nor should status given by "garden paths" be taken 4 as indicating general social approval of lying. Very different cultures emerge from this discussion as possessing saliently similar understandings both of lying and of the general power and morality dimensions of informational exchange. This similarity presumably stems from universal aspects of human communication. Where cultures differ appears to be in delimitation of basic "informational exchange" settings and in conventional use of the relevant power parameters. Folk models of knowledge and mformativeness (and the corresponding semantic domains) may universally involve strong shared elements. Conclusions A lie is simply a false statement - but cultural models of information, discourse, and power supply a rich context that makes the use of lie much

THE DEFINITION OF LIE

63

more complex than this simple definition indicates. Definitions of morally, informationally, or otherwise deviant speech acts follow readily from a definition of a simplified "default" Speech world. The cultural models in question not only underlie a whole sector of pur vpcab$a^ vate our social and moral judgments in these areas;,they .furtherAppear to have strong shared elements cross-culturally.•:.,: ;;,.•;;. Cultural models underlying linguistic systems are a fairly new area of analysis, though a few people were ahead of the rest bf us (Becker;1975 is a good example). However, collaboration among iiriguists,,anthropologists, and other social scientists in this area looks increasingly fruitful. My own preference for this approach stems from both its intuitive plausibility (ethnographers, if not grammarians, have long known that wordmeanings are interrelated with cultural models) and its explanation of a long-term paradox facing semantic analysts. Word-meaning has orderly aspects that make us feel that it ought to be simply formalizable; yet we all know from bitter experience how readily the complexities of meaning elude reductionistic formal analysis. If the analyst's intuitive feeling that definitions are simple is right, then perhaps much of the fuzziness and complexity lies in the context of meaning, rather than in the meaning itself. A better understanding of cultural models (aided by research such as that represented in this volume) is important to lexical semantics: Words do not mean in a vacuum, any more than people do. This paper leaves many unresolved problems. It is insufficient to discuss one cultural model or folk theory of speech (here, our default model of literal discourse as informational) as if it were largely independent of all the other models relevant to verbal interaction. Our folk understanding of knowledge also needs more investigation. On the linguistic front, in which cases can we expect the fuzziness of fuzzy semantics to be ultimately locatable in the sociophysical world (or in our perception of it), or in the fit between the world and a cultural model; and in which cases, if any, can we expect inherently fuzzy semantics? This last question can be answered only as we learn more about the relationship between linguistic and social (even metaphorical) categorization. Just now, I must be content with showing that a simpler semantics of lie follows from an analysis of the cultural models relevant to prevarication. Notes l. Only members of the Berkeley linguistic community will understand how much this work owes to their ideas and support. However, my intellectual debt to my advisors, Charles Fillmore and George Lakoff, should be evident. Linda Coleman and Paul Kay, original inspirers of this project, were patient and intelligent critics throughout. I have also benefited from the insightful comments of Susan Ervin-Tripp, Orin Gensler, David Gordon, John Gumperz, Dorothy Holland, Mark Johnson, Naomi Quinn, John Searle, Neil Thomason, Jef Verschueren, Jeanne Van Oosten, and the participants in the Princeton

64

« -"2.

3. 4.

5.

6.

7.

8.

9.

10.

11.

E V E E . SWEETSER

Conference on Folk Models. An earlier version of the paper was presented in the symposium Folk Theories in Everyday Cognition, organized by Holland and Quinn for the 80th Annual Meeting of the American Anthropological Association, 1981. The tetmfolk theory, which I originally used throughout, emphasizes the nonexpert status of such a theory or model; cultural model, which I am now adopting, stresses the fact that our cultural framework models the world for us. I have retained the word folk in contexts where Ifindit particularly useful. For a recent and complete survey of work on linguistic categorization, see G. Lakoff (in press). Coleman and Kay presented subjects with a series of short fictional scenarios, asking the subjects to judge in each case (I) whether a lie had been told in the interaction described and (2) how sure the subject felt about this judgment. The actions described in the scenarios varied independently with respect to deceptiveness, factual falsity of statements made, and speaker's belief of the content of the statements. Susan Ervin-Tripp has suggested to me that young children are simply "behaviorists," judging acts by result, not by intent. Before children can state their intentions, they are bound to get rewarded and punished behavioristically. Four- to nine-year-olds are certainly not insensitive to intentions but may remain behaviorists enough to class lies with other false statements. Paul Kay has brought to my attention a playful usage that seems odd in the context of either a feature or a prototype analysis of lie: "Do you know, I thought I told the truth the other day, but it turns out I lied to you: I'm so sorry." This usage seems to me parasitical on serious usage in that the speaker jokingly attributes to a past speech act his or her current mental knowledge-space (in Fauconnier's [1985] sense of mental space). Since past acts are not actually judged in the light of subsequently gained knowledge, we find this amusing. Lakoff (in press) comments that social lie and similar collocations pose problems for the theory of complex categories. A prototypical social lie is not necessarily a prototypical lie. Without proposing a new theory of complex categories, I feel it is clear that social lie is not an intersection of the categories lie and social act. Rather, it is viewed simultaneously (and perhaps somewhat contradictorily) as a member of two categories that we do not usually understand as interacting at all. Social rights and responsibilities are reciprocally arranged: If the Speaker has the right (authority) to say X, then the Hearer has a duty to believe it. If H has a special right to hear (to know) X, beyond the general right to information, then S has a correspondingly more important duty to tell X to H. Paul Kay has suggested to me that the rudeness of telling someone what they already know is best compared to the rudeness of giving an unnecessary or redundant gift. However, such gifts are only rude (/"they imply an unwarranted power-assumption. If I give you a paperback you own a copy of, I'm only rude if I thereby (unjustifiably) purported to extend your literary horizons; but if I pay for your bus ticket (which you are presumed capable of buying), then I'm rude unless you asked for help with change. All valuable resources, like information, confer power on their owners. Forman (n.d.), in a (somewhat astonishingly) still unpublished paper, "Informing, Reminding, and Displaying," elucidates the informational uses of apparently noninformative statements; he would categorize this as an example of informative reminding. Bok (1979) provides a treatment of the social issues involved in lying and decep-

THE DEFINITION OF LIE

65

tion. One case she analyzes is that of a woman who was the only likely kidney donor for her daughter and overtly willing. Perceiving severe repressed fears in her, doctors falsely told her that she was not physically compatible enough with her daughter to be a good donor. This deception robbed her of the chance to confront her fears and make her .own decision about giving the,kidney. Bok also notes that deception is less frightening if we ourselves have anthbrized the deceivers and are aware of their tactics. Unmarked traffic control cars voted into use by the community are less threatening than if the. police use them without citizens'input. .C-./^^U References Baker, C. •.<• 1975. This is Just a first approximation but. . . . In Papers from the Eleventh Regional Meeting of the Chicago Linguistic Society. Chicago: Chicago Linguistics Society. Pp. 37-47. Becker, A. 1975. A linguistic image of nature: The Burmese numerative classifier system. Linguistics 1975 (165): 109-121. Bok, S. 1979. Lying: A Moral Choice in Public and Private Life. New York: Pantheon Books. Coleman, L. and P. Kay 1981. Prototype semantics: The English verb 'He.' Language 57(l):26-44. Fauconnier, G. 1985. Mental Spaces. Cambridge, Mass.: M.I.T. Press. Fillmore, C. J. 1977. Topics in lexical semantics. In Current Issues in Linguistic Theory, R. Cole, ed. Bloomington: University of Indiana Press. Pp. 76-138. Forman, D. n.d. Informing, reminding, and displaying. Unpublished manuscript, Department of Linguistics, University of California, Berkeley. Gilsenan, M. 1976. Lying, honor, and contradiction. In Transaction and Meaning: Directions in the Anthropology of Exchange and Symbolic Behavior, A.S, A. Essays in Social Anthropology, Vol. 1, B. Kapferer, ed. Philadelphia: Institute for the Study of Human Issues. Pp. 191-219. Goffman, E. 1974. Frame Analysis: An Essay in the Organization of Experience. New York: Harper and Row, Publishers. Gordon, D. P. 1974. A developmental study of the semantics of factivity in the verbs 'know,' *think,' and 'remember.' Ph.D. dissertation, University of Michigan. (Published as Natural Language Studies, No. 15. Ann Arbor: University of Michigan Phonetics Laboratory.) Grice, H. P.. 1975. Logic and conversation. In Syntax and Semantics, Vol. 3, Speech Acts, P. Cole and J. Morgan, eds. New York: Academic Press. Pp. 113-127. Kay, P. 1983. Linguistic competence and folk theories of language: Two English hedges. In Proceedings of the Ninth Annual Meeting of the Berkeley Linguistics Society. Berkeley: University of California. Pp. 128-137.

66

EVE E. SWEETSER

Kay, P. and C. McDaniel , . i 1978. The Linguistic Significance of the Meanings of Basic Color Terms. Language 54(3):610-646. .^Lakoff, G ,

• • - .

••/•..

.;'

•

•

1972. Hedges: A study in meaning criteria and the logic of fuzzy concepts. In Papers from the Eighth Regional Meeting of the Chicago Linguistic Society. Chicago. Pp. 183-228. (Reprinted in Journal of Philosophical Logic 2:458508, 1973.) -")^'"i Lakoff, G. In press. Women, Fire, and Dangerous Things: What Categories Reveal about the Mind. Chicago: University of Chicago Press. Lakoff, R. 1973. The logic of politeness: Or minding your P's and Q's. In Papers from the Ninth Regional Meeting of the Chicago Linguistic Society. Chicago: Chicago Linguistic Society. Pp. 292-305. ""''. . Mervis, C. and E. Rosch 1981. Categorization of natural objects. Annual Review of Psychology 32:89115. Ochs Keenan, E. 1976. On the universality of conversational impEcatures. Language in Society 5(l):67-80. Piaget, J. 1932. The Moral Judgement of the Child. M. Gabain, trans. New York: Harcourt Brace. Rappaport, R. A. 1976. Liturgies and lies. In International Yearbook for the Sociology of Knowledge and Religion, Vol. 10, G. Dux, ed. Freiburg, Germany: Westdeutscher Verlag. Pp. 75-104. Rosch, E. 1978. Principles of categorization. In Cognition and Categorization, E. Rosch and B. Lloyd, eds. Hillsdale, NX: Lawrence Erlbaum Associates. Pp. 27-48. Searle, J. 1969. Speech Acts. Cambridge, England: Cambridge University Press. Thomason, N. R. 1983. Why we have the concept of 'lie' that we do. Unpublished Ph.D. dissertation. University of Oregon.

3

•

**

Linguistic competence and folk theories./••.•^•....•-^: of language •• v ; V ^ ^ ^ T : TWO.ENGLISH HEDGES1 • Paul Kay

.; ^" : ". ' ^ '

,.^@

'•.'•,-,..•.•: '.'•'*•;.']

In the ordinary sense in which we say that words like chair and table are ABOUT furniture, hedges are words about language and speech. There is nothing remarkable in this; language is part of our environment, and we have words about most things in our environment. The linguistically interesting aspect of hedges is that, although they are about language, they are not exactly used to talk about language as we would say that chair and table are used to talk about furniture or, for example, gerund and entailment are used to talk about language. When we use a word like chair or table or gerund or entailment, chairs, tables, gerunds, and entailments do not become ipso facto part of what is said. With hedges it is different; when we use a hedge like loosely speaking, the notion of "loose speech" which this expression invokes becomes part of the combinatorial semantics of the sentence and utterance in which it occurs. A familiar (if probably vacuous) combinatorial semantic rule is (SR) If adjective a denotes class A and noun n denotes class AT, then the denotation of the expression an is the intersection of the classes A and N. I wish to claim that the notion of "loose speech" is part of the combinatorial semantics of sentences containing the expression loosely speaking in the same way in which the notion of class intersection is claimed by proponents of (SR) to be part of the combinatorial semantics of an expression like red chair, A hedged sentence, when uttered, often contains a comment on itself or on its utterance or on some part thereof. For example, when someone says, Loosely speaking France is hexagonal, part of what they have uttered is a certain kind of comment on the locution France is hexagonal. In this sort of metalinguistic comment, the words that are the subject of the comment occur both in their familiar role as part of the linguistic stream and in a theoretically unfamiliar role as part of the world the utterance is about. Such metalinguistic reference seems unaccounted for (and perhaps unaccountable for) in standard theories of semantics that are based on a contextfree, recursive definition of truth for sentences, and in which linguistic

68

PAUL KAY

objects and world objects (or objects in a model) belong to disjoint realms. The problem, I believe, goes beyond that of indexicality as usually conceived, and although it would be interesting to investigate irt detail the relation between the kinds of facts to be discussed here and discussions of indexicality within model theoretic semantics (e.g.; Kaplan 1977), that comparison will not be attempted. The omission might be justified by appeal to limitations of space, but such a plea would be less than candid, as I suspect that the phenomena I will describe constitute a principled set of exceptions to any theory of natural language meaning that makes a rigorous separation between truth conditional meaning for linguistic types (i.e., sentences), normally called semantics, and other aspects of meaning, frequently called pragmatics (see, for example, Gazdar 1979:2f). The latter claim would, to be sure, require considerable clarification before a demonstration could be begun. In this chapter I must content myself with presenting a few facts and some timid empirical generalizations. The principal conceptual tool I will employ for stating these empirical generalizations will be that of folk theory. The term is borrowed from anthropology. In describing the system of knowledge and belief of another culture, an anthropologist speaks of that culture's folk theory of botany, the emotions, language, and so on. Anthropologists discover such folk theories by analysis of the use of words in the native language. The guiding idea is the familiar one that any natural lexicon implies a tacit, structured conceptualization of the stuff that the words of that lexicon are about. What the words we shall be concerned with here are about is language and speech, and the folk theory we shall be looking for is the tacit and mostly unconscious theory of language and speech we invoke when we employ certain parts of the lexicon of English. The present essay is thus in the first instance lexicographical. But we will see that in the domain of hedges, lexicography is inseparable from combinatorial semantics because the schemata or folk theories that constitute the semantic content of the hedges as lexical items serve as combinatorial structures for putting together the meaning of the sentences in which the hedges occur. Hence, world knowledge about language - what I have called folk theories of language - may at times be part of knowledge OF language. Knowledge of a language, linguistic competence, is commonly distinguished from knowledge of the world. Linguists do not generally consider it a matter of interest that the language we are competent in is also in our world and therefore a thing of which we have world knowledge, that is, a folk theory. Certainly linguists do not often ask whether world knowledge of language bears some special relation, that other sorts of world knowledge do not bear, to the knowledge that constitutes linguistic competence. Perhaps the question is not posed because the answer is considered obvious, namely No. The facts to be considered below suggest, however, that the folk theory of language presupposed by various hedges

LINGUISTIC COMPETENCE AND FOLK THEORIES OF LANGUAGE

69

should be interpreted both as world knowledge ABOUT language and as knowledge OF language - i.e., as part of linguistic competence. ;••••: The data to be considered in this chapter concern two hedges, loosely speaking and technically. The concept folk theory will figure in the analysis. of the meaning of each of these expressions. The comparison of the'two analyses Will reveal a not altogether obvious difference between the folk theories that constitute our tacit knowledge of the world (as realized "in word meanings) and consciously formulated theories: folk theories, like conscious theories, answer to a requirement of local consistency but, unlike conscious theories, folk theories answer to no requirement for global consistency. Loosely speaking The hedge loosely speaking may be employed in the service of a variety of semantic and/or pragmatic functions which, from a traditional point of view, appear disconcertingly diverse. Let us consider some of the possible semantic-pragmatic roles of loosely speaking in the response of Anthropologist A to Layman L in the following dialogue. (1) L: Where did the first human beings live? A: Loosely speaking the first human beings lived in Kenya. First, believing the evolutionary process to be inherently gradual, A may consider the expression the first human being to be semantically informed and hence devoid of the capacity for nonvacuous reference. If A had this problem, believing that the first human being could not possibly refer to anything, he might reply more fully (2) A: Strictly speaking, one can't really talk about "the first human beings," but loosely speaking, the first human beings . . . Secondly (and alternatively), A may think that the first human beings is a normal referring expression, but not the one that picks out exactly the entity about which he wishes to assert lived in Kenya. For example, A may consider it important to distinguish in this context the first human beings and the first human beings known to science. If this were A's reason for hedging with loosely speaking, his fuller answer might be along the lines (3) A: Strictly speaking, we can only talk of the first human population known to science, but loosely speaking, the first human beings . . . A's problem may be not with the first human beings but rather with in Kenya. A third motivation for loosely speaking could then be that A considers the unhedged sentence The first human beings lived in Kenya to have a reading which presupposes the modern nation of Kenya to have existed at the time the first human beings were alive. Such fastidious pedantry might motivate a longer reply along the lines

.70

rPAUL>KAY

(4) A: Loosely speaking, in Kenya; strictly speaking, in the place now called Kenya. Fourthly, and perhaps most typically, A may think that the unhedged sentence The first human beings lived in Kenya oversimplifies or otherwise distorts the pertinent facts^ but is nonetheless the best he can do given the exigencies of the conversational situation. Sometimes the demands of Gricean Quantity (Say no more than necessary) and Manner (Be brief) require a sacrifice in Quality (Teh the truth).Tn bur present example, the relevant facts might involve sites not only in Kenya but also in Uganda and Tanzania, fossils of uncertain relation to each other, and so on. Loosely speaking can be and probably often is used to apologize for this sort of deficiency in Quality, induced by the demands of Quantity and Manner. The fuller version of A's reply could be something like (5) A: Loosely speaking in Kenya. Strictly speaking, we are dealing here with a complex situation involving sites mainly in Kenya, but also in Tanzania and Uganda, and with a set of fossils which may not all represent the same species . . . Examples (2-5) illustrate four distinct kinds of "loose speech" that the hedge loosely speaking may reflect in (1): (i) the use of an incoherent description in an act of reference (2); (U) the use of a coherent but "wrong" description in an act of reference (3); (iii) the utterance of a sentence that (the speaker feels) permits an unintended interpretation that contains a false presupposition (4); and (iv) the utterance of a sentence that is defective in Gricean Quality, that is, in truth (5). What, then, does loosely speaking mean? George Lakoff (1972) gives the example (6) (a) A whale is a fish. (FALSE) (b) Loosely speaking, a whale is a fish. (TRUE) and argues that the semantic function of loosely speaking is that of a predicate modifier which, through selection of certain features of meaning internal to the intension of a category word like fish, maps it into another category-type intension. But we see that this cannot be correct, since in (1) loosely speaking does a variety of things that have nothing to do with the modification of a category word. Furthermore, it may do several of these things simultaneously: in uttering his part of (1), A might be bothered by any combination of the factors discussed in connection with (2-5) (except of course those combinations containing both (2) and (3), since these happen to be mutually exclusive]. Thus the semantic scope of loosely speaking must be at least as broad as the entire sentence it accompanies, for example, in (6)(b) the sentence A whale is a fish. Since presence or absence of loosely speaking in a sentence such as (6) may affect our judgment of its truth, the classical view holds that loosely speaking must make a contribution to the semantics of the SENTENCE in which

LINGUISTIC COMPETENCE AND FOLK THEORIES OF LANGUAGE

71

it occurs. Since the scope of loosely speaking must be at least as broad as the whole sentence it accompanies, its scope must be that whole sentence^ and one is tempted to conclude that the semantics of loosely speaking is a function from sentence intensions to sentence intensions - that is, from the set of worlds in which whales are fish (the null set in some theories) to the set of worlds that are like this one with respect to the fishiness of whales. But nothing of this sort can be right because, as we saw in connection with (2), loosely speaking sometimes functions to comment directly on the FORM of the sentence it accompanies. " .* Moreover, when (6)(a) is changed to (6)(b) by the addition of loosely speaking the reason that our judgment changes from false to true is not that a false proposition P( - A whale is a fish) has been changed into some true proposition P'. Rather, we abstain from judging (6)(b) false because we understand (6)(b) both to assert the sentenceA whale is a fish and to express a reservation regarding the adequacy ofthat assertion. If the dimension of adequacy is taken to be that of truth (tightness of "wordto-world fit") as seems to be the relevant dimension in the case of (6)(b), then we have no trouble accepting a judgment of true. In the general case, however, the dimension of adequacy directly addressed by the hedge loosely speaking need not be that of truth: the loose speech referred to may involve laxness in obedience to the rules of language, as in (2) and perhaps (3) or even looseness with respect to stylistic canons, as in (4). Of the four examples, (2-5), only (5) directly concerns truth, and even in this case, we do not experience (1) as expressing some proposition P', which is distinct from but closely related to The first hiimdn beings lived in Kenya, and which is exactly true. The empirical claim about loosely speaking that I have attempted to develop may be summarized as follows: (7) For any sentence 5 of the form loosely speaking P, where P is a declarative sentence, an utterance of S constitutes two acts: (i) an act of asserting P, (ii) an act of warning that (i) is in some way a deviant (loose) act of assertion. Probably the most typical way for an assertion to be deviant is in terms of Quality, but, as we have seen, an assertion may have other kinds of defects about which loosely speaking warns. If (7) is even approximately correct, expressions such as loosely speaking present an interesting challenge to current formal theories of semantics and pragmatics. If loosely speaking means what (7) says it means, this is surely its literal meaning (not figurative, ironic, et cetera). Although (7) specifies the literal meaning of loosely speaking, (7) does not consist of a specification of truth conditions of either S or P, but rather expresses a warning to the addressee that he should be wary in his acceptance of the assertion of P. If (7) is correct, literal meaning and truth conditions cannot always be the same thing, not even almost the same thing.

72

PAUL KAY

; It is not obvious how the meaning of an expression like loosely speaking is to be captured in a theory pf the generally accepted kind,/where. the truth conditional meaning of a sentence is established in terms of a possible world semantics independent of pragmatic considerations, [and no feedback from pragmatic reasoning to literal meaning is countenanced, But even supposing that with sufficient ingenuity we could develop art account of loosely speaking within this kind of framework, it is.not clear that we should wish to do so. If we look at the different kinds pf semanticpragmatic functions that may be accomplished by loosely speaking [illustrated, though by no means exhausted, in (2-5)], we find that they Constitute, from the traditional view, a disparate collection. Another way to view the same matter is to notice - as the reader may already have done that (7)(ii) is stated far too broadly. Loosely speaking doesn't point to just any kind of deviance in an act of asserting. For example, acts of asser: tion that deviate because they contain uninterpretable indexicals or because they fail to answer a question just posed are not examples of "loose speech." looset (8) (a) Jack and John were running y sPf^"S he j m w w s and j * ( one of them j (b) A: When did Mary get her car tuned up? B: *LooseIy speaking, because the engine was knocking.

down

I have spoken informally of the various kinds of "loose" speech represented by examples (2-5), and in this informal usage I think lies the key to the semantic unity of the expression loosely speaking. I suggest that what enables us to speak informallyjabout "loose" speech in connection with all of these examples is what constitutes the actual semantic unity of the expression loosely speaking. In every utterance of a sentence like (1), the linguistic act of asserting that the first human beings lived in Kenya is talked ABOUT (in the same familiar sense in which we say that in the utterance of a sentence like Trout eat flies trout are talked about). That is, when we say Loosely speaking P we bring to bear part of our world knowledge of what it is to assert something, or, as I would prefer to say, we bring to bear part of our folk theory of language and speech - the part that concerns assertion. We have knowledge, beliefs and schematizations of language and speech just as we have knowledge, beliefs and schematizations of everything else in our experience. When we use a hedge like loosely speaking in an utterance we use it to talk about some other part of that same utterance, and so at one level we use our world knowledge of language and speech in the same way we use our world knowledge about zoology when we employ the word trout or fly. Loosely speaking interprets the utterance in which it occurs as a world object according to a particular folk theory of utterance, which is part of our larger folk theory of language and speech. To speak loosely is to assert something not quite true. Typically, loose speech is speech that would be true in a world slightly different from the

LINGUISTIC COMPETENCE AND FOLK THEORIES OF LANGUAGE

73

one we are describing, but in some cases we characterize Our speech ^as loose if it fails to achieve precise truth because of some defect in jtscon-struction. Expert theories of language and speech normally make a strict distinction between locutions that don't (quite) state propositions and locutions that state propositions that aren't (quite) true; but not all parts •#? our unconscious folk theory of language and speech insists on this distinction; loosely speaking appears to invoke such an area of the folk theory. Technically

-^''^ .^Hrh-^w^Af-

Technically, used as a hedge, has a meaning that.may be roughly glossed "as stipulated by those persons in whom Society has vested the right to so stipulate." Thus when we say, Technically, a whale is a mammal, we appeal to the fact that systematic biologists have decreed that, whatever we common folk may say, whales are mammals. One line of evidence for this analysis of technically comes from pairs of synonyms - or near synonyms - of which only one member belongs to an authoritative jargon; in such pairs only the member from the jargon takes the hedge technically. (9) (a) Technically, that's a rodent, (order Rodentia) (b) Technically, that's a varmint. (10) (a) Technically, that's an insect, (order Insecta) (b) Technically, that's a bug. The (b) versions may be heard as attempts at humor, precisely because the words varmint and bug not only belong to no technical jargon, but, on the contrary, are markedly colloquial. Further, if we hear a sentence like (11) Technically, street lights are health hazards, our reaction is to wonder WHO has decreed that street lights are health hazards and BY WHAT AUTHORITY. If we learn that the Surgeon General of the United States has done so, even if we reject his arguments and therefore question the wisdom of the stipulation, we cannot legitimately deny the claim expressed in (11). If, on the other hand, we learn that an individual genius has proclaimed street lights to be health hazards on grounds we consider impeccable, we will surely agree that street lights are in fact health hazards, but we may well deplore that the claim expressed in (11) is not the case. Lakoff (1972) attributes to Eleanor Rosch a revealing example similar to the following, (12) Technically, a TV set is a piece of furniture. pointing out that the sentence can have different truth values in different contexts, if there exist in society two distinct bodies with the authority to make such stipulations about TV sets and furniture. For example, mov-

74

PAUL KAY

ing companies might designate TV sets as furniture, while the. insurance industry excludes TV sets from furniture.: .. ,-..:.,•;......:[/_,;L^-i Given this account of the meaning of technically, we may ask. whether technically displays the two properties of hedges, previously discussed, ' that provide problems for standard formal semantics.-These, it wili.be recalled, are (a) that the lexical meaning of a hedge may become one of ,the organizing schemata of the combinatorial semantics of the sentence in which the hedge occurs, and (b) that a hedged sentence may contain a metalinguistic comment regarding the way in which a word or phrase of the sentence is being used in that sentence. Regarding property (a), if we sketch the logical structure of (12) in terms of bur intuitive account of technically, we get something with the rough structure of (13), in which we find that the effect of the word technically is not confined to a single element but is distributed throughout the quantificational and predicational structure of the sentence. (13) There is an x such that Society has authorized x to stipulate the meaning of TV set, and Society has authorized x to stipulate the meaning of furniture, and x has stipulated the former to be included in the latter. The precise wording of (13) is not intended to be taken literally; the point of (13) is just that most of the "logical syntax" of (12) comes from the word technically. The lexical meaning of technically provides the structural skeleton of the meaning of sentences, like (12), in which it occurs. In this respect, technically acts like "logical" words (e.g., all, and, not) are supposed to act. But we noted that technically is a substantive, worldknowledge-embodying word; in fact it is precisely by virtue of the folk theory it embodies regarding language, society, and the social division of linguistic labor that technically achieves its organizing function in sentence like (12). Semantics and mere lexicography find themselves confounded. That technically displays property (b) - regarding metalinguistic comments in which the linguistic item(s) MENTIONED are simultaneously USED as regular linguistic counters - is not apparent from the examples so far given (9-12). One reason for this is that since the target words (e.g., TV set and furniture in (12)) appear with the generic indefinite article, the examples conduce to a straightforward interpretation in which these words are mentioned, but not also used. Consider, however, the following. (14) The movers have come for your furniture, which technically includes TV sets. Here the word furniture is both used and mentioned: furniture is used in the ordinary way as the lexical head of a definite noun phrase, your furniture, to pick out a set of world objects; furniture is simultaneously mentioned as the topic of a metalinguistic comment, which informs us that, by stipulation of relevant authorities, the extension of furniture includes TV sets.

LINGUISTIC COMPETENCE AND FOLK THEORIES OF LANGUAGE

75

Comparison of loosely speaking and technically In the case of each of the two hedges considered, 1 have sought to explain both its lexical meaning and its combinatorial semantic function in tejms, of an implicit folk theory of language and speech. The discussion of loosely speakinghinged on the notion of truth, impUcitly defined m terms of a metatheory in which there is a linguistic system disjoint from the world whose elements (words, sentences) may be combined to represent objects and states of affairs in the world via the meanings or intensions of those elements. The sentence Snow is white is true . . . . This general schematization of language is familiar as an informal sketch of the basic intuitions that lie behind the formidable accomplishments of that tradition of semantic theorizing descended from Frege via Tarski to the modern proponents of model theory, including in particular the various versions most relevant to linguists arising from the work of Richard Montague.! In this framework, words may refer to or represent world objects because the former have intensions that may be matched by the actual properties of the latter. This conscious theory of language, and particularly of reference, has recently been opposed by the baptismal-causal theory of Kripke (1972) and Putnam (1975). The reader may have noticed that in discussing the meaning of technically, I had recourse to Putnam's phrase "the division of linguistic labor" (1975:145ff). The part of the folk theory of language which technically invokes seems in its main lines to agree with the theory of Kripke and Putnam, especially Putnam's version. On this view, a word refers, not via an intension it contains, but on account of someone having once stipulated that henceforth this word shall designate some ostensively presented thing or thing-type. Putnam's idea that we have unconscious recourse, in using a word like gold, to the notion of some expert or official who has the right and the knowledge to diagnose real world gold in a presented sample is especially close to the account I have given above of that aspect of the folk theory of language which underlies the use of technically. Thus when we use loosely speaking, we are taking a Fregean view of language and, moreover, because of property (a), we are organizing the semantics of our utterance in accord with Fregean notions. On the other hand, when we use the hedge technically, we are taking a Putnamian view of language and are organizing the semantics of our utterance along Putnamian lines. If a natural language like English has a formal semantics that employs logical schemata such as conjunction, negation, etc., to compose the meaning of a sentence from the meaning of its parts, then we must number among that same array of structure-composing schemata such substantive folk beliefs about language as those implicitly underlying the explicit theories of reference associated with scholars like Frege

76

PAUL KAY

and Putnam. These are the combinatorial semantic schemata invoked by loosely speaking and technically respectively. Folk theories I have written throughout this chapter in terms of a "single folk theory of language and latterly pointed out that this "theory" differs from conscious theories in that it is not internally consistent. I could as easily have written that English encodes a variety of different folk theories of language. The distinction would have been rnsrely terminological and the same conclusions would have been reached. There are two points here: the first is that a folk theory does not present a globally consistent whole the way a conscious, expert theory does. This should surprise no one, since it is precisely the conscious reflection characteristic of expert theorizing that is generally considered to produce its global coherence. The second point is that folk theories are not "believed" in the way conscious theories are but are used or presupposed as the occasion of thought or communication demands. The penetration of these folk theories of language into the semantic structure of language, via hedges, appears to present several challenges to the generally accepted framework of much current semantic theory. Notes 1. Reprinted with permission of the Berkeley Linguistics Society from the Proceedings of the Ninth Annual Meeting(Bevkehy 1983). 2. The present paper is based on a much longer work on hedges which is still in progress but part of which has been made semi-public in a typescript ms. (Kay, n.d.) of which the subtitle was "hedges revisited." The word revisited referred to the well known paper of George Lakoff (1972). In Kay (n.d.) 1 discuss in detail Lakoff's approach to hedges and my own agreements with and divergences from that approach; space does not permit a recapitulation of that discussion here. Also in that (n.d.) paper there are references to personal communication and advice from many people whose contributions cannot be recited here, although all have helped shape my view of the subject. 1 must acknowledge, however, a very general intellectual debt to Charles Fillmore and George Lakoff. References Gazdar, G. 1979. Pragmatics. New York: Academic Press. Kaplan, D. 1977. "Dthat." In Syntax and Semantics, Vol. 9, Pragmatics, Peter Cole, ed. New York: Academic Press. Pp. 221-44. Kay, P. n.d. The role of cognitive schemata in word meaning: Hedges revisited. Unpublished manuscript, Department of Linguistics, University of California, Berkeley.

LINGUISTIC COMPETENCE AND FOLK THEORIES OF LANGUAGE

77

p, ; Kripke, S. •... „•,...•• '' 'v '•'•'*••''" '{•'"' 'r'-'-.!:;V:^;V;^ 1972. Naming and necessity. 7>i Semantics pf Natural Language, D. Davidson and G. Harman, eds^Dordrecht, Holland: P . Reidel Publishing Company. Pp. 253-355. ; ,-,:.;.,',;•—.; Lakoff, G. : • ••••,•:•.•" •''-!« . :1972,-Hedges: a study in meaning criteria and the logic of fuzzy concepts. In Papers from the Eighth Regional -Meeting,^Chibagd Linguistic Society. Chicago: Chicago Linguistic Society. Pp. 183-228. (Reprinted in Journal of Philosophical Logic 2:458-508,1973). "'''''!"'.',.' Putnam, H. ' '';; 1975. The meaning of meaning. In Mind, Language and Reality. Cambridge, England: Cambridge University Press. Pp. 215-271.

4 Prestige and intimacy

- ,• , •

THE CULTURAL MODELS BEHIND AMERICANS' TALK" ABOUT GENDER TYPES1

: v

Dorothy Holland & Debra Skinner

" . . . I can't believe we're talking about this!" Margaret, an informant in a study of college-age women, said this in the midst of a "talking diary" interview. Earlier, the interviewer had limited herself to questions that a friend or new acquaintance might ask: What's been happening since I talked to you last? How are your classes going? Who is this Alice that you're talking about? When did you join volleyball club? Then, at a point in the interview, Margaret began to describe a skit about "jocks," "frat guys," "Susie Sororities," and other campus types. For a time, Margaret answered the interviewer's questions about the different types and how they could be identified and then interrupted herseif: Margaret: . . . I can't believe we're talking about this! Interviewer: Why? Margaret: I don't know. You just don't sit around talking about it that much with anybody. It's just kind of there. Interviewer: So it's not the sort of thing you'd sit around in your dorm room and talk about to your roommates? Margaret: No, you allude to it more than anything else. Interviewer: What do you mean, allude? Margaret: You know, little things, like, "Oh, you're wearing your add-a-beads today." Things like that. Interviewer: And that's all you have to say? Margaret: Yeah, it's understood. As might be expected, our participant-observation and interview data from a group of college-age Americans shows such types to be a conventional way of talking about other people. One hears words like jock or hunk or freak in conversations about who John so-and-so is, what he's like, what he's likely to do, and why he treated Mary or whomever the way he did. One also hears arguments about whether specific individuals can be described accurately as a "chauvinist" or whatever category has been proposed and sometimes, caricatures of men in general— as in Margaret's skit—couched in these terms. Names like jerk and bitch are also popularly used as insults and others like honey and sweetheart appear in compliments and endearments.

PRESTIGE AND INTIMACY

m

A striking aspect of this talk about, other people is that a great deal of knowledge about gender-marked types is taken for granted. Women assume, for example, that telling another woman, "He's an asshole," will be taken as advice to avoid the male approaching them in a bar.1 They assume that other women know why calling a "jock"; an "ass" to his face would be a risky thing to do or why referring to someone as a "hick" is relevant to a description of him as insensitive. 2 -^: - ' Margaret and our other informants know implicitly what a number of scholars from a variety of disciplines (e.g., Agar 1980; Labov & Fanshel 1977; Rice 1980; Schank & Abelson 1977) have labored to make explicit; namely, with members of one's own cultural group, descriptions are constructed in conventional ways according to unspoken expectations and implicit common knowledge. The hearer is expected to infer missing information cued only by the information that is included and by the genre in which the information is presented. Margaret was chagrined by the interviewer's ignorance of types of men and women, knowledge that she, Margaret, had taken for granted in describing her skit to the interviewer. Not only was she startled by the interviewer's questions, but she also found them difficult to answer. It was hard work to make the information explicit. Our purpose in this paper is to describe the understandings of male/female relations that Margaret and the other American women in our study take for granted when they converse with one another. We refer to this body of shared implicit knowledge about gender-marked types and about ways to talk about these types as a cultural model. Focusing on the manner in which these cultural models of gender are grasped by individuals, we are also interested in how this knowledge of gender types is mentally represented. Does Margaret simply know a list of definitions of jock andfrat guy and other types of males and females, for example, or is her understanding organized in some other way? A partial account of what women know about the types of males they talk about STUDY A

In the first set of interviews we collected - the Study A-l interviews female informants were asked to list types of males. Male informants were asked to list types of females. Next, they were asked to describe the different types and to tell when someone might use such a term. Those 42 interviews revealed what is easily corroborated by listening to everyday conversation: Americans have an extremely rich vocabulary for talking about males and females. There are hundreds of terms for males and hundreds of terms for females. Furthermore, the vocabulary is colorful. Many of the words are derived by metaphorical extension from the domain of animals, the domain of foods, the domain of objects, occupations, or by

80

DOROTHY HOLLAND & DEBRA SKINNER

metonymic construction. New names are easy to make up and, as turkey, libber, and feminist, indicate, easily assimilated into common cultural knowledge (Holland & Davidson 1983; Holland & Skinner,1985). • -^ An obvious way to present this American cultural knowledge of gender types is simply to list and describe or define all of these different lands of males and females. We could even present the definitions in an economic fashion as in the ethnoscience tradition (see, for.example, Tyler 1969or Spradley 1972) by organizing and presenting the terms according to their taxonomic and paradigmatic relations. Tempting though this "dictionary definition" solution might seem, a decade of developments in cognitive anthropology, linguistics, and psychology suggests that this ethnosemantic approach cannot adequately describe how individuals organize their knowledge about gender types. As D'Andrade and associates have demonstrated, dictionary definitions often omit the very attributes of the topic that people think are the most important (D'Andrade et ah 1972). Studies of person and social types, in particular, show that what is important to people about these types is not what one must ascertain about persons to accurately classify them but rather what one must know in order to know how to behave toward them (Burton & Romney"1975; Harding & Glement 1980; White 1980).3 For Ixil-Maya speakers discussed in Harding and Clement, for example, the important things about social roles are associated wealth, local affiliations, and their relationship to the civic-ceremonial complexes4 - attributes that might not be included in dictionary-type definitions of the roles (see also Keesirig 1979). Rejecting dictinary-type definitions as ameans of describing the cultural model of gender, we turned first to the "cognitive-structure" approach used in the Burton and Romney, Harding and Clement, and White studies. In the Study A-2, interviews informants were asked to do more systematically and more comprehensively what they do on a limited scale in conversation: They were asked to compare and contrast types of males and types of females according to whatever criteria they considered important. If we could find out the bases for comparison arid contrast, then we would have an idea of the implicit propositions about gender types that organize women's thinking about men, and vice versa. From the Study A-l interviews, we selected 41 male types and 41 female types.5 We wrote each subtype on a card and asked the respondents to sort the 41 types according to similarity and then to describe the similarities they saw among the types they had put into each pile. The reasons they gave for their sortings were recorded verbatim. Important characteristics of gender-types. In the Study A-2 interviews, the respondents were allowed to compare and contrast the types according to whatever criteria seemed important to them. In most studies of gender stereotypes, the respondents are not allowed as much freedom; they are given a list of personality traits such as rational, warm, nurturant,

PRESTIGE AND INTIMACY

81

and independent and asked to say which traits are characteristic of males, which,of females (Rosenkrantz et al,;1968; Spence, Helmreich, &Stapp 1975). If we had been willing to assume that the cultural model of gender is; organized according to personality traits, we cotiW: have aske'd respondents to tell us which traits are associated with Iwhich types. However, since our Study A-l interviews showed no exclusive emphasis on personality traits, and since more general studies (e.g., Bromley 1977) show that other characteristics of persons are considered important, we wanted to give respondents freedom to emphasize whatever aspects^th^y considered important. (See Holland & Davidson 1983 for more discussion of the difference between'most gender stereotype research and our own.) As it turned out, a variety of characteristics were described, as the following examples show:6 1341 [jock] a male who is impressed by his own physical prowess - like a matinee i d o l . . . a physically attractive or physically impressive athlete. People also use it \Jock] to indicate a physically able and mentally deficient male 0931 [chauvinist pig] a guy who believes women are inferior 0131 [dude, athlete, jock, macho, stud, hunk, Don Juan, playboy, egotist, frattybaggerf guys that think they are real cool, woman-pleaser types, conceited type people 0831 [turkey, nerd, jerk, prick] all derogatory; terribly insecure 2231 [wimp, sissy, homosexual, queer, gay, hippie] they all seem queer. Seem like terms for homosexual except hippie doesn't fit. They're all strange, socially unacceptable. They're all fags. 0631 [man, guy, fellow, gentleman, boyfriend, fiance, lover, sweetheart] they connote a more positive image . . . the most positive image of all the cards. They connote a kind of "boy back home": a more traditional role of a male as I think of it ideally. Personality traits are mentioned frequently. Also mentioned are comments on looks, on specific attitudes, on the kind of a date the type makes, on sexual preference, and on many other specific mannerisms and background characteristics. Our next task was to identify any themes or dimensions that underlay the multifacted descriptions we had been given. We used a procedure that translates the measures of similarity from the sorting into a visual display. In the visual display that was created by a technique called multidimensional scaling, types that were often sorted together by the respondents were placed close together; types that were seldom sorted together were placed far apart.8 This multidimensional scaling procedure was used to produce Figure 4.1, which indicates how male types were sorted by females, and Figure 4.2, which indicates how female types were sorted by male respondents. Multidimensional scaling is primarily an aid to visualizing the patterns of comparison and contrast. It is also useful as a basis for estimating the

GO

•TYPESOF M E N '

Effective/Likable [Endearments]

X Macho (.09)

X Jock (.021

Hunk (.04) X

X Don Juan 1.06)

X X Chauvinist/Hustle X Egot

Guy/Fellow <-.09)/<-.09l Doesn't Stress

: JC Bastard X XDOBI-.14T"----: Pnck(-. Redneck (-.141 -, XJerk (-.237 XPunkUlSI' X>Jerdt~.22)

X Gay /Homosexual (.63m.63l X Queer (.56}

x wimp (.05)

Male/Female Differences X Sissy (.41)

X TuAayVj!!)) '" T

Figure 4.1. Three-dimensional representation of male social types (Stress = . 147)""

I "

• TYPESOFWOMEN. Low Status Possessions E Insults]

Overdemandmg/Unlikabie (Insults]

X Dumb Blonde [—.161 1

Airhead ( - .15)

X Southern Woman (—.24) X BBHeH.17)r-:|.

Figure 4.2. Three-dimensional representation of female social types (Stress = . 138)

"

84

DOROTHY HOLLAND & DEBRA SKINNER

number of conceptual dimensions of contrast that predominate in the sorts. For both the female and the male types, the sorting data were distorted if we allowed only two dimensions for the scaling; with three dijriensions, «the level of distortion measured as stress was acceptable. Thus, both figures show three dimensions/ (The third dimension on Figures ,4.1 and 4.2 is indicated by the numbers in parentheses.) '" / The multidimensional scalings are carried out according to a set of algorithms executed by a computer. The next steps are to examine the multidimensional scalings and interpret the dimensions of comparison and contrast the respondents seemed to be using when they sorted the types. To make these interpretations, we analyzed both the explanations given by the respondents in the A-2 interviews and the descriptions from the A~l interviews. These data led us to the conclusion that females type males according to whether they are: 1. likely to use their position or attractiveness to females for selfish purposes, 2. ineffectual and unlikable, and/or 3. unusual in their sexual appetites Chauvinists, playboys, and jocks, for example, are seen as types who use their social position as males or their attractiveness to women for selfish purposes; guys, fellows, boys, and gays, in contrast, are seen as being unlikely to try to capitalize on these advantages. Jerks, nerds, and turkeys are inept and unattractive, whereas boyfriends, financees, sweethearts, and, to a lesser extent, guys and fellows, are attractive and effective. The sexual appetites of gays and to a lesser extent playboys, are out of the ordinary and an important aspect of their lives. For nerds and skinheads, the opposite is true; sexuality is not a particularly notable aspect of their behavior. What we are arguing from the cognitive-structure analysis is that women focus on and organize their thinking about male types according to these three aspects of male characteristics and behavior. j Males do not compare and contrast females on the same three bases just described, although there are some complementary aspects. Males ; compared female types according to their: ,j 1. prestige as a (sexual) possession/companion 2. tendency to be overdemanding and engulfing 3. sexiness

; -•-->.

f|

1 'A

Bitches and scags, for example, are types that are overdemanding; 1] girlfriends and sweethearts and, to a lesser extent, women, are supportive ;] and helpful. Foxes and whores are sexually enticing; prudes and dykes ."J are sexually repelling. Foxes and dolls are high-status sexual companions, $ whereas whores and easy lays are low status. These characteristics are im~ v1 portant to males about females. Summing to this point: Study A has provided partial information on ;

PRESTIGE AND INTIMACY

.85

the cultural models of gender types for our samples. As a cognitive entity, a cultural model may be defined as learned mental representations of some aspect of the world - in this case, gender types. These omental representations or schemas actively guide attention to components of the world and provide inferences about these components and their/various states and form a framework for remembering, reconstructing,'" and describing experiences. (For a related conceptualization of schema, see, for example, Neisser 1976 or Rice 1980.) Study A provided the broad outlines of the characteristics that females are guided to look for in types of males, and vice versa.-These models tell females what to pay attention to about new males they meet, what to be on guard about in males they already know, and what questions to ask about newly identified types of males. Limitations, On the basis of Study A, we felt we had correctly grasped the characteristics of male and female types that were important to the respondents. We also felt that based on what we had learned about these important characteristics and what we had learned about the conventions for naming types, we could correctly predict how the respondents would react to types we had not included in the interviews and even how they would be likely to react to names of newly identified types. Using a method such as Burton's (1972), we could have undertaken a validation of our interpretations and predictions; however, we were concerned about certain limitations of our approach and decided instead to examine another source of data.9 The type of analysis we had done - cognitive-structure analysis - did not adequately present the total amount of information we had learned from the interviews. Cognitive-structure analysis is predicated on the idea of underlying "dimensions of meaning" as the organizing structure for the set of terms - in this case, gender types. The question of how these dimensions are mentally grasped by informants has received little explicit attention in the literature (see D'Andrade 1976 for an exception), but the implication is that the dimensions or characteristics of importance can be described accurately as single attributes or features of meaning. We had difficulty in finding and, as is discernible from the labels affixed to Figures 4.1 and 4.2, did not manage to isolate in every case, a single attribute or descriptor that seemed to capture the sense of the "characteristic" the respondents were talking about as the basis for their comparisons of the different types of males and females. Even when we did use two attributes or a descriptive phrase, we found it was not clear from our descriptions how women integrated "ineffective" and "unlikable" as co-occurring characteristics, for example, or why "exploitation of male/female differences" should not also be coupled with "unlikable." Furthermore, we realized that our descriptions of the characteristics as simple attributes were also limited because these attributes did not offer any

DOROTHY HOLLAND & DEBRA SKINNER

86

insight into the affect our respondents displayed when they discussed the different types. There was information from the interviews that illuminated these questions, but clearly that information was not being effectively cbn< veyed by the cognitive-structure analysis that presented the "characteristics" as though they were simple attributes. A more accurate way of describing the "characteristics" had to be found. A SECOND LOOK A T STUDY A

.•:;;.

! i

We returned to the A-l and the A-2 interviews to find out how'the respondents had communicated a sense of what jocks or wimps or broads are like. We reread the 460 or so descriptions of individual types in the A-l interviews and the 250 or so explanations for the groupings of types in the A-2 inerviews. In both sets of interviews, the descriptions were strikingly similar and included a variety of information. Some descriptions were limited to single descriptors reminiscent of the attribute-like features we had first lodked for to describe the "dimensions" from the multidimensional scaling. These single descriptors often had to do with character, mood, or personality: 0231 [sissy] a male who is effeminate 0431 [bastard] a male who is mean 0331 [turkey, nerd, jerk, frattybagger] these are people who are just plain stupid 1031 [stud] a guy who is horny Other single-focus descriptions contrasted with these in that they depicted not the type's inner state, but rather his acts or behavior: 0731 [hustler] a male who takes advantage of a person 0931 [pussy] a guy who doesn't stand up for what he believes in or who is a coward 0931 [pimp] a man who prostitutes women Another large set of the descriptions were unlike the ones just quoted in that they included more than one type of information. They contained information about the type's inner state and information about his behavior and other information, such as females' reactions to him. 0431 [boy, dude, dog, wimp, hippie, turkey, punk, nerd, jerk, prick, skinhead] these are losers - all the names that you call really queer dates. They're usually immature or ugly, or think they're cool, but aren't at all. They try to impress girls, but actually make fools of themselves. 2231 [redneck, dog, turkey, punk, nerd, jerk, skinhead, cowboy, brain] I think of little 98-pound weaklings - jerks. They're all ugly little jerks that you'd never want to be seen with, or never want to talk to. You cannot get rid of them. 1131 [couchwarmer] a guy who is too cheap to take you out so he takes you to his home all the time.

PRESTIGE AND INTIMACY

87

0431 [lover, athlete, jock, macho, stud, egotist, bastard, hunk, Don Juan, playboy] these are the typical jock-type, good looking but they know it. Can get any type of girl they want because girls usually go for •., them. They're popular. . 0331 [Don Juan, playboy] implies someone who likes to play around women are attracted to them but they don't set up a serious relationship^ , 0331 [ladies'man] a friendly man who is deceitful. Ladies'man and, ' \ macho mah ; are variations on the same theme - one tends to have larger biceps.' 1331 [boyfriend, fiance', lover, sweetheart] these are all subtitles of what we would call the man who is showing the romantic side of a man in relation to a woman. 0431 [sissy, homosexual, queer, gay] they're the type you find in my dancing class. They're just ail gay, pretty unmasculine, talk with a HSP. ;, .-:•,.., The respondents were clearly not limiting their thinking to a single characteristic of the male types they were describing. In order to convey their sense of the social types, they were providing, it might be surmised, the outlines of a social drama, or sometimes, a scene from the drama. In the scenes - which are sometimes described as though they were being visualized - the male type plays a role in an encounter or a relationship with another person, usually a female. He is her date or perhaps her friend or her would-be lover. His style of playing the role is different from how an ordinary male would play the role. In the descriptions, the unusual aspects of his style are communicated by an account of his actions or a description of his intentions, personality traits, or beliefs, He is friendly, but deceitful; he thinks he is cool, but actually makes a fool of himself. Sometimes, we also are told the female's reactions to such males (e.g., "women are attracted to them," or "they can get any girl they want"). 10 The recognition that the respondents were constructing their descriptions of gender types from social scenes or perhaps scenarios made it clear why trying to describe the characteristics of the types as single attributes was a difficult and perhaps impossible task. In trying to represent the information conveyed to us by the respondents as single attributes, we had undertaken the task of describing a gestalt of social information and action in a few words. Although the multidimensional scaling had assisted us in our identification of the key behaviors and characteristics of the different types, it did not help us in the identification and description of the taken-for-granted social world in which these characteristics are significant. The realization that the respondents were thinking of the types in terms of social dramas rather than single attributes prompted further study. The aim of Studies B and C - analyses of two other sets of data - was to uncover premises about the social worlds associated with the scenarios: What makes types such as jocks or wimps so special that they are labeled as different from ordinary males? What were the respondents assuming about

88

DOROTHY HOLLAND &DEBRA

SKINNER

normal relations between males and females that made these types stand out?

•••:;-'- . /

-.•:

Besides the implications of our reanalysis of Study A, the work of .linguist Fillmore also encouraged us to pursue the examination of this taken-for-granted world of male/female relations. We noted a similarity between the kinds of scenes that Fillmore (1975:124) argues are associated with linguistic frames and the descriptions given us by the respondents. Fillmore has presented his proposed frame semantics by elucidating the meaning of words such as orphan and widow. He (1975:129; 1982:34) has argued that the meaning of bachelor, for example, is integrally related to a conceptualization of a social world in which such things as bachelors exist. Bachelor cannot simply be defined as an unmarried male, for the role of bachelor is not relevant to all the social worlds in which unmarried males are found. Is Pope John Paul II a bachelor? Is a trice-married, presently divorced man a bachelor? Fillmore says that the category of bachelor is not relevant to these cases because the worlds of the Pope and the trice-married, divorced man deviate from the conceptualization we have of the social world in which bachelors exist. A complete analysis of the type suggested by Fillmore is given by Sweetser (this volume) for the word lie. She shows that the meaning of lie is not detachable from a conceptualized social world in which communication between individuals follows a culturally standardized, normative pattern. In this simplified world, the telling of false information has certain consequences, such as harm to the recipient of the lie, and thus is clearly a reprehensible act. Perhaps, wereasoned, the exploitation of male/female differences has particular poignancy to the women in our sample because of the implications of exploitation in the simplified world in which Don Juans, machos, hunks, and chauvinists are relevant characters. Perhaps females attach importance to ineffectiveness and insensitivity in males because this characteristic poses a difficult problem for what is taken for granted to be the normal course of male/female relationships. STUDIES B AND C

Studies B and C were conducted in the same locale as Study A with the same age population two years after the completion of Study A. Studies B and C consisted of participant-observation research and tape-recorded one-to two-hour interviews. The 23 informants of Study B and the 10 informants of Study C were each interviewed an average of 8 and 5 times, respectively. The participation-observation research was useful because it revealed the extent to which the same kind of talk about males and females that occurred in the interviews was also occurring in the everyday activities of the informants. The Study B interviews primarily consisted of "talking diary" interviews in which the informant was asked to describe what had been happening to her since the interviewer had last seen her. In the course of these inter-

PRESTIGE AND INTIMACY

89

views, which took place over a year-long period, the respondents frequently talked about encounters and relationships with males. As suggested, by the excerpt from the interview with Margaret at the beginning of. the paper,the "talking diary" interviewers tended to stay within the bounds of Questions the women were accustomed to answering and talking about ^withV their peers. •';:••*•-. ;--:-::^r '^•••;';'^,,-/-:-3i:^ The Study C interviews provided a more out-of-the-ordinary.lask for the informants. They were asked to tell about their first and then subsequent memories of someone, usually someone with whom they had had a relatively long relationship. After they had told about the memories as they wished, the interviewer asked in-depth questions about their impressions of and reasoning about the individual, often requiring the informant to make explicit information or beliefs about males and females that she would have otherwise taken for granted. In searching through these interviews for relevant passages, we looked for passages in which the gender of the other was of explicit significance to the informant's reasoning about the other person. Any passage which included reference to the gender-marked social types that had been identified in Study A-l was automatically consulted. Our guiding questions were: What do our informants assume about ordinary relationships between males and females? and What are the taken-for-granted worlds in which these male and female types interact? As it turned out, this takenfor-granted world is a world of prestige and intimacy gained and lost. The taken-for-granted world of male/female

relationships

In the taken-for-granted world of male/female relations, from the perspective of the women in our study, a male earns the admiration and affection of a female by treating her well. Intimacy is a result of this process. The female allows herself to become emotionally closer, perhaps as a friend, perhaps as a lover, perhaps as a fiancee, to those attractive males who make a sufficient effort to win her affection. Besides closeness and intimacy, the process of forming a relationship also has to do with prestige. When a male is attracted to a female and tries to earn her affection by good treatment, her attractiveness is validated and she gains prestige in her social group. For his part, the male gains prestige among his peers when he receives admiration and affection from and gains intimacy with females. Normally, prestigious males are attracted to and establish close relationships with prestigious females, and vice versa. Sometimes, however, a male can succeed in winning the affection of a female whose prestige is higher than his own. However, the more attractive she is, the more he must compensate for his lack of prestige by spectacular efforts to treat her well. Correspondingly, females sometimes do form close relationships with males who have higher prestige than they do. When the male is more

90

DOROTHY HOLLAND & DEBRA SKINNER

attractive or has higher prestige than the female, she often must compensate by giving her affection to him without his doing anything to earn it. Several aspects of this world can be illustrated by interpretations that informants made of their experiences with males. It should be noted that "although the interpretations included in the following sections pertain primarily to romantic relationships, our data indicate that friendships between a male and a female are interpreted in terms of the same taken-forgranted world. 'l THE INGENIOUS B E A U ; - •

.;;;.„':..•

Karla had an "official boyfriend," Christopher.11 Meanwhile, another. guy, Alex, was doing things like showing up at her door with the gift of an egg, dyed purple. For a visit to listen to records, he appeared dressed in a costume befitting the punk-rock genre, a costume he had creatively assembled from castaway clothes and a dead carnation. What meaning could this bizarre behavior have? Karla interpreted it as an effort to win her affection: Karla: . . . if you want to get down to brass tacks, the main crux of the problem [with the relationship with Christopher] right now is this new guy. Because I must say he fascinates me, he fascinates me more than anyone I've ever known and furthermore he's making the most interesting efforts to get me. Alex was treating Karla well. Being well treated by a male means being shown special considerations and courtesies; having one's values, desires, and feelings taken seriously; and being appreciated for one's qualities and accomplishments. Some other examples of such treatment besides the creative efforts of Alex included things, in Karla's eyes, like wearing a jacket on a casual date and being pampered when one is feeling sick. Another informant, Diana, gives additional examples in a life history interview. She begins by talking about how attractive females want to date attractive males and then switches to the kind of treatment she expects from males. Interviewer: How about dates . . . any more to add on dates? What was important to you? Diana: Well, if you were fairly nice looking you wanted to date a good looking guy, I mean, that was probably all part of our ego, we wanted to have the best looking date or things like that . . . of course you wanted to be attractive to them [males] you know. Like I said, you wanted them to think that you were pretty . . . . Interviewer: Did you want them to think anything else? Diana: Of course you wanted them to think you had a good personality, that you weren't just beauty and no brains. But it was important to me for someone to respect my values and most of my friends were the same. Of course, there's, in every community there's a few girls that don't have such strict moral values, but we wanted to make sure that the guys that we went out with did respect that or we wouldn't go out with them any more. AH my friends were about the

PRESTIGE AND INTIMACY

91

same in that respect; we wanted our dates to respect us and treat us like ladies, not like one of the fellows. >"" :••;••••.••' o r ' V v ' V fnterviewer: What would constitute being treated like a lady? • -:'; Diana; Well not only respecting our moral values, but to me, at least,- maybe it was because I was in the role of the female where you were old-fashioned and so on, but it was important for them to open the door for ;me, seat-me if we went out to eat, to open car doors for me, just common courtesy that. a,lot of times you don't even think about. ..::.;;.;: KARLA THE GERM .:.. ^ .!". ,,^.,.,Z.^.^Z^.^^p, Bad treatment is being ignored, being unappreciated or scorned, and being treated like an object rather than a person. In describing her relationship with Christopher, Karla recounted a phase of their relationship in which she became disgruntled with how Christopher was treating hen The situation came to a crisis when she returned to school after a holiday and promptly came down with a bad case of the flu. Instead of being solicitous, Christopher tried to avoid her: Karla: Well I was . . . feeling so horrible that night about nine o'clock that I put on my pajamas and went to bed, and Christopher comes by at 9:30 to see me. And he says, "What are you doing in bed?" Well [when I told him], he just kind of like turned pale. And I thought that it would be nice, very nice of him [laughs] to sort of well, you know, bring me a little chicken soup, tell me to have a nice day, send me a little card. I really wanted that, but instead he j u s t , . . . he wasn't exactly rude, but he sort of got out of the room as fast . as he could cause he's so scared he'd get it, and I can understand that, his prac-, tice schedule, he plays with a university group, if he got sick it would screw. him up a lot, but I don't like being treated like I have germs, whether I have them or not. . . . Later that week, Christopher took her to a play even though he was still afraid of catching the flu from her: He didn't say this, but he went out with me anyway, but he was just kind of like on edge all night long because of that and I think that's why he started making some nasty remarks. . . . So that made me angry and that's why we had our big fight. . . . Interviewer: How did uh, why did you . . . Karla: I said, "Why, how dare you treat me like a germ?" And he asked me to explain this, so I did and I told him that I had not had a good time that evening [laughs] and, I said that, w e l l . . . it was earlier in the evening in the restaurant that he had made the nasty comment about my family. I said, "Christopher, how can you sit there and say something like that, and act like your family and your family's background is so much better than mine when this very evening, Christopher, you have behaved with no class whatsoever?" . . . Karla interpreted Christopher's behavior as bad treatment, treatment that suggested that she was nothing more than an object - a germ. She thought that he had overestimated his own attractiveness or prestige rela-

92

DOROTHY HOLLAND & DEBRA SKINNER

tive to hers and that he had no right to expect her affection if he continued to treat her in a condescending manner. Eventually, she let him know she was angry. , H *. Although Karla changed her interpretation after another talk with ' Christopher, she initially took his behavior as meaning that he considered his prestige to be higher than hers. In subsequent passages, she described how she responded by challenging and bringing about a change in his definition of their relative positions. Evidence of the negotiation about relative prestige or attractiveness and its significance is also evident in the following passage. •••.••?.-.-••. THE UPPER HAND

"I just don't want him to get the upper hand on me. . . . "

' '. "'

•

Karen is describing a guy she has just started to date. She has been talking about the times he calls for dates and how much he likes her. She goes on to explain that she has not been completely straightforward with him: Karen: . . . I just don't want him to get the upper hand on me, you know. Like I play games with him . . . Interviewer: Could you give an example? In response to this question, Karen discusses in a very oblique way how northerners' (Hal, the new guy that she is dating, is a northerner) morals are different from those of southerners (Karen is from North Carolina). She describes northerners' sexual morals as being more "open and carefree." Interviewer: When you said that part about you didn't want him to get the upper hand, could you talk a little more about that? Karen: I didn't want him to think that I was really crazy about him and that he could just use me, you know, maybe if he knew I'd want to go out with him and stuff like that. So that's why I just sort of let him, in fact I was trying to get it with him, you know, get the upper hand with him, but it didn't work. He's the same way, you know. Interviewer: How did you try to do that and why didn't it work? Karen: Well, you know, I'd tell him - he'd say something about going out and I'd say, "Well just . . . we probably will, but it's a little early right now." I'd do stuff like that, and he'd ask me, he asked me if I had, urn, well the first night he asked me if I had a boyfriend back home and I didn't say anything, and he says, "Well, I figured you did." And, I said, "Well. . ."; I didn't say anything, you know. I just told him that I dated a couple of guys, you know. I didn't tell him if I still saw them or not, you know. She goes on to explain other ways in which she tries to give Hal the impression or allows him to infer that she has other boyfriends, including such subterfuges as sometimes leaving the dorm when she thinks he is going to call.

PRESTIGE AND INTIMACY

93

Here, Karen has read a lot into Hal's pattern of calling and asking her out and making overtures to .her. She has interpreted his behavior,as a, preliminary move in a negotiation in which she and he work put whether his prestige or attractiveness is higher than, equal to, or lower "than jiers.If she appears to have an active social life, then her prestige as an^attrac^ tive female is in evidence and he will not be able to treat her badly by taking her for granted and not calling when he says he will or, perhaps, although she does not directly say this, expecting her to become sexuiUy intimate faster than she would choose to. If, on the other hand, Jiis prestige is high, as evidenced by the fact that she finds him extremely attractive, then he will be able to exploit her and treat her badly if he wants to. She would rather that her prestige be seen as higher than his so she acts to, bring about that interpretation. Women judge whether their friends' relationships make sense in terms of the treatment they receive from males. Whether bad treatment is Understandable to other women comes up a number of times in our data. From the cultural model, a female may form a relationship with a very attractive male even though he treats her badly. An example of the application of this idea comes from Diana. Diana has been having a number of runins with Donny, her boyfriend, who attends a university in a nearby city. Their calls often end in recriminations and tears on Diana's part. Her friends on the hall constantly point out to her that Donny is being mean, that he's just a "jerk." Diana's reply to them can be summarized in her words: "It must be love." She implies that she finds him so attractive that she is willing to sacrifice good treatment for the sake of being around him. Her friends are not convinced. They think he's not worth the trouble he causes Diana; he does not seem all that attractive to them. Problematic males These stories describe experiences that the informants interpret according to a set of assumptions about normative relationships between males and females. In the taken-for-granted world constituted by these assumptions, arrogance in a male has special implications and getting involved with an "asshole" has predictable consequences. Arrogance, as elaborated below, has implications for a male's assessment of his own status relative to that of the females around him; the ineffectiveness or insensitivity of an asshole is problematic because of the way he is likely to treat females. This takenfor-granted world, in other words, provides the background against which several basic types of males pose a special challenge or problem for females. These types were foreshadowed by the dimensions identified in the multidimensional scaling. The problematic males are those who are arrogant and use their position or attractiveness as males for their own selfish purposes in interactions with females, are insensitive and unlikable, and have unusual sexual appetites.

94

DOROTHY HOLLAND & DEBRA SKINNER

ARROGANT AND SELFISH MALES

From our informants' interpretations, males who think they are "God's gift to the earth," "who think that anyone without a penis doesn't exist," "who are arrogant and out for themselves," and "who are good looking, but know it" are problematic because they are likely to assume that they do not have to earn a female's admiration and affection and intimacy; they are likely to expect these things from females simply on the basis of their looks or other claims to'high prestige.12 They are likely to'be able to exploit their attractiveness and prestige for their own selfish purposes, treating females in a bad or demeaning manner, and not suffer for their behavior. Examples of how informants see these jocks and other types are provided by the data. Annette and Sam. In one of her interviews, Karen told about an encounter she summarized as follows: Karen: A friend of mine [Annette] invited us, invited several of us to a party at a dorm. And, she told us that there'd be . . . , a couple of people there that she really liked a lot, guys, that is . . . well, they're on the basketball team, you know, big jocks and stuff like that, you know, and . . . when we got there, urn, the main one she wanted to see . . . I mean, he just, he didn't even hardly acknowledge her presence. He practically didn't even speak to her. . . . And, it just sort of messed up the whole party - mainly for her, and because of that, it messed it up for all of us. The remainder of the interview was devoted to questions about this episode: Interviewer: What were your expectations when you went to the party? Karen: . . . I expected to meet a couple of the players, and . . . I expected, you know, some real nice guys. And I thought that well, they'd be real glad to see her, and you know, just real friendly and everything. And, but, they, they didn't. Interviewer: What. . . how did they act when they came? I guess there were two of them that came at different times. Karen: Yeah. One of them [Robert] was real nice to her and glad to see her and all, you know. That wasn't the main one she wanted to see, the one she wanted to see [Sam] acted real stuck-up, you know, as if she wasn't even there'. Interviewer: Oh, how did he do that? Karen: He, he ignored her. And, I mean, he saw her several times . . . she'd be standing practically beside him, and he wouldn't say anything . . . this other girl came and he just talked to her practically the rest of the night. Later, the interviewer asks Karen why she thinks Sam acted as he did: Karen: I don't know if he was, if he, if, I don't know if maybe she just had it in her head that he liked her, or if he was just, if he didn't want her around, and he was just trying to talk to this other girl or something. But he did act, he acted sort of too good for her, you know?

PRESTIGE AND INTIMACY

95

The interviewer asked for more detail about why Sam acted as he did: Karen: . . . maybe because he was a big jock on campus or something, that he thought, that, you know, she was just an average girl, and he was too good for her, or something. But, um, I didn't look at him that way until he walked you know, he walked in, and then he just sort of carried himself like, you know, everybody's looking at me. And, I didn't like that at all about him. And, un£ he, he was just, it just seemed like he was standing there waiting for people to come and talk to him, you know. Instead of him acknowledging anybody else. -•. .

•

..,.

Karen talked about how upset Annette was and said she thought perhaps if Annette could talk to some other guy, she would feel better. Interviewer: What? Why would that make her feel better? Karen: Um, uh, she probably, I don't know, I guess just to boost her confidence back up, or to make her feel like she's really somebody. Instead of what he, I mean he made her feel like she wasn't even alive. . . . Annette continued to be upset about the incident, and Karen explained that Annette was trying to reason out why Sam acted as he did: Interviewer: What were some of the ways she reasoned it out? Karen: Um, well, she thought at first maybe because he was with that girl, he didn't want to talk to anybody else. And, but then, he was talking to other girls that were walking by, and urn, then she was thinking, maybe he was mad at her, but she didn't know why, you know, she was just thinking of different stuff likethat. Interviewer: Did you think of any things like that too? Karen: Uh, not really, I, I, it's gonna sound terrible. I thought, well he just didn't want to, didn't want to see her at all, cause he just didn't, I don't know, what I thought was that, he was like I said before, he was some big jock on campus, you know, and he just wanted the real, um, just certain girls around him, you know. Interviewer: What. . . what kinds? Karen: Real pretty, you know, real - (I think Annette's pretty, too) - and he just, you know, to make him look that much more better, you know. That's what I was thinking, after I, after I saw what he was doing to her. Interviewer: Why would, why do you think that he would want that, would want these girls? Karen: I guess to help his image, you know, make him look that much more better. In Karen's interpretation, Annette is treated badly. Her presence is not even acknowledged by Sam. The situation is an embarrassing one because Annette has revealed her attraction to Sam yet he has ignored her; she has been shown to be less attractive or of lower prestige than Sam. She's just an average girl. In Karen's eyes, Sam's attractiveness is diminished. She says, "I didn't

96

DOROTHY HOLLAND & DBBRA SKINNER

think he was attractive anymore." Her inclination is to demean him, to label him as a lower status male, an unattractive type: Karen: . . , I wanted to tell him, he, you know, what he did to her, yo'u know, ^ that he was acting like an ass. She does not call him an "ass," however, because she did not think Annette would have wanted her to and because he was around a group of people and she did not want to "make a fool" of herself either. Obviously, other people, such as the woman he was talking with, did not think Sam was an ass. Another part of this story contrasts Sam with Robert. Attractive males do not necessarily exploit their attractiveness Tor their own ends; they are not necessarily demeaning to females who have less prestige than they dp. Robert was nice. He treated Annette very well and even went out of his way for Annette's friend, whom he had just met: Karen: Oh, I like him [Robert] a lot, yeah. Cause he, cause he made her laugh and, he was just, so, he was real nice to all of us. And, um, well, one of my friends wanted a beer, you know, but there was this real long line, so he just walks right up. He takes a cup and goes and breaks in front of everybody, you know, and gets it and brings it. He takes a cup and goes back to her, you know. That really impressed me, there, cause he didn't know her, he didn't have to do that. You know. And um, that's just the type of guy he was, you know, just real friendly and nice to everybody. Robert was attractive and he treated Annette, his friend, in a way that earned her affection and admiration (and Karen's, too). Even though he was attractive and did not have to do things for Annette, he did. The difference between him and Sam, as Karen interpreted it, was that Robert just wanted to have a good time whereas Sam wanted to make himself look better. Not only was Sam guilty of demeaning Annette, he, in Karen's interpretation, was also using females to further his own ends. He was not sincerely trying to earn their affection and admiration and giving them good treatment in return. He was simply using them to get what he wanted in the case of the party, increased evidence of his own attractiveness: Karen: . . . you know it's just like they're [guys like SamJ they're out for themselves, you know, just to make, "1 just want to be seen." You know, it's like they're just using the girl or something. . . . INEFFECTIVE AND UNLIKABLE MALES

In contrast to males who are problematic because they are attractive but prone to treat women badly, there is a second type whose labels are used as insults. Karen, for example, fantasized insulting Sam by telling him he was acting like an "ass." Diana's friends claimed Donny was acting like a "jerk." Jerks, nerds, turkeys, and asses, among others (see Figure 4.1)

PRESTIGE AND INTIMACY

97

constitute this second typeof problematic male. These males are both unattractive and insensitive and thus unlikely to receive a female's admiration. and affection. Lacking sensitivity to What females want, !these low-prestige males are neither attractive to women ribr are they effective at pleasing , females and thereby earning their affection/Two accounts from the data illustrate interpretations of experience in which this type of male plays a p a r t , ..•,,., ...,,,....:.,

..,.. i ; ... ; -.,.,._,, : 7 ^, 1 ...,,i,.: .:.....,,.:...^^,.

Patty and the hick, : Patty was asked to describe someone whom she knew from her work. She picked Erve, a colleague, at the school in which she teaches. As a potential friend, she found Erve wanting: . fr; Patty: . . . And I mostly don't care for him very much. Part of it is the fact that he's a real Okie kind of person, in a derogatory sense, and I mean it that way, he's a real hick. . . . Patty goes on to list many things she dislikes about Erve, including his lack of a sense of humor, the strange things he says in the middle of conversations, the way he usurped the position of the coach, the tactless way he deals with the students; his disruption of the faculty lunches by his topics and styles of conversation, the fact that he asked her her age but did not tell his, and so forth. Furthermore, he did not seem to realize that she disliked him: . Patty: . . . and the other thing that now tops it off is for some reason he's decided I'm his friend and he will come and talk to me, and there's a period of the day, it's usually about twenty minutes of three . . . when everybody fades out and you can get something done, and he will come in there if he hasn't got anything to do and he will talk a blue streak, and I feel resentment about that. And I'm a passive-aggressive person so I never say anything. I just sit there and feel, and he's not long on sensitivity, so he never picks up the vibrations . . . I disagree with just about everything I seem to have noticed about him. At a later point in the interview, Patty further elaborates the idea that Erve is oblivious to her desires and feelings: Patty: . . . But in an annoying situation, I will put up with the situation rather than make waves. However, to someone who knows me, the air is absolutely thick with unharmonious vibrations. Interviewer: And he doesn't? Patty: No, he does not pick up on those things at all. There are people who will receive such feelings and ignore them and there are people who do not receive, and he is a nonreceiver. In describing his lack of attention to her feelings, she reiterated a situation in which she had complained to him about his treatment of a student and he had simply made a joke of her statement as though he did not understand she was angry:

; •'•"'..'. ;;

.

98

DOROTHY HOLLAND &DEBRA

SKINNER

Patty: . . . I was really peeved. I did it quite pleasantly, but anyone with a grain of sensitivity would have noticed that I was peeved. Carl [another teacher] knew that I was peeved; Alice [the principal] knew that I was annoyed and they heard ^me say the words in the same tone of voice that this guy did. , ••_'/ ;,.'.;. ..; Erve, in short, is remarkable in his lack of sensitivity. ": ";•'•" ::: • ;; Patty also points' but in the interview that Erve has a lower-class style about him, is "uncivilized," and has very parochial tastes in many aspects of life. Her comments, plus those of other informants and respondents in Study A, suggest that lower-class males are thought to be insensitive to females and therefore are not likely to treat a female well. Upper-class males are more likely to know how to respond to a female and thus are more likely to be able to earn the admiration and affection of females. Other sources of insensitivity are stupidity and meanness of character. Rachel and Edward. The main problem with the kinds of males who get classified as "jerks" and "nerds" and so forth is that they are often obnoxious. They are so insensitive that they cannot even tell they are unattractive to the female and so they often act as if the relationship were a closer one than what the female wants. Erve, for example, apparently could not sense Patty's negative opinion of him. He would come to her room and talk to her for long periods of time despite the fact that she did not want to talk to him. Another example of a male persevering in trying to get closer to a female is given by Rachel in her description of a painful and frustrating weekend with Edward. Rachel had been friends with Edward for many years. They had been planning to go on a weekend camping trip with two other friends. At the last minute, Edward casually phoned Rachel to tell her that the other friends had decided not to go and that he and Rachel should stay at his university instead of going to the mountains. Rachel was annoyed. She did not want to be alone with him for such a long time. However, because it was too late to arrange anything else and because she really wanted to go somewhere, she went to see him. The entire weekend turned out to be a frustrating struggle over the closeness of their relationship, with Edward indicating he wanted them to be closer and Rachel indicating she wanted the relationship to be less close. This struggle had been going on for quite a while: Rachel: . , . several periods during our relationship he's wanted to get closer than I wanted to get. I don't know, he's a really great guy and I feel real close to him, deep down, but personality-wise, we just have a lot of conflicts, and I don't know, he requires a lot of patience from me. To be around him I have to kind of say, "Okay. You're going to be around Edward, really put yourself down on his level." And he really needs me, as a friend, I feel like, and he tells me that. So I'm just not as enthusiastic about our relationship as he is. Lately, he's just, he's been, every time we've been together, which is several times a month, he'll bring up this stuff about, you know, he just can't help the way he feels

PRESTIGE AND INTIMACY

•m

and he can't stand it anymore, blah, blah, and I've told him how I felt,1 can't change my feelings and I'm really tired of talking about;frv> ..,;.'. ;,f;;, ; /i Rachel's assessment of Edward is that he speaks in a "hlcky way",arid is not "attractive to the opposite sex." His ideas have not developed much beyond what they were many years ago when they first became friehjs: Racket; . . . he seems very immature to me now .Sometimes I feel like he must have been dropped on his head when he was a baby,! mean, he's really slow • -sometimes. • • • •>:• •; '•-:•:/<*,•.*- y y ^ y " ^ ? ^ ^ - ^

:

That Rachel does not feel attracted to Edward is a problem in the,face of his efforts to win her affection: .'.• ""'.'•-'.. V Interviewer: How does it make you feel that he wants the relationship to be closer, whatever? Rachel: It makes me feel real sad because I don't feel that way at all, and I know how much it means to him and there's really nothing that can be done.about the situation, so it hurts me that he feels that way and it seems kind of like a hopeless situation right now, because he really can't get along too well right now without our friendship, but it's painful for him to have the friendship too. It also repulses me too because I can't stand the sight of us being more than friends. I'm just not attracted to him, and [then there's] our personality differences, it just never entered my mind at all. Despite his efforts to treat her well, Rachel is not attracted to Edward, and because of his perseverance in his attempts to get closer, she becomes irritated by his lack of acceptance of her feelings. Although he cares for her and does things for her, he is not attractive enough or sensitive, enough for her to want a closer relationship with him. Rachel attributes his disregard of her negative feelings to his family background and possibly his fundamentalist religious upbringing. She says he may have gotten the mistaken idea that males can earn a female's closeness, or at least that he can win hers, simply by dint of will power. Not only was Edward's attractiveness not sufficient for how close he wanted the relationship to be, but he also had the problem of not being able to accept that he was pursuing a lost cause. This made him even more unattractive in Rachel's eyes because she found his overtures obnoxious and irritating. Erve and Edward are problematic types for two reasons: (1) they are not very attractive or likable, and (2) they are also handicapped by a lack of awareness or lack of character to the point that they are, in some situations, at least, unable to tell what a female would like. For the more insensitive types of unattractive males who cannot tell what a female wants, there is little chance of earning an attractive female's affection, admiration, and intimacy by treating her well. Unattractive, insensitive males would not be a problem if they understood their situation and acted on that understanding, but they are often so "out of it" that they fail to understand their position. They act as though

100

DOROTHY HOLLAND & DEBRA SKINNER

they are attractive and capable of earning a female's admiration and affection. They hang around females who are more attractive than they are and are obnoxious because the female has to "put them off."/ •t In both Study B and Study C, we found cases in which such names zsjerk, ass, asshole, and creep were used as insults. In Study A;; as: well, respondents often indicated that these were insulting and derogatory names. It is now clear why this is the case: These names refer to types who are neither attractive nor adept at treating females well. They refer to men who are on the bottom of the prestige ranking; they are the least likely of males to earn females' admiration and affection, Their insensitivity makes them treat females as poorly as jocks, Don Juans and egotists are likely to do, but they do not have the redeeming quality of these latter types of being attractive in some way. Their prestige is especially low because they do not even know when they are disliked. They make "fools" of themselves by pursuing attractive females who are not at all interested in them. This factor of prestige is why Karen could not call Sam an "ass" and not look like a fool herself. By calling Sam an "ass," Karen would have been indicating that his prestige was low and that therefore it was unlikely he could earn the admiration of a female. Clearly, Sam did not have this problem. Not only was he known as a "big jock on campus," he was also at the party with a female. MALES WHO HAVE (UNUSUAL) SEXUAL APPETITES

From the Study B interviews, we know that talking about the sexual aspects of one's current relationships is an indication of intimacy or closeness. Most of our informants did not feel close enough to us to discuss sex and sexuality in their own personal relationships. The ones who did discuss these topics in the interviews had moments of embarrassment, and even the very articulate ones had difficulty in finding words to describe their interpretations. Where sex and sexuality are talked about on a more impersonal level as a topic of conversation or as a target for joking, however, the informants were less reticent. The "horny" and "oversexed" person, for example, was caricatured even in our presence for comic effect. Because of the informants' reticence and difficulty in talking about the aspects of their relationships that had to do with sex and sexuality, we have only a few in-depth accounts that present information relevant to informants' views on types of males who are problematic because they are sexually unusual or extraordinary. Because of this limitation, we include only one of the few relevant stories plus list a set of assumptions that have been pieced together from the data: 1. Males have a natural desire for sexual intimacy with females, and vice versa. 2. Besides desiring sexual intimacy for its own sake, males also want to demonstrate their sexuality. 3. It is the female's prerogative to decide the extent of sexual intimacy she has with a male.

PRESTIGE AND INTIMACY

101

4. As with affection, admiration, and other forms of intimacy, females are more prone to choose to be sexually intimate with a male if they find him attractive. • Males who have unusual sexual appetites are those who prefer to have intimacy with other males. Also unusual are heterosexual males who have an unnaturally high level of desire and/or a strong need to demonstrate their sexuality. These males are problematic in terms of the presupposed world of male/female relations because both types are likely to provide little prestige or intimacy,to females. Homosexual males focus on males and, although they can be friends with females, they are not suitable romantic partners for females. They are, in fact, competitors forother males. Sexually aggressive heterosexual males are even more problematic. They are prone to treat females in an uncaring manner because sex or their sexuality is of foremost importance to them, not the female and her concerns. They may become overly focused on sex and disregard aspects of the female that are important to her own self-identity. Or, the attractive ones may take advantage of their attractiveness, accepting intimacy from a woman with no intent to treat her well. Or, they may make overtures that cause her to have to make decisions about intimacy before she is ready. Karla, for example, in describing a "pass" a man made at her on their second date, recounts how she assessed what she considered to be a fast invitation to intimacy: Karla: I guess for about two weeks there, I was looking around for a surrogate for my old boyfriend. And {after this incident] I started looking «h him as somebody who would be more of a challenge, someone who'd be kind of fun to play with because I realized this attitude which would lead him to ask me that sort of question on the second date would also make him rather interesting to deal with, and so I was not put off from dating him at all, I just realized that I'd have to be rather clever about it. As Karla interprets it, males who have a strong sex drive or a high need to prove their sexuality are more of a challenge than males who do not. Relating to them is riskier because the pace is faster and more difficult to control than is the case in a normal relationship. Also, there is greater risk of being treated badly. Summary of the cultural model of gender types For American women, at least the ones in our samples, there is a standard, taken-for-granted way in which close male/female relationships both romantic and friendship - come about. The male demonstrates his appreciation of the female's personal qualities and accomplishments by concerning himself with her needs and wants, and she, in turn, acts on her attraction to him by permitting a close, intimate relationship and by openly expressing her admiration and affection for him. In the prototypical relationship, the two parties are equally attractive

102

DOROTHY HOLLAND & OEBRA SKINNER

and equally attracted to one another. However, if the discrepancy in relative attractiveness is not too great, adjustments are possible. A relatively unattractive male can compensate for his lesser standing by making extraordinary efforts to treat the woman well and make her happy. A rela" tively unattractive female can compensate by scaling down her expectations of good treatment. When sufficient compensation is not in evidence or when the more attractive partner seems to be the one who is compensating, the relationship does not make sense and people say to one another: What does he see in her? or, Why does she put tip with him? Don Juans, turkeys, gays, and other types of males are singled out and talked about in relation to this taken-for-granted world of male/female relationships. These types have characteristics that lead them to cause problems for women. Attractive, popular males who are arrogant or selfcentered, for example, take advantage of their attractiveness to women to gain affection and intimacy without intending to enact the friendship or romantic relationship that would normally follow mutual attraction in the taken-for-granted world. They treat the woman badly, which puts her in an uncomfortable position (like that of Annette) of being shown to be less attractive than the male. The woman has revealed her affection and admiration for the male with nothing to show in return. In contrast to Don Juan, jock, or chauvinist types, there are the jerks and nerds, who are not adept at pleasing a female. Unattractive males of this type, the "losers," are particularly problematic because they often pursue a female who is more attractive than they are. Since they are not only unattractive but also inept at earning her affection by treating her well, they are engaged in a futile pursuit. Yet they hang around, impervious to her disinterest and unaware that she is more attractive than they. Eventually, they become obnoxious. Sexually different males also create anomalies in the taken-for-granted world of male/female relationships. Both homosexual males and males who are heterosexual but overly focused on sexual activity render the meaning and value of physical or sexual intimacy between males and females problematic. Homosexual males do not want intimacy with females and therefore cannot be romantic partners for females. Relationships are a priori arrested at the level of friendship. Males who are overly sexually aggressive, on the other hand, force females to a decision about intimacy before the relationship has progressed very far. They are also unlikely to carry through with the relationship because for them there is less involvement with the female as an individual, a person. Discussion Two questions were posed at the beginning of the paper: What do Americans leave unspoken when they talk about gender types? and How is this implicit knowledge mentally organized? The preceding section sum-

PRESTIGE AND INTIMACY

103

marizes the implicit knowledge that the women in our samples take for granted about male/female relationships and about types of males who are likely to cause a relationship to go awry. Here, we outline the implications of our research for the cognitive organization of knowledge about gender types. The final section of the paper discusses the questions of the susceptibility of the cultural model to change and its likely distribution in the American population. - •' THE COGNITIVE ORGANIZATION OF CULTURAL KNOWLEDGE ABOUT GENDER TYPES . , r

I

Our studies indicate that individual Americans understand talk about jerks, wimps, he-men, chicks, broads, and their behavior by thinking of these characters in relation to a taken-for-granted relationship between males and females. In the prototypical sequence of events described in detail in the previous section, a male and a female are attracted to one another and develop a close relationship in which they become friends and/or romantic/sexual partners. Cognitively associated with this taken-for-. granted course of male/female relationships are scenarios of disruption in which one or another of the participants causes the relationship to abort or go awry. Most gender-marked types, it turns out, are types who cause such disruptions. This organization of knowledge according to prototypic events and scenarios is not what we had originally anticipated. At the start of our study, as explained, we rejected the hypothesis that knowledge about gender types is cognitively organized as a list of definitions 'of-.fox,- ddti, scag, and so forth. We turned instead to an analysis of the cognitive structure or similarity structure of the set of types. Researchers customarily assume that this type of analysis, which is usually carried out with the aid of multidimensional scaling, identifies key attributes of a domain (e.g., gender types) and that these key attributes organize and orient people's thinking about the domain. Such an analysis presumes that knowledge about gender types is basically organized according to a set of attributes. Our studies indicated, however, that such an approach was of limited utility for gender types. When the people in our sample were asked which types were similar, they did not perform this task by explicitly focusing on important attributes of the types. Rather, they related the types to a set of scenarios in which the prototypical male/female relationship is disrupted. We found, in other words, that respondents compared types of males and females according to their fit to scenarios. Furthermore, in order to explicate the scenarios, we found it necessary to consult additional data, from which we inferred the underlying taken-for-granted world of male/female relationships. The multidimensional scaling did assist us in identifying an important aspect of the cultural model, namely, the groupings of problematic types of males and females. Cognitive-structure analysis did not, however, provide us with a means of or a motiva-

104

DOROTHY HOLLAND & DEBRA SKINNER

tion for presenting the scenarios that seemed to be an integral part of our respondents' thinking about these different gender types. Nor did it provide us with the information about the taken-for-granted world we needed to understand the scenarios and their emotional poignancy. Nor did it * prepare us for the description of scenes that respondents sometimes seemed to picture as a means of capturing the nature of the type. Without knowledge of the scenarios, we would have been at a loss to explain why respondents thought some terms for gender types could be used as insults whereas others could not. The multidimensional scaling approach offered no indication why calling a male a "Don Juan" or a "playboy" or a "jock" or some other type who is likely to take advantage of a woman is usually not considered to be insulting whereas calling someone a "creep," a "turkey," a "jerk," or some other type of ineffectual male is. We had to learn more about the taken-for-granted world of male/female relationships to know that even though males who use their attractiveness to exploit women may be avoided because they are dangerous, they are not as low in prestige as males who are unlikable and ineffective. Males of the latter category are unattractive. A woman who refers to a male as an "asshole" is indicating that he is unattractive relative to herself. She is indicating that shefindshim "beneath her." The same is not necessarily the case for a woman who refers to a male as a "Don Juan." She is not necessarily insulting him. Although she may avoid him for fear that he will take advantage of her, she is attesting to his attractiveness, and, in fact, may be admitting that he is more attractive than she is. : ; ; In a similar vein, the present analysis illuminates rather than obscures, as does cognitive-structure analysis, the implications of categorizing someone according to a gender type. As Boltanski and Thevenot (1983) have pointed out, social-classification systems are different from nonsocial classification systems because, in applying them, one is also classifying oneself. In typing a male, a female is typing others and herself. This reflexive quality of categorization by gender type is both explained by the cultural model and apparent from the scenarios that include both the male and the female. Relationships between males and females reflect on both parties because of the assumptions in the cultural model that attractive males will choose to be with attractive females, and vice versa, and that attractive females can expect better treatment from males than can less attractive females. To classify a male is to make claims about the male and implicitly about oneself and about other women who have a close relationship with him. In calling Donny a "jerk," the women on Diana's hall were doing more than saying something about Donny. They were saying that they would not put up with his behavior and that they would not be associated with him. Classifications reveal one's standards and sensitivities and therefore one's assessment of one's own attractiveness and claims to prestige. In short, we argue that the cognitive organization of gender knowledge is insufficiently illuminated by an approach such as cognitive-structure

PRESTIGE AND INTIMACY

105:

analysis, which assumes that there is a set of key dimensions or attributes onto which sets of similar gender types are mapped. Our informants; associate type names with a prototypic male/female relationship arid with scenarios of interactions that they sometimes seem to visualize, riot just attributes, as would be expected from cognitive-structure analysis. :Fur~thermore, although people do know important attributes of the different gender types and can say which types are similar and which dissimilar, they also know more - a lot more. They have knowledge of a takeri-forgranted world to which these types are relevant and thus they know'bow: the various attributes ate interrelated. They know not only that'some gender names are insulting but also the basis for and the emotional intensity of the insult. Similarly, they appreciate the reflexivity of categorizing other people by these terms. It is possible that a neophyte (perhaps a child, an anthropologist, or a freshman) begins to learn gender types by memorizing what a cognitive-structure analysis reveals - similar types and their important attributes - but it is also likely that the neophyte would eventually infer the more fundamental parts of the cultural model. She would form an idea of the normal or prototypical course of male/female relationships and come to see the named gender types as actors in this prototypical world. She would go beyond the limited organization of knowledge revealed by the cognitive-structure analysis. INERTIA OF THE CULTURAL MODEL

If our argument is valid and the cultural model does organize the extensive amount of knowledge that we claim it does, then the difficulties of radically altering the model become apparent. The world posited by this cultural model is simply taken for granted as the world to which new experiences are relevant. Not only are new males seen as participants in this world, but they are also seen as possible variations on the small number of problematic types already identified. Gathering information on each new male or female one meets is unnecessary; one need check for only a small number of characteristics. However, the price of this cognitive economy is a bit of rigidity in interpreting the world and a certain slowness in recognizing or learning new models. Cognitive constraints are important forces for the inertia of the cultural model; even more important are the constraints that derive from the social nature of the model. Verbal descriptions of individuals as gender types are understood by listeners in light of the cultural model. Comments such as "Wearing your add-a-beads, eh?," are heard against the backdrop of the extensive implicit knowledge organized by the cultural model. Even new names for new types are interpreted according to this model; one guesses what the type is like - extensive explanation is unnecessary (Holland & Skinner 1985). The shared cultural model vastly facilitates communication; experiences can be rapidly communicated to other people if described according to the conventions of the cultural model. Again, however, economy has a price. It is easy to communicate about the familiar but

106

DOROTHY HOLLAND & DEBRA SKINNER

difficult to communicate about the unfamiliar. Even though it is certainly possible and perhaps even easy for some individuals to think up. px: recognize new gender types, communicating a concept of a truly new, radically different type of male or female to other people is a formidable '" task. Talk about types, new or old, is assumed to be talk that can be interpreted according to the cultural model- Even if the individual manages to think "outside" the cultural model (which may in fact happen^quite ,fre~. quently), hq or she still will face considerable difficulty in communicating the alternative models to other people. Because it provides the backdrop for interpreting and the conventions for talking about experiences, the cultural model is a social entity not easily altered by a single individual.; Along these lines, one wonders if truly radical changes can be made in the model by the mere introduction of a new type, such as & feminist, or, as the men in our samples labeled the type, a "libber.*' Elaborating or introducing a new type is a relatively mild attack since the bedrock of the model, the taken-for-granted, prototypic relationship of males and females is, at best, challenged only indirectly and the new type is easily distorted to fit the existing model. Because the conventions for talking about females and males as types are so much an integral part of the cultural model described here, it is likely that totally new ways of talking about or describing or representing male/female relations may be an easier means through which to introduce new models of these relations. The essays of feminist social scientists, for example, or more likely the selfanalysis talk learned in therapy could provide the new discourse genre.. THE DISTRIBUTION OF THE CULTURAL MODEL ACROSS SOCIAL SPACE AND HISTORICAL TIME

The samples consulted in this study can tell us little about the distribution of alternative cultural models of gender over time and across social groups in the United States. The samples were predominantly composed of respondents who are white, southern, and middle class. Furthermore, most of the Study B and Study C data come from women who are young and unmarried. Young, unmarried women attending universities such as those where our studies were conducted are usually participants in a social system that is closer to what Coleman (1961) and others such as Schwartz (1972) and Eisenhart and Holland (1983) have associated with adolescence or youth than it is to adult society. This youth society emphasizes social identities based on gender and, to a lesser extent, social class. Much of youth culture is devoted to the elaboration of gender relationships and gender types (see Davidson 1984 and Holland & Eisenhart 1981 for further detail on the peer groups of the women in our samples). The importance of gendermarked social types such as those described in this paper, in other words, may be a function of the age of the group studied. Similarly, the dynamics of attractiveness and intimacy posited by the

PRESTIGE AND INTIMACY

107

cultural model may be particularly stressed in youth as opposed to adult culture. Once the women in our samples have married and had children and/or permanently joined the labor force, they perhaps will learn a dif~' ferent perspective on male behavior. A male's potential behavior in and contribution to a household economy and family relationships or his behavior in the workplace may become more important to these women than the male's potential for providing intimacy and proof of one's attractiveness as a woman. Yet, we are reluctant to assume that the cultural model of gerider has no currency in the thinking and talking of participants in adult society. Our interviews with an albeit limited sample of older women suggest that the cultural model continues to be important in the interpretation of experiences that occur in the formation of friendship and (extra-1 marital) romantic relationships with males and in the interpretation of certain relationships in the workplace. Males who systematically treat fe• male co-workers differently from male co-workers are interpreted according to the cultural model described here. Perhaps even more intriguing than the question of the distribution of the cultural model across age groups is its relevance to our samples' male. counterparts. The cultural model of gender described may be "role-centric." Unlike scientific models, which are supposedly constructed from a detached perspective, the cultural model provides for the interpretation of males from the point of view of the female in a (potential) male/female relationship. It is also the case that when the women in our samples talk extensively about particular males, they are usually talking to other women. For these reasons, it might be expected that males' cultural models of gender could differ from that of the females and that males' models tend to take the perspective of the male in the relationship. Unfortunately, we lack in-depth interviews from the males and so have been restricted in our description of the males' perspective. The data from males in the Study A interviews do suggest, however, that males share with females a concern for attractiveness and intimacy although from a different vantage point. Complementary to females' concern about good treatment, for example, males are sensitive to their vulnerability to the demands of females; they worry about becoming involved with a female who is too demanding, too "bitchy." FUNDAMENTALS OF SOCIAL CATEGORIZATION

In our study, we ignored the question of alternative models of gender that may exist in different age, class, ethnic, and regional groups in the United States because we were concerned with a prior question; namely: How do individuals cognitively grasp the cultural models that inform their talk about gender types? Our work is an answer, from the perspective of cognitive anthropology, to the question of which aspects of the cultural model of gender are fundamental, basic, and stable versus which aspects are superficial and likely to be transient. In general, the process of identi-

108

DOROTHY HOLLAND & DEBRA SKINNER

fying a cultural model involves determining fundamental versus surface elements of the complex of beliefs and knowledge (see also Clement ,1982). The distinction between fundamental and surface elements of 0. belief ^system, is, of course, a necessary precursor to meaningful cross-(sub)cul' tural comparison. --•• :•••••:,••.{.•;••-Because of the limited availability of comparable research on social types, it is unwise to anticipate future generalizations about the cognitive fundamentals of cultural models of social types. Recognizing the perils, however, we would speculate that the organization of cultural knowledge will be similar for other social types, at least in American culture, to what we have found in the case of gender. We suspect that the implicit knowledge that informs the talk of other subgroups about gender types and the talk of all groups about other social types such as types of children, or types of hospital patients, or general role terms as described in Burton and Romney (1975) and in Harding and Clement (1980), may all conform to the pattern we have noted in the American cultural model of gender types. We suspect that the type names refer either to roles (e.g., boyfriend, fiance', bachelor, date) in the taken-for-granted world or, more likely, to styles of enacting these roles (e.g., Don Juan, jerk, wimp) that disrupt the prototypical course of the relationship to which they are relevant," On the other hand, we seriously doubt that the content of the prototypic relationship and therefore the problematic social types will be the same from one culture to another.

Notes 1. Because this paper draws on three separate research projects, there are many people and agencies to thank. First are the many individuals who participated. in the studies as respondents, as informants, and as interviewers. Second, several sources of funds made it possible to collect and analyze a large amount of data: a grant from the National Institute of Education (NIE-G-79-0108), a National Research Service Award (1-F32-MH08385-01), grants from the University Research Council of the University of North Carolina at Chapel Hill (1-0-101-3284-VP376, 1~(M0I-3284~VP497), and a Kenan leave from UNC. An earlier version of this paper was presented at the Princeton Conference on Folk Models; comments made by conference participants were extremely helpful. A version was also read in the departmental colloquium series at the University of North Carolina. Thanks go to the insightful comments of participants at that colloquium as well as to a helpful critique by Luc Boltanski. 2. Because the samples in Studies B and C (described below) included only two males, the paper focuses on women's perspectives of men. 3. Holland's pre~1982 publications are under the name of Clement. 4. The Ixil have suffered serious hardship and decimation in the recent and currently on-going government reprisals in Guatemala. As a result, the cultural and social systems of the Ixil have changed considerably since these data were collected (B. N. Colby, personal communication).

PRESTIGE AND INTIMACY

109

5. We selected types that were mentioned frequently. Otherwise, we attempted to include a wide range of types. . •;-••6. The codes indicate the following: The first two digits are unique identifiers for each respondent. The third digit refers to age, with code "3" indicating -18 to 25 year-olds. Codes "4" and above are older. A code of " 1 " in the fourth digit indicates a female respondent. The respondents for the A-l interviews included 13 females aged 18 to 25 and 13 females who were 40 or older. The corresponding figures for .the male respondents were 8 and 8, respectively. For the second interview; all the respondents, 16 males and 26 females, were all between the ages of 18 and 25. All the respondents were residing in North Carolina at the time of the interview. The implications of the sample limitations are discussed below. 7. In the A-l interviews, respondents described one gender type at a time. In the A-2 interviews, the sorting interviews, they described the grouping of types they had formed. •':.*...;• r v . 8. From the sorting data, we calculated similarity measures among all pairs of items using the formula suggested by Burton (1975). These similarities measures were then analyzed using a nonmetric multidimensional scaling program developed by Kruskal (1964a; 1964b) as modified by Napior. 9. Since completing the present paper, we have tested our ability to predict reactions to types not included in our A-2 interviews and to types whose names we created. Our predictions, which were based on the cultural model described in this paper and on conventions we had noted in the names for the types, were largely borne out (Holland & Skinner 1985). 10. In some cases, the description is not about the type of person to which the term refers, but rather is about a type of person who would use such a term or the kind of situation in which.the term would be used: 1231 [buck, macho, stud, chauvinist, egotist, bastard, prick, hunk, Don Juan, playboy] what a female chauvinist pig would think of males - stereotypical attitudes. (cm Case 6 in Fillmore 1982:34) Another example of this kind of assessment was given by a local professional who looked over the names collected in Study A - l . He said his clients would be disdainful of the terms we had been given for homosexuals. He summarized their opinion in a retort, "Only a wimp would call a fag, a *gay.'" 11. AH names are pseudonyms. Furthermore, a few details from the passages have been changed to protect the anonymity of the informants. 12. In the campus cultures of the two universities where these studies were carried out, one big source of prestige for males is participation in athletics, particularly on the University varsity squads. 13. Marilyn Strathem's (personal communication 1984) observations of the Hagen of New Guinea must be noted as a possible counterexample. Hagen males talk about females as gender-marked types, but they do so in the context of exchange and transaction, not in the context of interpreting problematic behavior in a cross-gender interpersonal situation.

References Agar, M. 1980. Stories, background knowledge and themes: Problems in the analysis of life history narrative. American Ethnologist 7(2):223-240.

110

DOROTHY HOLLAND & DEBRA SKINNER

Boltanski, L. and L. Thevenot ;'•.•.-•1983. Finding one's way in social space: A study based on games. Social Science Information 22(4/5):631~680. Bromley, D. B. , * 1977. Personality Description in Ordinary Language. New York: John Wiley and Sons. Burton, M. . 1972. Semantic dimensions of occupation names. In Multidimensional Scaling: Theory and Applications in the Behavioral Sciences, Vol. 2, Applications, A. K. Romney, R. N. Shepard, and S. B. Nerlove, eds. New York: Seminar Press. Pp. 55-71. 1975. Dissimilarity measures for unconstrained sorting data. Multivariate Behavioral Research 10(4):409-423. ;; .:. Burton, M. and A. K. Romney 1975. A multidimensional representation of role terms. American Ethnologist 2(3):397~407. Clement, D. 1982. Samoan cultural knowledge of mental disorders. In Cultural Conceptions of Mental Health and Therapy, A. J. Marsella and G, M. White, eds. Dordrecht, Holland: D. Reidel Publishing Company. Pp. 193-215. Coleman, J. 1961. The Adolescent Society. New York: Free Press. D'Andrade, R. G. 1976. A propositiona! analysis of U.S. American beliefs about illness. In Meanings in Anthropology, K. H. Basso and H. A. Selby, eds. Albuquerque: University of New Mexico Press. Pp. 155-180. D'Andrade, R. G., N. R. Quinn, S. B. Nerlove, and A. K. Romney 1972. Categories of disease in American-English and Mexican-r-Spanish. In Multidimensional Scaling: Theory and Applications in the Behavioral Sciences, Vol. 2, Applications, A. K. Romney, R. N. Shepard, and S. B. Nerlove, eds. New York: Seminar Press. Pp. 9-54. Davidson, D. 1984. Transmission of gender-specific knowledge in women's peer groups. Unpublished master's thesis. University of North Carolina, Chapel Hill. Eisenhart, M. and D. Holland 1983. Learning gender from peers: The role of peer groups in the cultural transmission of gender. Human Organization 42(4): 321-332. Fillmore, C. J. 1975, An alternative to checklist theories of meaning. In Proceedings of the First Annual Meeting of the Berkeley Linguistics Society, C. Cogen, H. Thompson,-G. Thurgood, K. Whistler and J. Wright, eds. Berkeley: University of California. Pp. 123-131. 1982. Towards a descriptive framework for spatial deixis. In Speech, Place, and Action, R. J. Jarvella and W. Klein, eds. New York: John Wiley and Sons. Harding, J. and D. Clement 1980. Regularities in the continuity and change of role structures: The Ixil Maya. In Predicting Sociocultural Change, S. Abbott and J. van WilUgen, eds. Athens, Ga.: University of Georgia Press. Pp. 5-26. Holland, D. and D. Davidson 1983. Themes in American cultural models of gender. Social Science Newsletter 68<3):49-60. Holland, D. and M. Eisenhart 1981. Women's peer groups and choice of career. Final report to National Institute of Education, Washington, D.C.

PRESTIGE AND INTIMACY

in

Holland, D. and D. Skinner : ;', 1985. The meaning of metaphors in gender stereotyping. North Carolina Working Papers in Culture and Cognition No. 3. Durham, N.C.: Duke University Department of Anthropology. Keesing, R. 1979. Linguistic knowledge and cultural knowledge: Some doubts and speculations. American Anthropologist 8!(l):14-37. '•'' '•''••'-.- ••; Kruskal, J. B. .......... 1964a, Multidimensional scaling by optimizing goodness of fit to nonmetric hypotheses. Psychometrika 29:1-27. 1964b. Nonmetric multidimensional scaling: A numerical method. Psychometrika 29:115-129. - •' ' : Labov, W. and D. Fanshel 1977. Therapeutic Discourse: Psychotherapy as Conversation. New York: Academic Press. '->.. , • Neisser, U. 1976. Cognition and Reality. San Francisco: W. H. Freeman and Company. Rice, G. E. 1980. On cultural schemata. American Ethnologist 7(1);152-172. Rosenkrantz, P., S. R. Vogel, H. Bee, I. K. Broverman, and D. M. Broverman 1968. Sex-role stereotypes and self concepts in college students. Journal of Consulting and Clinical Psychology 32:287-295. Schank, R. and R. Abelson 1977. Scripts, Plans, Goals and Understanding: An Inquiry into Human Knowledge Structures. Hillsdale, N.J.: Lawrence Erlbaum Associates. Schwartz, G. 1972. Youth culture: An anthropological approach. Addison-Wesley Module in Anthropology, No. 17. Reading, Mass.: Addison-Wesley Publishing Company. Spence, J. T., R. Helmreich, and J. Stapp 1975. Ratings of self and peers on sex role attributes and their relation to selfesteem and conceptions of masculinity and femininity. Journal of Personality and Social Psychology 32:29-39. Spradley, J., ed. 1972. Culture and Cognition: Rules, Maps, and Plans. San Francisco: Chandler Publishing Company. Tyler, S. A., ed. 1969. Cognitive Anthropology. New York: Holt, Rinehart and Winston. White, G. 1980. Conceptual universals in interpersonal language. American Anthropologist 82(4):759-781.

5 A folk model of the mind1

Roy'D'Andrade

A cultural model is a cognitive schema that is intersubjectively shared by a social group. Such models typically consist of a small number of conceptual objects and their relations to each other. For example, Rumelhart (1980), following Fillmore (1977), describes the schema - and cultural model - of buying something as made up of the purchaser, the seller, the merchandise, the price, the sale, and the money. There are several relationships among these parts; there is the interaction between the purchaser and the seller, which involves the communication to the buyer of the price, perhaps bargaining, the offer to buy, the acceptance of sale, the transfer of ownership of the merchandise and the money, and so on. This model is needed to understand not just buying, but also such cultural activities and institutions as lending, renting, leasing, gypping, salesmanship, profit making, stores, ads, and so on. Cognitive schemas tend to be composed of a small number of objects at most seven plus or minus two - because of the constraints of human short-term memory (Miller 1956; Wallace 1961). For example, to judge if some event is an instance of "buying" something, the person making the judgment must decide whether there has been a purchaser, seller, some merchandise with a. price, an offer, and an acceptance, along with the appropriate transfer. Since all these criteria must be held in mind simultaneously to make this judgment with any rapidity, the criteria cannot exceed the limits of short-term memory. The number of objects a person can hold in mind at any one moment is limited, but these objects may themselves be complex schemas (Casspn 1983). In the buying schema, for example, the part labeled bargaining is itself a complex schema that involves a. potential purchaser and seller, an initial price, a series of converging bids and counter offers, and possibly a. final agreement. Through hierarchical organization, human beings can comprehend a schema containing a very large and complex number of discriminations. The amount of work involved in unpacking a complex cultural schema can be quite surprising. One consequence of the hierarchical structure of schemas is that certain cultural models have a wide range of application as parts of other models.

A FOLK MODEL OF THE MIND

mm-

The cultural model of money, for example, has a wide range of appii<$-;'' tion, serving as a part of many other models. Although it is uidikely^hat\.;! anyone knows all the models of any culture, to have a treasonable;';:^l{: understanding of a culture, one must know at least those models that are V; widely incorporated into other models. -:,v, ;;••-..•; ;o v i K ^ : ^F:PK A schema is intersubjectively sharedwhen everybody in the group knows '"'••'.' the schema, and everybody knows that everyone else knows the schema, and everybody knows that everyone knows that everyone knows ;the schema (the third "knowing" is necessary because although you and Tinay' •' both know the money is hidden in the teapot, for example, and I may know that you know (1 saw you hide the money there), and you may know that I know (you caught a glimpse of me when I was spying on you as you hid the money), yet because I do not know that you know that I know, I cannot assume that your seeing me look at the teapot would tell you that I was thinking about the money. However, when everybody knows that everybody knows that everybody knows, then anyone's glance toward the teapot is understood by all, including the one giving the glance, as a potential reference to the money. One result of intersubjective sharing is that interpretations made about the world on the basis of the folk model are treated as if they were obvious facts of the world. The spectators at a baseball game all see that a particular pitch, thrown over the head of the catcher, was obviously a ball, and so obviously a ball, that one would have to be blind to miss it. Of course, those people who do not know the game of baseball, seeing only the catcher trying to catch something thrown to him, cannot make such an interpretation and do not experience any such fact. A second consequence of the intersubjective nature of folk models is that a great deal of information related to the folk model need not be made explicit. For example, in describing a game of baseball in which at the bottom of the ninth the score was tied, the bases were loaded, there were two outs, and the count was two and three, the narrator has only to say that the pitch was so far over the head of the catcher that he couldn't even catch it. People who know baseball do not need to be told the pitch was a ball, the ball gave the batter a walk, the walk forced a run home, the run gave the game to the team at bat, and the game was over. The narrator, speaking to someone who knows baseball, can reasonably assume that what obviously must happen (given the rules of baseball) does not need to be stated. One cultural model with a wide range of application in American and European culture is the folk model of the mind. This model can be called a "folk" model both because it is a statement of the common-sense understandings that people use in ordinary life and because it contrasts with various "specialized" and "scientific" models of the mind (see Keesing this volume). This model is widely incorporated in a variety of other cultural models, such as categories of criminal acts, the classification

114

ROY D'ANDRADE

system found in ordinary language character terms (D'Artdrade 1985), categories of speech acts (D'Artdrade & Wish 1985), and the cultural model of commitment involved in marriage (Quinn 1982) and so pn. : : : -,;f, ., An interesting characteristic of many kinds of cultural models is the * quality of awareness of the model displayed by informants.' In the ease of the model of the mind, for example, most informants do not have an organized view of the entire model. They use the model.but they.cannot produce a reasonable description of the model. In this sense, the model is like a well-learned set of procedures one knows how to carry out rather than a body of fact one can recount. This difference corresponds to the distinction made in artificial intelligence circles between "procedural" knowledge, such as knowing how to ride a bicycle, and "declarative" knowledge, such as knowing the history of France (Rumelhart & Norman 1981). However, the folk model of the mind does not seem to be a completely procedural system since informants can partially describe how the model operates when asked questions about specific examplesOne issue raised by the attempt to make explicit the folk model of the mind is the question of the empirical basis - the accuracy - of the model. At one extreme, it might be argued that this folk model of the mind is based on "obvious" facts of human experience. That is, one might argue that people can perceive their internal states and processes just as well as they can perceive trees and birds, and so the folk model is simply a description of what is there - perhaps it could not even be described differently. At the other extreme, one might argue that by their nature, internal states and processes are so difficult to perceive that the folk model has no more relation to reality than has the Azande model of witchcraft. Cross-cultural information about folk models of the mind in other cultures is potentially relevant to a resolution of this problem. Some comparison of the model presented here for American-European ("western") cultures and Lutz's Ifaluk material on ethnopsychology are presented in the last section of this paper. At this point, it is sufficient to note that this folk model cannot appropriately be applied under all circumstances; it generally is not thought to apply to such special conditions as "hypnosis," or to various mental disorders such as "psychosis" and "depression." Indeed, it seems that when the model does not apply to how someone is acting, people consider the person to be in an "abnormal" state. Thus, the model seems to act as a standard for determining "normality," I have found the work of linguistic philosophers, such as Anscombe, Vendler, and Searle, to be very helpful in developing a description of the western folk model of the mind, although sometimes it is difficult to decide if philosophers are describing how our folk model of the mind is or how it should be (see, for example, Ryle 1948, who did not like the western folk model of the mind at all). Also, philosophers are willing to criticize a folk model with respect to its internal consistency and its logical compatibility with other models in the same culture - a move anthropology

A FOLK MODEL OF THE MIND

115

has yet to make (but see White this volume). Work done by Edwin Hutchins in an unpublished paper on how people generate explanations of ongoing behavior has also been very helpful, although the'model developed by Hutchins differs considerably from the model presented here (Hutchins, n.d.), The initial model appears in the next section. It is followed by a summary of the major propositions of, the model and a set of interview questions designed to test these propositions, along with illustrative interview responses. The informants were five college and high school students who had never had courses in psychology. The interview material presented here has been selected on the basis of clarity and exphcitness. None of the interview material from the five informants contradicted the model, although some of the material could not be derived from just the model given here. In addition, some material from daily life and from literature that illustrates use of the model is presented. In the last section of this paper, this folk model is contrasted briefly to the scientific models of the mind found in academic psychology and psychoanalytic theory, and then related to a nonwestern folk model of the mind described by Catherine Lutz, with some concluding speculations about cross-cultural similarities and differences. The model of the mind The folk model of the mind is composed of a variety of mental processes and states. These processes and states, as indicated by English verbals, are: a. perceptions: i. simple state - see, hear, smell, taste, feel ii. achieved state - spot, sight, notice iii. simple process - look, observe, watch, listen, touch b. belief/knowledge: i. simple state - believe, know, remember, expect, assume, doubt, imagine, suspect, recall ii. achieved state - understand, realize, infer, learn, find out, discover, guess, conclude, establish, forget iii. simple process - reason, think about iv. accomplished process - figure out, plan c. feelings / emotions: i. simple state - love, like, fear, hate, blame, approve, pity, sympathize, feel sad, feel happy ii. achieved state - forgive, surprise, scare iii. simple process - enjoy, be frightened, be angered, be bored, mourn, emote d. desires/wishes: i. simple state - want to, desire, like to, feel like, need

116

ROY D'ANDRADE

ii. achieved state - choose, select -/'•••'•••. iii. simple process - wish, hope for . e, intentions: / , i. simple state - intend to, aim to, mean to, plan to r ii. achieved state - decide to f. resolution, will, or self-control: i. simple state - determined to, resolve to ii. achieved state - resolve to iii. simple process - force oneself to, make oneself, strive The distinctions of state and process and the subdistinctions of achievement and accomplishment are based on the time schema of the verb (Vendler 1967). When we inquire about a process, we ask, "What are you doing?" and the answer is, "I am looking/thinking/enjoying .". ."; that is, one is carrying out a repetitive set of internal actions. When we inquire about a state, we do not ask what the person is ". . . ing," rather we ask "Do you see/believe/like. . . ?" Outside idiomatic use, we do not say, "Iamseeing/believmg/liking. . . ." Both the state and process occur in time, but a process is something marked by an iteration of some action and thus admits continuous tenses. In many cases, one can treat the same internal events as either a process or state. "I have been thinking about the tie-up on the freeway" references the process of thinking, whereas "I believe we should avoid the freeway" places oneself in a particular state of belief. This semantic distinction indicates that the folk model has two different ways of regarding the mind - as a collection of "internal states" versus a set of "internal processes." A typical illustration of this distinction is the "sleeping person" example: Whether Joan is awake or asleep, we can say she"knows the multiplication table, fears nuclear war, probably intends to go shopping this weekend, and so on. But only if she is awake can we say she is calculating the answer to 11 times 15, worrying about nuclear war, planning to go on a trip, and so on. Thus, the mind is treated both as a container that is in various states and conditions, thereby having large number of potentialities simultaneously, and also as a processor engaged in carrying out certain operations, thereby being limited to a small number of concurrent actions, Further, states are linked to processes in that typically someone is in a particular state because some process has or is occurring. Thus, John sees Bill because he is observing Bill; Sally believes Lisa is her friend because she went through the process of assessing her relation to Lisa and finally concluded she was a real friend; and Roger has been frightening his cousin, which is why his cousin fears him. There is another relevant time distinction in English verbs based on the notion that certain processes and states are defined by a climax or terminal point that marks the end of the state or process. When such ter-

117

A FOLK MODEL OF THE MIND Table 5.1. Characteristics of internal states Perception

Belief

Feelings

Desires

cause outside . cause inside cause inside, cause inside mind and and : mind outside mind outside mind takes simple takes prop takes takes prop. either object objects object self usually agent

self usually agent

self usually object

self usually agent

not controllable

usually controllable

usually not controllable

count noun have many at once

count noun have one at a time

mass noun have many at once

Intentions " ''-.Resolutions' cause inside mind

cause inside mind..;"';-,.;:.

takes prop. " 'takes prop, "; object "object"';" self always " self always''"' agent "•

agent

not controllable

controls itself

control of control

count or mass perhaps have many at once

count noun

count noun

perhaps have many at once

perhaps have many at once

• •";;'.•.••-l'

minal points define a state, they are called achievements. When they define a process, they are called accomplishments. For both achievements and accomplishments, we ask, "How long did it take to . . . ." Generally, we do not ask how long a simple state or process takes - we do not say, "How long did it take to believe that. . . ." For the simple states and processes, the event is treated as homogeneous across the entire period through which it occurs. Once one begins the process, one is truly in the process even if it is concluded abruptly. Thus, even if one thinks for only an instant, one has been thinking. However, no matter how long one has been at it, one does not realize something until that very moment when the light dawns (Vendler 1967). There are a number of ways in which the various processes/states differ from each other. Table 5.1 summarizes a collection of these differences. In Table 5.1, the resolution category is almost indistinguishable from the intentions category. In general, what appears to distinguish resolutions from intentions is that resolutions are second-order intentions - intentions to keep certain other intentions despite difficulty and opposing desires. The first distinction in Table 5.1 involves the concept of cause: the idea that certain events are thought to bring about other events. Except in pathological cases, what one sees, hears, and/or senses is understood to be caused by various events and objects external to the mind. What one knows or believes is usually considered to be a creation from within, a result of the operation of the mind itself. What one feels emotionally is more problematic. Sometimes emotions are treated as something caused at least in the sense of being "triggered" - by external events ("E.T. is so charming I couldn't help liking him.") At other times, emotions are treated

118

ROY D'ANDRADE

as internally generated by the person ("Thinking about the game made Charley nervous.") Desires, like emotions, are also seen as both internally and externally caused. Intentions and resolutions, however,;are treated as directly caused only from within. \S' .:','...'. >" Whether caused from the outside or created inside, according to the folk model one is generally aware of what one perceives, thinks, feels, desires, and intends. Of course, sometimes one can see something and not be fully aware of what one saw, or have some feeling or desire about which one is confused, but these are treated rather like problems that can be resolved by turning one's full attention to the situation. Perceptions, thoughts, feelings, and so on in verb form vary in the kinds of objects they take. There appear to be two major kinds of objects: simple objects and propositional objects. Simple objects are objects like "cats" and "disasters" - they are things and events in the world, not thoughts about the world. Prepositional objects, on the other hand, are not "things" they are "thoughts" or "beliefs," such as the belief that there is likely to be a nuclear holocaust. Perception verbs usually take simple objects - we see John, hear about the war, notice a mistake. However, what one believes or knows, wishes or hopes for, aims to do or resolves to do normally involves some proposition about the world. In philosophy, states such as knowing or intending that take propositional objects are called "intentional states" (Kenny 1963). Stative verbs - that is, simple states and achieved states - of feeling and emotion can take either simple or propositional objects; for example, "Tom fears that Sue lost her wallet" versus "William is afraid of lightning." In the first case, it is a propositionaiized state of affairs (something imagined or thought) that is the object of Tom's fright; in the second, it is an external physical event that causes William's fear. It seems to be the case that feelings and emotions are sometimes treated in the folk model like perceptions that take simple objects and sometimes like cognitions that take propositional objects. Emotions also differ from the other internal states in that some emotions do not need an object of any kind: I may feel anxious or sad or happy not about anything, but just in general. Anscombe (1963) and Searle (1975; 1980) have pointed out that there are different "directions of fit" for various internal states. Perceptions and thoughts should fit the world, that is, should correspond to how the world is. But in the case of desires, intentions, and resolutions, it is the world that someone wants to bring to fit whatever state of affairs is represented. Perceptions, thoughts, feelings, desires, and intentions also differ in their relation to the self. With verbs of perception, thought, desire, and intention, the self is typically depicted as the active agent rather than the passive experiencer. However, one can say "the thought struck me," or "the urge to have a cigarette overwhelmed me," where the self is treated as something reacting to other parts of the mind. In the case of feelings and emotions, the typical verbal form is for the self to be a passive ex-

A FOLK MODEL OF THE MIND

m

periencer. Thus, we say that things bother, frighten, and bore us. Another common form is the use of the verb feel (e.g., "She feels happy5'); in which the emotion is treated as something that produces a sensation experienced • by the self. For many emotions, one can use either agentive or experiential verb forms: to fear versus to be afraid, to hate versus tofeelajrtgry/ and so on; v;z'^w-?:&yvs:i''i Even though the self can be treated as the experiencing object of most • internal states, the self is always the agent of intentions. Intentiohs do' not overwhelm us, or bother us - intentions are the very core of the active self. The folk model treats the self as an area of focus that can expand and contract, but the limit of its contraction lies outside the core act of • intending. ' -'•^•: ^/^•?'••> 'MV The self is also portrayed as able or unable to control various mental operations. One cannot directly control what one will perceive: One cannot turn the perception of blue to red or round to square under normal circumstances. Thoughts, on the other hand, are considered to be under control by the self: One can choose what one wishes to think about. However, it is acknowledged that sometimes it is difficult to stop thinking about something, especially if there are strong emotional promptings of some sort. Feelings, like perceptions, are not considered to be under one's direct control. One may be able to modify one's feelings by thinking of one thing rather than another, or by engaging in various activities, but according to the folk model, one cannot will one's self to hate or not to hate, to love or not to love someone, or even to enjoy something (but one can try). The situations seem iess clear with respect to desires; but overall, they operate with respect to self-control like emotions: There seems to be no way to make oneself not want something or to want something one has no desire for. With respect to intentions, the idea of self-control is redundant since intentions are self-control. In intending to do something, we (our self) decide what we shall do. An important aspect of emotions is marked in the folk model by the categorization of emotions by mass nouns rather than count nouns. In English, a count noun is something that can be numerically quantified one can have one house, two houses, and so on. A mass noun, on the other hand, does not have the defined edges that make counting possible one can have lots of money, sand, or anger, but in ordinary talk one does not have two monies, two sands, or two angers. In poetry, one can say "a grief ago," thus, treating "grief" as something countable; but in most discourse, emotions are usually not treated as discrete, quantifiable things one feels sad, not the third sadness today. Further, like water and color, emotions can blend together, so that one feels several feelings at the same time. This is not true of propositional thoughts - one can have only one at a time, and even though they can get mixed up, they do not blend. Desires, like feelings, can occur simultaneously, and perhaps in some way can blend, but this seems less clearly worked out in the folk model.

120

ROY D'ANDRADE

In the folk model of the mind, the different kinds of internal states and processes are organized into a complex causal system, described jn the next sections. "* ACTIONS AND INTENTIONS

Complex human actions are assumed to be voluntary unless something indicates otherwise. A voluntary action is one in which someone did something to accomplish some goal. Given the question, "Why did John raise his hand?" one can answer, "To get the teacher's attention," if it is understood that raising one's hand is a way of getting a teacher's atten-,, tion. It is unusual for someone to explain an act simply by saying that the act was intended: for example, the sentence "John raised his hand because he intended to" sounds odd unless there was some reason to suppose that John might have raised his hand involuntarily - perhaps because his hands were attached to strings that could be used to pick up his hand. Since in the foik model actions do not occur without intentions, and since, following the Gricean maxims, we do not say what is obvious, normally we do not explain an action by saying it was intended. Anscombe (1963) has pointed out that intentions may be formed either prior to the act or as the act is being carried out. When one turns the wheel of a car in an emergency to avoid an accident, one intends to turn the wheel. The action and intention occur together (See also Searle 1980). , INTENTIONS AND DESIRES

Why do people have one rather than another intention? The normal expectation based on the folk model is that people intend to do those things that they desire/want/need/wish to do. The term desire highlights the affective aspect of this state ("He felt no desire for a cigarette"); the term wish highlights the conceptual aspect ("He wished that he had told the truth"); the term need highlights the physical or emotional necessity of obtaining satisfaction ("He needed a drink in the worst way"); and the term want appears to light evenly each of these aspects. A desire may be directly satisfied by some action (e.g., "Susan kissed John because she wanted to") or the desire may be indirectly satisfied by the action (e.g., "Susan kissed John because she wanted to make Bill jealous"). In this example, we explain why someone did something by attributing some want or wish or desire or need to the actor without explicitly mentioning any intention. The intention can be assumed because it naturally follows from what is desired. Do people have intentions without any kind of wish, want, need, or desire as their cause? Not normally, but it is recognized that sometimes one does something intentionally without understanding why - without understanding what it could be one wants. "I told him I would go, but I don't know why I did - 1 certainly don't want to go." This is a puzzling

A FOLK MODEL OF THE MIND

121

state of affairs since intentions are supposed to be connected to desires. When the actor experiences intentions without wishes, it is as if there was a failure in perception. The connection should be there - why can't I see it? Sometimes people do things not because they want to, but because they have been coerced. "Bill gave the robber his money.because the robber. threatened to shoot him if he didn't." The conventional analysis of this situation is that although Bill did not want to hand over his money, he did want to continue living, and his desire to continue living was stronger, than his wish to keep his money. Thus, the intended act is still based on a wish, but one that is indirectly rather than directly related to the action. • Are desires really different than intentions? Or, are intentions just very, specific desires? According to the folk model, desires and intentions are different things, since I may have a wish to visit China without having formed any intention to visit China. One can have desires about which one intends to do nothing. Intentions are like desires in that both have as their objects desired future states of affairs, but in an intention the decision to act has been made. . Nevertheless, it would sound strange to talk about desires that do not become intentions even when all the conditions required to satisfy the desire are present - if I really want to go to China, and the means were available, and there were no drawbacks to going, would I not act on the wish? According to the folk model, I would if I really wanted to go to China. But then it would no longer be just a wish - it would also be my aim, goal, intention, decision, to go to China. According to the folk model, desires naturally become intentions under the right conditions. Desires also have an emotional component, and, as discussed, the self is often treated as the object acted on by a wish (e.g., "The desire for a cigarette overwhelmed me"), but the self is rarely if ever treated as the object of an intention. A sentence such as 'The intention to have a cigarette overwhelmed me" sounds wrong. There is considerable question in the philosophic literature about whether desires have a unique emotional component. Is there a distinct feeling that is desiring, Or is desiring simply the anticipation of some specific feelings, or is it a particular characteristic of certain feelings? If "John wants to see Susan," is there a distinct feeling of wanting involved, or is the wanting just the anticipatory enjoyment of Susan's company, the anticipation of not feeling lonely? The boundaries here do not seem to be clearly marked. One can answer a question about why someone wants something with a means-end formulation - John wants to see Susan because he wants to give her a present because he wants to impress her because he wants her to go with him to the dance because . . . . At some point in the means-end hierarchy, we come to such ultimate wants as staying alive, being happy, and/or avoiding unpleasant feelings. Are these ultimate wants

122

ROY D'ANDRADE

based really on feelings of some sort, or are they self-causing? The boundaries here are also not clearly marked. FEELINGS AND DESIRES

* Another answer to the question of why John wants to see Susan is "Because he misses her," or "Because he enjoys her company." In these explana-; tions, a desire is causally related to some feeling or emotion (The term feeling is somewhat more general than the term emotion. "Pain," for ex-. ample, is usually not called an "emotion," but it certainly is a feeling.) In general, feelings and emotions are thought to lead to desires. If John gets angry, we will wonder what he will want to do about whatever it is that is making him angry. If John is angry because Bill did not help him when he needed help, John's anger may result in his deciding not to speak to Bill, or in his wanting to telling Bill off, or in his intention to wait to get even with Bill (Lakoff & Kovecses, this volume). The emotion or feeling behind a desire need not be immediately experienced. John might want to see Susan because he thinks he would enjoy meeting her. Here, the feeling is anticipated. Is the anticipation of a feeling also a feeling (attached to a thought), or is it just a thought? Similarly, John might want to see Susan because he thinks one ought to visit old friends. Here, what seems to be anticipated is some feeling of guilt if the act is not done. In these cases, the folk model does not seem to be clear as to whether the anticipation also "carries" feeling. Feelings generally give rise to desires, but does every feeling give rise to a desire? Can one feel sad or angry or happy without it; leading to any identifiable desire? On this point, intuitions differ. However, we do expect that there will be a relation between the kinds of feelings a person has and the kinds of desires these feelings engender: Feelings of anger, for example, are expected to lead to desires that involve destruction or harm, whereas feelings of love are expected to give rise to desires that involve protection and care. The connection between feelings and desires does not seem to be as tight as the means-ends relation between intentions and wishes. Within broad constraints, there are many possible desires that can result more or less expectably from the same feeling. One reason the connection between feelings and desires is looser than the connection between desires and intentions is that the means-ends relations are located in different worlds. The means-ends relation between desires and intentions is located in the actor's understanding of the external world. If one wants to acquire a million dollars, certain intentions are reasonable - one might decide to buy a lottery ticket, apply for a job at Brinks, or study the stock market, for example. The constraints here are in the understood causal structure of the world - certain things might lead to acquiring a million dollars; other things would probably not. The assumption of the folk model appears to be that the causal structure of the external world affects a person's understand-

A FOLK MODEL OF THE MIND

123

ing of that casual structure - however imperfectly T- and thereby affects: what intentions will follow from what wants. .:-;lw.)j:Ukx-iv In the relation between feelings and desires, however, the causal struc-: ture is the mind of the individual. Why did John's anger at Bill lead him not to want to speak to Bill, rather than wanting to tell him off, or wantf ing to do any one of a number of other things? How will telling BUI off\ affect his feelings? Will he really feel better? The answer tpsuch gues-: tions lies in a causal structure that is John's mind. Someone who does not know John can only make a guess based on the assumption that John reacts the way other people do. John himself may not know the answers to any of these questions. :...-.i-v-.::---; In general, feelings do not seem to be clearly demarcated in the folk model. There are specific emotions, like love, amusement, irritation, and fright, that give rise to various desires. There also are general sentiments such as liking or enjoying something, or disliking something, or being pleased by something, or being made uncomfortable by something, which are given as explanations for desires (e.g., "He wants to go to the game because he likes to watch football.") How are these sentiments related to specific feelings? Some feelings are thought to be pleasant, others unpleasant - the so-called "hedonistic tone" of the various emotions seems well agreed on. Is the unpleasantness of fright a separate feeling that comes with being frightened, or is it simply a characteristic of fright, along with such other characteristics of fright as high arousal, and anticipations of disaster? If the unpleasantness of fright is just a characteristic of fright and not a separate feeling, how about the enjoyment of listening to music? Is that not a separate feeling? These questions have been much debated in philosophy. (For a review of these issues, see Kenny 1963.) . What seems to be the case with regard to the folk model is that sometimes "pleasure," "enjoyment," "liking," "displeasure," "dislike," "anticipation," and so on, are treated as feelings in their own right and sometimes they are treated as characteristics of other feelings. The equivocation of the folk model on this issue may be due to some innate difficulty that human beings have in perceiving the boundaries of feelings. The amorphous nature of feelings, indicated in the treatment of emotions as mass nouns rather than as count nouns, seems to lead to feelings being conceptualized in contradictory ways. This may be why the folk model is also equivocal with respect to whether wishes involve a unique kind of feeling, whether anticipations are also feelings, and whether there are wishes that are not based on feelings. (On the other hand, our experience of the "amorphous nature" of feelings may be due to the vagueness and ambiguity of the model we use to understand them, not to their actual lack of structure. It would be of psychological interest to know which hypothesis is true.) One interesting aspect of feelings' is that they are thought to cause various involuntary visceral responses - turning pale or flushing, trem-

124

ROY D'ANDRADE

bling, fainting, sweating, for example - although the degree of individual and situational variation in the manifestation of these responses is considered to be very great. / ...... --BELIEFS AND FEELINGS

• ..

:.

-;;;;,:

In the folk model, acts, intentions, desires, and feelings are connected in a simple causal chain. There are no direct feedback loops: Intentions do not lead directly to desires, nor do desires lead directly to feelings. We would not explain Tom's desire to go to Spain by saying it was his intention to visit Europe, nor would we explain Howard's hatred of Wimbledon by saying he wished to avoid seeing tennis matches. However, if reversed, these explanations sound sensible: We explain Tom's intention to go to Spain by saying he wants to visit Europe, and we explain that Howard wishes to avoid Wimbledon by saying he hates tennis. Beliefs, however, are expected to influence feelings, and feelings are expected to influence beliefs. Here, there is a two-way causal relationship. Someone who believes he or she has lost a friend is likely to feel sad. And someone who is sad is likely to think about the time he or she lost a friend and believe the world is a grimmer place. Even though there is a two-way causal connection between beliefs and feelings, the path from beliefs to feelings is not conceptualized exactly the same way as the path from feelings to beliefs. Feelings and emotions are considered reactions to the world, mediated by one's understanding of the world. These emotional reactions are treated as innate human tendencies, modified in each case by the-particulars of experience and character. The causal connection whereby experience - what one believes has happened - arouses feeling is considered to be strong and immediate. The effect of feelings and emotions on belief, however, is not considered to be as strong as the effect of belief on feelings. Feelings are portrayed as "coloring" one's thinking, "distorting" one's judgment, "pushing" one to recall certain things, confusing one, for example. The image here seems to be of a force which is a sort of perturbation of the medium. One imagines a swimmer caught in a current. By itself, just the process of thinking is not considered to have much power to arouse the emotions. "Just thinking" about nice things or bad things may have some emotional effect, but we expect such effects to be small except in pathological cases. It is only in its role as the formulator of what one believes or as the interpreter of perceived events that the process of thinking has major effects on feeling and emotion. Thinking is considered a part of how one comes to believe that things are a certain way, and it is to what is believed to be the case that people respond with emotion. In some mental states, feeling and belief blend together into a single entity. Thus, "approval" is a state that combines both belief and feeling. One cannot say that someone approves of something but has no feeling

A FOLK MODEL OF THE MIND

125

about it, or that someone approves of something but has no belief about it. Perhaps one can think something is good in some way without feeling anything, and perhaps one can like something without consideration or thought about it. But if one disapproves of something,'one does so-because of certain things one thinks and because one feels a certain way.: Like approval and disapproval, wonder and doubt also meld together feeling and belief. Related terms, like anticipation (discussed above) and surprise, may also' be used in the sense of a combined feeling and thought, although the affective component seems weaker here (Vendler 1972).•-<:': BELIEFS, DESIRES, AND INTENTIONS

Belief also has a two-way causal relationship with the perception of external objects and events. The major direction of causation runs from perception to belief: Seeing or hearing certain things leads me to believe certain things. I see the car go by, so I know (am justified in my belief) that a car went by, and I realize that traffic is still moving. However, belief is not considered just a reflex of perception. People can believe things to be true that they never experienced, and they can even believe they "saw" things happen that did not happen. Perception is not considered an errorfree process in the folk model, and belief is often thought to be one reason for an erroneous perception. For example, if I believe that Jim is a bad person, I may perceive his "bumping" into Tom as a deliberate attack although an unbiased observer would have seen only an accident. In the folk model, beliefs are also causally related to each other: One belief can give rise to another, inconsistency between different beliefs may bring about various attempts to escape from the dilemma and so on. The general interrelatedness of beliefs is indicated in the folk model concepts of inference, evaluation, and judgment, in which a particular proposition is finally accepted or rejected after searching among other propositions for confirming or disconfirming evidence. Thus, beliefs are treated in the folk model as having causally complex relations to both feelings and perception. The feedback loops in which belief affects feeling, which, in turn, affects belief, and in which perception affects belief, which then affects perception, give the portrayed machinery of the mind a complexity and flexibility it would not have if the causal chain were depicted as running solely in one direction. Even though the main line of causation in the folk model runs from perception to belief to feeling to desire to intention to action, belief also has a special direct relation to desire and intention. This relation is based on the fact that the states of intention and desire have propositional or intentional objects - that is, they are directed toward the world through the medium of thought, or through framing propositions. One wishes something or another were the case, and the formulation of something being the case is a thought. To want there to be a better world presupposes the mental formulation of the notion of a "better world."

126

ROY D'ANDRADE

Since what one can desire, wish for, or want depends on what one can think, thought enters directly into wishes, but not in a causal sense'. According to this account, cats can wish to catch birds because they can conceive of catching birds, but it is unlikely that cats wish to have souls because "it is unlikely that they can formulate the notion of having a soul. Thus, in the folk model the quality of one's wishes depends on the quality of one's thoughts - evil he who evil thinks. Intentions are, in this regard, like wishes: Any intention takes as its object a state of affairs formulated in a thought. However, there is a further relation between intentions and thoughts in the folk model, which is expressed in the notion of "planning." For example, suppose one wishes to visit Italy and decides to visit Rome during the coming summer. This intention cannot be carried out without further specification of action, which means planning. Such specifications involve working out what means of travel to take, where and when to make reservations, when to leave, where to stay and so on. Planning consists of thinking out a feasible set of actions to accomplish the intention or goal. Once the plan is made, each of the conceived actions becomes a subgoal or subintention, which itself may require more planning before the initiating intention can be accomplished. The folk model treatment of desire and intention as states that take propositionally framed objects or states of affairs means that what can be wanted, aimed for, and planned depends on what is known, or believed, or understood. There is a further effect here, and this is that since what is wanted, aimed for, and planned are things thought of, one may "deliberate" about these wants, aims, and plans. These deliberations may, in turn, lead to other feelings, such as guilt or doubt, or other wishes, which may counter the original wish, or may involve various second-order intentional states, such as resolution or indecision. Were this feedback loop, in which one can think about what one feels, desires, and intends, not present in the folk model, there would be no mechanism of self-control in the system, and hence we would have no basis for concepts of responsibility, morality, or conscience. Even though the normal situation is one in which a person can, through thought, intervene between the wish and the intention so that self-control is possible, according to the folk model there are abnormal situations in which either the wish is so strong or the capacity to think and understand what one is doing is so diminished (perhaps because of drugs, fatigue, strong feelings, etc.) that self-control cannot be expected. Since what one desires and intends are things about which one has a belief or thought, a thought potentially attached to some desire or intention can trigger that desire or intention. If a set of circumstances lead one to realize that one has a good chance of winning a million dollars, one may suddenly discover that one very strongly desires a million dollars.

A FOLK MODEL OF THE MIND

S127

Here, the causal relation is of a special kind. Thoughts are not considered to have the power of creating desires or intentions out of nothing, only the potential of "triggering" off a preexisting desire or intention-(Searie 1980). The chance of winning a million dollars could riot set off a great desire for money if one really did not care about money. ";;,?.; The difference between "creating*' and "triggering" appears to center on the contrast between making something'that did not exist versus activating something that is already present. The difference is not always clearly marked in the folk model: Sometimes emotions, for example, are treated as things "triggered" by experience, and at other times as things "created" by experience. The difference seems to depend on how the person's natural state is characterized: a tiny annoyance "sets off" the anger of people known to be irritable, although it might take an outrageous event to "make" a mild-tempered person angry. In sum, in the folk model, the cognitive processes of thinking, understanding, inferring, judging, and so on have extensive feedback relations with all the other kinds of internal states. By itself, the thinking process is considered to have only a small amount of power; but as the process by which beliefs are formed, and as the process through which different internal states interact, thoughts play a central role in the operation of the mind. According to the folk model, if the process of thinking or the capacity to think is badly disturbed, persons cannot be held accountable for their actions - they do not know what they are doing. This central role of thought also has the consequence that mental illness in the folk model is considered to be primarily a loss of cognitive capacity (C. Barlow, unpublished data); OTHER ASPECTS OF THE MIND

The description just presented does not cover all of the material included in the western folk model of the mind. No analysis has been given, for example, of kinds of ability, such as intelligence, creativity, and perceptiveness, or kinds of strengths, such as will power and stability. (A good start on the analysis of these aspects of the mind is presented in Heider's Psychology of Interpersonal Relations 1958.) What is attempted here is the description of the most basic elements of the model, elements needed before further analysis can be carried out. Thus, the concept of intelligence for example, assumes that the mind includes a process of thinking, and that people vary in the degree to which they can apply this process to certain kinds of problems to arrive at solutions. However, the specific ideas about intelligence held by Americans go considerably beyond the material presented here. Sternberg, et al. (1981), for example, studied folk concepts of intelligence and found that Americans distinguish three major kinds of intelligence, which might be glossed "knowledge about things," "problem-solving ability," and "social intelligence."

128

. ROY D'ANDRADE

Summary of major propositions and interview material

• /

1. Perceiving, thinking, feeling, wishing, and intending are distinctmental processes. .//-••'. : K ;( The best evidence for this proposition is the existence of the semanticaHy different verba! terms for these internal states arid processes. SOme of the semantic features of these terms are given in Table 5.1. 2. One is usually conscious of what one perceives, thinks, feels, wishes, and intends to do. However, many internal states and processes are indistinct and hard to delimit. Q. Could it be the case that someone sees something and isn't aware of what they see? A. Yes. You might see a situation and you think it is one thing and it is really something else. Q. Can you see something and not be aware that you're seeing anything at all? A. You'd better say it again. You lost me. Q. Can you see something and not be aware that you saw it at all? A. I don't know how. Q. Could someone think something and not be aware they thought it? A. Yeah. Q. How could that happen? A. Because your mind is so cluttered with all kinds of things. I'm not aware of half the stuff I think or things that are embedded in there. They sometimes come up and bother me later and I have to sit there and think about it and try to sort out what's the matter, why I can't do something. Q. Could you think something was true, believe it, but not know that you believed it? A. No, that sounds silly. Sorry. Q. Could you have a real feeling or emotion about something and not be aware you have that feeling? A. Yes. Q. Could you be angry at somebody and not know it? A. Yes. But it might come up later and you would realize it. Q. Could you be sad and not know it? A. You could be any kind of feeling and not know it. Q. Is that the way it usually works? A. No. Usually you know how you feel. At least a little. Q. Could you wish for something, desire something, and not know you wished for it? A. Yes, that is definitely true. Q. Can you give me an example of how that would work? A. Well, let's say I want to play really well in a concert, but it is so deep down that I don't know I want to play really well, but in fact that gets in my way, that wanting to play really well. I just don't let myself play naturally.

129

A FOLK MODEL OF THE MIND

Q. Could someone intend to do something and not be aware they intend to doit? A.

I think so.

• • - . - ; • ; .••••: •;

Q. You sound a little hesitant. Could you intend to go to France and not realize it? ';-•,.•'''-.••• W^'J";. •;<;,.. •'.'/•y: ( ',.,. A. No, not something concrete like that. .f Q. How about intending to get married to someone but you don't know it. A. No, that sounds silly. Maybe you 6ould have a very general intention like intending to do well and not know it. But that wbpidbe just like wanting to do well. Not something specific., A1: ^. ., Q. How come you can have specific feelings and not know you have them, but you can't have specific intentions and not know you have them? A. I don't know. 3. The process of thinking is controlled by the self in much the same way one controls any action. Q. Suppose somebody named John can't keep his mind on his homework. What might account for such a situation? A. He's got his mind on something else probably. Q. Why might he have his mind on something else? A. Because the something else is more appealing or more important at the time. Q. What can he do about it? A. Well, he could either go do something about the thing he's worried about or thinking about and do his homework, or he could force himself to get it out of his mind and then do his homework. Q. How do you force something out of your mind? A. You have to relax because you can't do anything about the other situation right then. You just have to relax and put your mind to what you are doing. Q, What does he have to do to put his mind to what he is doing? A. You have to focus it, you have to look at what you're doing, you have to be completely absorbed in what you're doing. You can't be floating around somewhere else. You can't be sitting apart and watch what you are doing, you have to do it. 4. The process of perception is not controlled by the self except in so far as one can direct one's attention toward or away from something. Q, If you don't like something you see, or something that you hear, like loud music, or you don't like what you're tasting, what can you do about it? A. You can either ignore it or try to change what you don't like. Q. If you were tasting something and didn't like the taste, could you just make it not taste so bad by will power? A. No, I don't think you could. I mean if it tastes bad, it just does. You either spit it out or you swallow it. Q. How about hypnosis? Could somebody hypnotize you so you would think "Oh, this tastes great." A. Yes, you could.

,130 ,•

ROY D'ANDRADE

Q. How does that work? A. I don't know how hypnosis works. Sorry. - •• 5. The process of feeling some emotion about something or desiring i
A FOLK MODEL OF THE MIND

:131

Yeah, I could think all the prime numbers in red. ": ; :•'::'£ Can you blend ideas about things? : ..; : , > .:.; 70 , What do you mean? , •••v/i; ,'••:,.•• „••,•.;-> ,:••/'.'••..;.; ::-~r A:'-'-' ' Well, you gave an example of prime numbers which are,red, right? Put them together in a picture. But could you do it just with thoughts? A. No. It would be a mix up. ,; ; ia>;\ :s A. Q. A. Q.

..J

8. In English, the self is typically treated as the object or experiencer of emotions (and also physicalsensations). T h e otli^rhientai processes typically treat the self as the subject or agent wh6'does the process, but, except in the case of intentions, it i i s ' p l ^ b t e T o r M e ' s e l f tb'^e the object of all the mental prpcesses. ; \; '"•';; :;''" v Q. I'm going to read some sentences and I want to know how they sound to you - tell me which ones sound normal and which ones do not. O.K.? A. O.K. Q. "John is often threatened by his feelings." A. Normal. Q. "John is often threatened by his thoughts." A. Normal. Q. "John is often threatened by his wishes." A. Yeah. Q. You sound a little h e s i t a n t . . . . A. Yeah, I was hesitating. Because I guess I think of wishes as desires and if you had said "desires," I would have said "yes" right away. Q. "John is often threatened by his intentions." A. That doesn't sound right. I can't make it click. 9. Most things that people do - outside of reflex actions like sneezing they do because of some intention or goal they have in mind. Q. When somebody does something, do they usually have an intention in mind? A. Yes. Q. Are there some things that people do that they don't have any intention in mind when they do them? A. Yeah, like sneezing or your heart beating; it just goes on. Q. Like buying a car? A. No. 10. Why does someone have certain intentions rather than others? One reason is that some intention is a subgoal considered necessary to reach another, more general goal. Another reason is that one wants or desires something, and that is why one intends to do something - to get what one wants. Q. Suppose John intends to buy a horse. What might be some explanations for that? A. He could want a horse, to ride a horse, or might want it for his farm for a work horse. Or he might want it for his kids.

11. Not every desire or wish gives rise to action, or the intention to do something. However, if one has an opportunity to do something, and

<' 132

ROY D'ANDRADE

there is nothing preventing one from doing it tike a conflicting desire or an outside force, and one does not even form an intention to try to do it, then one does not really desire it. -;•' _ Q. John says he wants to see Key Largo. He had a chance to go, but he *• ' didn't take it, although he didn't have any reason not to go. What could explain such a situation? A. I can understand that. I do it all the time. Q. What could explain such a situation? ,-]•-.,A. You just get obstinate. Even though you want to do something really badly, its like there's this part of you that thinks, "I don't' want to do it." Sort of like a mule; it justs sits there and doesn't want to go and fights you - I guess your intentions. ' ' Q. O.K., in that case some part of John didn't want to go. But if there wasn't a counterwish, could it be the case that he just didn't go even though he wanted to? A. That's like a contradiction. Because that doesn't make too much sense. There would have to be a reason why the person didn't do it if they wanted to do it. There'd have to be some reason like that or just a simple reason like they couldn't do it. It wouldn't be that they just wouldn't do it. 12. One often does things one does not wish to do because one has to, or because it is right, or because other people want one to, or because one is paid. In such cases, one wish prevails over another wish - the wish to stay alive, or be a good person for example. One does something one does not wish to do because there is something else one wishes for even more strongly. Q. Last night, John said he didn't want to study, but he did. What could explain such a situation? A. He probably had to. He probably had classes and things to do. I mean, nobody likes to study. So he made himself - he disciplined himself and did it. It had to be done. Q. O.K. So he doesn't want to study because that's work, but he wants to study to do something - to pass the course or something. So he has opposing wishes? A. Exactly. Q. Why. did one wish win over the other? A. I guess because it was stronger for him. 13. Sometimes - but rarely - one does something without knowing why. T h a t is, one does n o t k n o w what desire o r wish leads t o the action. Q. John stole Bill's socks. Now he says he doesn't know why he did it. Could John be telling the truth? A. That's an old fine. They're trying to get out of it. They know why they did it deep inside and they are trying to hide from it. Q. You think they really know? A. They probably have to really dig to find out, Q. So they might not be really aware of it when they say it? A. They're not really aware. Maybe they really believe they don't know why they did it.

A FOLK MODEL OF THE MIND

133

14. Is every nonreflex act the result of some wish or desire? Probably, but not surely. Q. Can one just do something for no reason at all - nothing intended or wanted? • -•.„•-•' •>•••': .•/•"• --,:•• A. Really no reason at all? I'd say there should be some reason somewhere. Otherwise, it's silly.1 Q. Could the reason be trivial? A. Could be trivial, could be anything. But there should be a reason, 15. Why does someone have certain desires rather than others? Some desires are for things that are needed in order to get something else one desires. Some things are desired because they make brie feel good, or one likes them, or they are pleasurable. Some things are desired because one is in some emotional state such as anger or love. Some things are desired because one thinks doing those things is right. Q. Why do people want things? A. They enjoy it, it gives them pleasure. Q. What are some other reasons? A. Some sort of honor they would receive. Something that makes them good either in their own eyes or makes them feel they're better in other people's eyes. Q. Could one be in love and not wish to do anything about jt? Not have it give rise to any kind of wish? A. Not in my movie. Q. Could you be angry and not have it give rise to some wish to do something? A. I guess not. Q. Could one be afraid and not wish to do anything? A. If you're afraid, you might just want to stay still and be safe and you wouldn't want to do anything. Q. But then you are trying to be safe, you want to be safe. A. Yes, so that's wanting something. Q. Could you be sad and not want to do anything? A. Yes. You're just all despondent. Just sitting there. I guess that is sort of doing nothing. 16. Most feelings are either pleasant or unpleasant. (Most events give rise to some feelings - so most events are either pleasant or unpleasant.) Q. Do people have feelings which are neutral - neither pleasant nor unpleasant? A. No. Q. Can you always tell if a feeling is either pleasant or unpleasant? A. Not at first. Sometimes it's unpleasant at first and then it changes. 17. Feelings and emotions are primarily reactions based on one's understanding of events. But sometimes there is a lack of fit between one's understanding and what one feels - either the amount of feeling is disproportiona! to the experienced event, or the kind of feeling is incongruous with the nature of the event. Q. What are some things that might make a person feel sad? sriVsTit Oak® Libra*?

; 134

ROY D'ANDRADE

;' . 1 A. Somebody dies. Or .you forget really important things you believe in, and suddenly it comes back to you, it can make you sad because you forgot it and you separated yourself from it. Q. What about anger? ,-:.• ,v.;ii A.\Frustrating kinds of things that you can't do anything about, like work or your boss is always picking on you. Q. What about fright? -• A. Well, anything can make you afraid. I mean, just a scary movie or ..,,,-.; A ., .something like that. ;.;.;- :_ Q. Could you feel sad even though nothing happened? .to:»;.-A. Yes..-, . . . . . . . . . . ,..] Q. Could you feel angry even though nothing happened? ".T"" TA.' No. ......•• Q. Could you feel happy even if nothing happened? A. Sometimes I read something and I'm happy, or I think about something that makes me happy. Does that count as something happening? Q. Yes. A. Well, then "no" for all of them. You can't just sit there and have a feeling. Q. Could someone feel sad if only a minor thing happened, like seeing a child drop a piece of candy? A. Sure. 18. What one believes and knows influences how one perceives the world. Q. Two people watch an argument between a policeman and a taxi driver. One of the watchers says it was almost a fight. The other onlooker says it wasn't serious at all. How could you explain this difference? A. They have different ideas about what serious is. Q. Suppose they both mean by serious that there was almost a real fight? A. Well, if it was obvious one way or the other, I don't know. That's like disagreeing on whether something is blue or red. Q. Well, suppose it wasn't that obvious? A. Well, maybe one of the watchers knew the taxi driver, and the other didn't. 19. One can affect one's feelings just by thinking about certain things rather than other things. However, the degree of influence here is weak. Q. If one wants to change one's feelings, say if one feels sad and wants to feel more cheerful, what can one do? A. If you're sad and you want to feel cheerful, you can go out and do something constructive or active or something you would feel cheerful about. Q. Could you just think about something and make yourself feel more cheerful? A. Yes. Q. Does that always work? A. No, sometimes it does. Q. How come it doesn't always work?

A FOLK MODEL OF THE MIND

135.

A. Because maybe your sad thing is too hard to get out of your mind by just thinking about something else. 20. What one feels also, influences how one thinks. Feelings may sometimes stimulate one to think in certain directions, or block thinking about certain things, or even completely wipe but the ability to think. Q. If you felt very angry, or very frightened, could it affect how you think? -•••: A. Yes';..' '.•i.:v;..=o.iJ;j.:,v.!-^.-.-.=-,,v^^.*.!:" • & ••• • Q. Would it make your thinking better or worse or what? A. Worse. It could affect how you think about a person for the worse so you just see one thing about the person, like if you are very angry. You don't even want to think about the good parts of them. Q. Is everyone the same about this? "';'- : A. I don't know. 21. Sometimes, what one thinks and what one feels fuse together into a single response, as in approving of something, or wondering about something. Q. Can someone approve of something, yet not have any feelings about it? A. No. If they approve, they approve, and that's a feeling. • Q. Could they approve of something and not have any thoughts or opinions about it? A. No, if they approve, they approve. Approve is an opinion and a thought. 22. What one believes is strongly influenced by what one perceives. One believes that what one perceives to have happened actually happened unless there are special reasons to think one is hallucinating, or led by ambiguity to imagine things. Q. John thinks that UFOs visit Del Mar, because he said he saw one land at the racetrack. What could account for John's opinion? A- He has an eye problem or he has a big imagination or maybe he really saw one. Q. Would it surprise you to know that John was a strong believer in UFOs even before he saw one land at the racetrack? A. No. He probably looks at UFO pictures in magazines and then thinks he sees one in real life. It could happen. Q. What could happen? A. You could imagine it. You could have an image so strong in your mind that you see maybe a plane or just a flash in the sky and suddenly your mind just inserts the whole picture there. That happens to me. When you have something strong, you can see just part of it and your mind sees the whole thing right there. 23. Thoughts are related to each other. Sometimes, one thought leads to another; sometimes one recognizes inconsistency between thoughts; sometimes one can figure out something from other things one knows or believes. Q. Sometimes someone says they didn't know something at first, but then they figured it out. What do they do when they "figure out" something? A. That's a hard question. They go over a problem in their mind, and

136

ROY D'ANDRADE

. .

somewhere there is something that will click. They go over it in their mind, and there's a bunch of little things over here that are just' maybe unconnected. And they see the connection. I can't explain it. , > - 24. In order to wish for something, or desire something, or intend to do ,.:,,something, one must be able to conceive of that something.; ,\ Q. Could a goldfish wish to discover the theory of relativity? A. I don't know. I doubt it. Because a goldfish isn't developed to the point where they could think thoughts like that. . Q. Is everything you wish for something you can think of? . . •

A . Yes.

Q. Could you wish for something you couldn't think of? A. It depends on what you mean by "think of." Maybe you could wish for something you couldn't remember very well. You can't wish for something you can't think about. 25. Thinking about something can trigger a wish or desire if the wish or desire is already there - either one already knew that one had the desire, or one realizes after thinking about it that one has the desire. Q. If you just think about eating something good, could it make you want to eat something even if you weren't really hungry? A. No, not if you really weren't hungry. But you might stimulate yourself by thinking about something if your were just a little bit hungry to really want to eat a certain thing. 26. Since one is usually aware of what one desires and what one intends to do, one can think about one's desires and intentions, plan things, change one's mind, select the better-rather than the worse course of action, and in general control one's self. Q. How come people have the ability to control themselves, at least some of the time? A. The brain sends a message to the body, like to your finger, and it moves. I don't know how. Q. How about self-control, like controlling oneself when one is on a diet. How does somebody keep from having ice cream for dessert? A. How can I keep myself from having ice cream tonight? I tell myself - my brain told my other brain that I didn't want it. I mean, I wanted to be thin more than I wanted the taste of ice cream in my mouth. Q. So it's like you spoke to yourself? A. Yes. My bad half was held in by my good half. 27. If one can't think clearly for any reason, one cannot control one's self very well, and one is not fully responsible for what one does. Q. What could account for the fact that there are some people who don't seem to be able to control themselves, even when they want to? A. They have psychological problems. Q. What does that mean? A. That means that there is something bothering them, I think. They are all mixed up. They have problems. Q. Could you expect someone to control themselves if they couldn't think clearly?

A FOLK MODEL OF THE MIND

137

A. No, not really, If you didn't know what was happening and you didn't know what you were doing, there would be no way to get back. Q. Should a person like that be punished if they did something wrong? A. No, it's not their fault if they didn't know what was happening:; The interview data collected so far support the major prbpositioiis'; presented here for the folk model of the mind. It should be understood ' that these propositions are a theory, not a simple description, of what. Americans - and probably most Europeans -believeabout. the mind.The usefulness and validity of such a theory will not be established on tjie basis. of one person's interviews of several informants, but rather on the results, obtained across a range of investigators, informants, and kinds Of data. ' Some idea about the historical depth of this folk model can be obtained from earlier novels and plays. Even though writers of novels and plays do not usually state the propositions of the folk model of the mind explicitly, they do use the model in constructing character and plot, and they sometimes comment on the reactions of their characters to events in very revealing ways. For example, in Emma, a novel by Jane Austen published in 1816, there is a description of Emma's and Emma's father's reaction to the recent marriage of Miss Taylor, who had been Emma's governess and companion (1969:17). She [Emma] had many acquaintances in the place, for her father was universally civil, but not one of them who could be accepted in lieu of Miss Taylor for even half a day. It was a melancholy change; and Emma could not but sigh over it, and wish for impossible things, till her father awoke, and made it necessary to be cheerful. The tacitly understood propositions here seem to be that "melancholy" is a natural reaction of the experience of loss, and that "sighing" is a natural expression of such a feeling, and further, that the experience of loss and the resulting sadness create a "wish" for something that will remove the sadness, along with thoughts about this "something." Austen (ibid.: 17} continues: His spirits required support. He was a nervous man, easily depressed; fond of everybody that he was used to, and hating to part with them; hating change of every kind. Matrimony, as the origin of change, was always disagreeable; and he was by no means yet reconciled to his own daughter's [Emma's sister] marrying, nor could he ever speak of her but with compassion, though it had been entirely a match of affection, when he was now obligated to part with Miss Taylor too; and from his habits of gentle selfishness and of being never able to suppose that other people could feel differently from himself, he was very much disposed to think Miss Taylor had done as sad thing for herself as for them, and would have been a great deal happier if she had spent all the rest of her life at Hartfield. Emma smiled and chatted as cheerfully as she could to keep him from such thoughts; but when tea came, it was impossible for him

138

ROY D'ANDRADE

not to say exactly as he had said at Dinner: 'Poor Miss Taylor! I wish ^he were here again. What a pity it is that Mr. Weston ever thought of her!' , Emma's father is also subject to the same emotional reaction'to the loss of Miss Taylor, but Austen treats him as a person who is emotionally 'predisposed to such reactions, so that Miss Taylor's marriage easily "triggers" his response. Because Emma knows her father is like this, she acts cheerful. We "fill in" the needed connections - Emma does not want her father to be unhappy, and believes (or at least hopes) that being "cheerful" will, by creating a happy environment for him, keep away his depression and anxiety, and so this wish of Emma's results in her intentionally acting in a cheerful manner. We also understand that Emma has the strength to keep to her intention despite her own sadness. Emma's father, on the other hand, lacks strength of character. His feelings and desires influence his thoughts inappropriately; his selfcenteredness leads him to think that other people feel the same about events as he does - even when this is obviously not the case - and his feelings and confused understanding lead him to think of his daughter's and Miss Taylor's marriages as unfortunate events even for them. Desires and emotions can, according to the model, influence belief, but they should not. A "strong" person does not let feelings and wishes distort reality, but a weak person is liable to. Overall, reading Jane Austen and other early English novelists, one is impressed with how little obvious change there is in the folk model of the mind in the past 200 years. But at much greater time depths, the implicit connections that knit together actions and reactions in stories are harder to discern, and it is difficult to tell if the difficulty lies in translation, or in a failure to appreciate the cultural understandings about the meaning of events, or in a change in the model of how the mind works (see, for example, the discussion of Achilles in Friedrich 1977). Another, more modern example of the use of the folk model of the mind: a 7-year-old child and her mother had the following conversation: Mother: Rachel: Mother: Rachel:

Rachel, you're making me mad! I didn't mean to make you mad. Well, you sure seem to be trying. But I didn't mean to. If I didn't mean to, how could I be trying?

Here, Rachel uses the connection in the folk model between intentions and actions. "Trying" is an action undertaken to bring about a particular intention - what one "means to do." Therefore, if there was no intention on Rachel's part to make her mother mad, by definition she could not have been "trying" to make her mother mad. (This example also illustrates nicely the ability of people - even young people - to reason effectively when using a well-understood cultural model. For a nonwestern example, see Hutchins 1980.)

A FOLK MODEL OF THE MIND

the folk model and science

139 ^ :c

It is not possible to contrast the folk model presented here with a single scientific model, since there is no one theory of the mind held by all psy-: chologists. There are, however, certain general trends within acaderhii; psychology with which the folk model can be compared. Based on aft ex-: amination of several popular undergraduate psychology texts, it seems that the current academic vocabulary is a blend of folk terms plus the addition of specialized terms. The typical text contains chapters on visidiy audition, taste and touch, cognition and memory, learning, motivation;" emotion, intelligence, personality, and mental disorders. The material on vision, audition, taste, and touch is heavily physiological, although various kinds of illusions are discussed in which conscious experience, is contradicted by physical facts. One major disagreement between the folk model and the academic model involves "motivation," Although the terra motivation has its roots in the folk model, it has come to have a specialized meaning in psychology. Motivation, unlike emotions, desires, and intentions, does not refer primarily to a phenomenological state or process - that is, it is not something primarily defined by the conscious experience of the person. Instead, motivation refers to a condition of deprivation or arousal of the "organism" that is only variably correlated with phenomenological experience. High motivation is likely to result in a person's thinking about the objects that would "satisfy" or "reduce" the motivation, emotional arousal (not necessarily of any specific kind), the experience of desire to do various actions that have led in the past to satisfaction, the formation of relevant intentions, and the carrying out of such actions if given the opportunity. Most psychologists consider motivation to be a real rather than hypothetical state of the person, but not a state that the person is necessarily aware of. The conscious mental states caused by motivational arousal may have some function in directing the final action the person takes, but these conscious mental states are typically considered to be neither necessary nor sufficient conditions for motivational arousal. The psychoanalytic theorists are also greatly concerned with motivation. Psychoanalytic theorists place more emphasis on motivational conflicts than do academic psychologists and are more interested in how the motivational situation influences thought and feeling through repression, isolation, displacement, denial, sublimation, and other mechanisms of defense. Psychoanalytic theory also differs from the folk theory in that it emphasizes unconscious states. The folk model allows that it is possible for someone to desire something or have some feeling of some kind but not know it, but such conditions are considered atypical. Psychoanalytic theory also distinguishes between two forms of thought - primary process thought and secondary process thought - but the folk model makes no such distinction.

140

ROY D'ANDRADE

Even though both the academic and psychoanalytic models modify the folk model, it is clear that these are modifications of an already existing conception of the mind. The general tenor of the academic model is. to place emphasis on what can be described physically - hours of deprivat i o n , the neural pathways, peripheral responses and so on - with the hope that the mental states and processes of the folk model will eventually be. reduced to a physical science vocabulary and simply ignore those.parts of the folk model that cannot now be physically described. For example, until recently, there was a complete avoidance in modern psychology pf the term consciousness - a process that is difficult to handle within a physical science model. In the past decade, this has begun to change. Sperry (1982:1225), for example, states: . . . one of the most important indirect results of the split-brain work is a revised concept of the nature of consciousness and its fundamental relation to brain processing. The key development is a switch from prior noncausal, parallelist views to a new causal, or 'interactionist' interpretation that ascribes to inner experience an integral causal control role in brain function and behavior. . . . The events of inner experience, as emergent properties of brain processes, become themselves explanatory causal constructs in their own right, interacting at their own level with their own laws and dynamics. The whole world of inner experience (the world of the humanities), long rejected by 20th century scientific materialism, thus becomes recognized and included within the domain of science. Sperry's position does not appear to be the majority position of research psychologists, who continue to carry the hope that the folk model eventually can be completely physicalized without the use of "emergent properties." However, with the rise of modern cognitive psychology, much greater attention has been given to the problem of consciousness, its function, and physical bases (Mandler 1982; Natsoulas 1978). The situation is quite different with regard to the psychoanalytic model, which considers consciousness, intentions, and the self as things of interest in their own right. However, the conscious mental states and processes are considered to be only a small part of the picture - and not the part where the main action is. Despite the shifts in psychoanalytic thinking from its early days, it has not changed in considering unconscious states and processes to be the center of the causal system. Thus, even though the academic and psychoanalytic models have their origins in the folk model, both are deeply at variance with the folk model. That is, the folk model treats the conscious mental states as having central causal powers. In the folk model, one does what one does primarily because of what one consciously feels and thinks. The causal center for the academic model is in the various physical states of the organism - in tissue needs, external stimuli, or neural activation. For the psychoanalytic model, the causal center is in unconscious mental states. Given these dif-

141

A FOLK MODEL OF THE MIND

ferences in the location of the casual center of the operations of the mind, the three models are likely to continue to diverge. . ; -v1:.,. The west versus Ifaluk

J(

?V

The American-European folk model also contrasts with the folk models recorded by anthropologists for nonwestern peoples. Recently, Catherine Lutz presented a summary of the ethnopsychological knowledge system of the people of Ifaluk (Lutz 1980; 1982; 1983; 1985; see iaIso4his volume). Ifaluk is a small atoll, only one-half square mile in area, located in the Western Caroline Islands of Micronesia. The island was previously studied by Burrows and Spiro (1963). The present population is 430 persons. Most of the islanders are monolingual speakers of aMalayo-Polynesian language. The culture of this small society is distinctive for its strong values on nonaggression, cooperation, and sharing. The folk model used on Ifaluk contrasts with the model presented here for American-European - or "western" - culture in a variety of ways. However, the general framework of both models is similar. In both models, there seems to be a similar division of internal states into thoughts, feelings, and desires. In the model used on Ifaluk, there is a distinct class of emotion terms, for which a general correspondence to English emotion terms can be found, although the particular blends of affective tone differ from what we find in English.-For example, the term fago refers to feelings of "compassion," "love," and "sadness"; and although it involves caring about someone, it is also judged by native informants to be semantically similar to words involving loneliness and loss (Lutz 1982). A similar affective blend is found in Samoan for the cognate term alofa (Gerber 1975). This particular blend is different from the American English term love and its cognates, which do not prototypically involve sadness and loss (but note the sadness of many love songs and stories). Even though there appears to be an overall similarity between the models in the division of mental states and processes into thoughts, feelings, and wishes, on Ifaluk the distinctions are made much less sharply. The term nunuwan, one of the two major terms used to describe mental states (niferash, "our insides"), refers to "mental events ranging from what we consider thought to what we consider emotion" (Lutz 1985:47). The meaning of nunuwan appears to be somewhat like the special meaning of English of the word feel when used in the sense of "to think," as in, "I feel it is likely we will succeed." (As mentioned, several terms in English also blend thought and feeling, such as approval and doubt.) The other primary term used on Ifaluk to describe internal states is tipwhich Lutz translates "will/emotion/desire." When asked the difference between nunuwan and tip-, people say that the two are very similar. The distinction is that tip- has connotations of desire and movement toward

142

ROY D'ANDRADE

the object: An informant said "Our tip- is what we want, like to chat with someone or to go visit another village" (Lutz 1985:48). It appears that Zip-: always takes a propositional object, unlike nunuwan. However, like nunuwan, emotion is held to be inherent in the experience of tip-: It is likely . ,that intentions are also included within the semantic range of tip-, since there appears to be no separate term for intentions as part of ."our. insides." In general, it would appear that the people of Ifaluk regard emotional experience as a central feature of the mind and emphasize the affective elements in the experience of both thinking and wishing. Lutz has traced out how the values of nonaggression, cooperation, and sharing are sup-. ported by the various conceptions of emotion. For example, One term, metagu, glossed "fear/anxiety," which is the feeling that occurs when one must be in the midst of a large group of people, or when one encounters a ghost or a shark, or when someone is justifiably angry with one, is considered a necessary part of socialization. A person who does not experience metagu is like a "shameless" person in English - that is, someone who will not have the proper constraints on his or her behavior. A child who does not experience metagu is considered to lack a primary inhibitor of misbehavior, and such a deficiency would indicate that parents failed to socialize the child properly - to display song, "justifiable anger" at the child's misdeeds, which is thought inevitably to elicit metagu in the person to whom the anger is directed (Lutz 1983). The people of Ifaluk considered feelings to be natural responses to particular events, typically interpersonal situations of various kinds. Such eliciting events are considered a basic part of the definition of the emotion (Lutz 1982). Emotions are also thought to give rise to particular behavior;/ago, for example, is thought to give rise to talking kindly, giving food, and crying. In portraying emotions as natural reactions to experience and also as causes of behavior, the folk model of the people of Ifaluk is similar to the western model. However, the model used on Ifaluk appears to give more consideration to the dyadic aspect of emotion, where if A feels emotion X and expresses it, then these actions will cause B to feel emotion Y. Thus, if A feels song, B feels metagu, whereas if A feels tang (frustration/grief), B feels fago (Lutz 1982). The model used on Ifaluk also agrees with the western model in distinguishing between emotions and physical sensations. Lutz (1985:49) states: Other aspects of 'our insides,' and ones which are distinguished from both nunuwan and tip-, are the states of hunger (pechaiy), pain (metagi), and sexual sensations (mwegiligil). These latter states are considered to be universal and unlearned human proclivities. Although their occurrence can lead to thoughts and feelings, they are considered an entirely different class of events from the latter. The Ifaluk further distinguish between these three states of physical sensation and the corresponding desires or drive-like states that follow upon the sensations. These include

A FOLK MODEL OF THE MIND

:H3

•wanting food (or a particular food)' (mwan), 'wanting pain to end' (gar), and 'horniness' (pashua). In the western model, this distinction between the physical state and the mental state for hungers pain, and sex is not lexicalized nordoes it seem to be a distinction that most people make in ordinary discourse.- The model used on Ifaluk also differs from the present western model in considering the mind to be located primarily in the gut, which includes the stomach and abdominal region. Thus, thoughts, feelings, desires, hunger, pain, and sexual sensations are all experienced in the gut. .When people eat well, they say "Our insides are good," which means they have both good physical sensations and good emotions. Loss of appetite is typically regarded as a symptom of either physical or emotional distress. In extreme grief, people say "my gut is ripping," and others advise them not to "hate" their own "gut" (Lutz 1985). According to the model used on Ifaluk, unpleasant emotions that are not expressed may cause illness. Individuals are advised to "throw out" their feelings in order to avoid illness. At funerals, people are advised to "cry big" in order to avoid illness. Expressing one's feelings (except angry feelings) is considered a sign of maturity and social intelligence as well as a way of staying healthy. Further, one's bad feelings can make other people ill. This is especially likely in the case of a mother and infant. It is said, "It is like the baby knows the 'thoughts/emotions' of its mother and becomes nguch 'sick and tired/bored' of the mother" (Lutz 1985:55). This connection between emotionality and illness is also found in the western folk model: For example, it is thought people who are homesick or sad about the loss of a loved one sometimes "pine away," and that chronic anger can lead to a heart attack. The model used on Ifaluk, however, appears to make the connection between emotions and illness much more generally and explicitly, perhaps reinforced by the attribution of both physical and mental sensations to a location in the gut. The model used on Ifaluk, like the western model, gives a central role to "thought" in the control of behavior. The concept bush, "crazy, incompetent," which is considered the opposite of reply, "social intelligence," is widely used to refer to behavior that is deviant and appears to be due to a failure to perceive the nature of the situation correctly. Ail infants and children to about the age of 6 are considered bush. People we would label -QS psychotic are called bush; on Ifaluk this is manifested by their being unable to work and engaging in inexplicable behaviors, such as shouting or eating without table manners. Lutz reports the case of such a person whose "crazy" behavior consisted of saying repetitively "my knife, my lighter, my basket," etc. On Ifaluk sharing is strongly stressed as proper behavior, and the use of first person singular pronoun is felt to be rude in many contexts - and "crazy" in this one (Lutz 1985). The ability to think correctly, especially on the part of children, is con-

144

ROY D'ANDRADE

sidered to be influenced by instruction. Children are given lectures in which a rule of proper behavior is gone over quietly and repeatedly. Lutz (1985:61) states: . . . children are believed to obey when and because they listen and understand language; intention and knowledge become virtually synonymous in this system. It is assumed that correct behavior naturally and inevitably follows from understanding, which should follow from listening. Although the concept of independent will is not absent (this is represented in the \'_ concept of tip-,) the greatest stress is placed on the connections between language, listening, understanding, and correct behavior. Here, the connection between thought and desire found in the western model is reversed. In the western model, if one desires or intends to do what is good, then one must be able to conceive of what is good. In the model used on Ifaluk, if one can and does conceive of what is good, one must do what is good. However, there have been theologians in the western tradition who also argued that if one truly understood what was good, one would desire it. Based on indirect evidence, there appears to be another difference between the model used on Ifaluk and the western model. In his interviews with a psychotic man, Spiro found that his assistants became disgusted with this man's reports of his hallucinations, saying he "talk lie, only talk lie" (Spiro 1950). Based on these reactions, it seems likely that the notion that someone might really see and feel what is not actually there is not part of their model of the mind. Overall, however, the model used on Ifaluk and the western model seem to have similar frameworks. Thoughts, feelings, and desires are distinguished. Feelings are considered a natural response to experience, not under self-control, and also to have the power to move the person toward action. The emotions are distinguished from physical sensations. Understanding is required for appropriate behavior, and lack of understanding results in loss of control. On the other hand, there are significant differences between the two models. The one used on Ifaluk fuses thought and feeling with regard to the upper-level term nunuwan and apparently does not distinguish desire from intention. In this model, the gut is thought to be the site of feeling and thinking rather than the head. The emotion terms blend affects in somewhat different ways than the western model. The interpersonal role of emotion is more distinctly conceptualized than in the western model, as is the role of emotion in physical illness and the therapeutic use of catharsis. An understanding of hallucinatory experience may be absent from this model. Finally, understanding what is right is treated as a necessary and sufficient condition for doing what is right, rather than being treated as simply a necessary condition. Based on these two cases, it seems likely that the folk model of the

A FOLK MODEL OF THE MIND

-145

mind will turn out to be like the folk raodel for colors as described by Berlin and Kay (1969). That is, certain salient areas of the experiential field will be universally recognized, although the degree to which the total field is differentiated and the exact borders and boundaries between areas will vary cross-culturally. However, at this point no simple ordering of basic concepts like the ordering found for color terms has been found for the model of the mind. In some areas, the people of Ifaluk do not make . distinctions we do (e.g., the distinction between desire and intention), but in other areas they make more distinctions that we do (e.g.* they commonly distinguish between the physical sensations and the emotional desires concerning sex, hunger, and the cessation of pain, but this distinction is rarely made by us). r Speculations about cultural differences and similarities Logically, it might have been the case that the Ifalukan materials could not even be translated into the western model. Suppose they had an extremely different model of the mind, one that made none of the distinctions made in the western model. Since internal states and processes are private, how could we ever learn anything about their model? However, this is not what we find. The model used by the people of Ifaluk can be translated. How is this possible? If it were the case that an ethnographer could not learn the model, one would wonder how the children on Ifaluk could learn the model. This raises a more general question: If these models are models of private experience, how are they ever learned, either here or on Ifaluk? Even if everyone's private experience is highly similar, how can someone else's words be matched to anyone else's private experience? What in fact is the case is that neither model is only a model of private experience. Both models use similar external, public events as identifying marks in their definitions of internal states. Thus, thinking is like speech, and speech is public. What are thoughts? One can say that thoughts are like things one says to oneself, or images of what one sees with one's eyes. Feelings are like those sensations that do have public elicitors; we know how to tickle each other. Furthermore, as human beings, we have what appears to be an innate communication system for emotions, signalled by patterns of facial expression (Ekman 1971). Various autonomic responses are also available as public events for the definition of feelings. Feelings are typically aroused by relatively specific external events. To understand what wishes are, we have the public expression of requests and commands: Wanting is the feeling that gives rise to the child's saying "gimme, gimme." Intentions are related to such speech acts as promises and threats; that is, to the accomplishment of events to which one has given a commitment. The tight connection pointed out by Vendler (1972) between speech acts and internal states is not fortuitous; the thesis pre-

146

ROY D'ANDRADE

sented here is that speech acts are one of the major classes of public events used as identifying marks of internal states and processes. /'••-•. This cannot be the full answer to how we learn about internal processes, since even though types of speech acts and facial patterns may offer a **• means of identifying internal events, they do not account for our beliefs about the causal relations among these interna! events, such as our belief .that we can think what we want to but that we cannot make ourselves fee! what we want to, or our belief that desires influence intentions but not the reverse- One answer to this issue is to say that these are universal of experience. Once one has categories such as "feeling" and "thought," identified by their relationship to various public events, one cannot escape noticing that one cannot decide what to feel but one can decide what to think. Such a hypothesis has a ring of plausibility but seems completely untestable. Finally, one speculates about what generally might account for cultural differences in folk models of the mind. Perhaps differences in the social and interactional conditions of life give differentia! salience to some of the identifying public marks of internal states. The emphasis on emotional mental states in the model used on Ifaluk would seem to be related to the strong salience of such emotion-linked actions as aggression and sharing in daily life. However, such differences in salience would not explain why there are differences in the conceptualization of causal relations between various mental states, such as the notion that lecturing on what is good causes the hearer to understand what is good thereby causing the hearer to be well behaved. Nor would these differences in the salience of emotion linked actions explain why the people of Ifaluk believe the verbal expression of feelings, especially depressive feelings, keeps one from being made ill by those feelings. It seems likely that some part of this folk model, like most folk models, cannot be explained by variation in current social or ecological factors. Parts of most folk models are legacies from the past, and the information needed to discover whatever causes once operated to create these models is often not obtainable. Note I. Support for research reported in this paper was provided in part by a grant from the National Science Foundation (BNS 8005731). The author wishes to thank Paul Kay and Laurie Price for critical commentary on an earlier draft of this paper and Susan Lindner and Ronald Langacker for discussions concerning the semantics of mental states and processes. References Austen, I. 1969. Emma. London: Collins. (First published 1816.) Anscombe, G. 1963. Intention. Ithaca, N.Y.: Cornell University Press.

A FOLK MODEL OF THE MIND

-147

Berlin, B. and P. Kay 1969. Basic Color Terms: Their Universality and Evolution. Berkeley: University of California Press. , Burrows, E. and M. Spiro ,;• 1963. An Atoll Culture: Ethnography of Ifaluk in the Central Carolines. New Haven, Conn.: H.R.A.F, Press. Casson, R. 1983. Schemata in cognitive anthropology. Annual Review of Anthropology 12:429-462. D'Andrade, R. G. 1985. Character terms and cultural models. In Directions in Cognitive Anthropology, J. Dougherty, ed. Urbana: University of Illinois Press. Pp. 321-344. D'Andrade, R. G. and M. Wish 1985. Speech act theory in quantitative research on interpersonal behavior. Discourse Processes 8:229-259. Ekman, P. 1971. Universals and cultural differences in facial expressions of emotion. In Nebraska Symposium on Motivation Series, J. K. Cole, ed. Lincoln: University of Nebraska Press. Pp. 207-283. Fillmore, C. 1977. Topics in lexical semantics. In Current Issues in Linguistic Theory, R. W. Cole, ed. Bloomington: Indiana University Press. Pp. 76-138. Friedrich, P. 1977. Sanity and the myth of honor: The problem of Achilles. Ethos 5(3): 281-305. Gerber, E. 1975. The cultural patterning of emotions in Samoa. Unpublished Ph.D. dissertation. University of California, San Diego. Heider, F. •; 1958. The Psychology of Interpersonal Relations. New York: John Wiley and Sons. Hutchins, E. 1980. Culture and Inference. Cambridge, Mass.: Harvard University Press. n.d. An analysis of interpretations of on-going behavior. Unpublished manuscript, Department of Anthropology, University of California, San Diego. Kenny, A. 1963. Action, Emotion, and Will. London: Routledge and Kegan Paul. Lutz, C. 1980. Emotion words and emotional development on Ifaluk atoll. Unpublished Ph.D. dissertation. Harvard University. 1982. The domain of emotion words on Ifaluk. American Ethnologist 9:113-128. 1983. Parental goals, ethnopsychology, and the development of emotional meaning. Ethos II(4):246-262. 1985. Ethnopsychology compared to what? Explaining behavior and consciousness among the Ifaluk. In Person, Self, and Experience: Exploring Pacific Ethnopsychologies, G. White and J. Kirkpatrick, eds. Berkeley: University of California Press. Pp. 35-79. Mandler, G. 1982. Mind and Body. New York: W. W. Norton and Company. Miller, G. 1956. The magical number seven, plus or minus two: Some limits on our ability to process information. Psychological Review 63:81-97. Natsoulas, T. 1978. Consciousness. American Psychologist 33(10):906-914.

148

ROY D'ANDRADE

Quinn, N. • ' • r ": .•./ •• 1982. "Commitment" in American marriage: A cultural analysis. American Ethnologist 9(4):775-798, ^V'; Rumelhart, D. \/".--."."!''' ~4 1980. Schemata: The building blocks of cognition. In Theoretical Issues in Reading Comprehension; Perspectives from Cognitive Psychology, Linguistics, Artificial Intelligence and Education, R. Spiro, B. Bruce, and W. Brewer, eds. Hillsdale, N.J.: Lawrence Erlbaum Associates. Pp. 33-58. •-•. 1: Rumelhart, D. and D. Norman . \ 1981. Analogical processes in learning. In Cognitive Skills and Their Acquisition, J. Anderson, ed. Hillsdale, N.J.: Lawrence Erlbaum Associates. Pp. 335-359. Ryle, G. 1948. The Concept of Mind. London: Hutchinson House. Searle, J. 1975. A taxonomy of illocutionary acts. In Language, Mind, and Knowledge, Minnesota Studies in the Philosophy of Science, Vol. 7, K. Gunderson, ed. Minneapolis: University of Minneapolis Press. Pp. 344-369. 1980. The intentionality of intention and action. Cognitive Science 4:47-70. Spiro, M. A. 1950. A psychotic personality in the South Seas. Psychiatry 13(2):189-204. Sperry, R. 1982. Some effects of disconnecting the cerebral hemispheres. Science 217:1223-1226. Sternberg, R., B. Conway, J. Ketron, and M. Bernstein 1981. People's conceptions of intelligence. Journal of Personality and Social Psychology 41(l):37-55. Wallace, A. 1961. On being just complicated enough. Proceedings of the National Academy of Sciences 47:458-464. Vendler, Z. 1967. Linguistics in Philosophy. Ithaca, N.Y.: Cornell University Press. 1972. Res Cognitans: An Essay in Rational Psychology. Ithaca, N.Y.: Cornell University Press.

PART I I Reasoning and problem solving from presupposed worlds

AN AMERICAN PSYCHOLOGY OF PROBLEM SOLVING 1 t.-^p:'•:-. h:/

.h;h.<:.A.i

-X;

•\-;>;tf.!jV3

Proverbs_aiejenerally regarded as reposjtpiiej^OQjloyisiojn,, As stylized Mayings that presume to represent the coramonsensical in everyday life, they are a topic of special interest for this volume's focus on cultural models. The dictionary defines a proverb as "a short, pithy saying in frequent and widespread use, expressing a well-known truth or fact." Attention to just what "well-known truths" are, in fact, expressed by proverbs and how, cognitively and linguistically, they obtain their particular brand jjLmejmingjg^^ models thatjmderlie them. Proverbs are especially interesting because, like much of ordinary language, they accomplish both conceptual and pragmatic work (see Briggs 1985). On the one hand, proverbs offer succinct ("pithy?) descriptions of events. A f^m^iajre^pxessionsuch as*Tt gidyjajce^rje^a^ppletojeoa the barrel" brings a mmiber HJallent alSdlreU-known propositions about people and sociaTlife to beaxoj^j>arti<S^ soTTKislproverb provides an interpretation of specific actions o£ events lOelrMOlf^igej^^ more-than economical-descriptions?-Xheylare-eslelntiaT^concerjng^h rhoralitv. with evaluating and shajgngjgujsjgjaf action and thus arjnjjji-. quentiyused in contexts of legal and moral argumentation (see, for example, Messenger T959rSalam6hei 1976). In the proverb just quoted, the evaluative claim is explicit: one b^.app^lt^^^nii^d^OTi^pXHE^ g^iod^pj^s. In other sayings, evaluative implications may rest just beneath the surface^jucbT as "Ygujcan't judge a book by its jgv^.^n_thisl say-, ing. a prior evaluation (eitprgood or bad)is correctedJ)yjg;mmding therjUjstenerjoXjsiUjL^^ , reality. V?fiether explicit or implicit, the evaluative assertions expressed in proverbs lend them directive force as recommendations for a desired course of action. The saying about bad apples ruining good ones may imply that some action should be taken to spare the threatened good apples, even though the overt form of the saying is that of a simple description of a state of affairs. This form of "indirect directive" is typical of many_pjoy-

/ J

152

GEOFFREY M. WHITE

erbsJhat_overtly takejfo^orm of descriptions fautjhat have the effect of suggestions, recommendations, or c^dmmanoTs and thejike. ••';•?/ -; ^iiKgj^ccgceptual and pragmatic functions^ of proverbs are also handled routiner7l>j[c>Kh^ languageTand ,-raltllr^s^rolmTTheworld have a^iicognizaBe^Ia^^ InlpiiT^TIminrelOh'a^^ Bir^Hqrland partly in their coimmurucatiy&ejy^^ pressions of imlpoltaiit culpral knowledge, proyerb^c^mbme a cognitiye economy "of reasoning with pragmaticrfgr^aimej^^ people. To understand why proverbs rather.than.lesslorjoiulaic laliguagT51' are used, it is usefulJo. ask,,JWhat isihe.speakerir^ns^o^^tfcpr^^--erbs?' rather than simply 'What is he or she trying.to say?' Tot4o so, one rmghT examine the situationslhlvmch proverbs are invoked to have.spme"'"' jfoaal '^^;~sudhJ|jf^^]Etu^tal. Quarrels (Salamone 1976) or Yoruba child rearing'(Arev/a &Tlhmdes 1964), where focused observation might record repeated uses of proverbs in particular contexts. However, understanding the social uses of proverbs also requires knowing something about the interpretive work done by both speaker and listener. This paper is concerned primarily with the conceptual processes that underlie proverb meaning rather than with questions of social usage. The analysis is based on the assumption that certain key understandings make up a kind of kernel of proverb meaning, even though such meaning may be shifted or elaborated in particular contexts of use. The fact that proverbs represent generalized knowledge, applied to the interpretation of particular events, suggests that they may tell us something about enduring cultural models of experience. Dyer (1983) has noted that the abstract advice encoded in familiar sayings or "adages" plays an important role in understanding stories. Narrative comprehension frequently proceeds by using existing knowledge structures to process hew information and draw inferences about the social and moral implications of what is said; in other words, to get the point. A closer examination of proverbial understanding as a cognitive process, then, may illuminate the organization of global knowledge structures. Interlocutors comprehend proverb meaning through a process of inference that allows them to link the saying with prior understandings and to fill in unstated propositions. Even though this is so in much of natural discourse, proverbial sayings tend to be particularly figuratiy^jiartial, and indirecV Tojm^^ a proverbH5~uttered re^ quires going beyradjhjs uttera^ to draw apjprpj?naie^ So, for example, understanding abatement like "It only takes one bad apple to spoil the barrel" involves both a translation of metaphorical imagery as well as a cultural theory of moral .corruption. The fact that certain proverbs are frequently used suggests that they express key understandings about everyday life. If so, proverbs may pro-

PROVERBS AND CULTURAL MODELS

153

vide a source of insight into cultural models in particular areas of com-;'; mon experience. This paper pursues this idea by examining a set of; American English proverbs that may all be used in a similar way: to give advice to someone in a problematic situation, broadly conceived. By selecting sayings that can be used to counsel someone dealing with a.;$&$?'.;< sonal quandary, the analysis focuses on certain American understandings about persons, problems, and purposeful behavior. ••.,.: .. Proverbial understanding Before looking more closely at proverbs of problem solving, it is necessary to consider briefly the linguistic and conceptual processes that enter into proverb meaning. Certain aspects of the linguistic form of proverbial sayings mark them as distinctive from other types of ordinary language, lending them their particular aura of veracity. For example, by using verb forms not qualified or marked by number or tense, proverbs acquire a timeless, enduring quality, seemingly not subject to the vicissitudes of circumstances or change. And, by using such quantifiers as all, every, and no, proverbs do not allow exceptions or hedges. Thus, one finds "Time heals all wounds" rather than, say, "Time heals some wounds," which would hardly provide a comforting bit of advice. Allowing exceptions or hedges would deny proverbs their claim to universal validity. Some proverbs also draw on hyperbole as a device for underscoring the obvious, commonsensical quality of an assertion. Thus, we have "Rome wasn't built in a day" rather than, say, "Rome wasn't built insjx mojithslpi^ perbjpj^Sjfern .village' wasn't b u i l f ^ V". ^ ^ ^ ^ S g ^ s t ^ S r a c t e r i s t i c feature qf.PI8Xfir.toUg.their extensive useZ^r^^horicajlimagery to concj^pWailze ^dexpresssojal roessalgs~.

It is^jdficajU..thatrnost

pro^ms^ly^S^^V^^^]t^^'mra----

jx>§i£igjL(but there are.jBXcepfid^QucL.as "Where there's a will, there's jnvaj£'). Tne factthat most proverbs are constructed in this way suggests an important complementarity of function between the conceptual role of metaphor and the pragmatic uses of proverbs. If one views metaphor as a device for expressing abstract concepts in terms of other concepts more closely grounded in physical experience, then metaphorical imagery / would seem to be an excellent vehicle for proverbial sayings that seek to express propositions taken to be self-evident on the basis of shared experience and that can thus be used to give advice, make recommendations, and so forth. When seen in this light, proverbs appear as a special case of the more general process of metaphorical understanding. As described j by Lakoff and Johnson (1980:115): . . . metaphor pervades our normal conceptual system. Because so many of the concepts that are important to us are either abstract or not clearly delineated in our experience (the emotions, ideas, time, etc.), we need to

154

GEOFFREY M. WHITE

get a grasp on them by means of other concepts that we understand in clearer terms (spatial orientation, objects, etc.)-; This need leads to metaphorical definition in our conceptual system. .Xakoff and Johnson speak of metaphorical understanding as a way of interpreting abstract and loosely structured experiences by conceptualizing them in terms of other, more concrete and clearly formulated types of experience. Although neither type of experience is more "basic," the latter is more closely grounded in the physical realm of the body and environment. Carbonell and Minton (n.d.) and others have described metaphorical understanding as a process of common-sense reasoning. They suggest that simile, analogy, and metaphor are all based on the same type of cognitive process (analogical reasoning) used to interpret new situations in terms of other, previously encountered and understood situations. The .essential process in this type of reasoning is one of mapping aspects of a previously known and well-delineated ("source") domain to a newer and less well structured ("target") domain (see Collins & Gentner, and Lakoff & Kovecses, this volume). This model of metaphor may also be extended to the process of proverbial understanding. Asnoted, most proverbs assert their truths about social~and.mQ.ral matters by linking feaKire'soTl^Ql^ more mundane_domains with widely known and clearly defined concept^"lmIaTlments. v ln~^ proverblaiunderstanding that some published collections of proverbs organize their contents in terms of types of source domain, such as "animals," "natural environment," "food," "fishing," "travel," and the like (see Brown 1970; Schultz 1980).2 A key question in models of analogical reasoning is 'How are mappings between domains constructed?' or, 'How are the relevant cross-domain similarities identified?' The ultimate answers to these questions will have to draw from pragmatic and contextual information not yet dealt with in cognitive theories of metaphor. However, for many metaphors in frequent use, the mapping is well known and hence does not have to be reconstructed each time any metaphor is used. This notion of "frozen" metaphors applies well to proverbs, which are among the most formulaic and standardized types of metaphorical usage. The fact that people are readily abie to paraphrase proverbs out of context, to render their meanings in nonmetaphorical language without reference to particular denotata or instances of usage, strongly indicates the "frozen" quality of proverbial inference and the important role of prior cultural models in their interpretation. Noting the prepackaged association of abstract, social meanings with concrete metaphors gives only a partial picture of the process of reasoning underlying proverb meanings. Proverbs are also used to pick out and communicate salient aspects of a social situation in terms of prior knowl-

PROVERBS AND CULTURAL MODELS

•im-

edge about similar situations. As in the use of metaphor general!y,;;uncer^ tain or ambiguous events can thus be understood and evalpat^injjeri^:"^;--':';'.^ of existing models of social experience. However, unlike the process;^;,';>:• •-; •$ metaphorical understanding, most of the "action" in the process; pfjpro^. ;^ :^ ;:;: verbial understanding is concerned with drawing put ,behayiprai;^aftd|^:\.j--':;V evaluative implications, with distilling a particular interpretation of a situa-s -, tion, rather than with constructing a mapping to link source domain vvjth target domain. Thus, understanding what is meant by the assertion "The;; squeaky wheel gets the grease" does not concern interpreting the notionA. of "squeaky wheel" in terms of vocal assertiveness so much as making the inference that if such behavior leads to positive outcomes (getting j "grease"), it is worth pursuing. .: Proverbs function as effective communicative devices because they set up the listener to draw such practical inferences by expressing one or more: key propositions embedded in a cultural model with known entailments. By instantiating certain elements of an existing model, other, related propositions are invoked through inference. In this way, the proverb user is able to formulate and communicate a point of view without verbally articulating all of its elements. Behavioral directives need not be stated overtly since any listener with common sense will draw the appropriate conclusions, given the premises asserted and/or implied by the proverb. At the same time, its metaphorical form brings those conclusions into sharper focus by formulating them in a domain in which propositions and their behavioral entailments are more tightly and obviously connected. ; The interpretation of proverbs may be viewed as an interactive construction in which the speaker (1) perceives and evaluates a social situation in terms of an abstract cultural model, (2) articulates that point of. view in a proverb expressing one or more interlinked propositions, which is then (3) interpreted by the listener, who expands on those propositions by locating them in the relevant cultural model and drawing appropriate inferences. Just what inferences are drawn will depend on the context of use, the abstract propositions expressed by a particular proverb, and the cultural model(s) in which they participate. Insofar as proverbial inference follows from the instantiation of pieces of a knowledge structure, proverbs offer a window onto the organization of generalized models of experience. The analysis in the next section examines the _w.ay informants interpret proverbs pertaining to human action and problem soIving-JJy asking, 'What does one need to know or assume in order to interpret the meaning of a proverb?' one may begin to identify some of the key prop- ./ ositions and inferences that enter into the ethnopsychology of American problem solving. An ethnopsychology of problem solving My interest in proverbs began in the context of research prrramjnfinAense reasoning about personalXsocialrpsy^hologicaj)i^probjejns. In the course

156

GEOFFREY M. WHITE

of an earlier study of the ways different cultural groups explain and deal with adjustment problems (White 1982), we noticed that informants oc~; casionally used idiomatic and proverbial^sa^mgslucn as^pnieigaTs^all 'v/dunds^roiOT l£then occurred to us that t h ^ J ^ a j u ] ^ ^ sayings-that-ex-pressxuiturally constituted understandings about how-to: respond to problematic circumlitanceSrGuid^ decided.tQ^suryeyxhgJange of j-veJMcnown sayings that, in ourjuclgment, 'could be used as advic^TrTdealing with a p r o f i K ^ c s ^ ^ ^ n T W e then seTouf to~e~xanrine^ese^^ ^c^tualizationsjEhey^ejm^Qdy^ > --" TQie„nQtipri^f "gersona|^rob of ~j everyday quandary oT^avefsiTy that is of some social ,or psychological v. significance for the person or persons involved. Hence, cultural knowledge about such problems is general ratherJl^ijpemllcjltnp^tams to the nature of persons and jt]fi|I^rj^ Sgngs"such as• _"_Every cloud has a silver UningJ^are^vvidely known_and used because thexniayl)^^ andjituations, rather than to one speci^f^domain of experience. The understandings Ffiat uriderfie'sucfi proverbs make up scmieof^the^most^is^remtses c^f^rnerican ethribpsychology. These understandings are quite different from the sort of conceptualizations |tudie^(3^c^mtive psychologists doing research on problem sojvi3g,lnjparticularjon.^,ask.e^iro^^[its," where problems and solutions^are well specified in the form of winning gaimesTsolvTrlg^]^^^ Simoh"1972)."f he common-sense-reasonmg-abouXprjDblems expressed in proverbs is primarily concerned with r^rson-problem^ejalidm^tlierlhan problemfsolution algorithms. Assuch, they draw"6h a ricKTiody oTknowh ""edge about persons and social action. Sayings such as "The grass is always greener on the other side of the fence" represent conceptualizations of problem situations with an implicit agenda about how to evaluate and respond : to them. In probing the meanings of proverbs such as this, the following s analysis is led into a consideration of the ethnopsychological understand1 ings required to interpret them. METHODS

Assuming that certain proverbs pertain to the way people respond to everyday problems,.we^began-by^selecting-a-numbej- o£pjoverbs that could be used in rougjiJxthe;Sl^~el^ a personal Hilemma or quandary. ^jismg,this_^ge^eralcontext as a frame, we were able to select a„set of proverbs that ccojJ^e^onu^a^dln^BeT^ to draw out common or contrasting tfreniesfin cuItuxalJcnawJ^dg&abput ~ pji^oniland-prob'lerns^ ~~ -• The approach taken here draws on both elicited data obtained from informants as well as the investigator's (and reader's) intuitive knowledge

PROVERBS AND CULTURAL MODELS

:

&

•

•

of proverb meanings. This strategy combines several types of formal and V nonformal data that are constrained in different ways and that thus shedL; light on proverb meanings from different angles/ : ;;\, 'o^^-V; Although proverbs may seem to be simple, direct, and obvious when ^usejHTOT when considered in light of the unspokjnj^^ fon^icm^hffi~givrthe^^ hecoln^slhteresHng'To" ) ask a number of informants to paraphrase proverbs in order td see how I they render their meanings in less metaphorical language dnd the extent i to which they agree in doing so. We did this, by selecting a s'iet of 11 prov- \ erbs and asking 17 informants to explicate them. The resulting paraplirases \ carve out a range of possible interpretations for each proverb and point / to similarities and differences in meaning among them. / To pursue the hypothesis that the proverbs we selected derive their meaning from a common underlying ethnopsychology, we next asked our informants to look for similarities among the sayings and to sort them into groups. We also asked informants to state briefly their rationale for grouping certain proverbs as similar, thus forcing the kind of abstract , speculation in which we ourselves were engaged in interpreting proverb \ meanings. Because the paraphrases and rationales given for grouping proverbs J together are quite varied and complex, it is useful to examine the overall pattern of similarities among the sayings in order to identify those judged . most similar or different. I have used multidimensional scaling as a way of graphically representing this pattern of proverb similarities in a visual model. We mayjtheuiojjiy^jt^ rationales glvelfTor grouping proverbs together in order to reconstructthe con^eJi|^CoislOQr3udging simiTarities^mong~tKe~rn;"''- —*••—— SOME PROVERBIAL SAYINGS

In searching for candidate sayings, we discovered that it is quite difficult simply to retrieve proverbs from memory at will. They resist introspective recall. However, given the right set of circumstances, the appropriate proverb seems almost to leap to mind.4 Our approach was to draw up. a list of problem-solving proverbs by searching through published collections of English proverbs (Collins 1959; Ridout & Witting 1967; Stevenson 1948; Wilson 1970; see also Simpson 1982) and to supplement that list using ourselves and acquaintances as informants. Based oh the criteria that a proverb be widely known, frequently used, and pertain (at least potentially) to personal adversity, we selected the i 1 sayings listed in Table 6.1. This corpus is not intended to be either exhaustive or representative of the full range of American sayings about problem solving. The only claim is thatjJie.statements„hLTable,6.,l ,are..a subsejL0f,,sayings_relevant "to the ways_Americans construe responses to problematic j;in;umstances. WTiaTcjahe^aiiabo^ proverbs at first glance? As expectedr'

158

GEOFFREY M. WHITE

Table 6.1. Eleven American English proverbial sayings

;;>-./

1. Every cloud has a silver lining. •'" " ' . ' v ^ •;/;;:';:: 2^ God.helps those who help themselves. -•-• ;• •'•/£'•,'.:'•, 3. The grass'is always greener on the other side of the fence. . '•:'/ v.; y/.;^ 4. There's rio use crying over spilt milk. 5 ? ,Where; there's .a.,will there's a way. ,., r ..,,:... ; ,., ..,,,J:rr,._',.;..sv 6. Necessity, is the mother of invention. .,.,;,;. 7. Rome wasnt built in a day. V -J-..,. 8. The squeaky wheel gets the grease. \ „ ,'.*"•".,;..,J '1 9. You can't have your cake and eat it too. ' 10. Don't make a mountain out of a mole hill. 11. Time heals all wounds.

most of them are phrased in figurative language. With the possible exception of two of those listed ("God helps those who help themselves" "and "Where there's a will there's a way"), all are overtly metaphorical. Common objects and events such as clouds, green grass, spilt milk, squeaky wheels, and eating cake are used to characterize problem situations in terms of more immediate, physical imagery. Although diverse, these images represent several more general types of metaphor: notions of mechanics and construction (squeaky wheels, building Rome), food (cake, milk), and visual imagery (green grass,;sjlver linings). The proverbs listed were chosen because they say something about rela-„ tions between a person and a problem or goal. They presupposeji discrepancy between the state of the world and the state oj the person (intentions, desires, actions and the like). Each saying evaluates the likelihood of achieving a goal or changing a problematic situation, and, in so doing, carries an implied recommendation about the appropriate response that will bring person and world back into alignment, creating a better fit between personal outlook and worldly circumstances. As a preview of the following analysis, note that the proverbs in Table 6.1 span at least two distinct types of recommended response: those encouraging an active attempt at changing the world (e.g., "The squeaky wheel gets the grease") and those calling for adjustment of the person (e.g., "There's no use crying over spilt milk"). PSYCHOLOGICAL INFERENCE

In order for these proverbs to carry implications for appropriate action, they require certain background assumptions about human psychology and action (see Kirkpatrick & White 1985). They acquire their meaning against a backdrop of cultural understandings about the organization of perception, feeling, and thought that mediate the interrelation of person and world. By drawing on a cultural model of the person, informants make

PROVERBS AND CULTURAL MODELS

" ^ 5 ? / ^ - ^ ••'•;•

specific inferences about the actions that follow from proy^bJi^^^;^5 , v- : ::^ , l tions about a problem/goal, or a person's perception of it..The. analysis ; of proverb interpretation developed here indicates that certain ;elenients of the American cultural model of the mind described by D'Andrade (this volume) and other notions about personal action described -by iHeider (1958) and Hutchins (n.d.) underlie proverb meanings. In particular, D'Alidrade's assertion that "the main line of causation in the cultural mode!" runs from perception through thought and feeling to intention and action captures much of the structure of reasoning in these proverbs about personal processes that mediate the fit between person and world,::.? Despite considerable variability in the specific propositions asserted by different proverbs in Table 6.1, they draw on similar understandings about human psychology and action to obtain their full meaning and force. Some of these understandings surface in the paraphrases and judgments of similarity, such that inferences about feelings and intentions are made explicit as informants seek to articulate proverb meanings. These data, discussed below, indicate that proverbial reasoning involves an inferential process that moves from (1) an assertion about some aspect of the person or problem, to (2) an expansion of its psychological implications based on a cultural model of the person, to (3) inferences about an appropriate response or course of action. A consideration of how informants paraphrase proverb meanings illustrates these different facets or levels of proverbial reasoning. Seventeen native speakers of English, all students at the University of Hawaii, were asked to paraphrase each of the 11 proverbs in Table 6.1. The proverbs were presented written on 3" x 5" index cards, one to a card. Informants were asked first to look at all the proverbs and ask questions about any that were unfamiliar. Except for three people who said they did not know "The squeaky wheel gets the grease," all of the proverbs were well known. Once having reviewed the set of 11 sayings, informants were asked to paraphrase each one by briefly writing out its meaning in plain language. The 17 distinct paraphrasings obtained in this way represent a range of interpretations that capture several levels of inference associated with each proverb. Depending on the particular proverb, the paraphrases span all or some of the levels of proverbial reasoning outlined: (1) description of the problem situation, (2) its psychological implications, and (3) a recommended response or course of action. It appears that the paraphrases are mostly pitched at levels (1) and (2), whereas the more abstract judgments of similarity tend to be made on the basis of (3), the implied recommendation, as seen in the following section. This type of variation in the level at which explications of the same proverb may be phrased is illustrated by examining all of the paraphrases given for a single proverb. The saying "The grass is always greener on the other side of the fence" provides a particularly good example of varia-

160

GEOFFREY M. WHITE

tion in paraphrases that reflect the structure of proverbial reasoning. Because this saying is less explicit about the relations between the. state of the person and the problem situation than some of the other proverbs (for example, "Where there's a will there's a way"), it requires the listener to make inferences in order to draw implications for behavior (there is no use trying to move to the other side of the fence) from its basic proposition about the perception of a problem or goal (it only appears better on the other side of the fence). Like the proverbs "Every cloud has a silver lining" and "Don't make a mountain out of a mole hill," the saying about grass being greener on the other side of the fence uses the notion of visual perception as a metaphor for thought. By asserting that a person has misperceived a problem or goal (has not seen the silver lining; has mistaken a mole hill for a mountain; has the illusion of grass being greener than it really is), these proverbs are in fact saying that a person's judgment or thinking about the problem/goal is flawed. In this way, the metaphor does its work of taking a potentially complex and ambiguous process (such as faulty reasoning) and describing it in terms of events that are more clearly delineated and accessible to public demonstration (such as determining what things look like). As might be expected, then, the greatest number of paraphrases of the proverb "The grass is always greener on the other side of the fence" focus on the act of misjudgment or misperception, saying that things either "seem better," "look better," or "appear more attractive" on the other side of the fence. Some informants extend the metaphor of sight into their paraphrase: 1. 2. 3. 4.

Other people's situation sometimes look better than they actually are. Things look better with other people. Things appear to be more attractive or better when you are not involved. No matter what one has, he can always see something better he doesn't have if he looks for it. 5. People tend to focus on their own problems and on their neighbor's assets. Other informants simply assert that things "seem better": 6. 7. 8. 9.

Things which are unobtainable always seem better, That which you cannot have always seems better. Our own condition always seems bleaker than what others have. Someone else's things will seem nicer than yours not because they are better, just because they aren't yours. 10. Once we make choices, the choice not taken always seems better. 11. Fantasy of what we have or have not. The paraphrases listed here all speak directly to the person's perception of problems or goals. In other words, they say something about the relation between person and problem situation that approximates the proposition asserted in the proverb itself. Other informants, however, chose

PROVERBS AND CULTURAL MODELS

•;I61

to paraphrase the proverb's meaning by going beyond the information given in the proverb to say what such a situation would imply about the person's feelings and desires. Specifically, six informants rendered the meaning of the proverb by noting that people who continually see people or things elsewhere as better will not be satisfied, content, or happy; they will be envious. '••'• ••-.;>"•' ?••;•. —.•;..•" ,: One is rarely satisfied with what one has. .-./.rr-r^ix. The common dissatisfaction one has with one's own state of affairs,;- ;. One is never satisfied with what one possesses or situation in which one is in. Contentment is seldom achieved. •.,, You will always be envious of what the other person has - used in a situation where someone is never happy. 17. It won't help to be envious; the other person's blessings may only look that way to you.

12. 13. 14. 15. 16.

By drawing on a cultural model of the person that links certain kinds of perceptions or thoughts with specific feelings and desires, people are readily able to characterize the emotional state of a person for whom "trie grass is always greener on the other side of the fence." The direction of inference here, from perception of situation to emotion is consistent with the general direction of inference in American ethnopsychology (D'Andrade, this volume) and in Ifalukian knowledge about the situational antecedents of emotion (Lutz, this volume): PERCEPTION/THOUGHT

> FEELING/DESIRE

However, note that feelings such as "satisfaction," "contentment," or "envy" point beyond emotional responses to things a person may want, "possess," or "achieve." In other words, they are also about goals and desires. When seen as desires (which, in D'Andrade's scheme, mediate feelings and intentions), it becomes more apparent that our informants' psychological inferences play an important role in reasoning about the intentions and actions expected to follow from a particular problem, perceived in a certain way. These inferences about the feelings or desires of someone who sees greener grass on the other side of the fence imply, in turn, certain kinds of intentions and actions that follow from dissatisfaction. It is these further implications for behavior, also derived from an underlying ethnopsychology, that give these proverbs their directive force as sources of advice about a recommended course of action. Although informants did not, in general, refer directly to these behavioral implications when they paraphrased the proverbs, they did frequently point to this level of proverb meaning when stating reasons for similarity among them, as seen below. These different levels of meaning, then, extend the underlying chain

162

GEOFFREY M. WHITE

of inference further in the direction postulated by D'Andrade to link perception and feeling with intention and action: PERCEPTION/THOUGHT • —-> FEELING/DESIRE /-.;•';. .. : '^INTENTIO^/^Ty)N This type of inference chain gives ah indication of drie way by which people draw behavioral implications from statements about a problem situation. When depicted in this way, the course of reasoning underlying some of the proverbs in Table 6.1 resembles quite closely Hutchins's (n.d.) description of American common-sense reasoning about behavior. In his analysis, ordinary interpretations of action move backward from behavior to the attribution of intentions to inferences about some background problem that gives rise to those intentions: BEHAVIOR

> INTENTION

> PROBLEM

If we postulate that a person's perception of some event as a problem (PROBLEM) leads to a desire (WANT) for change and ultimately to an attempt (TRY) to bring about change, it is possible to see how a proverb that questions perception can have implications for action. Proverbs asserting that a person's perception of a problem is flawed (such as "Every cloud has a silver lining," "Don't make a mountain out of a mole hill," and 'The grass is always greener on the other side of the fence") discourage emotions and actions aimed at changing the situation by negating the premise that it is in fact a problem at all (-PROBLEM): -PROBLEM > -WANT = > -TRY PERCEPTION/THOUGHT > FEELING/DESIRE > INTENTION/ACTION By asking what must be assumed in order to understand the behavioral implications of the remaining sayings, it is possible to identify a small number of ethnopsychological inferences that link a proverb's overt assertion with its implied recommendation for action. Two of the remaining proverbs that are similar in meaning to those mentioned and were judged so by our informants ("You can't have your cake and eat it too" and "There's no use crying over spilt milk") also have the effect of discouraging attempts to change a problem situation. In these examples, however, the effect is achieved through a different course of reasoning. By asserting that a situation cannot be changed, these sayings imply that further attempts to do so are futile. They appear to draw on the underlying belief that, for a person to try to reach a goal or change a problematic situation, he or she must believe it is possible to do so. Here again the proverbs rely on a basic element of American ethnopsychology for their intended meaning. In his analysis of "naive psychology," Heider (1958) observed that ordinary explanations of behavior and predictions of successful outcomes generally infer both ability

PROVERBS AND CULTURAL MODELS

^^63;:

(CAN) and effort (TRY) in addition to desire (WANT) as ihgredieMm'^:purposeful action. As many subsequent writers in attribution research:,have/> V.vVv:; noted (e.g., Schmidt & D'Addamio 1973), the negation of any of these ; ^ :'elements will affect inferences about the probability of success.: Thus,.by negating ability (~ CAN), these proverbs imply that one will not TRY to; change a problem situation:

-CAN

>-TRY .

.;,;:;;,;;^;^i:,;.;^:

Informants' paraphrases of these two proverbs indicate that they do, in fact, rely on some such notion of inability. More than one-third of the informants (6 and 7, respectively) used the expression cannot in describing their meanings. In the case of "spilt milk," most informants point out that one cannot change something that is in the past; whereas "having one's cake and eating it too" is a matter of one choice's excluding another. By denying the possibility of attaining some desired end, both of these sayings discourage active striving. Another significant subset of proverbs in Table 6.1 appears to rely on the same underlying belief that an active attempt (TRY) to do or change something presupposes belief in ability (CAN). However, rather than negating the possibility of changing a situation, sayings such as "God helps those who help themselves," "The squeaky wheel gets the grease," "Necessity is the mother of invention," and "Where there's a will there's a way" all assert that some goal is within reach, that a certain desired outcome is possible. By affirming the actor's ability (CAN), these sayings have the opposite effect of those just described. They lead to a recommendation for an active attempt (TRY) at goal attainment or problem resolution: CAN

> TRY

Most paraphrases of these sayings refer to the possibility of doing or getting something, given some antecedent condition. In addition, a significant proportion of the paraphrases for several of the sayings include ethnopsychological inferences about the person. For example, in the case of "Necessity is the mother of invention," 5 informants made reference to either "ingenuity" or "creativity" in times of need. And, in explicating "God helps those who help themselves," 7 people mentioned variously "initiative," "self-reliance," "responsibility," or "independence." And, in the saying "Where there's a will there's a way," which makes overt reference to a psychological disposition ("will"), nearly all informants mentioned an internal state of desire ("desire" (N = 4), "want" (N = 4), "determination" (N = 3), "perseverance" (N = 2), "motivation" (N = 2), or "believe that you can" N - 1). I have ordered this discussion of informants' paraphrases of the 11 sayings listed in Table 6.1 according to the sayings' implications for action. However, the paraphrases themselves do not make frequent reference to the recommendations for action implicit in all of them. The relevance of

164

GEOFFREY M. WHITE

such implicit recommendations for proverb meanings is evident in the uses to which these sayings are put in everyday interaction. A brief examination of informants' judgments about similarities among the proverbs, and -itheir reasons for them, indicates that these effects are recognized arid can be articulated. IMPLICATIONS FOR ACTION

After paraphrasing each proverb (written one to a card), the 17 students were asked to group them into piles of any size according to similarity in meaning. Informants were also asked to write brief reasons for the groupings they created. The sorting data were analyzed by first computing an overall measure of similarity for all pairs of proverbs, taking into account the number of informants who placed each pair together in the same pile, and the size of the pile in each case (see Burton 1975). The resulting matrix of similarity scores among all pairs of proverbs can be represented in visual form using multidimensional scaling (MDS) (Kruskal et al. 1977). MDS depicts similarities among the proverbs in terms of spatial distance, such that sayings judged more similar to one another in meaning are placed closer to one another in the spatial mapping.5 The MDS model of judged similarities among the 11 proverbs is depicted in Figure 6.1. The configuration in Figure 6.1 aids in the interpretation of proverb meanings by directing attention to groups of proverbs that informants judged as similar. I do not assume that the horizontal and vertical axes underlying the MDS model will necessarily reflect dimensions of meaning common to all of the proverbs.6 However, the most notable characteristic of the configuration is the overall right-left distribution of proverbs along the horizontal axis. Six sayings .are arrayed vertically on the right and four along the left, with "Rome wasn't built in a day" occupying a more intermediate position. The diagram locates "The grass is always greener . . . ," "You can't have your cake . . . ," "Every cloud has . . . ," "Don't make a mountain . . . ," "There's no use crying . . . ," and "Time heals . . ." in opposition to "The squeaky wheel . . . ," "God helps . . . ," "Necessity is the mother of , . . ," and "Where there's a will. . . ." This arrangement indicates that in the sorting task few people grouped the proverbs on one side of the diagram together with those on the other and reflects the contrasting recommendations for action embedded in these proverbs. The proverbs on the left encourage some kind of goal-oriented action; those on therightrecommend against such striving. Or, put another way, the proverbs on the right encourage adjustment of the person rather than the situation. If this interpretation is correct, the horizontal dimension captures the divergent inferences about whether to TRY to change a problem situation. The reasons stated by informants who sorted the proverbs along these lines give some support to this interpretation. Consider first the kinds of

PROVERBS AND CULTURAL MODELS

3^5 : ;,,

'(3)GRASSISG^EEMR^

(9) CANT .HAVE YOUR $ A k r l 3 ; -:.

(8) SQUEAKY WHEEL .'.• :|2} HELP THEMSELVES (1) SILVER LINING

(10) MOUNTAIN/MOLE HILL (6) MOTHER OF INVENTION (5) WHERE THERE'S A WILL

|4)SPILT.MtLK

,

..

(11) TIME HEALS

(7) ROME WASN'T BUILT IN A DAY

Figure 6.1. Two-dimensional model of similarities among 1 i American proverbs .(Stress = .092) .. rationale given by those who grouped together the proverbs "The squeaky wheel. . . ," "God helps. . . , " "Necessity is the mother . . . ," and "Where there's a will. . . ."Seven informants (out of the total of 17) placed these 4 proverbs together as a set or a subset of a larger group. The reasons given for their similarity are: 1. These give positive suggestions. 2. Positive reaction. 3. These tell you to go out and do something, 4. These are exhorting one to help themselves, they are motivators. 5. It's those people who initiate some solution that get it accomplished. 6. These imply the value of self-help, keeping at it, plugging away. 7. Concerned with self-determination and getting ahead. Notice that a number of the informants articulate the rationale for their grouping by pointing to the proverbs' performative value. These sayings are said to be similar because they variously "suggest," "tell," or "exhort" one to take a particular course of action. In describing their meanings in this way, these informants are referring to the force of the inference about TRYING, analyzed above as an implicit recommendation. Other people simply describe the recommended goal-seeking behavior itself (in

166

GEOFFREY M . W H I T E

terms such as "initiate," "self-determination," "help themselves," or''keeping at it") and/or its positive outcome ("getting ahead," "get it. accomplished''). The two informants who characterized these sayings as "positive" < may be expressing the fact that their implied recommendation is an affirmation rather than a negation of a basic proposition in the cultural model. The reasons given for grouping proverbs located on the right side of Figure 6.1 contrast sharply with the positive rationales just listed. Among these proverbs, the most commonly grouped together were "The grass is always greener. . . ," "You can't have your cake . -.-•.," "Don't make a mountain . . . ," and "It's no use crying . . . ." Five people grouped these four together as a set or subset. The reasons stated were: 1. 2. 3. 4. 5.

Negative reaction. These are no-nos. "Don't. . ." may not be said but is implied. These are telling one to quit looking at things from a negative perspective. These all imply acceptance of your situation. Things take care of themselves; individuals must adapt to the situation.

The directive force pf the proverbs is again made explicit by some of the informants who note that they take a command form by implying "Don't" or by "telling" someone to do something. The proverbial implication not to do something is noted by informants who describe them as "negative" or "no-nos." Rather than attempting to change the situation, individuals must variously "accept" or "adapt" and the like. ., A number of informants made finer discriminations in their sortings of the proverbs located on the right side of Figure 6.1. Inspection of the diagram shows that the two proverbs "The grass is always greener . . . " and "You can't have your cake . . ." were judged quite similar to each other, as were "Don't make a mountain . . ." and "It's no use crying . . ." Five people grouped just the former two proverbs together and gave the following reasons for their judgments: 1. Sayings counsel one that he should be happy where he is. 2. These comment on the fact that people are not usually content with what they have. 3. Both for people who want more than they have. 4. Point out a human tendency toward dissatisfaction. 5. You'd say these to people who moan and groan too much. It is apparent that, at this lower level of specificity, the reasons for placing just two similar proverbs together draw on ethnopsychological inferences about the state of the person, just as did the paraphrasings discussed earlier. Since "The grass is always greener . , ." is one of the two proverbs in this grouping, it is not surprising that informants refer to some of the same feelings or desires mentioned previously: "dissatisfaction," not "content" or "happy." These are emotions associated with desires that, for one reason or another, cannot be fulfilled.

PROVERBS AND CULTURAL MODELS

167:

The reasons given for grouping the two proverbs "The grass is always greener . . ." and "You can't have your cake . . ." indicatethat someinV; formants perform the sorting task more on the basis of an inference 4bbut- ~ the state of the person (e.g. the person is "not content'^) than on the^basis of the specific proposition expressed metaphorically bythe^pr^erb^JeithCT' ;".: that the goal is not realistic, in the case of the former, or that it £annbt be reached, in the case of the latter). The ethnopsychological .basis.for judgments of similarity among the proverbs is also evident in the reasons given by the six informants who placed together the proverbs "Don't iriake a mountain . . . " and "It's no use crying . . . , " distinct from "The grass is always greener" and "You can't have your cake": :v ^ 1. These would be said to complaining or depressed individuals. 2. Both directed toward someone who's feeling sorry for themselves in one way or another. 3. These remind us not to be too concerned with little problems as they will pass. 4. Advising getting things in perspective. 5. These are negative - tell you not to do something. Don't make something more serious than it is. 6. Negative proverbs with a reprimanding attitude used to comfort. In addition to noting the directive force of these proverbs with terms like "advising," 'telling," "reminding," or "reprimanding," and characterizing them as "negative," several reasons again refer to the emotional state of the person. However, these emotion attributions differ somewhat from those described earlier. Terms such as "depressed," "feeling sorry," and "too concerned" have a different tenor than do "dissatisfaction" or "discontent." The difference between these two sets of emotions follows from a distinction between seeking goals that cannot be had, on the one hand, and coping with present difficulties, on the other. Here again, the basis for informants' judgments of similarity among the proverbs rely more on inferences about emotional responses than on the specific propositions about perception of a problem (—PROBLEM) or ability to change a situation or reach a goal (-CAN). The proverbs "Time heals all wounds" and "Rome wasn't built in a day" are also shown in Figure 6.1 to have been judged somewhat similar to one another. Insofar as they are generally aligned on the right side of the MDS diagram, they may be seen to advise adjustment of the person rather than an attempt to change the situation. The saying "Rome isn't built in a day" is more ambiguous in this respect since it advises patience in the short term, but persistence over the long term. Perhaps for this reason it is located in a more intermediate position on the horizontal dimension in Figure 6.1. Both sayings pertain to the perception of time. They seek to resolve a discrepancy between person and situation by adjusting the person's perspective on time: In the long run, things will get better; in the long run, goals will be attained. The reasons given by four informants

168

GEOFFREY M. WHITE

who grouped these two sayings together refer to this perspective on iime, as well as to the personal response (patience) that may be inferred from a lengthening of time perspective: -: / '::••'•*•** 1. Both insinuate that things take time. 2. Things will be better, spirit inducers. 3. These are a reminder to be patient, things take time. 4. Counsel patience.

/. -.---:;•••%-

The reasons given for judging certain proverbs as similar are stated at different levels of inference, just as the paraphrases reflect different parts of the reasoning process used to interpret proverbs. This ability of informants t o describe different aspects of the interpretive process may account for the considerable diversity found in t h e paraphrasing and sorting data. Because t h e task of grouping proverbs together requires a greater level of generality t h a n simple paraphrasing, informants appear t o have based their judgments more o n t h e proverbs' implications for action than o n the specific propositions that conceptualize a problem situation. Thus, some based their judgments of similarity o n t h e basic opposition of a n active attempt at change versus adjustment of the person. Others appear to have made judgments of similarity based on inferences about the specific emotional responses that mediate a certain kind of situation and the implied recommendation for adjustment. I have argued that b o t h kinds of judgment a r e based o n inferences about h u m a n psychology a n d action drawn from an underlying cultural model of the person. It is through a process of ethnopsychological reasoning that people link descriptions of a problem situation with recommendations for an appropriate response. Conclusion Proverbs appeal to reason. In particular, they appeal to common-sense reasoning based on cultural models of experience. Each proverb examined here represents a point of view, a way of looking at problems and persons that, because of our shared knowledge about such things, carries certain inevitable implications for action. By characterizing a problem situation in a certain way, as a matter of spilt milk or squeaky wheels, proverbs interpret that situation by identifying it as an instance of a more general model. Instantiation of part of an existing knowledge structure (such as the proposition that a certain, event has been misperceived as a problem) then creates the basis for further inferences about emotion and action. The inferential structure of proverb meanings - from problem description through psychological inference to implication for action - reflects the pragmatic work done by proverbs. Indeed, it seems likely that these peculiar bits of formulaic language are widely used precisely because they carry directive force.7 As indirect directives, they are strategic linguistic devices for evaluating and shaping the course of social experience through appeals to common sense. The fact that proverbs are recognized as ex-

PROVERBS AND CULTURAL MODELS

pressions of common sense or folk wisdom is indicative of their frecjuerii use in attempting to clarify uncertainty. -V ;;>;%^|M ^ An appeal to folk wisdom is a useful way of attempting to resoly^per-: ;; sonal conflict or ambivalence. The sayings examined here all^tenippp; lift a person out of a personal quandary or, in the : .lerms : pff^o^iye ;•/•:problem-solving, a "blocked condition," by suggesting a.point ofview that resolves a discrepancy between person and worldly circumstances ^cf. Hutchins & Levin 1981). They introduce a new perspectivei'jiiy variously altering one's perception of time, of the problem, or of oneV ability ^to do something about it. Once located in the framework of ah underlying ethnopsychology, each type of assertion carries specific implications; for emotion and action. Analysis of the processes of inference underlying proverb interpretation reveals the operation of specific cultural understandings about persons and action that have been identified previously by other students of American ethnopsychology (D'Andrade, this volume; Heider 1958; Hutchins n.d.; Schmidt & D'Addamio 1973). It is a reassuring affirmation of the flexibility of language and culture that even this small corpus of proverbs represents contradictory ways of construing problem situations. The contrast between sayings such as "Where there's a will there's a way" and "You can't have your cake and eat it too" indicates that cultural models provide alternative (and sometimes mutually inconsistent) ways of interpreting experience. In this instance, these two sayings rely on the same ethnopsychology, which asserts that purposeful action typically presupposes belief in the ability to do or achieve something. One saying provides a way of affirming ability and recommending positive action; the other can be used to negate ability and discourage an active response. Such diverse, and even contradictory, devices for conceptualizing experience suggest that American proverbs and cultural models are readily adapted to a wide variety of purposes and occasions. The attempt here to analyze proverbs through several types of interpretive data, including paraphrasing and similarity judgments, has been aimed at teasing out specific propositions and inferences that contribute to proverb meanings. This approach would be augmented by additional, complementary types of data, such as examples of discourse obtained through interviewing or natural observation. Alternatively, more structured, experimental elicitation could be devised to test the accuracy of the model sketched here. Both types of data would supplement, and probably correct, the account rendered here. But then, "You can't have your cake and eat it too." Or is it, "Where there's a will there's a way"? Notes I. This paper has profited from the contributions of a group of graduate students at the East-West Center who worked together with the author in collecting and analyzing the data discussed here. Jonathan Gurish, Joyce Kahane, and

no

GEOFFREY M. WHITE

Russell Young made especially important contributions but should not be held responsible for the arguments made here. I would also like to thank Paul Kay, and Lynn Thomas, as well as Willett Kempton and other participants in the . Princeton Conference on Folk Models, especially the organizers,Dorothy >"< Holland and Naomi Quinn, for helpful comments on an earlier version of : this paper. The paper was first presented in a symposium organized by Holland and Quinn for the 80th Annual Meeting of the American Anthropological; •'• Association, entitled Folk Theories in Everyday Cognition. : . ;v 2. In American English, proverbs appear to represent one end of a continuuni of linguistic forms that vary in their degree of standardization and formaliza-1 tion, with no sharp distinction between them and other types of idiomatic or colloquial expressions. In contrast, Chinese language sayings form several distinct types, including two written forms, one of which is distinguishable by its four-character composition. 3. An important question for future research concerns the extent to which the relation between specific types of source domain and target domain are arbitrary, or sometimes associated with particular types of conceptualization. Thus, proverbs about human temperament frequently draw from the domain of animals (For example, "Curiosity killed the cat," "You can't teach an old dog new tricks," and "You can lead a horse to water, but you can't make him drink"), and those concerned with events beyond human control frequently use environmental imagery (such as "It never rains but it pours" and "The calm comes before the storm"), 4. The fact that proverbs are difficult to recall from memory without an eliciting context or situation raises questions about the form in which they are stored in memory. It is clear that proverbs themselves are not stored in a distinct or bounded domain. There is no taxonomy of wise sayings. Rather, they are tied to cultural knowledge about types of situation or action-scenario. Such knowledge may be actively generated or assembled in the course of understanding specific events. Each proverb condenses a set of interlinked propositions that have general relevance for social life and can be used recurrently to intepret a range of events. This view of proverbs resembles Schank's (1980) reformulation of the notion of "script" as a reconstructive process that relies on various generalized sources of information called "Memory Organization Packets" (MOPs). These memory structures are the generalizations and abstractions from experience that are used to make predictions about future events, just as proverbs are used to make recommendations about a course of action. 5. MDS will represent a set of similarity scores in terms of any number of dimensions. In general, the investigator selects the MDS solution that most accurately displays the similarities while using the least number of dimensions. Analysis of our proverb-sorting data with MDS indicates that these data may be adequately represented in two dimensions ("stress" - .092). 6. Even in semantic domains where word meanings may be meaningfully scaled in relation to a few bipolar oppositions, such as adjectivelike terms used to describe personal traits (see White 1980), MDS will not provide much help in discovering the meanings of dimensions produced by complex inferential processes. For example, in their chapter for this volume, Holland and Skinner clearly show that a scaling model of terms for gender types gives few clues about informants' knowledge of male/female interactions that produced that model. In using MDS to represent relations among proverbs, there is even less reason to expect the dimensions of a scaling model to have specific semantic significance. The inferential processes that underlie proverb meanings are unlikely to map directly onto a few bipolar dimensions. There would have to be some components of meaning pertinent to all of the proverbs for the

PROVERBS AND CULTURAL MODELS

m:.-'

MDS axes to have significance as dimensions of meaning. As seen in the foJe-^ going discussion, most of the propositions and inferences that contribute to proverb meanings are relevant to only a subset of the sayings m Tib&£.l;.;.v • / Except for the fundamental contrast in their implied recommeridaiiomiforf action (the opposition between sayings that encourage action and those ^ihat; ^ encourage adjustment, reflected in the horizontal dimension of Figure\'&Af&.;; :: .. there do not appear to be any components of meaning common to all sayings^' in the corpus. .:,••'.•-'•y*::-.:J\ lc7. The explanation for why proverbs are used rather than other kinds of1 ordinary: '• language, or why they are used on certain occasions and hot others! 'requires : recourse to social and contextual information not discussed in this paper i:The fact that proverbs are used at all may carry implicit social meaning concerning the nature of the relationship between speaker and listener. For example^ some participants in the Folk Models conference argued that the use of proverbs such as those in Table 6.1 frequently indicates a bid for dominance in interaction. References Arewa, E. O. and A. Dundes 1964. Proverbs and the ethnography of speaking folklore. American Anthropologist 66(6,2):70-85. Briggs, C. L. 1985. The pragmatics of proverb performances in New Mexican Spanish. American Anthropologist 87(4):793-810. Brown, R. L. 1970. A Book of Proverbs. New York: Taplinger Publishing Company. Burton, M. 1975. Dissimilarity measures for unconstrained sorting data. Multivariate Behavioral Research 10:409-423. Carbonell, J. G. and S. Minton n.d. Metaphor and common-sense reasoning. Paper presented at conference, Folk Models, May, Institute for Advanced Study, Princeton, N.J. Collins, V. H. 1959. A Book of English Proverbs with Origins and Explanations. London: Longmans. Dyer, M. G. 1983. Understanding stories through morals and remmdings. In Proceedings of the Eighth International Joint Conference on Artificial Intelligence. Pp. 75-77. Heider, F. 1958. The Psychology of Interpersonal Relations. New York: John Wiley and Sons. Hutchins, E. n.d. An analysis of interpretations of on-going behavior. Unpublished Manuscript, Department of Anthropology, University of California, San Diego. Hutchins, E. and J. Levin 1981. Point of view in problem solving. Center for Human Information Processing Technical Report #105. San Diego: University of California. Kirkpatrick, J. and G. M. White 1985. Exploring ethnopsychologies. In Person, Self and Experience: Exploring Pacific Ethnopsychologies. G. M. White and J. Kirkpatrick, eds. Berkeley: University of California Press, Pp. 3-32.

172

GEOFFREY M. WHITE

Kruskal, J., F. Young and J. Seery --•'>••.;/ 1977. How to Use KYST-2, A Very Flexible Program to do Multidimensional Scaling and Unfolding. Murray Hill, N.J.: Bell Laboratories. • . / • . Lakoff, G. and M. Johnson / V• 4980. Metaphors We Live By. Chicago: University of Chicago Press. • Messenger, J. C. •• •::.-•:.:;•' - - ^ M . : •'•;-.:-:-'•'._• 1959. The role of proverbs in a Nigerian judicial system. Southwestern Journal of Anthropology 15:64-73. Newell, A. and H. Simon 1972. Human Problem Solving. Englewood Cliffs, N.J.: Prentice-Hall,... Ridout, R. and C. Witting 1967. English Proverbs Explained. London: Heinemann. Salamone, F. A. 1976. The arrow and the bird: Proverbs in the solution of Hausa conjugalconflicts. Journal of Anthropological Research 32:358-371. Schank, R. C. 1980. Language and memory. Cognitive Science 4:243-284. Schmidt, C, and J. D'Addamio 1973. A model of the common-sense theory of intention and personal causation. In Proceedings of the Third International Conference on Artificial Intelligence. Stanford: Stanford University. Pp. 465-471. Schultz, E. 1980. Samoan Proverbial Expressions. Auckland: Polynesian Press. (First published in 1953.) Simpson, J. A., ed. 1982. The Concise Oxford Dictionary of Proverbs. London: Oxford University Press. Stevenson, B. 1948. The Home Book of Proverbs, Maxims and Familiar Phrases. New York: Macmillan Company. White, G. M. 1982. The role of cultural explanations in 'psychologization' and 'somatization.' Social Science and Medicine 16:1519-1530. 1980. Conceptual universals in interpersonal language. American Anthropologist 82(4):759-781. Wilson, F. P., ed. 1970. The Oxford Dictionary of English Proverbs. 3rd ed. Oxford: Clarendon Press.

7 Convergent evidence for a cultural mode! of American marriage1 Naomi Quinn

This paper analyzes two passages excerpted from a longer interview, illustrating the utility of a method of discourse analysis elsewhere (Quinn 1985a; 1985b; n.d.) applied to much more extensive interview material of the same sort. In the larger study from which these excerpts have been borrowed, husbands and wives in 11 marriages were interviewed, separately, over an average of 15 or 16 hours each, on the topic of their marriages.2 Interpretation of the passages at hand draws on the more extended analysis of this entire body of material. The full analysis depends for its convincingness on its ability to account for features of discourse about marriage in many passages such as those examined here. Of course, the entire analysis cannot be presented in this brief paper, but the examples provided suggest its range. The two segments of discourse scrutinized here, one of which followed the other about midway through the first hourlong interview with a woman whom we call Nan, are reproduced below:3 3 W-l: I think Tom and I both were real naive about each other. I mean, I think that we got married on the strength of a lot of similar tastes and a lot of love and appreciation but not much real sense of who each other were. I really don't think that we, either of us, had examined each other and said - 1 mean, I don't think I had said, "Really, who is this Tom Harper, how can I describe him, what is he? What . . . ." You know, "Is that the kind of person I need to be married to?" I don't think I had ever consciously done that - examined my needs and to see if Tom'd fit them. I think it was an intuitive kind of thing and I look at it now and I don't think I necessarily could have done that. I mean, the things that have been strengths of our marriage are the same things that got us married - I think being comfortable with each other, the similar tastes, the same kind of - ways of dealing with a lot of things. And the things that have been difficult in the marriage I couldn't have foreseen; I don't think now but I have sometimes thought back, you know, "Gee, people really do go into marriage, with their eyes, closed." I just find it - how amazing that many marriages get to stay together, when you consider the way they do it. 3 W-l: I think during some of Tom's and I - during some of the most difficult passages that we had when we have really despaired in a sense and thought, "This - we are going to be driven apart by all our problems," including, you know, our problems with each other, and one of the things we have both thought

174

NAOMI QUINIS

is that, "If I know Tom as well as I know him and love him as much as l love him and still have this much trouble being married to him, what in the work} chance would I have of finding anybody else who would be any easier, to be married to and I wouldn't know that person any better when I got - married j. him than I knew Tom." /: Right, right and that would be the whole thing all over again. 3 W-l: Exactly and never having learned or worked through what actually you need to learn and work through to make the first marriage stick. And I think that's one of the things that - almost laziness in a sense or unwillingness to put out effort for nothing. Why in the world would you want to stop and not get the use out of all the years you've already spent together? /; A sense of investment, ha? 3W-1: Yeah, really, A sense of, well, through the good and the bad. We have learned a lot about each other. We've learned a lot of ways of working with each other. If it took seven more years before you learned that much with the; next person. Where - you know, where would you go? The object of the analysis to be demonstrated on these interview excerpts is reconstruction of the cultural understandings of marriage that must be assumed to underlie discourse about marriage in order to make such discourse comprehensible. The reader will find it useful, in following this demonstration, to refer to these interview passages.

Metaphors of marriage An extremely helpful feature of the discourse, and hence a departure point for this analysis, is the metaphor in which talk of marriage is cast. Metaphors are rich clues partly because they are ubiquitous (Lakoff & Johnson 1980). It is possible to talk about marriage in the technical terms that American English provides - to speak of one's spouse rather than one's "partner" or the person to whom one is "hitched"; of being married rather than "getting tied down" or "jumping into marriage"; of getting divorced rather than "splitting up" or "bailing out" of a marriage that is "falling apart." As Nan's discourse illustrates, people sometimes use the nonmetaphorical alternatives. However, they do not sustain such nonmetaphorical language for long probably because the technical language is neutral toward the marital experience it describes. The range of available metaphorical language, by contrast, allows the speaker to make a variety of points about that experience. The metaphors for marriage provide a first set of clues to the cultural model of marriage underlying discourse of the sort examined here. Superficially varied, these metaphors fall into a few classes. For example, Nan casts marriage in several different metaphors that have in common the expectation that it is to be enduring. In one metaphor, an extremely popular one in talk about marriage, MARRIAGE IS A MANUFACTURED PRODUCT. In Nan's words, such entities have "strengths" and "stay to-

, CONVERGENT EVIDENCE FOR A MODEL OF MARRIAGE

W?5'

gether," but they take work to produce — "And never having leariied or worked through what actually you need to learn and. work through to make the first marriage stick," she observes. She and other interviewees •show a great deal of creativity in exploiting this metaphor. They speak of marriages that "last" and "work" as well-made products should,^and they-char-, acterize a marriage that does so as a "good thing" and a "strong marrifge," much as one would say of a manufactured productmade to last ih#:it was "good" and "strong." They pursue other entailments of the ni6|apnpr, recognizing, in comments such as the following, that the manufacture Of such a product requires, not just work, but also craftsmanship, durable material, good components that have been put together well, and a whole that is structurally sound and substantially constructed: : :; • 2H-3: It's just that our relationship is extremely important to each of us and, you know, we want to work hard at making it so and making it better. 4H-4: When the marriage was strong, it was very strong because it was made as we went along - it was sort of a do-it-yourself project. 1H-2: I think that maybe I have an appreciation for the fact that a happy marriage is not entirely problem-free and that probably means that you really have to start out with something that's strong if it's going to last. 4H~4: And I suppose what that means is that we have both looked into the other person and found their best parts and used those parts to make the relationship gel. 9H-3: I guess stacking that up against what I saw in this other marriage, I guess that, you know, it seems like it was stuck together pretty good. 4H-2: They had a basic solid foundation in their marriages that could be shaped into something good. 2W-2; Each one [experience] is kind of like building on another, that our relationship just gets more solid all the time. Moreover, this same metaphor of marriage as some kind of manufactured product can also characterize marriages that fail to endure: Such marriages may be "weakened" or "ruined" under a variety of circumstances; they are "broken" ones that are not "working" anymore; they are "shaky" or they have "strains" in them; or they may be only "the facade of a good marriage." In another metaphor of enduringness that recurs in this discourse, MARRIAGE IS AN ONGOING JOURNEY. As Nan puts it, spouses go "through the good and the bad" together; and they make progress, as reflected in her objection to "stopping" one marriage and starting over with somebody new: "Where would you go?" As this last comment also suggests, a journey has a final destination, arrival at which provides another way of expressing the idea of marital enduringness. Another wife states this point more optimistically:

176

NAOMIQUINN

-4W~1; That 1 have changed so much and that we have changed so much and that we have been able to work through so many basic struggles in oiir marriage and be at a place now where we trust each other, we love each other, we .« like each other. We appreciate each other. And feel pretty confident about being '" able to continue that way and continue working any.other stuff that comes up. Just seems pretty amazing to me. It could have gone in so many different directions and that it didn't is incredible. But I think both of us take a whole lot of credit for the direction it went in, that we worked at this really hard. This passage exploits a further entailment of the ongoing journey metaphor: the directionality of such a journey. Interviewees sometimes use this metaphor also to suggest how marriages fail to endure: not only do they "stop" or come to a point at which they are "unable to continue," but spouses also find themselves "in a place where they don't want to be," or they "split and start going in a different direction." ? A third common metaphor of enduringness, MARRIAGE IS A DURABLE BOND BETWEEN TWO PEOPLE, also appears in one passage under analysis. The secureness of this bond is often reflected in metaphors picturing married couples as "cemented together," "bound together," "tied to each other," or, in the words of one husband quoted earlier, in a relationship that has "gelled." Here, Nan conveys the same notion obliquely, by the forceful means required to sever such a bond: "We are going to be driven apart by all our problems." Again, a marriage in danger of ending can be characterized in the same metaphorical terms as an enduring one. These, then, are some metaphors in which Nan and other interviewees express their expectation that marriage is an enduring relationship. Another expectation about marriage reflected in metaphor is that it be beneficial. In the two passages from Nan's interview this expectation is reflected first in the metaphor of one spouse's "fit" to the other's "needs" - "I don't think I had ever consciously done that. Examined my needs and to see if Tom'd fit them." In this metaphor, A SPOUSE IS A FITTING PART. Again, this is not a lone example of such metaphorical usage; other interviewees make such comments about their spouses as, "I couldn't find a replacement. I couldn't find another woman to replace Beth"; "The best thing about Bill, for me, is that he fits me so well"; and "We've kind of meshed in a lot of ways." In a second metaphor for marital benefit, Nan conceptualizes the years she has been married as time "spent" - "Why in the world would you want to stop and not get the use out of all the years you've already spent together?" she says. Here, MARRIAGE IS AN INVESTMENT. In American English, time is a resource, which, like money (Lakoff & Johnson 1980:7-9), can be invested. Other interviewees also regard the benefits of marriage as resources that spouses derive from the marriage, as reflected in such comments as, "And that was really something that we got out of marriage"; "We did a lot more talking about what we did or didn't want in our own marriage"; "I'm sure they must have something good in their

CONVERGENT EVIDENCE FOR A MODEL OF MARRIAGE

<#77

marriage or they wouldn't still be together"; "I think she sort of felf that . she would get those same things from marriage." Again, these metaphors of spouses as irreplaceable or well-fitting parts and of marriage as a! con- "... tainer of resources illustrate but do not exhaust the metaphors interviewees use to express the idea that marriage is beneficial.;::;?;^ ; : r ; ; : ; h ^ ^ y ; v : Nan's interview excerpt provides a good illustration of how interviewees frequently exploit the entailments of a given metaphor to.make multiple points about marriage. Here, in the remark, "through the good and the bad," she makes the ongoing journey metaphor do double duty, characterizing marriage as something that is both enduring - a progression "through" successive experiences - and beneficial - some of these experiences are "good" ones. In her metaphor, marriage is also potentially costly, entailing encounters with bad as well as good. Other interviewees also make the point that marriage may entail costs as well as benefits. They talk, for instance, of "being short-changed in this relationship," or of the possibility of divorce "when the effort is more than the reward." Implicit in this last example is a folk social psychology4 of voluntary relationships that, like its counterpart in academic exchange theory, assumes that the parties to such a relationship will not continue in it unless their benefits outweigh their costs, to render the relationship rewarding in net terms. Elsewhere in her interviews, Nan herself develops the implications of this assumption: 3W-12: Because I think it costs me a lot and I don't think he's measuring that cost. And I'm scared it's going to cost me too much and leave me without being able to stay in the relationship. A further implication of this exchange model of relationships such as marriage is that their continuation depends on both parties experiencing net benefit. Again, Nan makes this explicit when she says, at the beginning of the first passage, "Tom and I both were real naive about each other. . . . I really don't think that we, either of us, had examined each other" for one's fit to the other's need. Such assertions about mutual needs to be met and mutual benefit to be realized are common in this discourse; interviewees frequently emphasize that they are speaking for "both of us," or add provisos such as "and I for her," or "and vice versa" to their descriptions of the benefits they derive or anticipate from marriage. A third presupposition, that marriage is unknown at the outset, is vividly captured in Nan's metaphor of people who "go into marriage with their eyes closed." This is an instance of a general-purpose metaphor in American usage - KNOWING IS SEEING (Lakoff & Johnson 1980:48). In it, a lack of knowledge about marriage is cast as a failure to see, an equation reflected also in Nan's description of how she and her husband married without having examined each other, so that she did not know who he was, really. If initial ignorance of marriage can be captured in these metaphors of sightlessness, lack of observation, and nonrecognition, it is equally

178

NAOMI QUINN

-well reflected in another set of common metaphors picturing the manner of entry into marriage as precipitous and unprepared - "We had no idea what we were getting into"; "We sailed right into marriage*'; "Hf jumped ,'from one marriage into another"; or, in one particularly vivid example, people "falling into marriage like king pins at the bowling alley." These modes of entry contrast with the more considered ways people talk about leaving marriage - "walking away," "stopping and getting off," or/'having to bail out," for example. A fourth expectation reflected in the two passages with which we began is that, once experienced, marriage turns out to be difficult. Nan speaks directly of "the things that have been difficult in the marriage," and she also casts these things in a metaphor, commonly used to describe marital difficulties, of the "problems" that threaten to drive her and Tom apart. In their metaphorical characterizations of marriage, interviewees exploit the difficulty inherent not only in problems, but also in all kinds of mentally, psychically, and physically demanding situations: thus, marriage may involve struggle, trial, or conflict, for example. A favorite description of marital difficulties, probably because it is so conveniently conjoined with the notion of enduringness in the metaphor of marriage as an ongoing journey, speaks of the hardships endured in the course of that journey; this metaphor appears in Nan's interview excerpts as the "difficult passages that we had." Elsewhere, interviewees elaborate on this metaphor, speaking, for instance, of the uphill stretches or the rocky road to be traveled in a marriage. One husband uses a ship metaphor to capture, at once, the necessity for a marriage to be structurally sound in order to endure and the further understanding that it must be so built in order to withstand marital difficulties - the stormy weather through which it will sometimes be required to sail: 3H-6: The self-righting concept that, you know, the marriage has enough soundness and equilibrium that it will take steps to right itself in any kind of stormy situation. A final expectation about marriage revealed in Nan's remarks is that it takes effort. This expectation follows from the understanding that marriage is difficult: In our folk physics of difficult activities, with its basis in experience of the physical world, such activities require effort to perform. Nan alludes to this effort directly when she speaks of the "laziness" implicated in her "unwillingness to put out effort for nothing." She also describes effort in a metaphor of "working through" what "you need to learn and work through to make the first marriage stick." Here, the problem metaphor used earlier is extended by allusion to the kind of effort, "working through," required for problem solution. Other interviewees speak in terms of other kinds of effort: of the "searching" required to discover "where each of us were"; of the necessity, entailed by the journey metaphor, to "fight our way back almost to the beginning"; or that entailed

CONVERGENT EVIDENCE FOR A MODEL OF MARRIAGE

179

by the manufactured product metaphor to "redo the whole thing.". Frequently, also, by extension of this latter metaphor, interviewees allude to the "hard work" they have had to put in "for a good relationship." In the words of one husband quoted, "we want to work hard a t . . . making it better." ; These metaphors for marriage thus appear to be organized by five schemas for propositions about marriage, which can be glossed as follows: MARRIAGE IS ENDURING MARRIAGE IS MUTUALLY BENEFICIAL MARRIAGE IS UNKNOWN AT THE OUTSET MARRIAGE IS DIFFICULT MARRIAGE IS EFFORTFUL In the following section, we see how Nan draws on these propositionschemas to construct a reasoned argument about marriage. The variant of the schema notion adopted here is Hutchins's (1980:51), although what he calls simply a schema is here alluded to as ^propositionschema in recognition that mental schemas may organize other than propositional material (Lakoff 19S4; see Quinn & Holland, in the Introduction to this volume). In Hutchins's terms, such a proposition-schema is a "template" from which any number of propositions can be constructed. The centraiity of these five schemas to Americans' understanding of marriage is evidenced by the recurrence of propositions cast in metaphors of the enduringness of marriage, the mutual benefit to be derived from it, initial lack of knowledge about it, its difficulty, and the effort it requires, along with other propositions in which these same concepts are nbnmetaphorically represented throughout the discourse under analysis in the present study. These five classes of metaphor, together with three others, virtually exhaust the metaphors people adopt in their talk about marriage. The three additional categories of metaphor occurring in this talk delineate three further proposition-schemas that appear to play a roie in the American cultural model of marriage: MARRIAGE IS JOINT MARRIAGE MAY SUCCEED OR FAIL MARRIAGE IS RISKY None of these three proposition-schemas figure in the two interview excerpts that are the focus of this analysis. Therefore, these schemas and the evidence, in metaphorical usage, for their role in Americans' understanding of marriage are sketched only briefly and partially here and treated no further (a full discussion of the metaphors for marriage appears in Quinn 1985a). The notion that marriage is a joint arrangement is reflected in a rich variety of metaphors. The marital relationship, for instance, is described as a "unit" or a "pair," as being "together in this" or presenting a "united

180

NAOMI QUINN

front." Some of these metaphors, for example, of marriage as a "partnership," or married life as "teaming up," convey at once the jointness of marriage and the effortfulness of this joint enterprise. Other metaphors v bear the dual entailments of jointness and enduringness, wher/the Meta. •* phorical link between spouses is, by its nature, an enduring one, asm the examples cited earlier in this section, of spouses "bound together" or • "cemented together" or "tied to each other" or using the "best parts" of each (1o make the relationship gel." Another metaphor already encountered, that of a spouse as a fitting part, simultaneously carriesthe entailments of benefit and jointness. ',".',, The most frequent metaphors of success, and conversely of failure, exploit an entailment of the manufactured product metaphor. They add another layer of meaning to that metaphor to characterize the successful marriage as one that "works" and the failed marriage, by contrast, as one that is no longer working. Another popular metaphor, this one building on that of marriage as an effortful activity, characterizes marital success in terms of some difficult task brought to completion - a marriage, like a problem, "worked out," or an unsuccessful one that perhaps "doesn't work out." Two of the varied metaphors of risk used to talk about marriage characterize it as a matter of chance, such as gambling - "there's so many odds against marriage," for instance - or as being in danger of survival - as in the comment, "the marriage may be in trouble." The journey metaphor, which so aptly combines the concepts of enduringness, difficulties encountered along the way, and the effort of overcoming those obstacles to progress, can also bear the additional entailment of risk to survival, as the danger inherent in an arduous journey. Like that for effort, the schemas involving success (or failure) and risk derive not directly from our understanding of marriage but from our folk physics of difficult activities, of which marriage is one. Not only do we recognize that such activities require effort for their execution, but we also know that in spite of such effort, they may or may not be successfully completed: The difficulties may be insurmountable, so that undertaking to overcome them carries the risk of failure. This folk physics of difficult activities, then, like the folk social psychology of voluntary relationships, is a cultural model within a cultural model. We can only understand why marriage should be cast in metaphors of effort, success or failure, and risk if we know about difficulty. Reasoning about marriage The demonstration that Nan's metaphors for marriage, and those of other interviewees, are organized by a small number of schemas for propositions about marriage sets the stage for the next part of our analysis. This requires that we return to Nan's interview excerpts for a more fine-grained examination of her discourse. We now take advantage of another feature

CONVERGENT EVIDENCE FOR A MODEL OF MARRIAGE

iSl;:

of such discourse: the reasoning people do in the course of their explanaf. tions of marriage. This reasoning lends convergent support for #he five proposition-schemas identified on the basis of Nan's metaphors and supplies evidence for how these five schemas articulate with one another in Americans' cultural model of marriage. In this reasoning; ipropositiohs about marital enduringness, benefit, difficulty, and so on serve as building blocks for composite proposition-schemas. The more complex schema is created by conjoining two such propositions in a causal relation.,.:,; In order to uncover the logic of this reasoning, however, some preliminary decoding is required. It is necessary to decode the metaphors for marriage in which such reasoning is frequently couched to reveal the common schemas underlying these metaphors. It is also necessary to recognize regularity beneath another feature of the discourse - the varied syntax and semantics of causality in American English. A further syntactic feature of the discourse of particular relevance to this analytic task is the referencing of propositions developed earlier in a reasoning sequence in order to invoke these propositions again later in the same sequence. Making sense of such reasoning requires that these allusions be traced to their original referent so the concept reinvoked can be identified. One way speakers mark their references to earlier assertions is to repeat the metaphor in which the original proposition was cast. Thus, in the first of the two passages at hand, we see that Nan uses the metaphor KNOWING IS SEEING to talk about how married life begins. This metaphor is to tie together the argument of the entire passage, and its separate instantiations must be decoded and traced to their common referent. Nan opens her argument in this passage by establishing that she and her husband were naive about each other, not having "much sense of who each other were" at the time they got married. This is the first use of the KNOWING IS SEEING metaphor; Nan means not that they literally did not recognize each other but that she and her husband did not have much knowledge about each other at the outset. She plays out this use of the metaphor to dramatic effect when she goes on to note, "I really don't think that we, either of us, had examined each other" and " . . . I don't think I had said, 'Really, who is this Tom Harper, how can I describe him, what is he?'" In the next sentence, Nan makes clear exactly what about her husband she did not notice at the beginning: Tom's fit to her needs. In this comment - "I don't think that I had ever consciously done that. Examined my needs and to see if Tom'd fit them" - "examining" and "seeing" are derived, once more, from the KNOWING IS SEEING metaphor and stand for the processes of understanding involved in evaluating and deciding about a situation. The analysis provided in the last section shows that the other metaphor introduced here, A SPOUSE IS A FITTING PART, is but one in a larger category of metaphors reflecting the propositionschema, MARRIAGE IS MUTUALLY BENEFICIAL. In this latter meta-

v. :

182

NAOMI QUINN

phor, the fit of one spouse to the needs of the other allows each.tp fulfill these needs and hence derive the expected benefit from the relationship. Nan argues, more particularly, that some amount of misfit inevitably results from failure to examine goodness of fit to needs before'getting married. As pointed out in the previous section, that Nan means this argument to hold mutually for herself and her husband is indicated by her liberal use of reciprocals - "both," "either of us/'^each other"— to talk about their initial failure to examine the goodness of one's fit to the other's needs. • ^"-...•-•l,; i:_r That Nan means her assertions to be generalizations about marriage, not something peculiar to her own marital experience, is brought home by her summary, near the end of the passage: "Gee, people really do go into marriage, with their eyes, closed." This comment can only be interpreted once the KNOWING IS SEEING metaphor is understood to refer to its initial application to her own marriage; If your eyes are closed, you cannot see and hence you will fail to observe any misfit of the person you are marrying to the needs you have. Thus, that many people get married in this manner bears the inference that not only her marriage to Tom but also many other marriages result in a misfit of one spouse to the other's needs. Misfit to needs, then, represents mutual lack of marital benefit; having one's eyes closed and not examining one another and not looking to see who the other person is at the time one gets married all stand for lack of knowledge about this important aspect of marriage at its inception. Having made these two substitutions, we'can see that the argument so far, made explicit in the assertion, "I don't think I had ever done that. Examined my needs and to see if Tom'd fit them," takes the form: UNKNOWN AT THE OUTSET

> -MUTUALLY BENEFICIAL

A homegrown notation has been adopted to depict the reasoning embedded in passages of natural discourse such as this one. In this notation, the proposition-schemas that constitute terms in longer reasoning sequences continue to be represented in capital letters but in abbreviated form (i.e., ENDURING, MUTUALLY BENEFICIAL, UNKNOWN AT THE OUTSET, DIFFICULT, and EFFORTFUL). For ease of recognition, this abbreviation preserves the English sense of each proposition-schema rather than converting that schema into arbitrary symbols. The negation of a proposition is represented by a logical symbol commonly used for negation (~) at the front of that term. A right arrow represents a causal link connecting a proposition derived from one of the five propositionschemas, or its negation, to another proposition or its negation to create a complex schema. The direction of causality is from the left term to the right term, with the direction of the arrow. In the remark at hand, "I don't

CONVERGENT EVIDENCE FOR A MODEL OF MARRIAGE

1&3

think I had ever consciously done that. Examined my needs ;and td:s£e if Tom'd fit them," the direction of causality is revealed by.the syntax) X (in order) to Y, one of many syntactic devices for expressing causality in English. --••;.:>;•••• '•{:/•>. ^ v i * -b^w • The next step in Nan's argument rests on a further extension, of the KNOWING IS SEEING metaphor, in the identification of .things ;3I couldn't have foreseen." Again, the metaphor marks this comment as an allusion to the unexamined needs described earlier: Thus, to interpret the statement, "The things that have been difficult in the marriage I couldn't have foreseen," we must recognize that the unforeseen things stand for the unexamined needs Nan and her husband Tom turned out not to fit. We have already identified this misfitting part metaphor as belonging to a class of metaphors that stands for the benefits of marriage. No decoding is required of "the things that have been difficult in the marriage," a phrase that introduces the proposition-schema MARRIAGE IS DIFFICULT in nonmetaphorical language. Substituting the referent of "things unforeseen" and decoding the misfitting part metaphor, we see that "The things that have been difficult in the marriage I couldn't have foreseen," bears the interpretation: -MUTUALLY BENEFICIAL

> DIFFICULT

In this case, the direction of causality must be arrived at by inference. The sense of "foresee" requires that the difficulties in question were temporally preceded by the unexamined lack of fit to needs. Temporal order supports an inference of causal order; as Linde (this volume) observes, "the natural order of English is post hoc, ergo propter hoc.'" Marital benefits that were not forthcoming at the outset of this marriage led to subsequent difficulties. Finally, Nan concludes that it is "amazing that many marriages get to stay together, when you consider the way they do it." "The way they do it" is a clear reference to the assertion in the previous sentence, that people go into marriage with their eyes closed. As we have seen, the metaphor in this latter statement is one of marriage unknown at the outset. But "the way they do it" should lead to divorce; Nan uses a counterfactual construction to dramatize the seeming anomaly that so many marriages in fact do endure - as captured in the common metaphor of marriage as a well-made product that "stays together." Once the two metaphors in this conclusion are decoded to reveal it to be a statement about the relationship between initial ignorance of marriage and its ultimate enduringness, it remains only to reverse the counterfactual and specify the direction of causality. The logic of the assertion, "how amazing that many marriages get to stay together, when you consider the way they do it," is revealed to be:

UNKNOWN AT THE OUTSET

> -ENDURING

184

NAOMIQUINN

Here, causality is inferred from the syntax, Y when X, where X stands for the causal agent, and Vis what is caused. '-""'".^'''"..! What remains unexplained is the larger organization of the argument, vwhich allows this speaker to go on from her first assertion concerning what people do not know about each other when they marry to a conclusion about marital enduringness. To make sense of this leap and to reconstruct the full sequence of reasoning that could account for the final conclusion she reaches, it is necessary to assume that Nan has in mind a further proposition she does not make explicit. This additional proposition derives from a schema in which marital difficulty is a proximate cause for the failure of marriages to endure: DIFFICULT =

>

-ENDURING

Inserting this proposition-schema into the chain of argument, we see, gives an account of how the speaker must have reasoned to have produced this discourse sequence. Another strong ground for granting Nan's implicit assumption of this causal schema is that propositions of this form are articulated in other reasoning in the discourse under study. Nan herself makes this relation between difficulty and enduringness explicit in the next passage, when she says, in metaphorical language we have analyzed earlier, "We are going to be driven apart by all our problems." Further illustrations appear in the discourse of other interviewees; for example: 7H-1: I don't know, we just reached a kind of crisis in the relationship. At this point, there were a lot of tears and that was either make or break at that point. 4W-1: I think it's amazing that anybody stays married. I really have - that for people to live together day in and day out is an amazing struggle. The metaphors in which marital enduringness and difficulty are cast, in these two comments, are already familiar to the reader. These brief examples suggest that inserting an unstated assumption at this point in the analysis of Nan's reasoning is not arbitrary; there is plentiful evidence in the remainder of the discourse under study that speakers do make such a connection between marital difficulty and marital enduringness, The full sequence of reasoning that must be assumed, then, in order to allow the conclusion Nan reaches, is as follows: UNKNOWN AT THE OUTSET > -MUTUALLY BENEFICIAL -MUTUALLY BENEFICIAL > DIFFICULT [DIFFICULT « « = » -ENDURING] UNKNOWN AT THE OUTSET > -ENDURING Here, two final notational conventions are introduced. A line drawn below any set of proposition-schemas indicates that taken together, these schemas allow the further proposition-schema below the line. The proposition derived from this final schema, then, is the conclusion reasoned to: in this case, "How amazing that many marriages get to stay together when

:;

CONVERGENT EVIDENCE FOR A MODEL OF MARRIAGE

185^;

you consider the way they do it." Square brackets around"Wi^SJjS&i^'•.'.;-.' tion-schema indicate that it has not been made explicit in the argun#it|> " but that it must be assumed in order to arrive at the conclusion to ^rnch fe:; the speaker has reasoned. '''•'•':-'-:^':^iMhV^-'-y,-{ Thus, initial lack of knowledge about marriage leads to f^^^'M-l^h'y perience marital benefit, which leads to marital difficulty, which leads to divorce. The schematic structure allowing this longish causa! erminis, v readily available for reasoning about marriage. Moreover, chains of prpp^; ositions violating this structure would not make sense to us. No one would be likely to argue, for instance, that a marriage that was mutually beneficial was therefore difficult, or that one in which mutual benefit was not forthcoming was therefore likely to endure. Such chains of reasoning do not, in fact, occur in this discourse. The sequence of causally related propositionschemas displayed in the preceding paragraph seems to represent a widely shared understanding of how American marriage works. That other interviewees invoke the same chain of reasoning, or segments of this chain, in reasoning tasks similar to the one Nan has set herself, and that they, like Nan, may reason through to their conclusions without explicitly stating one or more of the propositions required to link together their argument suggests that not only are the separate proposition-schemas for each causal link in this chain available for reasoning about marriage but also that the sequence of linked proposition-schemas is itself a stable composite schema, available in its entirety. As observed earlier, in the second excerpt the complex causal propositionschema linking propositions about marital enduringness and marital difficulty is explicitly stated. A further proposition-schema, MARRIAGE IS EFFORTFUL, is introduced into Nan's argument in this passage. The excerpt illustrates the articulation of this new schema with the rest of a cultural model of marriage. Here, Nan sets about repairing the untenable conclusion she was left with in the first passage: Given the way people enter into it, how is any marriage to endure? Nan's dilemma stems from a centra! contradiction in how she and other Americans think about marriage. The analysis of metaphors for marriage appearing in these passages reveals that the statement "We are going to be driven apart by all our problems" contains two such metaphors: Marital difficulty is characterized as problems that must be "worked through," and marital enduringness is captured in a metaphor of two people attached to each other so securely that they must be "driven apart" to be separated. Causality is handled, in this assertion, by the syntactic construction, YbyX. Thus decoded, the statement reads: DIFFICULT

> -ENDURING

Nan goes on to argue, however, that marital difficulties need not be allowed to drive a couple apart. She proceeds by first disposing of one

186

NAOMI QUINN

possible solution: Leaving one marriage for another, she demonstrates, does not eliminate such difficulties. This is true because lack of knowledge about the person you are marrying inevitably leads to marital difficulties, -4 so that one marriage is likely to be no easier than the other: "If I know Tom as well as I know him and love him as much as I love him and still have this much trouble being married to him, what in thb world chance would I have of finding anybody else who would be any easier to be married to and I wouldn't know that person any better when t got - married him than I knew Tom." The schema underlying this assertion i s : . . : UNKNOWN AT THE OUTSET = = = > DIFFICULT / Causality is somewhat complex in this sentence, depending as it does on both the syntax of the sentence and the logical equivalence (her knowledge of a new husband would be the same as her knowledge of Tom when she married him) expressed at the end of it. Using upper-case letters to indicate logical relations and italics to indicate syntactic items, the causal structure of the argument can be seen to be: IF [ifX,

Y} AND [X' = X] THEN {ifX,

Y]

The X in this argument, that she did not know Tom when she married him, was asserted in the earlier passage and does not need restating here. This conclusion, that initial ignorance of marriage leads to subsequent marital difficulties, depends on a piece of reasoning also drawn from the preceding excerpt, which, once again, is-not explicitly restated minis one:, Since initial ignorance about each other leads to lack of one's fit to the other's needs, and thus to lack of mutual benefit, and since lack of benefit causes marital difficulty, then the consequence of this initial ignorance is subsequent difficulty. Represented in notation, the full sequence of reasoning on which Nan's assertion relies is: [UNKNOWN AT THE OUTSET = = > -MUTUALLY BENEFICIAL] [-MUTUALLY BENEFICIAL = > DIFFICULT] UNKNOWN AT THE OUTSET > DIFFICULT Because the notation that has been adopted is designed to represent causality as determinate, it is too crude to capture another feature of Nan's thinking that emerges at this point. A folk theory of probability, only hinted at here, enters into her argument by way of the likelihood assumption ". , . what in the world chance would I have . . . " - that any new husband she finds will fit her needs (and she his) equally as imperfectly as Tom. The strategy of remarrying is rejected ~ one might as well stay in the first marriage. The argument now goes on to establish how an enduring marriage can be achieved: through effort. You must, concludes Nan, learn or work

CONVERGENT EVIDENCE FOR A MODEL OF MARRIAGE

187

through "what actually you need to learn and work through to make the first marriage stick." Here, effort is cast in the metaphor of problem solving, and enduringness is captured in another familiar metaphor of marriage as a well-made product - one that "stick?." \; EFFORTFUL —

> ENDURING

•

In this statement, causality is made explicit in the syntax, X (in order) to Y, and rests on an entailment of the well-make, product metaphor that specified processes are requisite to the manufacture of any such product; It is necessary to, do X (in order) to make Y. ^••"-C-^'r.^,:':--^ . The last part of the passage restates this relation between effort and enduringness in the negative, with the purpose of clinching the argument against remarriage as an alternative strategy. To divorce and remarry someone new is to throw away the accumulated effort you have put into the first marriage. Moreover, an implication of the earlier assertion that one marriage is likely to be as mismatched and hence as difficult as another is that the effort required of any marriage is the same; no advantage is to be gained from starting over from the beginning. It follows that sufficient effort to make a marriage endure will never be accumulated: "If it took seven more years before you learned that much with the next person. Where - you know, where would you go?" (Nan has been married for seven years.) Learning "that much," in this remark, refers to "We have learned a lot," a reference, in turn, to "what actually you need to learn and work through" to make a marriage "stick," a few sentences earlier. This metaphor, as we saw in the previous section, invokes the effort entailed by learning and problem solving. Thus, the first term in Nan's conclusion, "if it took seven more years before you learned that much with the next person," stands for the lack of accumulated effort at the outset of a second marriage. As also shown in the last section, the second term, "Where would you go?" adopts a journey metaphor of indeterminate destination to suggest lack of marital enduringness. People who waste effort, we are cautioned, never get anywhere, a dictum as applicable to marriage as it is to problem solving or travel. The logic of this remark is: ~ EFFORTFUL

> -ENDURING

The syntax of causality here is if X, Y. Again, we ask how the two propositions about marriage developed in this passage, the first asserting an inverse relation between marital difficulty and marital enduringness and the second asserting a relation between enduringness and effort, make sense as a whole. Why does one lead the speaker to assert the other? Again, a term in the argument has been left implicit. This is a schema for the relation between effort and difficulty. That difficulty cannot be overcome without effort is so well understood, we speculate, that Nan takes it for granted. By inserting this proposi-

188

NAOMI QUINN

tion-schema, we see how she must have reasoned to her final conclusion. The overall pattern of her argument, then, is: [-EFFORTFUL DIFFICULT = -EFFORTFUL

>

> DIFFICULT] -ENDURING > ~ENDURING

Reasoning in these passages tells a story about marriage that is confirmed by the reasoning sequences in the larger body of discourse under analysis. A large part of this story has now emerged. Even though people ordinarily do not know "what they are getting into" when they marry, they do have certain powerful expectations. They expect marriage to be an enduring relationship, but at the same time they view it as a voluntary relationship, the continuance of which is contingent on its benefit.These two assumptions, one about marriage and one drawn from our shared understanding of voluntary relationships, pose a contradiction. The contradiction is realized when, as is typical, the expected benefits of marriage do not automatically materialize. This problem is reconceptualized in the manner Americans think about many things: Lack of marital benefit becomes a difficulty to be overcome in the enterprise of making a marriage endure. Succeeding at this as at any task is largely a matter of effort. Taken together, the two reasoning sequences presented here reflect a widely held set of expectations about how the prototypical American marriage goes. A notable feature of the reasoning embedded in this and other talk about marriage is the emptiness of the causal connections posited between terms of this cultural model of marriage. Effort is required for enduringness. Difficulties result in a marriage that does not endure. Lack of mutual benefits leads to marital difficulties. And so forth. However, the nature of causality in each case goes unspecified. The simple arrow used in the notation would seem to be an accurate representation of the causality of reasoning. (An exception was the inability of the notation to handle Nan's assertion about equal likelihood.) It is as if speakers invoke these causal connections to reason with, abstracting for this purpose a kind of nonspecific causality out of a lot of more detailed knowledge about how the world works. We may speculate that the speaker assumes the hearer to share this latter knowledge of why lack of benefits might lead to marital difficulty or why effort might overcome such difficulty, for example, This "intersubjective sharing," in D'Andrade's term (this volume), would explain how such knowledge can be dropped out of the argument under construction without affecting its intelligibility or persuasiveness. However, the analyst intent on reconstructing the full cultural model of marriage may wish to fill in its details. To do so, we must pursue still a third trail of evidence in this body of discourse: scattered commentary in which the implicit assumptions nested within sequences of reasoning about marriage are addressed more explicitly and spelled out more fully.

CONVERGENT EVIDENCE FOR A MODEL OF MARRIAGE

>189

Nested cultural models Some causal connections are so well understood that they rarely if ever bear comment. Such, ifor example, seems to be the nature of the connection between difficulty and effort. Understanding of this causal link is transported into the world of marriage from bur understanding of the physical world. Perhaps because it is based on direct and repeated physical experience, the knowledge that performance of difficult activities requires effort seem perfectly obvious to us - as Whorf (1941:85) speculated about the more general idea 6n which this one rests: that th^expenditu^^f energy produces effects. Other causal connections ''between the terms of the cultural model are not so taken-for-granted, however.;;.Such, for example, is the nature of the link between MARITAL DIFFICULTY and MUTUAL BENEFIT. Why should difficulties arise over the'attainment of such benefits? The discourse at hand has already offered a clue. An important kind of benefit people expect out of marriage is need fulfillment. This expectation'is reflected in Nan's observations, in the first passage, about examining her own needs and her husband's fit to them. In our folk psychology of human needs, certain needs, such as those for sex, love, companionship, support, understanding, can only be fulfilled by other people. Americans expect that the person one marries should fulfill most, if not all, of these kinds of needs. This expectation is sometimes stated explicitly, as in the comments these two husbands make about their wives: 7H-2: I haven't met a single woman since Beth, at all, who would ever come close to matching her in terms of, what she can do for me. What another woman could - for how she could fulfill me. And I - and understand me, particularly. Beth understands me very well. She knows what makes me tick. 5H~9: Maybe it's the combination that there is a - there's an intellectual stimulation with one another, there's an emotional stimulation with one another, there's a child-bearing stimulation with one another, or wrestling with great issues of the world, and so I think Eileen encapsulates for me an ongoing growth potential for me and all that gambit and vice versa, I believe so. And I think we have found parts of that in many other people many times, but no one who we felt could replace in that sense. It is this expectation of need fulfillment that makes sense of the final theme in Americans' story about marriage - that it is to be jointly lived. Fulfillment of needs in marriage supposes a substantial amount of physical proximity, emotional intimacy, and coordination of daily activities. It becomes clear why dual metaphors of a spouse as a fitting or irreplaceable part - both joined to one and beneficial to one - are such apposite ones for characterizing the marital relationship. By this folk psychology of needs and their fulfillment, people have different needs and are endowed with differential capabilities for fulfilling these needs in other people. Although such capabilities are, to a certain

190

NAOMI QU1NN

extent at least, learnable, individuals' differing natural endowments arid divergent histories insure that at the outset of a marriage each spouse's needs and the other's capabilities will almost certainly be mismatched. This likelihood is heightened, as Nan explains, if people enter marriage, as they * are apt to do, unobservantly or precipitously, without prior knowledge about each other's needs. Some interviewees say they began marriage ignorant even of the idea that it involves need fulfillment. Moreover, as other interviewees point out, individuals may change over the course of a marriage, often developing or discovering new needs that a spouse's capabilities cannot easily be stretched to meet. They speak of "growing out of touch with each other," "growing apart," or "going in a different direction," of "holding each other back," or a wife who is "hoIcUrig; me up," of coming to "a place where we have to separate," or being "at a point in growth and who we are that says, 'Okay, we need not and we probably should not perpetuate this.'" Some clear statements of the model of need fulfillment underlying such observations are: •.-->,_ 4W-3: I think we are committed to making our marriage work. Making the effort to do the best we can until - unless at some point doing the best we can doesn't work, simply doesn't work. Doesn't meet our needs, doesn't make anybody happy and that kind of thing. 7H~1: I don't think - when a marriage gets to the point where you're really holding down the other person, you're really restricting them, it's not worth sticking together because life's too precious to waste your time, with another person. Unless they're really fulfilling you on an emotional level. Thus, people expect marriage to be mutually beneficial, but, as we have seen, not automatically so. Some "misfit" of each spouse to the needs of .. the other, either at the outset of marriage, or later in its course, is to be anticipated. What is difficult about marriage then, by this folk psychological theory, is fulfillment of a spouse's needs. The larger body of interviews from which these passages are drawn contains many other passages in which the difficulties of need fulfillment are elaborated. It is difficult, interviewees say, to communicate one's needs and to understand the needs one's spouse is communicating. It is difficult, sometimes impossible, to learn to fill these needs even when they have been comprehended. It is also difficult to sacrifice one's own desires, as the fulfillment of another's needs often necessitates. Because of these difficulties, deriving mutual benefit out of a marital relationship so the relationship will succeed is not an easy task. Thus, only by deciphering certain American cultural understandings of the self can we fathom the connection in Americans' thinking about marriage between its benefits and its difficulties. The passages we analyze , in this paper give only a sampling of that folk psychology; discussions of needs and their fulfillment arise naturally in the course of talk about marriage and are scattered throughout the entire body of discourse under ^-

CONVERGENT EVIDENCE FOR A MODEL OF MARRIAGE

191

study. This sporadic evidence must be drawn together to pemit reconstruction of the cultural model of needs and their fulfillment sketched here '•- jso that the application of this folk psychology to marriage can be appreciated. •••••' Americans' model of one piece of the world, marriage, contains within it assumptions drawn from models of other domains, some df which; like the folk physics of difficult activities, the folk social psychology pf voluntary relationships, the folk theory of probability, and: the folk psychology -of human needs, are of wide.applicability,.available,for recpmbination with more special-purpose models to structure not only the domain of marriage but also multiple domains of our experience. Because our cultural knowledge is organized in this hierarchical way (D'Andrade this volume), models nested within models, we must follow the explanatory trail left in discourse, which leads us from understandings about marriage to understandings about need fulfillment, for example. We must then retrace our steps to establish the implications of the nested cultural model for the cultural model under investigation.

Notes \. An earlier version of this paper, under the title "What Discourse Can Tell about Culture: Convergent Evidence for a Cultural Model of American Marriage," was delivered at the 82nd Annual Meeting of the American Anthropological Association, November, 1983, in a symposium organized by Susal Gal entitled Making Conversation: Culture, Discourse Style, and Linguistic Structure. The revised version has benefited from the suggestions of Dorothy Holland. The research project on which this paper is based has been made possible by National Institute of Mental Health research grant No. 1 ROl MH330370-01, National Science Foundation research grant No. BNS-8205739, and a stipend from the Institute for Advanced Study, Princeton, New Jersey. People who made the project successful are Rebecca Taylor, a talented research assistant who conducted a large portion of the interviews, and Laurie Moore, who also interviewed, as well as Phyllis Taylor, Donna Rubin and Georgia Hunter, who transcribed the interviews with skill. I am particularly indebted to Georgia, whose dedication to the enormous transcription task was heroic. I cannot adequately thank "Nan" and the other 21 anonymous wives and husbands who participated in the long interview process and left me with a lasting appreciation for their unique and creative ways of understanding their marriages. 2. All interviewees were native-born Americans who spoke English as a first language. All were married during the period of their interviews, all in their first marriages. Beyond these commonalities, they were selected to maximize diversity with regard to such obvious differences as their geographic and ethnic origins, their occupational and educational backgrounds, and the age of their marriages. No claim is made for the statistical representativeness of the people interviewed, nor would representativeness with respect to various sociological characteristics of the middle-sized southern town in which all interviewees resided even have been feasible for a sample so small. The study aimed to investigate how people organize knowledge rather than how any particular

192

3. ^ '* ; ,

4.

NAOMIQUINN

feature of this knowledge varies across sociological categories such as gender, ethnicity, religion, or class. ,, This is a Active name, of course. The code at the beginning of this and later interview segments contains, in order, an interviewee identification number, a W or an H to indicate a wife or a husband, and the number of the interview from which that segment was drawn in the sequence of interviews with' that person. Husbands do not have the same identification numbers as their wives. As in the second segment from Nan's interview, comments or questions interjected by the interviewer are prefaced by an /, and resumption of the interviewee's part of the conversation is indicated by his or her identification number and letter. This and other interview segments reproduced in this paper have been regularized for stammers, stutters, elisions, slips of the tongue, and hesitations. . To characterize a given cultural model as "folk social psychology" or (later in this paper) "folk psychology," "folk physics," or "folk probability theory," is to invite the observation that our ordinary everyday ideas about a given phenomenon may not correspond, although they may be related, to their counterpart in current scientific theory. Although this relationship between folk and scientific models is not pursued in this paper, other papers in this volume discuss how widely shared cultural understandings may be 'incorrect" from the stance of scientific explanation and evidence (Collins & Gentner, Kempton), may draw on existing social scientific models (Linde), and, as is likely the case with the folk social psychology of exchange in voluntary relationships discussed here, may contribute unanalyzed assumptions to those social scientific theories (Kay).

References Hutchins, E. . . . . . . . 1980. Culture and Inference: A Trobriand Case Study. Cambridge, Mass.: Harvard University Press. Lakoff, G. 1984. Classifiers as a reflection of mind: A cognitive model approach to prototype theory. Berkeley Cognitive Science Report No. 19. Berkeley: University of California Institute of Human Learning. Lakoff, G. and M. Johnson 1980. Metaphors We Live By. Chicago: University of Chicago Press. Quinn, N. 1985a. American marriage through metaphors: A cultural analysis. North Carolina Working Papers in Culture and Cognition No. 1. Durham, NC: Duke University Department of Anthropology. 1985b. American marriage and the folk social psychology of need fulfillment. North Carolina Working Papers in Culture and Cognition No. 2. Durham, NC: Duke University Department of Anthropology. n.d. American marriage: A cultural analysis. Unpublished manuscript in preparation, Department of Anthropology, Duke University, Durham, N.C. Whorf, B. L. 1941. The relation of habitual thought and behavior to language. In Language, Culture, and Personality: Essays in Memory of Edward Sapir, L. Spier, ed. Menasha, Wise. Sapir Memorial Publication Fund. Pp. 75-93.

PART I I I The role jof metaphor and analogy .: in representing knowledge of presupposed worlds

8 The cognitive model of anger inherent in American English1

George Lakoff & Zolian Kovecses •-.

Emotions are often considered to be feelings alone, and as such they are viewed as being devoid of conceptual content. As a result, the study of emotions is usually not taken seriously by students of semantics and conceptual structure. A topic such as The Logic of Emotions would seem oh this view to be a contradiction in terms, since emotions, being devoid of conceptual content, would give rise to no inferences at all, or at least none of any interest. We would like to argue that the opposite is true, that emotions have an extremely complex conceptual structure, which gives rise to wide variety of nontrivial inferences. The conceptualization of anger At first glance, the conventional expressions used to talk about anger seem so diverse that finding any coherent system would seem impossible. For example, if we look up anger in, say, Roget's University Thesaurus, we find about three hundred entries, most of which have something or other to do with anger, but the thesaurus does not tell us exactly what. Many of these are idioms, and they too seem too diverse to reflect any coherent cognitive model. Here are some example sentences using such idioms: He lost his cool. She was looking daggers at me. I almost burst a blood vessel. He was foaming at the mouth. You're beginning to get to me. You make my blood boil. He's wrestling with his anger. Watch out! He's on a short fuse. He's just letting off steam. Try to keep a grip on yourself. Don't fly off the handle. When I told him, he blew up. He channeled his anger into something constructive. He was red with anger.

196

t

GEORGE LAKOFF & ZOLTAN KOVECSES

He was blue in the face. He appeased his anger. He was doing a slow burn. He suppressed his anger. She kept bugging me. When I told my mother, she had a cow.

./

:

*..

What do these expressions have to do with anger, and what do they have to do with each other? We will be arguing that they are not random. When we look at inferences among these expressions, it becomes clear that there must be a systematic structure of some kind. We know, for example, that someone who is foaming at the mouth has lost his cool. We know that someone who is looking daggers at you is likely to be doing a slow burn or be on a short fuse. We know that someone whose blood is boiling has not had his anger appeased. We know that someone who has channeled his anger into something constructive has not had a cow, How do we know these things? Is it just that each idiom has a literal meaning and the inferences are based on the literal meanings? Or is there something more going on? What we will try to show is that there is a coherent conceptual organization underlying all these expressions, and that much of it is metaphorical and metonymical in nature. METAPHOR AND METONYMY

The analysis we are proposing begins with the common cultural model of the physiological effects of anger: •• THE PHYSIOLOGICAL EFFECTS OF ANGER ARE INCREASED BODY HEAT, INCREASED INTERNAL PRESSURE (BLOOD PRESSURE, MUSCULAR PRESSURE), AGITATION, AND INTERFERENCE WITH ACCURATE PERCEPTION. AS ANGER INCREASES, ITS PHYSIOLOGICAL EFFECTS INCREASE. THERE IS A LIMIT BEYOND WHICH THE PHYSIOLOGICAL EFFECTS OF ANGER IMPAIR NORMAL FUNCTIONING. We use this cultural model in large measure to tell when someone is angry on the basis of their appearance - as well as to signal anger, or hide it. In doing this, we make use of a general metonymic principle: THE PHYSIOLOGICAL EFFECTS OF AN EMOTION STAND FOR THE EMOTION Given this principle, the cultural model given above yields a system of metonymies for anger:

THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

197

/•;• BODY HEAT: -• ... h~:^^\y ^ t^^V^.W:Don't get hot under the collar. :-: Billy's a hothead, >..:.-;o:i.';"J' ^ M ; y , They were having a heated argument. ';.:^V;,.:/-••. . When the cop gave her a ticket, she got allhoi'and'pothered and .,. , .„• started cursing. h . ...•,-: '.^••-••••••'••••. -^ ,•.'•.,.',••.•' ^,;;::..:JNTORNAL.PRESSU^ ,. Don't get a hernia! v; ' : V• , ^•!•' •ur--:r,\ru'i-:--.:,'-.!^r,-:,' •:'•••-h^l'" • W h e n I f o u n d p u t , I .almost burst a blood vessel.:;'. ;;•;••, He almost had a hemorrhage. ; ' ^ ( V • , iv" •'"•i'"'I. Increased body heat and/or blood pressure is assumed to cause redness in the face and neck area, and such redness can also metonymically indicate anger. REDNESS IN FACE AND NECK AREA: She was scarlet with rage. He got red with anger. He was flushed with anger. AGITATION: She was shaking with anger. I was hopping mad. He was quivering with rage. He's all worked up. She's all wrought up. -.*';-.v INTERFERENCE WITH ACCURATE PERCEPTION: She was blind with rage. I was beginning to see red. I was so mad I couldn't see straight. Each of these expressions indicate the presence of anger via its supposed physiological effects. The cultural model of physiological effects, especially the part that emphasizes HEAT, forms the basis of the most general metaphor for anger: ANGER IS HEAT. There are two versions of this metaphor, one where the heat is applied to fluids, the other where it is applied to solids. When it is applied to fluids, we get: ANGER IS THE HEAT OF A FLUID IN A CONTAINER. The specific motivation for this consists of the HEAT, INTERNAL PRESSURE, and AGITATION parts of the cultural model. When ANGER IS HEAT is applied to solids, we get the version ANGER IS FIRE, which is motivated by the HEAT and REDNESS aspects of the cultural theory of physiological effects. As we will see shortly, the fluid version is much more highly elaborated. The reason for this, we surmise, is that in our overall conceptual system we have the general metaphor:

•198

... *• . . .

GEORGE LAKOFF& ZOLTAN KOVECSES

THE BODY IS A CONTAINER FOR THE EMOTIONS , He was fitted with a n g e r s .n -:• ;•>•.",-i VM\ X"\ V'^: / She couldn't contain her joy. -...,>/;;\\ v ,s / She was brimming with rage. .^ \ .-:•;; ; . < ; - , , /Try'to getyour angsvput of your.system. •-._ „.•/;

The ANGER IS HEAT metaphor, when applied to fluids, combines with the metaphor THE BODY IS A CONTAINER FOR THE EMOTIONS to yield the central metaphor of the system: ANGER IS THE HEAT OF A FLUID IN A CONTAINER You make my blood boil. Simmer down! •-. *. .-.„_ -:\ I had reached the boiling point. Let him stew. A historically derived instance of this metaphor is: She was seething with rage. Although most speakers do not now use seethe to indicate physical boiling, the boiling image is still there when seethe is used to indicate anger. Similarly, pissed off is used only to refer to anger, not to the hot liquid under pressure in the bladder. Still, the effectiveness of the expression seems to depend on such an image. When there is no heat the liquid is cool and calm. In the central metaphor, cool and calmness corresponds ..to lack of anger. Keep cool.. Stay calm. As we will see shortly, the central metaphor is an extremely productive one. There are two ways in which a conceptual metaphor can be productive. The first is lexical. The words and fixed expressions of a language can code, that is, be used to express aspects of, a given conceptual metaphor to a greater or lesser extent. The number of conventional linguistic expressions that code a given conceptual metaphor is one measure of the productivity of the metaphor. In addition, the words and fixed expressions of a language can elaborate the conceptual metaphor. For example, a stew is a special case in which there is a hot fluid in a container. It is something that continues at a given level of heat for a long time. This special case can be used to elaborate the central metaphor. "Stewing" indicates the continuance of anger over a long period. Another special case is "simmer," which indicates a low boil. This can be used to indicate a lowering of the intensity of anger. Although both of these are cooking terms, cooking plays no metaphorical role in these cases. It just happens to be a case where there is a hot fluid in a container. This is typical of lexical elaborations.

THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

W

Let us refer to the HEAT OF FLUID IN A CONTAINER as the source domain of the central metaphor, and to ANGER as the target domain. We usually have extensive knowledge about source domains. A second WayJin which a conceptual metaphor can be productive is that it can carry over details of that knowledge from the source domain to the target domain. We will refer to such carryovers as metaphorical entailments. Such ; entailments are part of our conceptual system. They constitute elaborations of conceptual metaphors. The central metaphor has a rich system of metaphorical entailments. For example, one thing we;know about hot fluids is that, when they start to boil, the fluid goes upward. This gives rise to the entailment: .•'••-, WHEN THE INTENSITY OF ANGER INCREASES, THE FLUID RISES His pent-up anger welled up inside him. She could feel her gorge rising. We got a rise put of him. My anger kept building up inside me. Pretty soon I was in a towering rage. We also know that intense heat produces steam and creates pressure on the container. This yields the metaphorical entailments: INTENSE ANGER PRODUCES STEAM She got all steamed up. Billy's just blowing off steam. I was fuming. . INTENSE ANGER PRODUCES PRESSURE ON THE CONTAINER He was bursting with anger. I could barely contain my rage. I could barely keep it in anymore. A variant of this involves keeping the pressure back: I suppressed my anger. He turned his anger inward. He managed to keep his anger bottled up inside him. He was blue in the face. When the pressure on the container becomes too high, the container explodes, This yields the entailment: WHEN ANGER BECOMES TOO INTENSE, THE PERSON EXPLODES When I told him, he just exploded. She blew up at me. We won't tolerate any more of your outbursts.

GEORGE LAKOFF & ZOLTAN KOVECSES

200

This can be elaborated, using special cases: Pistons: He blew a gasket. Volcanos: She erupted. Electricity: I blew a fuse: '"••- Explosives: She's on a short fuse. Bombs: That really set me off.

,'*•• ,

' -

'

In an explosion, parts of the container go up in the air. "'

WHEN A PERSON EXPLODES, PARTS OF HIM GO UP IN THE AIR. I blew my stack. I blew my top. She flipped her lid. He hit the ceiling. I went through the roof.

When something explodes, what was inside it comes out. WHEN A PERSON EXPLODES, WHAT WAS INSIDE HIM COMES OUT His anger finally came out. Smoke was pouring out of his ears. This can be elaborated in terms of animals giving birth, where something that was inside causing pressure bursts out: WHEN A PERSON EXPLODES, WHAT WAS INSIDE HIM COMES OUT She was having kittens. My mother will have a cow when I tell her. Let us now turn to the question of what issues the central metaphor addresses and what kind of ontology of anger it reveals. The central metaphor focuses on the fact that anger can be intense, that it can lead to a loss of control, and that a loss of control can be dangerous. Let us begin with intensity. Anger is conceptualized as a mass, and takes the grammar of mass nouns, as opposed to count nouns. Thus, you can say: How much anger has he got in'him? but not: *How many angers does he have in him? Anger thus has the ontology of a mass entity, that is, it has a scale indicating its amount, it exists when the amount is greater than zero and goes out of existence when the amount falls to zero. In the central metaphor, the scale indicating the amount of anger is the heat scale. But, as the central metaphor indicates, the anger scale is not open-ended; it

THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

201

has a limit. Just as a hot fluid in a closed contamer can drily takei'sb much heat before it explodes, so we conceptualize the anger scale as having a; limit point. We can only bear so much anger, before we explode', that; is) lose control. This has its correlates in our cultural theory of physiological effects. As anger gets more intense the physiological effects, increase and ,, those increases interfere with our normal functioning. Body.heat, blood.... pressure,, agitation, and interference with perception cannot increase without limit before our ability to function normally becomes .seriously; impaired and we lose control over our functioning. In the; cultural model of anger, loss of control is dangerous, both to the angry.person and to those around him. In the central metaphor, the danger of loss' of control is understood as the danger of explosion. The structural aspect of a conceptual metaphor consists of a set of correspondences between a source domain and a target domain. These correspondences can be factored into two types: ontological and epistemic. Ontological correspondences are correspondences between the entities in the source domain and the corresponding entities in the target domain. For example, the container in the source domain corresponds to the body in the target domain. Epistemic correspondences are correspondences between knowledge about the source domain and corresponding knowledge about the target domain. We can schematize these correspondences between the FLUID domain and the ANGER domain as follows: Source: HEAT OF FLUID IN CONTAINER Target: ANGER Ontological Correspondences: The container is the body. The heat of fluid is the anger. The heat scale is the anger scale, with end points zero and limit. Container heat is body heat. Pressure in container is internal pressure in the body. Agitation of fluid and container is physical agitation. The limit of the container's capacity to withstand pressure caused by heat is the limit on the anger scale. Explosion is loss of control. Danger of explosion is danger of loss of control. Coolness in the fluid is lack of anger. Calmness of the fluid is lack of agitation. Epistemic correspondences: Source: The effect of intense fluid heat is container heat, internal pressure, and agitation. Target: The effect of intense anger is body heat, internal pressure, and agitation.

202

GEORGE LAKOFF & ZOLTAN KOVECSES

Source; When the fluid is heated past a certain limit, pressure increases to the point at which the container explodes.. , . , , ;; •>;•_.•• •_••. ; Target; When anger increases past a certain limit, pressure increases to •t the point at which the person loses control. . .j n: Source: An explosion is damaging to the container and dangerous to bystanders. ' Target; A loss of control is damaging to an angry person and dangerous to other people. Source: An explosion may be prevented by the application of sufficient force and energy to keep the fluid in. Target: A loss of control may be prevented by the application of sufficient force and energy to keep the anger in. Source: It is sometimes possible to control the release of heated fluid for either destructive or constructive purposes; this has the effect of lowering the level of heat and pressure. Target: It is sometimes possible to control the release of anger for either destructive or constructive purposes; this has the effect of lowering the level of anger and internal pressure. The latter case defines an elaboration of the entailment WHEN A PERSON EXPLODES, WHAT WAS INSIDE HIM COMES OUT:

ANGER CAN BE LET OUT UNDER CONTROL He let out his anger. I gave vent to my anger. Channel your anger into something constructive. He took out his anger on me. ' So far, we have seen that the cultural theory of physiological reactions provides the basis for the central metaphor, and that the central metaphor characterizes detailed correspondences between the source domain and the target domain - correspondences concerning both ontology and knowledge. At this point, our analysis enables us to see why various relationships among idioms hold. We can see why someone who is in a towering rage has not kept cool, why someone who is stewing may have contained his anger but has not gotten it out of his system, why someone who has suppressed his anger has not yet erupted, and why someone who has channeled his anger into something constructive has not had a cow. Let us now turn to the case where the general ANGER IS HEAT metaphor is applied to solids: ANGER IS FIRE Those are inflammatory remarks. She was doing a slow burn. He was breathing fire. Your insincere apology just added fuel to the fire.

THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

yA}$^0^$$]':

After the argument, Dave was smoldering for .6ays^^.-i^ip^^fM0^>'y That kindled my ire. *'''•''••• ••'•'•^r'M&M$ffi£$£'-PBoy, am I burned up! ; -.-.; ;••••.;•. :HI&&$*£££?''•{ He was consumed by his anger. ,•-•• ; :,-•./- ••-.: :.•; ^soffiffiQ.,, This metaphor highlights the cause of anger (kindle, intlarne), tfre interim:;;>; sity and duration (smoldering, slow burn, burrieci up), the danger to others '"•'. (breathing fire), and the damage tq the angry person (consumed). The corresponderices in ontology are as follows: .>',,,:, Source: FIRE Target: ANGER

'"".''

The fire is anger. The thing burning is the angry person. The cause of the fire is the cause of the anger. The intensity of the fire is the intensity of the anger. The physical damage to the thing burning is mental damage to the angry person. The capacity of the thing buring to serve its normal function is the capacity of the angry person to function normally. An object at the point of being consumed by fire corresponds to a person whose anger is at the limit. The danger of the fire to things nearby is danger of the anger to other people. ... . The correspondences in knowlege are: Source: Things can burn at low intensity for a long time and then burst into flame. Target: People can be angry at a low intensity for a long time and then suddenly become extremely angry. Source: Fires are dangerous to things nearby. Target: Angry people are dangerous to other people. Source: Things consumed by fire cannot serve their normal function. Target: At the limit of the anger scale, people cannot function normally. Putting together what we have done so far, we can see why someone who is doing; a' Slow burn has not hit the ceiling yet, why someone whose anger is bottled up is not breathing fire, why someone who is consumed by anger probably cannot see straight, and why adding fuel to the fire . might just cause the person you are talking to to have kittens. THE OTHER PRINCIPAL METAPHORS

As we have seen, the ANGER IS HEAT metaphor is based on the cultural model of the physiological effects of anger, according to which increased body heat is a major effect of anger. That cultural model also maintains

204

GEORGE LAKOFF & ZOLTAN KOVECSES

that agitation is an important effect. Agitation is also an important part of our cultural model of insanity. According to this view, people who are insane are unduly agitated - they go wild, start raying, flail their,arms, foam at the mouth, and so on. Correspondingly, these physiological efJfiScts can stand, metonymically, for insanity. One can indicate that someone is insane by describing him as foaming at the mouth, raving, going wild, for example. The overlap between the cultural models of the effects of anger and the effects of insanity provides a basis for the metaphor: ANGER IS INSANITY I just touched him, and he went crazy. You're driving me nuts! When the umpire called him out on strikes, he went bananas. One more complaint and I'll go berserk. He got so angry, he went out of his mind. When he gets angry, he goes bonkers. She went into an insane rage. If anything else goes wrong, I'll get hysterical. Perhaps the most common conventional expression for anger came into English historically as a result of this metaphor: I'm mad! Because of this metaphorical link between insanity and anger, expressions that indicate insane behavior can also indicate angry behavior. Given the metonymy INSANE BEHAVIOR STANDS FOR INSANITY and the metaphor ANGER IS INSANITY, we get the metaphorical metonymy: INSANE BEHAVIOR STANDS FOR ANGER When my mother finds out, she'll have a fit. When the ump threw him out of the game, Billy started foaming at the mouth. He's fit to be tied. He's about to throw a tantrum. Violent behavior indicative of frustration is viewed as a form of insane behavior. According to our cultural model of anger, people who can neither control nor relieve the pressure of anger engage in violent frustrated behavior. This cultural model is the basis for the metonymy: VIOLENT FRUSTRATED BEHAVIOR STANDS FOR ANGER He's tearing his hair out! If one more thing goes wrong, I'll start banging my head against the wall. The loud music next door has got him climbing the walls! She's been slamming doors all morning.

THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

2Q5

The ANGER IS INSANITY metaphor has the following correspondences: Source: INSANITY / Target: ANGER '. The cause of insanity is the cause of anger. Becoming insane is passing the limit point on the anger scale. Insane behavior is angry behavior. Source: An insane person cannot function normally. Target: A person who is angry beyond the limit point cannot function normally. Source: An insane person is dangerous to others. Target: A person who is angry beyond the limit point is dangerous to others. At this point, we can see a generalization. Emotional effects are understood as physical effects. Anger is understood as a form of energy. According to our cultural understanding of physics, when enough input energy is applied to a body, the body begins to produce output energy. Thus, the cause of anger is viewed as input energy that produces internal heat (output energy). Moreover, the internal heat can function as input energy, producing various forms of output energy: steam, pressure, externally radiating heat, and agitation. Such output energy (the angry behavior) is viewed as dangerous to others, in the insanity metaphor, insanity is understood as a highly energized state, with insane behavior as a form of energy output. All in all, anger is understood in our cultural model as a negative emotion. It produces undesirable physiological reactions, leads to an inability to function normally, and is dangerous to others. The angry person, recognizing this danger, views his anger as an opponent.

ANGER IS AN OPPONENT (IN A STRUGGLE) I'm struggling with my anger. He was battling his anger. She fought back her anger. I've been wrestling with my anger all day. I was seized by anger. I'm finally coming to grips with my anger. He lost control over his anger. Anger took control of him. He surrendered to his anger. He yielded to his anger. I was overcome by anger. Her anger has been appeased.

206

GEORGE LAKOFF & ZOLTAN KOVECSES

The ANGER IS AN OPPONENT metaphor is constituted by the follow? ing correspondences: ' - ' . , . . . ' : / . ' :.. Source: STRUGGLE :, ^/^ ,v , ;:farget; ANGER The opponent is anger." -• Winning is controlling anger. . / ; Losing is having anger control you. Surrender is allowing anger to take control of you. The pool of resources needed for winning is the energy heeded to control anger. • 'r-rr••-.••.-, One thing that is left out of this account so far is what constitutes "appeasement." To appease an opponent is to give in to his demands. This suggests that anger has demands. We will address the question of what these demands are below. The OPPONENT metaphor focuses on the issue of control and the danger of loss of control to the angry person himself. There is another metaphor that focuses on the issue of control, but whose main focus is the danger to others. It is a very widespread metaphor in Western culture, namely, PASSIONS ARE BEASTS INSIDE A PERSON. According to this metaphor, there is a part of each person that is a wild animal. Civilized people are supposed to keep that part of them private, that is, they are supposed to keep the animal inside them. In the metaphor, loss of control is equivalent to the animal getting loose. And the behavior of a person who has lost control is the behavior of a wild animal. There are versions of this metaphor for the various passions - desire, anger, and so forth. In the case of anger, the beast presents a danger to other people. ANGER IS A DANGEROUS ANIMAL He has & ferocious temper. He has & fierce temper. It's dangerous to arouse his anger. That awakened my ire. His anger grew. He has a monstrous temper. He unleashed his anger. Don't let your anger get out of hand. He lost his grip on his anger. His anger is insatiable. An example that draws on both the FIRE and DANGEROUS ANIMAL metaphors is: He was breathing fire. The image here is of a dragon, a dangerous animal that can devour you with fire.

^

THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

207

- The DANGEROUS ANIMAL metaphor portrays anger as a sleeping animal that is dangerous to awaken; as something that can grow and thereby become dangerous; as something that has to beheld back; and as something with a dangerous appetite. Here are the correspondences that'' constitute'the metaphor, . . . ''..'f : "'':^^:li'\ Source; DANGEROUS ANIMAL Target; A N G E R ;; •-•••^-.-.-•- • ••.-;-••;

... -.-••;:•••••,---I

;;;••:,•,,=-v ^vc;"

The dangerous animal is the anger. The animal's getting loose is loss of control of anger. \. " „ The owner of the dangerous animal is the angry person.'.''['•"• Sleeping for the animal is anger near the zero level. Being awake for the animal is anger near the limit. Source: It is dangerous for a dangerous animal to be loose. Target: It is dangerous for a person's anger to be out of control. Source: A dangerous animal is safe when it is sleeping and dangerous when it is awake. Target: Anger is safe near the zero level and dangerous near the limit. Source: A dangerous animal is safe when it is very small and dangerous when it is grown. Target: Anger is safe near the zero level and dangerous near the limit. Source: It is the responsibility of a dangerous animal's owner to keep it under control. Target: It is the responsibility of an angry person to keep his anger under control. Source; It requires a lot of energy to control a dangerous animal. Target: It requires a lot of energy to control one's anger. There is another class of expressions that, as far as we can tell, are instances of the same metaphor. These are cases in which angry behavior is described in terms of aggressive animal behavior. ANGRY BEHAVIOR IS AGGRESSIVE ANIMAL BEHAVIOR He was bristling with anger. That got my hackles up. He began to bare his teeth. That ruffled her feathers. She was bridling with anger. Don't snap at me! He started snarling. Don't bite my head off! Why did you jump down my throat? Perhaps the best way to account for these cases would be to extend the ontological correspondences of the ANGER IS A DANGEROUS ANIMAL metaphor to include:

208

GEORGE LAKOFF & ZOLTAN KOVECSES

..:„,;,' The aggressive behavior of the dangerous animal is angry : : r behavior. '•:'•' '• A-•:•.,. . If we do this, we can account naturally for the fact that these expressions > indicate anger. They would do so via a combination of metaphor,'and metonymy, in which the aggressive behavior metaphorically corresponds to angry behavior, which in turn metonymically stands for anger. For example, the snarling of the animal corresponds to the angry verbal behavior of the person, which in turn indicates the presence of anger. Aggressive verbal behavior is a common form of angry behavior, as snap, growl, snarl, and so forth indicate. We can see this in a number of cases outside of the animal domain: : AGGRESSIVE VERBAL BEHAVIOR STANDS FOR ANGER She gave him a tongue-lashing. I really chewed him out good! Other forms of aggressive behavior can also stand metonymically for anger, especially aggressive visual behavior: AGGRESSIVE VISUAL BEHAVIOR STANDS FOR ANGER She was looking daggers at me. He gave me a dirty look. If looks could kill, . . . . He was glowering at me. All these metonymic expressions can be-.used to indicate anger. As in the case of the OPPONENT metaphor, our analysis of the DANGEROUS ANIMAL metaphor leaves an expression unaccounted for - "insatiable." This expression indicates that the animal has an appetite. This "appetite" seems to correspond to the "demands" in the OPPONENT metaphor, as can be seen from the fact that the following sentences entail each other: Harry's anger is insatiable. Harry's anger cannot be appeased. To see what it is that anger demands and has an appetite for, let us turn to expressions that indicate causes of anger. Perhaps the most common group of expressions that indicate anger consists of conventionalized forms of annoyance: minor pains, burdens placed on domestic animals, and so forth. Thus we have the metaphor: THE CAUSE OF ANGER IS A PHYSICAL ANNOYANCE Don't be a pain in the ass, Get off my back! You don't have to ride me so hard. You're getting under my skin. He's a pain in the neck. Don't be a pest!

. THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

209

These forms of annoyance involve an offender and a victim. The offender is at fault. The victim, who is innocent, is the one who gets angry. . , jThere is another set of conventionalized expressions used to speak of, or to, people who arein the process of making someone angry. These are ^expressions of territoriality, in which the cause of anger is, viewed as a trespasser.!; .-••-.:• •••?. •-.-'.' •'••^>^ ' ::v-;- ^-vi-b^ii v ? • CAUSING ANGER IS TRESPASSING You're beginning to get to me. '". This is where I draw the line! """" Don't step on my toes!

'

'•\'-^''''','..2, " ' "•"•

l

;'":

'•'"'

Again, there is an offender (the cause of anger) and a victim (the person who is getting angry). In general, the cause of anger seems to be an offense, in which there is an offender who is at fault and an innocent victim, who is the person who gets angry. The offense seems to constitute some sort of injustice. This is reflected in the conventional wisdom: Don't get mad, get even! In order for this saying to make sense, there has to be some connection between anger and retribution. Getting even is equivalent to balancing the scales of justice. The saying assumes a model in which injustice leads to anger and retribution can alleviate or prevent anger. In short, what anger "demands" and has an "appetite" for is revenge. This is why warnings and threats can count as angry behavior: If I get mad, watch out! Don't get me angry, or you'll be sorry. The angry behavior is, in itself, viewed as a form of retribution. We are now in a position to make sense of another metaphor for anger: ANGER IS A BURDEN Unburdening himself of his anger gave him a sense of relief. After I lost my temper, I felt lighter. He carries his anger around with him. He has a chip on his shoulder. You'll feel better if you get it off your chest. In English, it is common for responsibilities to be metaphorized as burdens. There are two kinds of responsibilities involved in the cultural model of anger that has emerged so far. The first is a responsibility to control one's anger. In cases of extreme anger, this may place a considerable burden on one's "inner resources." The second comes from the model of retributive justice that is built into our concept of anger; it is the responsibility to seek vengeance. What is particularly interesting is that these two responsibilities are in conflict in the case of angry retribution: If you take out your anger on someone, you are not meeting your responsibility to control your anger, and if you do not take out your anger on someone, you

210

GEORGE LAKOFF & ZOLTAN KdVECSES ;

are not meeting your responsibility to provide retribution. The slogan "Don't get mad, get even!" offers one way out: retribution without anger. The human potential movement provides another way but by suggesting that letting your anger out is okay. But the factis thit neither of these 'solutions is the cultural norm. It should also be mentioned in passing that the human potential movement's way of dealing with anger by sanctioning its release is not all that revolutionary. It assumes almost all of pur standard cultural model and metaphorical understanding, arid .makes one change: sanctioning the "release."' ' "'''"" '"-" ""'.'.l SOME MINOR METAPHORS

There are a few very general metaphors that apply to anger as well as to many other things, and are commonly used in comprehending and speaking about anger. The first we will discuss has to do with existence. Existence is commonly understood in terms of physical presence. You are typically aware of something's presence if it is nearby and you can see it. This is the basis for the metaphor: EXISTENCE IS PRESENCE His anger went away. His anger eventually came back. My anger lingered on for days. She couldn't get rid of her anger. After a while, her anger just vanished. My anger slowly began to dissipate. When he saw her smile, his anger disappeared. In the case of emotions, existence is often conceived of as location in a bounded space. Here the emotion is the bounded space and it exists when the person is in that space: EMOTIONS ARE BOUNDED SPACES She flew into a rage. She was in an angry mood. He was in a state of anger. I am not easily roused to anger. These cases are relatively independent of the rest of the anger system, and are included here more for completeness than for profundity. THE PROTOTYPE SCENARIO

The metaphors and metonymies that we have investigated so far converge on a certain prototypical cognitive model of anger. It is not the only model of anger we have; in fact, there are quite a few. But as we shall see, all of the others can be characterized as minimal variants of the model that the metaphors converge on. The model has a temporal dimension, and can be conceived of as a scenario with a number of stages. We will call

T H E COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

211

this the "prototype scenario"; it is similar to what De Sousa (1980) calls the "paradigm scenario." We will be referring to the person who gets angry as St short for the Self. Stage 1; offending event . . . . . . . ';'.'" " There is an offending event that displeases S. There is a wrongdoer who intentionally does something directly to S. The wrongdoer is at fault and Self is innocent. The offending event constitutes an injustice and produces anger in"5. The scales of justice can only be balanced by some act of retribution. That is, the intensity .'_ of retribution must be roughly equal to the intensity of offense. S has the responsibility to perform such an act of retribution. Stage 2; anger Associated with the entity anger is a scale that measures its intensity. As the intensity of anger increases, S experiences physiological , effects: increase in body heat, internal pressure, and physical agitation. As the anger gets very intense, it exerts a force upon S to perform an act of retribution. Because acts of retribution are dangerous and/or socially unacceptable, S has a responsibility to control his anger. Moreover, loss of control is damaging to S's own well-being, which is another motivation for controlling anger. Stage 3: attempt at control S attempts to control his anger. Stage 4: loss of control Each person has a certain tolerance for controlling anger. That tolerance can be viewed as the limit point on the anger scale. When the intensity of anger goes beyond that limit, S can no longer control his anger. S exhibits angry behavior and his anger forces him to attempt an act of retribution. Since S is out of control and acting under coercion, he is not responsible for his actions. Stage 5: act of retribution S performs the act of retribution. The wrongdoer is the target of the act. The intensity of retribution roughly equals the intensity of the offense and the scales are balanced again. The intensity of anger drops to zero. At this point, we can see how the various conceptual metaphors we have discussed all map onto a part of the prototypical scenario, and how they jointly converge oh that scenario. This enables us to show exactly how the various metaphors are related to one another, and how they function together to help characterize a single concept. This is something that Lakoff and Johnson (1980) were unable to do. The course of anger depicted in the prototype scenario is by no means the only course anger can take. In claiming that the scenario is prototypical we are claiming that according to our cultural theory of anger, this is a

212

GEORGE LAKOFF & ZOLTAN KOVECSES

normal course for anger to take. Deviations of many kinds' are both recognized as existing and recognized as being noteworthy and not the norm. Let us take some examples: ; /••;;.;•; "V -

-

'

• -

'

Someone who "turns the other cheek," that is, who does not/get angry or seek retribution. In this culture, such a person is considered virtually saintly. Someone who has no difficulty controlling his angdr is especially praiseworthy. A "hothead" is someone who considers more events offensive than most people, who has a lower threshold for anger than the norm, who cannot control his anger, and whose acts of retribution are considered out of proportion to the offense. Someone who is extremely hotheaded is considered emotionally "unbalanced." On the other hand, someone who acts in the manner described in the prototypical scenario would not be considered abnormal at all. Before turning to the nonprototypical cases, it will be useful for us to make a rough sketch of the ontology of anger: the entities, predicates, and events required. This will serve two purposes. First, it will allow us to show in detail how the nonprototypical cases are related to the prototypical model. Second, it will allow us to investigate the nature of this ontology. We will include only the detail required for our purposes. It is part of our cultural model of a .person that he can temporarily lose control of his body or his emotions. Implicit in this concept is a separation of the body and the emotions from the S. This separation is especially important in the ontology of anger. Anger, as a separable entity, can overcome someone, take control, and cause him to act in ways he would not normally act. In such cases, the S is no longer in control of the body. Thus, the ontology of anger must include an S, anger, and the body. A fuller treatment would probably also require viewing the mind as a separate entity, but that is beyond our present purposes. Since anger has a quantitative aspect, the ontology must include a scale of anger, including an intensity, a zero point, and a limit point. The basic anger scenario also includes an offending event and a retributive act. Each of these has a quantitative aspect, and must also include an intensity, a zero point, and a limit. In the prototypical case, the offending event is an action on the part of a wrongdoer against a victim. The retribution takes the form of an act by an agent against some target. The ontology of anger also includes a number of predicates: displeasing, at fault, exert force on, cause, exist, control, dangerous, damaging, balance, and outweigh. There are also some other kinds of events: the physiological effects; the angry behaviors; and the immediate cause of anger, in case it is not the same as the offending event.

THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

SUMMARY OF THE ONTOLOGY OF ANGER V Aspects of the. person: S

•

• • -

•

Body.: • Anger

Offense and retribution: Scales of intensity: -••

End points: Predicates:

Other events: J

••

:•/:-••••:••:•:•[•

.213

. . .

':>:• : J M •,..' rJ, .,;•

Offending Event v:;:;V .Retributive Act Intensity of Anger v Intensity of Offense - . Intensity of Retribution Zero ' • Limit -•• Displease At Fault Cause Exist • Exert Force o n Control Dangerous Damaging Balance Outweigh Physiological Reactions Angry Behaviors Immediate Cause .

RESTATEMENT OF THE PROTOTYPICAL SCENARIO

Given the preceding ontology and principles of the cultural model, we can restate the prototypical anger scenario in terms that will facilitate showing the relationships among the wide variety of anger scenarios. We will first restate the prototypical scenario and then go on to the nonprototypical scenarios. PROTOTYPICAL ANGER SCENARIO Constraints: Victim = S Agent of Retribution = S Target of Anger = Wrongdoer Immediate Cause of Anger - Offending Event Angry Behavior = Retribution Stage 1: Offending Event Wrongdoer offends S Wrongdoer is at fault The offending event displeases S

214

GEORGE LAKOFF & ZOLTAN KOVECSES

, The intensity of the offense outweighs the intensity of the retribution (which equals zero at this point), thus creating an imbalance. The offense causes anger to come into existence. : j."Stage2: Anger ' ••> -Anger exists. .•;:.::.\L••••••^i „•:••••. • / ; , • • ' ' - - ? *' • S experiences physiological effects (heat, pressure, agitation). Anger;exerts force on the S to attempt an act of retribution. Stage 3: Attempt to control .anger ' J S exerts a counterforce in an attempt to control anger. Stage 4: Loss of control The intensity of anger goes above the limit. Anger takes control of S. S exhibits angry behavior (loss of judgment, aggressive actions). There is damage to S. There is a danger to the target of anger, in this case, the wrongdoer. Stage 5: Retribution S performs retributive act against W (this is usually angry behavior directed at W). The intensity of retribution balances the intensity of offense. The intensity of anger drops to zero. Anger ceases to exist. THE NONPROTOTYPICAL CASES

We are now in a position to show how a large range of instances of anger cluster about the above prototype. The examples are in the following form: " a nonprototypical anger scenario with its name in boldface, followed by an informal description; an account of the minimal difference between the given scenario and the prototype scenario; finally, an example sentence. Iiisatsafele arager: You perform the act of retribution and the anger just does not go away. In Stage 5, the intensity of anger stays above zero and the anger continues to exist. Example: His anger lingered on. Frustrated anger: You just cannot get back at the wrongdoer and you get frustrated. It is not possible to gain retribution for the offensive act. S engages in frustrated behavior. Option: S directs his anger at himself. Examples: He was climbing the walls. She was tearing her hair out. He . was banging his head against the wall. He's taking it out on himself. ' Redirected anger: Instead of directing your anger at the person who made . you angry, you direct it at someone or something else.

THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

215

The target of anger is not the wrongdoer. Examples: When I lose my temper, I kick the cat. When you get angry,punch a pillow until your anger goes away. When something bad happened at the office, he would take it out on his wife. ,, . ^ Exaggerated response: Your reaction is way out of proportion to the offense. The intensity of retribution outweighs the intensity of offense. , Examples: Why jump down my throat? You have a right to get angry, but not to go that far. Controlled response: You get angry, but retain control and consciously direct your anger at the wrongdoer. ; S remains in control. Everything else remains the same. Example: He vented his anger on her. ; -.-.'. Constructive isse: Instead of attempting an act of retribution, you put your anger to a constructive use. S remains in control and performs a constructive act instead of a retributive act. The scales remain unbalanced, but the anger disappears. ; Example: Try to channel your anger into something constructive. Teraninatimg event: Before you have a chance to lose control, some unrelated event happens to make your anger disappear. Anger doesn't take control of S. Some event causes the anger to go out of existence. Example: When his daughter smiled at him, his anger disappeared. : , Spontaneous cessation: Before you lose control, your anger just goes away. Anger doesn't take control of S and the intensity of anger goes to zero. Example: His anger just went away by itself. Successful suppression: You successfully suppress your anger. S keeps control and the intensity of anger is not near the limit. Example: He suppressed his anger. Controlled reduction: Before you lose control, you engage in angry behavior and the intensity of anger goes down. S does not lose control, S engages in angry behavior and the intensity of anger goes down. Example: He's just letting off steam. Immediate explosion: You get angry and lose control ah at once. No Stage 3. Stages 2 and 4 combine into a single event. Example: I said "Hi Roundeyes!" and he blew up. Siow burn: Anger continues for a long time. Stage 2 lasts a long time. Example: He was doing a slow burn. Nursing a grudge: S maintains his anger for a long period waiting for a chance at a retributive act. Maintaining that level of anger takes special effort.

216

GEORGE LAKOFF & ZOLTAN K6VECSES

Stage 2 lasts a long time and requires effort. The retributive act does not equal angry behavior. Bos't get mad, get even: This is advice (rarely followed) about the >'* pointlessness of getting angry. It suggests avoiding Stages 2, 3, and 4, and instead going directly to Stage 5. This advice is defined as an alternative to the prototypical scenario. Indirect cause: It is some result of the wrongdoer's action, not the action itself, that causes anger. •• ,; The offense is not the immediate cause of anger, but rather is more indirect - the cause of the immediate cause. Consider the following case: Your secretary forgets to fill out a form that results in your not getting a deserved promotion. Offending event == secretary forgets to fill out form. Immediate cause = you do not get promotion. You are angry about not getting the promotion. You are angry at the secretary for not filling out the form. In general, about marks the immediate cause, at marks the target, and for marks the offense. Cool amger: There are no physiological effects and S remains in control. Cold anger: S puts so much effort into suppressing the anger that temperature goes down, while internal pressure increases. There are neither signs of heat nor agitation, and there is no danger that S will lose control and display his anger. In the prototypical case, a display of anger constitutes retribution. But since there is no such display, and since there is internal pressure, release from that pressure can only come through retribution of some other kind, one that is more severe than the display of emotion. It is for this reason that cold anger is viewed as being much more dangerous than anger of the usual kind. Expressions Hke Sally gave me an icy stare are instances of cold anger. This expression implies that Sally is angry at me, is controlling her anger with effort, and is not about to a lose control; it suggests the possibility that she may take retributive action against me of some sort other than losing her temper. Auger with: To be angry with someone, S has to have a positive relationship with the wrongdoer, the wrongdoer must be answerable to S, the intensity is above the threshold, but not near the limit. Perhaps the best example is a parent-child relationship, where the parent is angry with the child. Righteous indication: The offending event is a moral offense and the victim is not the S. The intensity of anger is not near the limit. Wrath: The intensity of the offense is very great and many acts of retribution are required in order to create a balance. The intensity of the angeris well above the limit and the anger lasts a long time. There appears to be a recognizable form of anger for which there are no "^ conventional linguistic expressions, so far as we can tell. We will call this

(

THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

217

a manipulative use of anger. It is a case where a person cultivates his anger and does not attempt to control it, with the effect that he intimidates those around him into following his wishes in order to keep him from getting angry. This can work either by fear of by guilt. The people manipulated can either be afraid of his anger or may feel guilty about what the arigef; does to him. This form of anger is fairly distant from the prototype arid f it is no surprise that we have no name for it. Interestingly enough, there is a linguistic test that can be used to verify that what we have called the prototypical scenario is indeed prototypical. It involves the use of the word but. Consider the following examples (where the asterisk indicates a semantic aberration): . •'. Max got angry, but he didn't blow his top. *Max got angry, but he blew his top. Max blew up at his boss, but the anger didn't go away. *Max blew up at his boss, but the anger went away. Sam got me angry, but it wasn't him that I took my anger out on. *Sam got me angry, but it was him that I took my anger out on. The word but marks a situation counter to expectation. In these examples, the prototypical scenario defines what is to be expected. The acceptable sentences with but run counter to the prototypical scenario, and thus fit the conditions for the use of but. The unacceptable sentences fit the prototypical scenario, and define expected situations. This is incompatible with the use of but. Thus we have a linguistic test that accords with our intuitions about what is or isn't prototypical. Each of the nonprototypical cases just cited is a case involving anger. There appear to be no necessary and sufficient conditions that will fit all these cases. However, they can all be seen as variants of the prototypical anger scenario. Prototypes often involve clusters of conditions and the prototypical anger scenario is no exception. The clustering can be seen explicitly in identity conditions such as: Victim = Self, Target = Wrongdoer, Offending Event = Immediate Cause, and so forth. When these identities do not hold, we get nonprototypical cases. For example, with righteous indignation, Victim does not have to equal Self. In the case of an indirect cause, Offending Event does not equal Immediate Cause. In the case of redirected anger, Target does not equal Wrongdoer. Usually the act of retribution and the disappearance of anger go together, but in the case of spontaneous cessation and insatiable anger, that is not the case. And in the Don't-get-mad-get-even case, angry behavior is avoided, and is therefore not identical to the act of retribution. Part of what makes the prototypical scenario prototypical is that it is sufficiently rich so that variations on it can account for nonprototypical cases, and it also has a conflation of conditions that are not conflated in nonprototypical cases, The point is that there is no single unified cognitive model of anger. Instead there is a category of cognitive models with a prototypical model in the center. This suggests that it is a mistake to try to find a single

218

GEORGE LAKOFF.& ZOLTAN KOVECSES

cognitive model for ail instances of a concept. Kinds of anger are not all instances of the samg model; instead they are variants on a prototypical model. There is no common core that all kinds of anger have in £dmmon. Instead, the kinds of anger bear family resemblances to one another, METAPHORICAL ASPECTS OF THE PROTOTYPE SCENARIO

-'tr.'^b

The analysis we have done so far is consistent with a certain traditional view of metaphor, namely;. . . ,, ;:;:: The concept of anger exists and is understood independently of any metaphors. The anger ontology and the category of scenarios represent the literal meaning of the concept of anger. Metaphors do no more than provide ways of talking about the ontology of anger. This view entails the following: The elements of the anger ontology really, literally exist, independent of any metaphors. A brief examination of the anger ontology reveals that this is simply not the case. In the ontology, anger exists as an independent entity, capable of exterting force and controlling a person. This is what Lakoff and Johnson (1980) refer to as an "ontological metaphor." In this case, it would be the ANGER IS AN ENTITY metaphor. A person's anger does.not really, literally exist as an independent entity, though we do comprehend it metaphorically as such. In the ontology, there is an intensity scale for anger, which is understood as being oriented UP, by virtue of the MORE IS UP metaphor. The intensity scale has a limit associated with it - another ontological metaphor. Anger is understood as being capable of exerting force and taking control of a person. The FORCE and CONTROL here are also metaphorical, based on physical force and physical control. The anger ontology also borrows certain elements from the ontology of retributive justice: offense and retribution, with their scales of intensity and the concept of balance. These are also metaphorical, with metaphorical BALANCE based on physical balance. In short, the anger ontology is largely constituted by metaphor. Let us now examine these constitutive metaphors. Their source domains - ENTITY, INTENSITY, LIMIT, FORCE, and CONTROL all seem to be superordinate concepts, that is concepts that are fairly abstract. By contrast, the principal metaphors that map onto the anger ontology - HOT FLUID, INSANITY, FIRE, BURDEN, STRUGGLE appear to be basic-level concepts, that is, concepts that are linked more directly to experience, concepts that are information-rich and rich in conventional mental imagery. Let us call the metaphors based on such concepts "basic-level metaphors." We would like to suggest that most of our

THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

219

understanding of anger comes via these basic-level metaphors. The HOT ^ FLUID and FIRE metaphors, give us an understanding of what kind of entity anger is. And the STRUGGLE metaphor gives us a sense of. what .is involved in controlling it. Without these metaphors, our understanding; of anger would be extremely impoverished, to say the least, One |s.Jenipted; to ask which is more primary: the constitutive metaphors or the';'basic? level ones. We dp not know if that is a meaningful question, All we know is that both exist, and have their separate functions: The ;,basic-leyel; metaphors allow us to comprehend and draw inferences about anger^ using our knowledge of familiar, well-structured domains. The. constitutive metaphors provide the bulk of the anger ontology. •>.: .\\ The embodiment of anger We have seen that the concept of anger has a rich conceptual structure and that those who view it as just a feeling devoid of conceptual content are mistaken. But the opposite view also exists. Schachter and Singer (1962) have claimed that emotions are purely cognitive, and that there are no physiological differences among the emotions. They claim that the feeling of an emotion is simply a state of generalized arousal, and that which emotion one feels is simply a matter of what frame of mind one is in. The results of Ekman, Levenson, and Friesen (1983) contradict the Schacter-Singer claims with evidence showing that pulse rate and skin termperature do correlate with particular emotions. Although the kind of analysis we have offered does not tell us anything direct about what the physiology of emotions might be, it does correlate positively with the Ekman group's results. As we saw, the conceptual metaphors and metonymies used in the comprehension of anger are based on a cultural theory of the physiology of anger, the major part of which involves heat arid internal pressure. The Ekman group's results (which are entirely independent of the analysis given here) suggest that our cultural theory of the physiology of anger corresponds remarkably well with the actual physiology: when people experience anger their skin temperature and pulse rate rises. Although the cultural theory is only a "folk" theory, it has stood the test of time. It has made sense to hundreds of millions of English speakers over a period of roughly a thousand years. The Ekman group's results suggest that ordinary speakers of English by the million have had very subtle insight into their own physiology. Those results suggest that many of the details of the way we conceptualize anger arise from the autonomic nervous system, and that the conceptual metaphors and metonymies used in understanding anger are by no means arbitrary; instead they are motivated by our physiology. The Ekman group's results, together with our hypothesis concerning conceptual embodiment, make an interesting prediction. It predicts that

220

GEORGE LAKOFF & ZOLTAN KOVECSES

if we look at metaphors and metonymies for anger in the languages of the world, we will not find any that contradict the physiological results that they found. In short, we should not find languages where the basic emotion of anger is understood in terms of both cold and freedom from * pressure. The nonbasic case of cold anger discussed above is irrelevant, since it is a special form of anger and not an instance of the normal basicanger emotion, and since it does involve internal pressure. If Schachter and Singer are right and the Ekman group has made a mistake, then the English metaphors arid metonymies for anger arS arbitrary, that is, they are not embodied, not motivated by any physiological reality. The heat and internal pressure metaphors should thus be completely accidental. If there is no physiological basis for anger at all, as Schachter and Singer suggest, we would then expect metaphors for anger to be randomly distributed in the languages of the world. We would expect metaphors for cold and freedom from pressure to be just as common as metaphors for heat and pressure; in fact, on the Schacter-Singer account, we would expect that metaphors based on shape, darkness, trees, water - anything at all - would be just as common as metaphors based on heat and pressure. The research has not been done, but our guess is that the facts will match the predictions of the Ekman group. If those predictions hold up, it will show that the match between the Ekman group's results and ours is no fluke, and it will give even more substance to our claim that concepts are embodied. Review We have shown that the expressions that indicate anger in American English are not a random collection, but rather are structured in terms of an elaborate cognitive model that is implicit in the semantics of the language. This indicates that anger is not just an amorphous feeling, but rather that it has an elaborate cognitive structure. However, very significant problems and questions remain. First, there are aspects of our understanding of anger that our methodology cannot shed any light on. Take, for example, the range of offenses that cause anger and the corresponding range of appropriate responses. Our methodology reveals nothing in this area. Second, the study of the language as a whole gives us no guide to individual variation. We have no idea how close any individual comes to the model we have uncovered, and we have no idea how people differ from one another. Third, our methodology does not enable us to say much about the exact psychological status of the model we have uncovered. How much of it do people really use in comprehending anger? Do people base their actions on this model? Are people aware of the model? How much of

THE COGNITIVE MODEL OF ANGER IN AMERICAN ENGLISH

221

it, if any, do people consciously believe? And most intriguirigly, does the model have any effect on what people feel? Certain things, however, do seem to be clear. Most speakers of American English seem to use consistently the expressions we have described and make inferences that appear, so far as we can tell, to be consistent with our model. We make this claim on the basis of our own intuitive observations, though to really establish it, one would have to do thorough empirical studies. If we are right, our model has considerable psychological reality, but how much and what kind remains to be determined. The fact that our analysis meshes so closely with the physiological study done by the Ekman group suggests that emotional concepts are embodied, that is, that the actual content of the concepts are correlated with bodily experience. This is especially interesting in the case of metaphorical concepts, since the correlation is between the metaphors and the physiology, rather than directly between the literal sense and the physiology. It provides confirmation of the claim by Lakoff and Johnson (1980) that conceptual metaphors are not mere flights of fancy, but can even have a basis in bodily experience. What does all this say about cultural models? First, they make use of imaginative mechanisms - metaphor, metonymy, and abstract scenarios. Second, they are not purely imaginative; they can be motivated by the most concrete of things, bodily experience. Third, linguistic evidence is an extraordinarily precise guide to the structure of such models. Note i. This is a shortened, slightly revised version of the chapter entitled "Anger" that appears as Case Study 1 in Women, Fire, and Dangerous Things: What Categories Reveal about the Mind, by George Lakoff, University of Chicago Press, 1987.

References DeSousa, R. 1980. The rationality of emotions. In Explaining Emotions, A. O. Rorty, ed. Berkeley: University of California Press. Pp. 127-151. Ekman, P., R. W. Levenson, and W. V. Friesen 1983. Autonomic nervous system activity distinguishes among emotions. Science 22I(4616):I208-I210. Lakoff, G. and M. Johnson 1980. Metaphors We Live By. Chicago: University of Chicago Press. Schachter, S. and J. Singer 1962. Cognitive, social and physiological determinants of emotional states. Psychological Review 69:379-399.

Two.theories of home heat .control Wittett Kempton

Human beings strive to connect related phenomena and make sense of the world. In so doing, they create what I call folk theory. The word folk signifies both that these theories are shared by a social group arid that they are acquired from everyday experience or social interaction. To call them theories is to assert that they use abstractions that apply to many analogous situations, enable predictions, and guide behavior. I contrast folk theories with institutionalized theories, which are used by specialists and acquired from scientific literature or controlled experiments. Thus, a folk theory is one type of cultural model. This chapter analyzes folk theories for home heating control, particularly thermostat control. From interviews with Michigan residents, folk theories were inferred using methods developed by Lakoff and Johnson (1980; also see Lakoff & Kovecses this volume) and Quinn (1982; this volume). The inferred folk theories were compared with behavior guided by the theory, using both observed behavior and self-reported behavior. The interviews also elicited lists of devices analogous to thermostats and a history of use in present and past residences. The concept of folk theory Anthropological interest in folk theory has germinated over the past few years in the form of two recent volumes (Dougherty 1985; Marsella & White 1982) and the conference leading to the present volume. Present work on folk models and folk theory continues the expansion of cognitive anthropology from folk categories to more complex structures, such as sets of propositions (D'Andrade 1981; Kay 1966), inference rules (Cole & Scribner 1974; Hutchins 1980), cognition in everyday activities (Holland 1985; Lave & Rogoff 1983; Murtaugh, Faust, & de la Rocha unpublished), and connections in discourse (Agar 1980; Rice 1980). Recent discoveries by psychologists and educators provide the most precisely defined example of folk theory to date. Related cognitive structures have been called "naive theory" (DiSessa 1982; McCloskey 1983a; McCloskey, Caramazza & Green 1980), "mental models" (de Kleer &

TWO THEORIES OF HOME HEAT CONTROL

223

Brown 1983; Gentner& Stevens 1983; Johnson-Laird 1981), "naive prob-. lem representation" (Larkin 1983; Larkin et al. 1980), or "intuitive theory" (McCloskey 1983b; see DiSessa 1985 for a contrasting view).* I describe one study to provide an example of folk theory and to show how my perspective differs. The study compared folk theories of motion with physicists' theories of motion: . . . People develop on the basis of their everyday experience remarkably well-articulated naive theories of motion . . . theories developed by different individuals are best described as different forms of the same basic theory. Although this basic theory appears to be a reasonable outcome of experience with real-world motion, it is strikingly inconsistent With the fundamental principles of classical physics. In fact, the naive theory is remarkably similar to a pre-Newtonian physical theory popular in the fourteenth through sixteenth centuries. (McCloskey 1983a:299) To paraphrase this quotation, the folk theory of motion (1) is based on everyday experience; (2) varies among individuals, although important elements are shared; and (3) is inconsistent with principles of institutionalized physics. McCloskey makes a fourth finding, that the folk theory persists in the face of formal physics training. Students simply reinterpret classroom material to fit their preexisting folk theory (McCloskey 1983a:318). The physics instructors would not usually even be aware their students had a preexisting folk theory.3 To an anthropologist, the nonrecognition of conflicting systems and the persistence of the folk system both resemble phenomena at a cultural boundary. Isolated elements and terms diffuse across the cultural boundary, but they are incorporated into the prior folk theory rather than inducing change. There is also a parallel in that one culture seems dominant: When the folk and institutional theories differ, the folk theory is considered wrong. The research just cited does not question the correctness of the institutional theory, since a major objective of the classroom studies has been to improve teaching. Wiser and Carey caution that we cannot understand folk theory by simply diagnosing its failure to solve problems in the domain of the expert. We must find the problems it does solve correctly and examine the explanatory mechanisms it uses to do so (Wiser & Carey 1983:295). Anthropologists' naturalistic proclivities have led them away from folk theory in the classroom and toward the environments in which folk theory earns its keep in everyday use; many examples are found in this volume. Residential thermostats This study deals with home heating thermostats for several reasons. Home heating systems are fairly simple and well understood (at least in comparison with marriage, disease etiology, gender roles, divination, the mind, and other topics of this volume). Information about them is communicated

224

WILLETT KEMPTON

almost entirely through folk channels - no one must study thermostats in high school or pass examinations; there is no widespread institutionalized dogma. Yet, many people adjust their thermostats, typically more .than once per day, so they must have some principles or theories guiding this "behavior. Also, since the range of behavior affected by the theory Is restricted - turning a single dial - behavioral records can be collected automatically. By comparing behavior patterns with interviews, we can better infer how folk theory guides the behavior we observe, Finally, this domain was selected because an improved understanding of what people do with thermostats would have significant practical consequence for national energy programs. The following discussion draws from two sets of interviews and one set of behavioral data. The first set of interviews, with people in 30 Michigan households, elicited general information about energy management. Since these interviews were exploratory, they often did not go into much detail on thermostats. The second is a set of 8 interviews, with 12 Michigan informants, that focused on thermal comfort and especially dealt with thermostat control. I use the interview data to infer folk theory, which in turn is compared with behavioral patterns. The data on behavior derive from automatic recordings of thermostat settings in 26 houses in New Jersey. In addition, my analysis draws from discussions with heating specialists and energy conservation analysts and from technical energy articles. This study reports the first phase of our research on this topic. The second phase, now being analyzed, combines both interview and behavioral data from the same households. Although the combined method is conceptually appealing, the expense of combined data - particularly when using automatic recorders - has necessitated smaller samples than we are able to draw on here. I hypothesize that two theories of thermostats exist in the United States (and perhaps throughout the industrialized world). One, the feedback theory, holds that the thermostat senses temperature and turns the furnace on and off to maintain an even temperature. The other, which I call the valve theory, holds that the thermostat controls the amount of heat. That is, like a gas burner or a water valve, a higher setting causes a higher rate of flow. Technically knowledgeable readers of this paper have commented that the feedback theory is correct and the valve theory is wrong. However, as I will demonstrate, both folk theories simplify and distort as compared to a full physical description, each causes its own types of operational errors and inefficiencies, and each has certain advantages. Records of thermostat use

%

Behavioral records of thermostat settings have been collected by Princeton University's Center for Energy and Environmental Studies (Dutt, n-

TWO THEORIES OF HOME HEAT CONTROL

225

Eichenberger, & Socolow 1979; Socolow 1978). During a 2-year period', automatic devices recorded hourly thermostat settings (and many other energy variables) of 26 upper middle-class families in identical townhouses. Here, I examine 2 of those homes to illustrate 2 patterns of thermostat adjustment, which I link to the 2 folk theories. '. Figure 9.1 shows hourly measurements of thermostat settings in one house over a 3-day period in the winter of 1976. The solid line shows the hourly thermostat setting, and the dotted line shows the room temperature. We see that the, thermostat usually is changed at times when occupants and activities change: 8 A.M., noon, and 5 to 8 P.M. During other periods, not shown in Figure 9.1, the thermostat may be left at the same setting for several days. I conclude that the thermostat setting is changed when the desired temperature changes: when waking or going to sleep, when entering or exiting the house, and around mealtimes. Figure 9.2 shows hourly thermostat settings for a second house, also during the winter of 1976. In this house, the thermostat is often changed between each hourly datum. In fact, the only times on the figure when the thermostat is not changed are probable sleeping times, for example, from 1 A.M. to 7 A.M. Monday, and from 10 P.M. Monday to 8 A.M. Tuesday. It appears that whenever someone is awake in this house, the thermostat is adjusted at least hourly. Examination of the full 2 years of data (not shown here) also shows many thermostat adjustments, not at regular times, and a wide range of settings (from 61° to 85° F; 16° to 29°C). I hypothesize that the frequent thermostat settings of this second household result from the residents' having a valve theory of their thermostat.4 Although I could not interview the people in the households shown in Figures 9.1 and 9.2, informants in my interviews do report following similar patterns (though rarely as extreme as Figure 9.2). The next two sections discuss interview evidence for the two theories. The feedback theory According to the feedback theory, the thermostat turns the furnace on or off according to room temperature. When the room is colder than the setting on the thermostat dial, the thermostat turns the furnace on. Then, when the room is as warm as the setting, it turns the furnace off. Since the theory posits that the furnace runs at a single constant speed, the thermostat can control the amount of heating only by the length of time the furnace is on. Thus, if the dial is adjusted upward only a little, the furnace will run a short time and turn off; if it is adjusted upward a large amount, the furnace must run longer to bring the house to that temperature. Left at one setting, the thermostat will switch the furnace on and off as necessary to maintain approximately that temperature. Heating engineers are fairly comfortable with the folk theory described

to to

INSIDE TEMPERATURE

85 i

THERMOSTAT SETTING

°F .^-»

60 - " - - i — i — « — > — i — i 12

2

4

6

1—i—f—i—i

8 10 12 2

am

4

6

8

1—i—i—i—i—T 10 12 2

4

6

8

i—i—!—«—i—i—i 10 12 2

6

pm

pm Monday

4

Tuesday

(9 January 1983)

Figure 9.1. Pattern of thermostat adjustments consistent with the feedback theory

8

*..

, /

1—i—i—r~i

10 12 2 am

4

6

3

1—r—i—i— 10 12 2

;4

;. , j- .'pm • '. - Wednesday; :< !•'

"F

85 •

INSIDE TEMPERATURE

80 -

THERMOSTAT SETTING

75 ' 70 65 eg J — , — , — , — , — r 10 12 2

4

6

"i—i—i—i—f—i—l—I—I—l—i—i—i—i—i—i—i—i—i—i—r—[—i—i—r^—i—i—i— 8 10 12 2

4

6

8

10 12 2

4. ,6

8

10 12 2

pm Monday

4

pm Tuesday

{2 February 1976)

Figure 9.2. Pattern of thermostat adjustments consistent with the valve theory

6

8 10 12 2 am

4

6 8 . 10 12 2 . . » Wednesday

4 pm

228

WILLETT KEMPTON

here - they consider it simplified, but essentially correct. As I show, however, their evaluation of correctness may be based on irrelevant criteria. In interview segments such as the following (from a Michigan farmer), •< I infer that the feedback theory is being used: You Just turn the thermostat up, and once she gets up there [to the desired temperature] she'll kick off automatically. And then shell kick on and off to keep it at that temperature. From anthropomorphic statements such as this and others about the thermostat "feeling it is too cold," I infer the following metaphor (Lakoff & Johnson 1980) for the feedback theory: The thermostat is a little person with a switch controlling the furnace. The little person turns the switch on and off, based on perceived temperature. Thermostats and the mathematical description of self-regulating devices (Wiener 1948) are new to this century. These devices trace their ancestry to Watt's steam engine governor. Thermostats are the only self-regulating devices whose operation is visible in the average home (most homes have visible thermostats not only for the furnace control but also in the refrigerator, oven, portable heater, etc.).

The valve theory I elucidate the valve theory first through interview material from a single informant then present evidence that it is held by a substantial proportion of Americans. The valve theory was most clearly articulated by Bill, a well-educated man in his thirties. Bill was raised in California but has lived in Michigan for three years. As shown in the following quotation, Bill described the thermostat dial as not just switching on and off, but as controlling the rate of heat flow. (In dialogues, Relabels my question and B labels Bill's response.) W: What would the system do at 68° versus 85°? How did the numbers on the thermostat dial relate to anything that was going on with the furnace? B: Well, you could feel the heat coming up the vent. There'd be less, less warm air, less hot air at the different setting. I mean - that was clear. Similarly, in describing the house in which he grew up:

^

B: I do remember, as a kid, the heat vents were more or less hot to touch - to the hand as a sensor near the vent - at different settings. . . . When I asked Bill for a description of how the thermostat works, he said he could not describe it technically. However, when I asked him to explain how it works in relation to what he did with it, he immediately ... explained: ^

TWO THEORIES OF HOME HEAT CONTROL

229

B: I think it's pretty simple really. Um, I assume, um, that there is some kind of linear relationship between where the lever is and the way some kind of heat r generating system functions. And, um, that it's like stepping on the gas pedal; that there I have a notion of hydraulics, you know, the harder you push there is,'the more fluid gets pushed into the engine, and the more explosions there are, and the faster it goes. And so here, the, the harder or the more you push the lever or twist the lever in - there is a scale which indicates, you know, regular units - the . . . more power the system puts out to generate heat. . . . Bill's analogy of an automobile gas pedal describes the continuously varying flow of heat he belieyes he is regulating. .( , To elicit operational practices, I posed hypothetical situations. Bill's practices were consistent with his theory and his metaphor: W: Let's say you're in the house and you're cold. . . . Let's say it's a cold day, you feel cold, you want to do something about it. B; Oh, what I might do is, I might turn the thing up high to get out, a iot of air out fast, then after a little while turn it off or turn it down. W: Un-huh B: So there are also, you know, these issues about, um, the rate at which the thing produces heat, the higher the setting is, the more heat that's produced per unit of time, so if you're cold, you want to get warm fast, um, so you turn it up high. Um, my feeling is, my, my kind of Calvinist or Puritan feeling is that that's sinful. That, that really ought to turn it to the setting, the warmth setting which you think you'll eventually be comfortable and just bear the cold until the thing slowly heats up the house to that level. Although Bill believes that a higher setting would cause the house to heat up faster, he may refrain from doing this because he considers it wanton. Thus, for reasons other than the theory of the thermostat itself, he sometimes operates the device in the same way as someone who holds the feedback theory. His conceptualization of why the house maintains a steady temperature is radically different from that predicted by the feedback theory. W: OK, so how would you use the thermostat and, and how would it be different from the way you used that oven? B: Um, well, I guess you'd find some kind of moderate, steady, steady setting, setting for the thermostat to maintain a comfortable temperature in a steady state. W; OK, and that would be something intermediate between having it cranked all the way up for heating up quickly, which would be too much, and having it all the way down, which would not be enough? B; Right, yea. W: So what, what's this, what's steady state about? You're trying to balance off what against what? B: Humm. Uh, well I guess basically, the amount of heat that comes into the system has to equal the amount that's somehow disappearing. In this passage, Bill states that an even temperature is not maintained by the thermostat itself; rather, it is a balance set by the human operator.

230

WILLETT KEMPTON

The operator adjusts the rate of heat entering the system to equal the amount leaving or dissipating. The operator's balance of energy -input against dissipation is also captured by the gas pedal analogy tb which Bill Referred frequently. When Bill says the thermostat is like a gas'ped^.he is using what Gentner (1983) calls analogy or structure mapping ~'fhV|as pedal is analogous to the thermostat because they both have the sanie re)a-; tions between objects and actions, not because they have similar attributes or appearance. A person who is attempting to balance heat against dissipation would reasonably change the thermostat setting frequently. The result would be a pattern of many adjustments like that seen in Figure 9.2. Since our environment contains many tnore valves than feedback devices, everyone is likely to have a valve theory (applied to water faucets, etc.). People who have a feedback theory will also have a valve theory and will apply the feedback theory to only a small set of devices. To elicit devices Bill saw as similar to thermostats, I suggested that the burner on a gas stove operates like Bill's description of the thermostat; the higher you turn it, the more heat you get. He agreed, so I asked which other devices would be similar: B: I just flashed to electric mixers. The higher you turn them, the faster they go . . . the harder you push on the gas, the faster the car moves . . . turning on the faucet. . . you can see the water squirting out in greater volume at a greater rate, you know, as the lever is increased to turn it up. By contrast, he discards on-off controls as different:

»:•[.'•

B: What other analogies might I think of? Uh, turning on and off the lights just, uh, that's a more binary kind of process. Bill's device analogies are consistent with his abstract description, his reported operational practices, and the metaphors he uses to describe the thermostat. / Bill recognized that his model of thermostat operation may be different from that of specialists. In fact, early in the interview he described himself as not having any knowledge about thermostats. He thought someone had given him a formal explanation, but he could not remember it. Since he could not reproduce the institutionally sanctioned explanation, he describes himself as "ignorant": B: Now um, I really am ignorant about the functioning of these devices . . . once or twice, somebody might have described to me how the thermostat operated, but, my reaction at the time was maybe comprehension, but the sense of feeling out of control. And, maybe partly because of anxiety or lack of familiarity with the device, I've since forgotten what I was told. W: That's fine. B: Yea, so, this really would be, if I tried speculating on how they worked, it would really be just that.

TWO THEORIES OF HOME HEAT CONTROL

231

W: Speculation. B: Sort of de novo. Bill's denial of a solid understanding here is belied by the other quotations, which show a detailed and complete theory of thermostat operation, including a set of predictions, past perceptions, and operating rules corisistent with his theory. Further, his theory is stable enough to resist change. Later in the interview, he hypothesizes a possible second "level of sophistication" that some thermostats might have,' in order to explain why his parents' furnace seemed to turn on and off without human intervention: B: . . . I can imagine the devices at varying levels of sophistication and you'd have this, uh, kind of, feedback arrangement on, that exists for my parents'. W: Un-huh. B: Uh, where, uh, you know, it's not the human, the human operator isn't the only person that - the only agent that - turns the system on and off. The system itself is self-regulating. So, hm, in the course of the night, you know, it could turn on and, and off according to some measurement that it's making of the temperature, in the room. And um, um, you know, in my parents' house you could hear that thing going on and off in the course of the night making those irritating, windy sounds. W: Irritating, windy sounds. This is the drafts you were talking about before. B: Right. Um, now, I can't remember whether the thermostat in this other house had that self-control, kind of procedure. I don't think it did and I think that niight have been one of the reasons why we couldn't leave it on all night.' ••' Bill proposed this partial feedback theory to explain what he could not explain with his valve theory, but he seemed to consider the feedback thermostat (in his parents' house) to be a special case. In discussing my analysis with Bill 7 months later, he described this partial feedback theory as transitory: B: I didn't have the feedback theory until your questioning forced it upon me . . . I discovered it for the moment, but later forgot it. When I went back to using the thermostat, I probably went back to doing it the same way. Bill's reported return to the valve theory suggests that folk theory is resistant to change. Informal data suggest even more that the valve theory is resistant to change. In casual discussions of this material with proponents of the feedback theory, three individuals told me that their spouse turned up the thermostat to heat the house faster, that this practice was ineffective, and that their repeated attempts to convince their spouse of this had failed. Their failures may be attributed to the proselytizers' working to convey an individual belief when they needed to convey an entire theory. Worse, they had to supplant a theory that was already working satisfactorily and that was not explicitly acknowledged as existing.

232

WILLETT KEMPTON

To summarize the valve theory as described by Bill, the thermostat controls the rate at which the furnace generates heat. People maintain a constant temperature in their houses by adjusting the setting so that the amount A?of heat coming in just balances the amount being dissipated. Devices that operate similarly include the water faucet and the gas pedal on an automobile. Bill allows that some sophisticated systems may consider temperature and automatically adjust the heat from the furnace. Even though Bill's theory resists change, he devalues his own theory relative to the institutionally sanctioned one. Despite having a fully elaborated conceptualization and complex operating rules, he describes himself as "ignorant about the functioning of these devices." Is Bill's valve theory unusual? No national survey data exist on this issue, but estimates can be made. Of the 12 informants in my focused interviews, the valve theory was given in full form by 2, and in partial form by at least 3 more; data were indeterminate in some of the more brief interviews. Another study, in Wales, interviewed residents of 38 new thermostatically controlled rural houses. When asked whether a cold house would warm up quicker to the desired level if the thermostat were turned up full, past its normal setting, 62% said it would (O'Sullivan & McGeevor 1982:104). A third study asked 43 college students in Cambridge, Massachusetts, to assume they entered a cold house and wanted it to heat up as quickly as possible. Those who would turn it above the desired temperature for faster heating ranged from 24% to 46%, the percentage varying with previous questions about analogous situations (Gentner & Tenney, unpublished 1983 data). Therefore, from those limited studies I estimate that 25% to 50% of the population of the United States uses at least part of the valve theory. My sense, which I do not yet have formal data to support, is that the majority of the population holds a combination of these two theories. Functionality of the two theories In many anthropological studies, it would be nonsensical or impractical to evaluate the functionality of a folk theory. In this study, the domain and the goals of the folk are fairly straightforward. Known scientific theories describe home heating systems, and the major goals of using them are presumably to be comfortable without spending too much money. Thus, the functionality question can be addressed by asking how well each folk theory meets the goals of its users (this approach is advocated in more detail by Rappaport 1979:98; and by Kempton & Lave 1983). MANAGEMENT EFFORT AND THERMAL COMFORT

The valve theory does rather well in many situations. For example, if one assumes that thermostats do not have feedback mechanisms, a house would become colder when the weather is colder. The corresponding manage-

TWO THEORIES OF HOME HEAT CONTROL

233

merit rule is: When it is colder putside, you must turn up the heat. This practice is rationalized by the device model itself. Conversely, with a feedback device model, this adjustment would not seem necessary. In fact; the valve theory leads to correct management, since in many houses infiltration and distribution asymmetries will cool marginal rooms rnoreih cold weather/ ^i.p. "•1;.-.-; r. ::•: ."•:-.: -'•'v' - ;• A second jssue is whether a higher setting provides faster warmup. The feedback theory denies faster warmup because the furnace runs at a constant rate. However, due to human comfort factors and characteristics:: of interacting systems, a person entering a cold house from outside will not feel warm when the air first reaches the correct temperature.5 Thus/ greater comfort would be realized if the thermostat were set high, to raise the air temperature above normal initially, then returned to the normal setting. Again, the correct action would logically follow frorh the valve model, but not from the feedback model. ENERGY USE

The valve theory correctly predicts a third fact of considerable importance to the user: More fuel is consumed at higher settings than at lower ones. The prediction is correct, even if the explanation is wrong (higher fuel use does not occur not because of a valve opening wider but because higher inside temperatures cause more heat loss through the shell of the house). Nevertheless, higher use is a direct prediction of the valve theory, not of the feedback theory. Consequently, some interviews suggest that valve theorists are more likely to believe correctly that night setback saves energy. The following quote illustrates how the informant's husband follows the logic of the feedback theory to an erroneous conclusion: /: Now, my husband disagrees with me. He, he feds, and he will argue with me long enough, that we do not save any fuel by turning the thermostat up and down. . . . Because he, he feels that by the time you turn it down to 55 and all the objects in the house drops to 55°, and in order to get all the objects in the house back up to 65°, you're going to use more fuel than if you would have left it at 65 and it just kicks in now and then. This error is due to having the feedback theory without having additional theories of the interacting systems, particularly a theory of heat loss at different indoor temperatures. The feedback theory would work if it were augmented. But the necessary augmentations would complicate the model considerably - perhaps beyond the level of complexity most people are willing to bother learning about home heating. By contrast, if one considers the thermostat to be like a valve, these problems are solved with little effort, These points favor the valve theory. However, it seems to encourage frequent unneeded adjustments of the thermostat. These adjustments may induce considerable waste of time and human effort. Frequent adjustments

t ^ C — . . =«. .

..~S^

234

WILLETT KEMPTON

will not necessarily increase energy consumption as long as the thermostat is turned back down as soon as the house becomes warm. If the occupant forgets, energy will be wasted. * /

THE EXPERT'S PERSPECTIVE

.

'

-

Another problem with the valve theory is that it does not correspond to the mechanism inside the device. Inside, one sees a temperature-activated switch that can be on or off but cannot control the amount of heat from the furnace. This fact will seem a decisive failure to the few technically minded people who might actually look inside, but it is of little consequence for normal use of the thermostat. Why do heating experts consider the valve theory incorrect when it provides its users with about the same number of useful predictions as the feedback theory? After their training, experts possess a full, institutionally sanctioned theory. This full theory can be arrived at from the feedback theory by simply adding details and adjacent systems. By contrast, arriving from the valve theory requires a conversion - at some point, the learner will say something like, "Oh, I see, it's not a valve, it's an automatic switch." The technical experts will evaluate folk theory from this perspective""- not by asking whether it fulfills the need of the folk. But it is the latter criterion on which the anthropologist will rely due to her methodological training, and on which sound public policy must be based. Change in folk theory

!

By examining folk theory as it changes, further clues are provided about the consistency between the theory's metaphor and the inferences and practices derived from it. Such data were fortuitously provided in one interview, in which the informant initially held parts of each theory and later shifted predominately to the feedback theory. This is shown in the following series of quotes, which also illustrate the difficulties in definitively identifying a person's theory without an extensive interview. This informant, Peggy, is a language teacher who grew up in Michigan. In the beginning of the interview, she described the thermostat as sensitive to temperature:

*

P: I guess, what I always thought was when you turn the, the temperature, you turn the thermostat to 65, the furnace works to keep the room at 65 and then as soon as it's 65, the furnace stops working and then when it starts to get a *little bit cold again the furnace will work again. And I think, that the, the temperature on the thermos -1 think it keeps it at that temperature in that area. So maybe on the other side of the room, there might be window or a door and it could be draftier over there. But where the thermostat is, it will always be 65. Peggy is clearly aware of the feedback nature of the thermostat. Correspondingly, when I elicited the list of similar devices, Peggy choose a different set from Bill's:

i;

TWO THEORIES Of HOME HEAT CONTROL

235

P;: I guess it might be like an oven. You know, when you want to cook something at 350? You turntheoven on to 350 and it, a, the, fire or the electricity.or whatever, or the gas, works to keep the temperature at 350 and to not get hot-/ J ; ter and to not get colder. • v •:;; She also mentioned the refrigerator, saying that it has a number or letter for coldness, and the refrigerator is "working to maintain an.even temperature"; another example was the dial on the. toaster, which selects "lighter or darker." i;-v ^•.f^-,,.'"> •."'".•' . s-ii^.•;;?': -!>•: ' • '•-:''•'.)•",• After these portions of th£ Interview, I was certain that Peggy had a feedback theory. It was thus a surprise when I asked for her operating rules: W; Let's say it's very cold . . . you come into the house and it's very cold, and you want to heat the house up. Let's say you want to heat the house up to 65. What would you do . . . P: If it's very, very cold? W: Un-huh. P: I might turn it up to 70, for maybe 20 minutes, half an hour and then turn it back down to 65 to see if I can get it warmer faster. W: Uhh [trying to maintain composure] You wouldn't turn it to 80 or something, 85? P: I wouldn't feel comfortable doing that, no. W: Because of danger, or because of just cost, it would be wasteful, or? P: Yeah, I think it would be wasteful. I don't know, I just, I, even 70 seems immoral, somehow, you know. W: Yeah [laugh]. Well, I don't consider [it] a moral issue, I don't inject that, I just try to get an idea of what you think would work best for you. Let's, you could take two cases, maybe. In one case, there's no more energy crisis and you don't have to worry about it any more. P: Yeah, maybe 80. Of the 12 informants interviewed in depth about the thermostat, Peggy was one of 3 who described the thermostat with feedback theory but nevertheless in practice turned it higher to heat faster. Such complications make it difficult to determine which theory is being used from the answer to a single fixed question (as we might like to do in a survey to estimate national rates of these theories). In elaborating on this practice, Peggy compares the house thermostat with the oven: P: But I think people have the general idea, and I guess I do too, that the higher you put it [the house thermostat], the faster and harder it's going to work. W: Right, the harder it will work. P: . . . I know the oven, when you turn it on 450, well we have a gas oven, the flame is higher than it is if you turn it on at 200. During this description of the oven, she hesitates and says "But it might not work that way at all," meaning perhaps turning the house thermostat up further does not make it run faster; she says "It could be either way."

236

WILLETT KEMPTON

In thinking over the issue, she notes that her immediately previous description of turning the thermostat up to heat the house faster . . . P: . . . contradicts what I said at the beginning. That a thermos - how a ther, '"• mostat works, that it works up to a certain point and then when it gets to that point it turns off until, or it goes down until it needs to go oh again/Sb I thought, well if it does that, then if I turn it at 80, it's going to get to 65 after ttri minutes. If 1 turn it on 65, it's going to get to 65 . . .after ten minutes. And if I forget about it [at 80], it's going to keep going up to 80 unless I turn it back, v ; P; So, . . . I thought that it probably contradicted what I said before,: but th^n I kept thinking about the oven and how I know [unintelligible] it gets hotter faster. W: Un-huh. P; So I thought, well maybe some furnaces are different. W; Yeah, yeah. P: Maybe some are very well regulated and maybe some do just produce a little more heat. The concrete, observable case of the oven brings Peggy to use the valve theory in practice (whether she is correct about the oven is irrelevant here), but some internal reason seems to draw her to the feedback theory in her abstract description. I asked her how she arrived at her initial description of the thermostat. P: I don't know. I don't think I learned it from my parents because I never paid much attention to the thermostat. . . . [later] Oh, I know, one time when I was living in the same house [with friends] and I lived in one of the bedrooms that didn't have a register. It was an added-on, it was an addition to the house and they, they didn't bother to put, you know, to connect the furnace , . . there was no register or, yeah, so there was a little space heater in there and I know you turned the knob on that space heater, to a certain temperature and then the little, you know, the coils would light up and make a lot of noise, and then when it got to be that temperature, the whole thing would just turn off. And then a few minutes later it came back on. W: So you could see that happening. P: I - you could see that happening and feel that happening. And I may have thought, yea, that must be how the furnace, the regular furnace works too. And then maybe from things people said. This striking example shows how an immediately visible device can display its operation and thus influence folk theory. This experiential background, as Peggy points out, may also have dovetailed with someone's explanation, thereby reinforcing her recollection of both. As a result of integrating this information during the interview, Peggy said that the interview made her realize that turning the setting above the desired temperature was not effective. Another event that makes system operation highly salient is a change in the type of system. A rural couple who switched from nonthermostatic space heaters to a furnace describe the change:

TWO THEORIES OF HOME HEAT CONTROL

237

/; With the other ones, the space heaters, there was just no regulating to it. You know, if the temperature dropped outside, if you didn't wake up and go;out and turn it up, it drops, you know, it would cool down. But with, naturally: with the thermostatically controlled fuel oil furnace, it would kick on automatically by itself when it got down below that temperature. : •%;,,-.:..;; ,,,;•;;:, The amount of adjustment needed for a true value-controlled system is highly salient. Accordingly, this couple, and another who had grownup with wood heat, had a fully elaborated and consistent feedback theory.

Conclusions In studying residential heat control, I found that two folk theories were applied to thermostats. Of the two, the feedback theory is more closely related to expert theory. Home heating, like many other areas of knowledge in our society, has a "correct" set of theories defined by experts and their institutions. Informants who held the valve theory were insecure about it - they denied understanding the device, even when they had complete descriptive models and elaborate procedures for using it. This insecurity has also been manifested on several occasions of giving this paper as a talk. Some questioners seem to be inhibited by feeling that they may have a "wrong" theory and they do not want to be embarrassed by their questions. One might suppose that choice of the correct theory for thermostats concerns straightforward technical facts and presume that the experts must know how thermostats really work - after all, they design them! We have seen, however, that the folk theory endorsed by the experts may not work as well in practical day-to-day application. A theory that is useful for designing thermostats is not guaranteed to be a good theory for using them. More needs to be known before this research can have practical applications. In earlier work on household methods for measuring energy, my colleagues and I argued that many folk measurement methods are counterproductive and that individuals would benefit if folk methods were made more similar to expert ones (Kempton et al. 1982; Kempton & Montgomery 1982). In the case of folk theory for thermostats, the jury is still out. If people converted from the valve to the feedback model, they would save management effort by not having to adjust the thermostat so often and they would occasionally save energy by not forgetfully leaving it set high. However, widespread conversion to feedback theory would risk eliminating the theoretical rationale for night setback - an immensely larger penalty. A simple valve theory always directly predicts less use from lower settings since the valve is partially shut at lower settings. This prediction of lower use is not made by a simple feedback theory alone, but requires an accompanying theory of heat loss. The problem is seen by the quote on page 233, in which a feedback theorist argued that savings from night setback

238

WILLETT KEMPTON

of the thermostat would be cancelled by subsequent turning up of the thermostat. //,:: More research is needed, since two problems with the valve theory do argue for conversion. First, the problems with the first order feedback -theory can be solved by simply adding on components, whereas problems with the valve model require ad hoc repairs (as in Bill's Calvinism) or replacement of the entire model. A second argument is that when tfee operation of the system is made visible (either by a miniature model - the space heater - or by system conversion), the folk, on their own, choose feedback theory.

Postscript; thermostat management as an industry „...,'... . I close with a down-to-earth question: What is thermostat management worth? Although no national data exist on thermostat energy savings, rough estimates can be made. During the 1982 heating year, households in the United States spent $85 billion on direct energy purchases, averaging $1,022 per household (DOE, EIA 1983). Since the 1973 oil embargo, households have decreased their energy use by about 15% (Crane 1984; Williams, Dutt, & Geller 1983), which represents a current savings of $15 billion per year. Heating accounts for roughly half of residential energy cost; since consumers know more conservation methods for heating than for appliances, I estimate that heating accounts for two-thirds of the savings, or $10 billion annually. The only reliable estimate of actual thermostat savings has been made by Fels and Goldberg (1984), who analyzed New Jersey residential gas consumption. By a statistical procedure that compared monthly gas use with weather fluctuations, they were able to separate the effects of thermostat setting from other factors, such as home improvement (e.g., insulation) or more efficient appliances. They estimate that more than half of the natural gas savings were due to lower thermostat settings. If we assume the same proportion applies generally to heating fuels,6 $5 billion is saved annually due to changes in home thermostat use since the oil embargo. To put this number in perspective, Socal's recent agreement to purchase Gulf for $13 billion was the largest corporate acquisition in history (Cole 1984). With three years of thermostat savings, American households could have outbid Socal and purchased Gulf Oil for themselves. Although the dollar figure is an approximation, one can safely conclude that thermostat management provides American households with annual savings in the billions. Yet little reliable data link my aggregate national estimate to specific behavioral changes (e.g., nighttime setback versus constant lower settings), or to the cognitive and social systems that generate the behavior. Whatever the cause, increased household thermostat

TWO THEORIES OF HOME HEAT CONTROL

'239

management now provides disposable income for other spending. Thus, thermostat management can be considered a multibillion dollar cottage industry. Further study of this industry's production methods would seem warranted.

Notes 1. For comments on this paper, I am grateful to Dan Bobrow, Roy D'Andrade, Gautam Dutt, Peter Gladhart, Dedre Gentner, Dorothy Holland, Charlotte Linde, Ann Millard, Bonnie Morrison, William Rittenberg, Jeff Weihl, and an anonymous reviewer for Cognitive Science. Other helpful questions were raised following my presentations of this material at University of California, Irvine, the Princeton Conference on Folk Models, Xerox PARC, the MSU Families and Energy Conference, and at Bolt Beranek and Newman, Inc. Unpublished data were kindly provided by Gautam Dutt at Princeton University and Jim Barnett at the National Bureau of Standards (thermostat behavior records); and by Dedre Gentner and Yevette Tenney at BBN (Cambridge survey of college students). This work is supported by the National Science Foundation, under grant BNS-82 10088, and by the Michigan State University Agricultural Experiment Station, as project 3152. This paper is Michigan Agricultural Experiment Station Journal Article No. U141, and also appears in Cognitive Science 10(l):75-90 (1986). 2. DiSessa (1985) argues that the things McCloskey calls single coherent theories are in fact data-driven collections of heterogeneous "phenomenological primitives." These primitives originate in superficial interpretations of reality ap. plied to common situations via recognition. The data in this paper argue ibr more connectedness than DiSessa sees, although I acknowledge the possibility that the word theory conveys more consistency and coherence than is appropriate. 3. Nonrecognition of cognitive variation within a culture, and even within a single family, seems common (Kempton 1981). 4. Although I propose folk theory as an explanation of the pattern in Figure 9.1, many factors contribute to thermostat setting, and frequent shifts could be due to non-folk-theoretical causes such as domestic conflict over desirable setting. 5. This can be demonstrated by a simple thought experiment. On entering a warm building with a point-source of heat (say, a wood stove), a person feels cold although the building is warm. That person will choose to stand near the point source, despite the higher-than-normal temperature. After the person has "warmed up," he or she will choose a normal temperature. In the case of turning the thermostat up on entering a cold house, there appear to be two physical causes: (1) The cold near-body masses - clothing, skin surface, and trapped surface air - are heated more rapidly by warmer ambient temperature; and (2) When air temperature rises but (slower-heating) wall and furniture surfaces are still cold, a person will feel colder than air temperature because of infrared radiation losses. A more complete analysis of these effects would require quantitative analysis of their relative magnitudes and time constants. 6. This figure probably underestimates the thermostat's proportion, since Fels and Goldberg calculated it to be 50% of all gas conservation, whereas I consider it only 50% of heating conservation.

240

WILLETT KEMPTON

References Agar, M. , . . -.''. / 1980. Stores, background knowledge and themes: Problems in the: analysis of „ life history narrative. American Ethnologist 7(2):223-240. -'- ;—-'-'••••-• -* Cole, M. and S, Scribner /.;./;, 1974. Culture and Thought: A Psychological Introduction. New York: John Wiley and Sons. Cole, R. J. 1984. Socal Agrees to Buy Gulf in Record Deal. New York Times, 6 March 1984, p. 1. •---•"'•• ; Crane, L. T . 1984. Residential energy conservation: How far have we progressed and how much farther can we go? Committee print 98-R, Committee on Energy and Commerce, U.S. House of Representatives. Washington, D.C.: U.S. Government Printing Office. D'Andrade, R. G. 1981. The cultural part of cognition. Cognitive Science 5(3):179-195. de Kleer, J, and J. S. Brown 1983. Assumptions and ambiguities in mechanistic mental models. In Mental Models, D. Gentner and A. L. Stevens, eds. Hillsdale, N.J.: Lawrence Eribaum Associates. Pp. 155-190. DiSessa, A. A. 1982. Unlearning Aristotelian physics: A study of knowledge-based learning. Cognitive Science 6:37-75. 1985. Final report on intuition as knowledge. Unpublished manuscript, M.I.T. Laboratory for Computer Science, M.I.T., Cambridge, Mass. Dougherty, J. W. D., ed. 1985. Directions in Cognitive Anthropology. Urbana: University of Illinois Press. Dutt, G. S., A. Eichenberger, and R. H. Socolow 1979. Twin Rivers Project, public data set documentation. Report PU/CEES No. 78, Princeton, N.J.: Princeton University Center for Energy and Environmental Studies, : Department of Energy, Energy Information Administration 1983. Residential energy consumption survey: Consumption and expenditures April 1981 through March 1982, Part 1: National Data. DOE/EIA-0321/I(81). Washington, D.C.: U.S. Government Priming Office. Fels, M. F. and M. L. Goldberg 1984. With just billing and weather data, can one separate lower thermostat settings from extra insulation? In Families and Energy: Coping with Uncertainty, B. M. Morrison and W. Kempton, eds. East Lansing: Institute for Family and Child Study, Michigan State University. Pp. 195-206. Gentner, D. 1983. Structure-mapping: A framework for analogy. Cognitive Science 7:155170. Gentner, D. and A. L. Stevens, eds. 1983. Mental Models. Hillsdale, N.J.: Lawrence Eribaum Associates. Holland, D. 1985. From situation to impression: How Americans use cultural knowledge to get to know themselves and one another. In Directions in Cognitive Anthropology, J. W. D. Dougherty, ed. Urbana: University of Illinois Press. Pp. 389-411. Hutchins, E. 1980. Culture and Inference: A Trobriand Case Study. Cambridge, Mass.: Harvard University Press.

TWO THEORIES QF HOME HEAT CONTROL

\M

Johnson-Laird, P. N. •?^A'ioV 1981. The form and function of mental models, In Proceedings of Th|£d Annual Conference of the Cognitive Science Society, Berkeley: University of ' California. Pp. 103-105. '/..••^w;' Kay.P. -.. • :• .,.-v; '..- / r ; ^ - v - ; K / ^ S K : V "/ 1966, Ethnography and theory of culture. Bucknell Review 14:106-113.;^': • Kempton, W. ;•: i; /;nv/-' • 1981. The Folk Classification of Ceramics: A Study of Cognitive Prototypes. New York: Academic Press. SJi••':'••}' Kempton, W., P. Gladhart, D. Keefe, and L. Montgomery, .;.; ;.;,.. v. !:,;rV. 1982. Willett Kempton, letter and papers.//i Fiscal Year 1983ipept, of Energy Budget Review (Conservation and Renewable Energy). Hearings before the Committee on Science and Technology, U.S. House of Representatives, No. 112, Vol. 2. Pp. 1015-1068. Washington, D.C.: U.S. Government Printing Office. Kempton, W. and J. Lave 1983. Review of Mental Models, D. Gentner and A. L. Stevens, eds. American Anthropologist 85(4): 1002-1004. Kempton, W. and L. Montgomery 1982. Folk quantification of energy. Energy - The International Journal 7(10): 817-827. Lakoff, G. and M. Johnson 1980. Metaphors We Live By. Chicago: University of Chicago Press. Larkin, J. H. 1983. The role of problem representation in physics. In Mental Models, D. Gentner and A. L. Stevens, eds. Hillsdale, N.J.: Lawrence Erlbaum Associates. Pp. 75-98. Larkin, J., J. McDermott, D. P. Simon, and H. A. Simon 1980. Expert and novice performance in solving physics problems. Science 208:1335-1342. Lave, J. and B. Rogoff, eds. 1983. Everyday Cognition: Its Development in Social Context. Cambridge, Mass.: Harvard University Press. Marsella, A. J. and G. M. White, eds. 1982. Cultural Conceptions of Mental Health and Therapy. Dordrecht, Holland: D. Reidel Publishing Company. McCloskey, M. 1983a. Naive Theories of Motion. In Mental Models, D. Gentner and A. L. Stevens, eds. Hillsdale, N.J.: Lawrence Erlbaum Associates. Pp. 299-324. 1983b. Intuitive physics. Scientific American 248(4):122~130. McCloskey, M., A. Caramazza, and B. Green 1980. Curvilinear motion in the absence of external forces: Naive beliefs about the motion of objects. Science 210:1139-1141. O'Sullivan, P., and P. A. McGeevor 1982. The effects of occupants on energy use in housing. In Energy Conservation in the Built Environment. Proceedings of CIB W67 Third International , Symposium, Vol. 5. Dublin: An Foras Forbartha. Pp. 5.96-5.107. Quinn, N. 1982. "Commitment" in American marriage: A cultural analysis. American Ethnologist 9(4):775-798. Rappaport, R. A. 1979. On cognized models. In Ecology, Meaning, and Religion. Richmond, Calif.: North Atlantic Books. Pp. 97-144. Rice, G. 1980. On cultural schemata. American Ethnologist 7(1):152-171.

242

WILLETT KEMPTON

Socolow, R. H., ed. ' 1978/Savmg Energy in the Home: Princeton's Experiments at Twin Rivers. 'Cam! ; bridge,; Mass.: BalUnger.;;: .. ' -"' \ •/!•; Wiener, N. J • ^\-_rl'T: „*<: 1948, Cybernetics, or Control and Communication in the Animal and the Machine. New York: M.I.T. Press arid John Wiley and Sohs/<2hd ed; 1961) Williams, R. H . , G. S, butt, and H. S. GeHer ;;•!,:• • 1983.^Future energy savings in U.S. housmg. Annua! Review of Energy 8:269-332. Wiser, M. arid S. Caxvy'J:^:r-'u''': 1983. : When heat and temperature were one. In Mental Models, D. Gentner and A. L. Stevens, eds. Hillsdale, N J . : Lawrence Erlbaum Associates. Pp. 161-291: ' : • : ) . • " : .'.'; " ;;."":.-.,:::

10 How people construct mental models^

Allan Collins & Dedre Gentner

Analogies are powerful ways to understand how things work in a new domain. We think this is because analogies enable people to construct a structure-mapping that carries across the way the components in a system interact. This allows people to create new mental models that they can then run to generate predictions about what should happen in various situations in the real world. This paper shows how analogies can be used to construct models of evaporation and how two subjects used such models to reason about evaporation. As Lakoff and Johnson (1980) have documented, our language is full of metaphor and analogy. People discuss conversation as a physical transfer: (e.g., "Let's see if I can get this across to you" (Reddy 1979). They analogize marriage to a manufactured object: (e.g., "They had a basic solid foundation in their marriages that could be shaped into something good" (Quinn this volume). They speak of anger as a hot liquid in a container (Lakoff & Kovecses this volume); and they describe their home thermostat as analogous to the accelerator on a car (Kempton this volume). Why are analogies so common? What exactly are they doing for us? We believe people use them to create generative mental models, models they can use to arrive at new inferences. In this paper, we first discuss the general notion of a generative mental model, using three examples of artificial intelligence models of qualitative physics; second, we lay out the analogy hypothesis of the paper, which we illustrate in terms of the component analogies that enter into mental models of evaporation; and finally, we describe how two subjects used these analogies in reasoning about evaporation. The notion of running a generative model can be illustrated by an example from Waltz (1981). People hearing "The dachshund bit the mailman on the nose" spontaneously imagine scenarios such as the dachshund standing on a ledge, or the mailman bending down to pet the dachshund. Similarly, if you try to answer the question, "How far can you throw a potato chip?" your thought processes may have the feel of a mental simulation. Examples such as these suggest that simulation and generative inference are integral to language understanding (Waltz 1981). However, such

244

ALLAN COLLINS St DEDRE CENTNER

Table 10.1. Examples of liquids in different states (after Hayes 1985)

Still, in Bulk Moving, in Bulk Still, Divided Moving, Divided

Supported on surface

Supported in space

Liquid on a wet surface Liquid flowing on a surface, e.g., a roof Dew drops on a surface Raindrops on a window

Liquid in a container Liquid pumped in pipe

Unsupported •

•

'

'

-

'

• • • ' • • "

:

' - '

:

"

: K

•"••

Liquid pouring from a container

Mist filling a valley?

Cloud

Mist rolling down a valley?

Rain

imagistic descriptions have a magical quality, which we try to resolve in terms of the formalisms of mental models research. Inference and qualitative simulation become possible when the internal structure of a model is specified in terms of connections between components whose input-output functions are known. Hayes (1985) and de Kleer (1977) have independently tried to characterize how people decompose different systems in order to reason about the world. Both came up with tacit partitions of the world in order to simulate what will happen in a particular situation. Hayes attacked the problem of how people reason about liquids and de Kleer how they reason about sliding objects. Forbus (1981) later extended the de Kleer analysis to bouncing balls moving through two dimensions. Understanding these ideas is central to our argument about the role of analogies in constructing mental models, so we briefly review the way these three authors partition the world in order to construct qualitative simulations. Hayes (1985) partitions the possible states of liquids into a space with three dimensions: (1) whether the liquid is moving or still; (2) whether it is in bulk or divided (e.g., a lake vs. mist); and (3) whether it is on a surface, supported in space, or unsupported. For example, rain is liquid that is moving, divided, and unsupported, whereas pouring liquid is moving, in bulk, and unsupported. Spilled liquid is still, in bulk, and on a surface except when it is first moving on the surface. Hayes gives examples for most of the possible states in these three dimensions (see Table 10.1), except some that are impossible (e.g., bulk liquid that is both still and unsupported). Hayes shows how one can construct transitions between different liquid states using a small number of possible transition types in order to construct "a history" of some event. He illustrates this with the example of pouring milk from a cup onto a table. InitiaUy, the milk is contained in

HOW PEOPLE CONSTRUCT MENTAL MODELS

/

A

D

V E

C

•245

\

B J

Figure 10.1. Partitioning of loop-the-loop track (after de Kleer 1977) bulk in the cup as it is tipped. When the surface reaches the lip of the cup, there begins a falling from the cup to the table, which extends through time until the cup is empty, except for the wetness on its surface. Beginning at the point when the falling liquid hits the table, there is a spreading of the liquid on the table top until the pool of liquid reaches the table edge. At that point, another falling starts along the length of the edge that the spreading reaches. This falling continues until there is only a wetness on the table top. The falling also initiates a spreading on the floor, which lasts until there is a nonmoving wetness covering an area of the floor. Because people can construct this kind of history out of their knowledge of liquid states and the transitions between them, they can simulate what will happen if you pour a cup of milk on a table. Alternatively, if they find liquid on the floor and table, they can imagine how it got there. In the roller coaster world that de Kleer (1977) has analyzed, he partitions the kinds of track a ball might roll along into concave, convex, and straight tracks, de Kleer uses a small number of allowable transitions for example, slide forward, slide back, and fall - to construct a simulation of the behavior of a ball on a loop-the-loop track such as shown in Figure 10.1. In the figure, the track starts in segment A and continues up and around through segments B, C, D, E, and F. By constructing all possible continuations for a given input, one forms a directed graph, which de Kleer calls the envisionment. Each alternative transition is a branch that can be followed out. Suppose one starts a ball rolling at the end of segment A, It will slide forward to segment B, From B, it can slide forward into C or slide back to A. If a slide forward occurs, then the ball can either slide forward into D or fall. Sliding backward from B into A leads to oscillation. The same sort of branching of possible states occurs in Hayes's physics of liquids. For example, in the milk-pouring episode, the spread of milk on the table may never reach the edge, in which case the episode ends with wetness on the table and in the cup. Thus, using qualitative models can allow a person to generate all the different possible events that might happen. Forbus (1981) has made a similar analysis of bouncing balls in twodimensional space. To do this, he developed a vocabulary for partition-

246

ALLAN COLLINS Sc DEDRE GENTNER

Figure 10.2. Partitionirtg of two-dimensional space above a well (after Forbus)

ing space into places in which significantly different behaviors can occur. In this model, the qualitative state of an object at a given moment consists of a place, an activity (e.g.: fly or collide), and a direction (e.g., left and up). As in the other two models discussed, simulation rules specify the allowable transitions between qualitative states. Figure 10.2 shows his example of how a space above a well would be partitioned into regions. Table 10.2 shows the allowable state transitions for a ball moving in a region. The table shows the region in which the ball will be next (including "same") and its next direction. What the ball will do next - its next activity - depends on the kind of place it enters next. If the next place is a surface, the next activity is a collision; otherwise, the next activity will be a continuation of its present motion. If we begin with a ball in region A that is headed left and down, it can either go downward and collide with the surface below A, or go into the middle region B, where its direction will also be left and down. From B, going left and down, it can either go into the lower region D or into the left region C, still heading left and down. If it goes into D3 it can collide either with the left side wall or with the bottom. Collisions with vertical walls reverse the left-right direction; collisions with horizontal walls reverse the up-down direction. As should be evident, it is possible to imagine a number of different paths for the ball to travel, only some of which end up with the ball caught in the well. The point of these examples is that they illustrate how people might be able to construct mental models that have the introspective feel of manipulating images (Kosslyn 1980). Put in the terms of Hayes, de Kleer, and Forbus, there is no magic to starting a ball moving or a liquid flowing mentally and seeing what happens. You do not have to know in advance what happens, and you do not have to store a mental moving picture of the events. Indeed, you may decide that a particular candidate event

HOW PEOPLE CONSTRUCT..MENTAL MODELS

247.

Table 10.2. Transition rules for a ball moving from region to region in Forbus's (1981) bouncing ball world showing next place and next direction for the case of a ball moving income region of space •• ; --

Direction none LD.; L LU

U RU

R RD

Next Place .same down down left same same left left up same up same right right up same right down

Next"'--Direction D

p".. LD LD LD L L LU LU D U R R RU RU RD RD RD

Note: Regions are represented by words and directions by capital letters. Each line represents a next-place, next-direction pair. could not ever have occurred. All you have to know is what inputs lead to what outputs for each state transition and how those kinds of states are connected together. The analogy hypothesis It should by now be clear that qualitative-state models provide a powerful, versatile way for people to reason about familiar domains in which the states and transitions are known. But what happens when people want to go beyond familiar physical situations and reason about domains, such as evaporation, in which the states and transitions may be unfamiliar or even invisible? Here, we come to the central proposal of this paper, the analogy hypothesis. According to this hypothesis, a major way in which people reason about unfamiliar domains is through analogical mappings. They use analogies to map the set of transition rules from a known domain

248

ALLAN COLLINS & DEDRE GENTNER

(the base) into the new domain (the target), thereby constructing a mental model that can generate inferences in the target domain. Any system whose transition rules are reasonably well specified can serve as an analogical model for understanding a new system (Collins & Gentner, 1982; Gentner 1982). So far, this analogy hypothesis is a special case of Gentner's (1982; 1983; Gentner & Gentner 1983) more general claim that analogy is a mapping of structural relations from a base domain to a target domain that allows people to carry across inferences from the base to the target. To construct a mental model in a new domain, a particularly powerful set of relations to map across is the transition rules between states. These rules allow one to generate inferences and create simulations in the target domain analogous to the ones that can be performed in the base domain. However, the situation is often more complicated. Often, no one base domain seems to provide an adequate analogy for all the phenomena in the target domain. In these cases, we find that people partition the target system into a set of component models, each mapped analogically from a different base system. As we demonstrate, people vary greatly in the degree to which they connect these component models into a consistent whole. An extreme case of inconsistency is the pastiche model, in which a target domain model is given by a large number of minianalogies, each covering only a small part of the domain and each somewhat inconsistent with the others. At the other extreme, some people connect together their component models into a consistent overall model. Thus, they can combine the results of their mappings to make predictions about how the overall target system will behave. The remainder of this paper illustrates the analogy hypothesis by showing how analogies can be used to construct different versions of a molecular model of evaporation (see Stevens & Collins 1980). We then show how two subjects used these analogies to reason about evaporation. A molecular model of evaporation involves a set of component subprocesses; 1. How molecules behave in the water 2. How molecules escape from the water to the air 3. How molecules behave in the air 4. How molecules return to the water from the air 5. How molecules go from liquid to vapor, and vice versa Notice that this analysis bears some similarity to the Forbus analysis of bouncing balls. There are two regions, the water and the air, and the transitions (i.e., escape and return) between them. The behavior of the molecules in the water and air describes the transitions that keep the molecules in the same region. There is also a second kind of transition, from liquid to vapor and vice versa. Thus, there are two types of state transi-

HOW PEOPLE CONSTRUCT MENTAL MODELS

•2491

SAND-GRAIN MODEL

EQUAL SPEED MODEL

RANDOM SPEED MODEL

MOLECULAR ATTRACTION MODEL

/VN Figure 10.3. Component models of the behavior of molecules in water

tions that occur for evaporation processes: from one region to another, and from one phase to another. We contrast different possible views of each of these component processes. Some of these views we have clearly identified in subjects' protocols; others are only alluded to. Where the protocols were unclear as to which of two alternative models was implied, we have generally included both models. This is not an exhaustive set of all possible component models but only of those that were suggested in subjects' protocols. However, they do show how people can derive their views from different analogies. BEHAVIOR IN WATER

Figure 10.3 shows four different analogical models that subjects might have of how molecules behave in water. The first view we call the sandgrain model - the molecules just sit there like grains of sand, moving and slipping when something pushes on them. The temperature of the water

250

ALLAN COLLINS & DEDRE G.ENTNER

HEAT THRESHOLD MODEL

BELOW-BOILING

ABOVE-BOfLING

ROCKETSHIPand MOLECULAR ESCAPE MODELS

Figure 10.4. Component models of how water molecules escape from water to air is the average temperature of the individual molecules. This is a very primitive model. The next two views assume that the molecules are bouncing around in the water like billiard balls in random directions (Collins & Gentner 1983). In both views, the speed of the molecules reflects the temperature of the water. The difference is that in one version - the equal speed model - all the molecules are moving at the same speed. The other version is a random speed model, which allows for differences in speed for different particles. On this view, temperature reflects the average speed of a collection of molecules. The fourth view, called the molecular attraction model, incorporates attraction between molecules into the random speed model. In it, molecules move around randomly, but their paths are highly constrained by the attractive (or repulsive) electrical forces between molecules. This view is essentially correct. ESCAPE FROM THE WATER

Figure 10.4 shows three possible component models for escape (pictorially, two are the same). What we labeled the heat-threshold model is a threshold view of escape: The molecules have to reach some temperature, such as the boiling point of the liquid, and then they pop out of the liquid, the way popcorn pops out of the pan when it is heated. The remaining two models focus on molecular velocity, rather than on the incorrect notion of molecular temperature. The rocketship model is based on the assumption that the molecules in the water are moving in random direc-

HOW PEOPLE CONSTRUCT MENTAL MODELS

251

CONTAINER MODEL

VAFUA3LE-SIZE-R00M MODEL COLD AIR

WARM AIR

EXCHANGE-OF-EfMERGY MODEL COLD AIR

WARM AIR

Figure 10.5. Component models of the behavior of water molecules in air (water molecules are filled circles, air molecules are open circles)

tions. To escape from the water (like a rocketship from the earth), a molecule must have an initial velocity in the vertical direction sufficient to escape from gravity. The third view, the molecular escape model, posits that the initial velocity must be great enough to escape from the molecular attraction of the other molecules. Both latter models are in part correct, but the major effect is due to the molecular attraction of the water. BEHAVIOR IN THE AIR

Three component models of how the water molecules behave in the air are depicted in Figure 10.5. The container model posits that the air holds water molecules and air molecules mixed together until it is filled up (at 100% humidity). The variable-size-room model is a refinement of the container model to account for the fact that warm air holds more moisture than cold air. In this model, molecules in warm air are further apart and

252

ALLAN COLLINS & DEDRE GENTNER

CROWDED ROOM MODEL

AGGREGATION MODEL

*<*

RECAPTURE MODEL o~*^

Figure 10.6. Component models of how water molecules return from air to water so are less dense than molecules in cold air." That leaves more space to put water molecules in warm air than in cold air. In the exchange-of-energy model, the chief reason that cold air holds less moisture than warm air is that its air molecules are less energetic. When water molecules in the air collide with air molecules, they are more likely to give up energy if the air is cold (and hence less energetic) than if it is warm. If the water molecules become less energetic, they are more easily captured by the molecular attraction of other water molecules (or a nucleus particle) in the air. When enough water molecules aggregate, they will precipitate. This latter view is essentially correct.2 RETURN TO THE WATER

Figure 10.6 shows three models of how water molecules return to the water. The crowded room model assumes that when all the space in the air is filled, no more water molecules can get in. This is more a prevention-ofescape model than a return model. The aggregation model assumes that water molecules move around in the air until they encounter a nucleus or particle (which could be another water molecule) around which water accumulates. The less energetic the molecule, the more likely it is to be caught by the molecular attraction of the particle. As these particles accumulate water, gravitational forces overcome the random movement of

HOW PEOPLE CONSTRUCT MENTAL MODELS

253 -:

the particles and they precipitate. The recapture model assumes that par-? tides are attracted by the surface of the water (or other surfaces), fThe less energy they have, the more likely they are to be recaptured.'The ac-; tion in this view takes place near the surface, unlike the aggregation view, A fourth possibility is to ignore return processes altogether.;Some of ;oiiir subjects described evaporation solely in terms of water leaving the liquid state and appeared unaware of any need to consider the other direction,. of water vapor returning to the liquid state. Both the aggregation and the recapture models are essentially correct, but the aggregation model takes place over a long time period with relatively high humidities, whereas "the recapture model is applicable in any situation in which evaporation is occurring. LIQUID-VAPOR TRANSITION

Figure 10.7 shows four different views we have identified for the transition from liquid to vapor and from vapor to liquid. One view, the coterminus model, is that the transition occurs when the molecules leave the

T

COTERMINUS MODEL

INTRINSIC STATE MODEL

0,

0"n

Jh

DISASSEMBLY MODEL

BINDING MODEL

Figure 10.7. Component models of the liquid-vapor transition

254

ALLAN COLLINS & DEDRE GENTNER

water and escape to the air, and vice versa. In this view, the two transitions, between water and air, and between liquid and vaporware the same transition. In other words, whether a molecule is in the vapor or liquid state depends solely on the location: All molecules beneath the surface -- of the water are liquid, and all molecules above the surface of the water are vapor. A second view, the intrinsic state model, treats the liquid or gas state as an intrinsic property of the molecule. If the molecule becomes hot enough, it changes from liquid to vapor, and if it becomes cold enough, it changes from vapor to liquid. Location is correlated with state in that molecules in the vapor state tend to move into the air, whereas molecules in the liquid state remain in the water. A third view, the disassembly model, is based on a little chemistry: In it, liquid water is thought of as made up of molecules of H 2 0, whereas the hydrogen and oxygen are thought to be separated in water vapor. The expert view, which we call the binding model, is based on molecular attraction: Water molecules in the liquid state are partially bound together by electrical attraction of the neighboring molecules, whereas molecules in the gaseous state bounce around rather freely. The bubbles in a boiling pan of water are thus water molecules that have broken free of each other to create a small volume of water vapor, and clouds and mist are microscopic droplets of liquid water that have condensed but are suspended in the air.

COMBINING COMPONENT MODELS

Table 10.3 summarizes all the component models described here. Subjects can combine these component models in different ways. The following section shows two subjects, RS and PC, who had different models. RS had a model constructed from the random speed model of water, the rocketship or molecular escape model of escape, the variable-size-room-model of the air, the crowded room model of return, and the coterminus model of the liquid-vapor transition. The other subject, PC, had a less consistent and less stable model of evaporation. His view included something like the heat-threshold model of escape, the container model of the air, the recapture model of return, and the intrinsic state model of the liquidvapor transition. In contrast, as we have indicated, the expert view is made up of the molecular attraction model of water, the rocketship and molecular escape models of escape, the exchange-of-energy model of the air, the aggregation and recapture models of return, and the binding model of the liquid-vapor transition.

An experiment on mental models Four subjects were asked eight questions (shown in Table 10.4) about evaporation. They were asked to explain their reasoning on each question. The

HOW PEOPLE CONSTRUCT MHNTAL MODELS

<&£$-:•

Table 10.3. Component models of evaporation Models of Water " • • ' : ; •:-:'.^;v„ •';• v'-; ;•:*::•••••;•< :.: Sand-Grain M o d e l ; '•.-••-.r//6 ; Equal Speed Model , . .=;.:• Random Speed Model Molecular Attraction Model . Models of Escape , -Heat-Threshold Model...^ Rocketship Model . ,.; Molecular Escape Model ,.. Models of Air • • :. ; ;^ Container Model Variable-Size-Room Model Exchange-Of-Energy Model Models of Return Crowded Room Mode! Aggregation Model Recapture Model Models of Liquid-Vapor Transition Coterminus Model Intrinsic State Model Disassembly Model Binding Model

Table 10.4. Evaporation questions Question 1. Which is heavier, a quart container full of water or a quart container full of steam? Question 2. Why can you see your breath on a cold day? Question 3. If you put a thin layer of oil on a lake, would you increase, decrease, or cause no change in the rate of evaporation from the lake? Question 4. Which will evaporate faster, a pan of hot water placed in the refrigerator or the same pan left at room temperature and why? Question 5. Does evaporation affect water temperature, and if so how? Why or why not? Question 6. If you wanted to compress some water vapor into a smaller space but keep the pressure constant, what would you do? Why? Question 7. On a hot humid day, you must sweat more or less or the same amount as on a hot dry day at the same temperature. Why? Question 8. If you had two glasses of water sealed in an air-tight container, and one was half filled with pure water, while the other half was filled with salt water, what would you expect to happen after a long period of time (say about a month)? Why?

256

ALLAN COLLINS & DEDRE CENTNER

subjects were two male Harvard undergraduates, one female secretary with a college degree, and one female doctoral student in history. All were reasonably intelligent but were novices about evaporation processes. Ouf analysis centers on the first two subjects because they best illustrate the kind •- of reasoning we see in novice subjects. ' ••••••. ; '-r•*• ANALYSIS OF SUBJECT RS'S MODEL OF EVAPORATION PROCESSES

?

'--;

The first subject (RS) has a model of evaporation processes based oh (1) the random speed model of water, (2) some variation of either the rocketship or the molecuikr^escape model of escape, (3) the variable-size-rbom model of the air, (4) the crowded room model of return, and (5) the coterminus model of the liquid-vapor transition. His view includes notions of the energy needed for molecules to escape from a body of water and the difficulty water molecules have in entering a cold air mass because of the higher density. He seems to share a common misconception with the second subject: namely, that visible clouds (such as one sees coming out of a boiling kettle) are made up of water vapor rather than recondensed liquid water. This misconception forced him into several wrong explanations, even though his reasoning powers are impressive. As we show, he seems to check out his reasoning by running different models of evaporation processes (Stevens & Collins 1980) and to try to account for any differences in results when he finds them. We present the most relevant portions of his responses to three of the questions and our analysis of his reasoning processes (omitted portions of his response are indicated by dots). Q2: On a cold day you can see your breath. Why? RS: I think again this is a function of the water content of your breath that you are breathing out. On a colder day it makes what would normally be an invisible gaseous expansion of your breath (or whatever), it makes it more dense. The cold temperature causes the water molecules to be more dense and that in turn makes it visible relative to the surrounding gases or relative to what your breath would be on a warmer day, when you don't get that cold effect causing the water content to be more dense. . . . So I guess I will stick with that original thinking process that it is the surrounding cold air - that the cold air surrounding your expired breath causes the breath itself (which has a high water content and well I guess carbon dioxide and whatever else a human being expels when you breathe out), causes the entire gaseous matter to become more dense and as a consequence become visible relative to the surrounding air. What in fact happens on a cold day is that the invisible water vapor in one's breath condenses, because it is rapidly cooled. This condensed liquid water is visible as clouds or mist. However, the subject did not know that the clouds were liquid rather than water vapor, since he seems to think of water molecules in the air as vapor by definition (i.e., the coterminus model), so he needed to find some other account of what happens. For

HOW PEOPLE CONSTRUCT MENTAL MODELS

•2SJ

o o

COLD DAY

WARM DAY

Figure 10.8. RS's model of why you see your breath on a cold day that, he turned to his knowledge that cold air masses are more dense than warm air masses - his protocol includes a clear statement of the variablesize-room model of the air. Our depiction of the process he imagined is shown in Figure 10.8. What his view seems to involve is a vapor-filled gas cloud being emitted from a person's mouth. On a cold day, the surrounding air cools the emitted breath, causing it to compact into a.small visible cloud. The denser the breath, the more visible it is. This latter inference has its analogue with smoke or mist: The more densely packed they are, the more visible they are. In fact, as the breath disperses, the particles become less visible. This suggests that the subject may be invoking an analogy, not explicitly mentioned, to the behavior of smoke or mist in constructing his model. The next protocol segment clearly illustrates his belief in the variablesize-room model of the air and the crowded room model of return. Q4: Which will evaporate faster, a pan of hot water placed in the refrigerator or the same pan left at room temperature? Why? RS: When I first read that question, my initial impression, that putting a pan of hot water in the refrigerator you suddenly have these clouds of vapor in it, threw me off for a second. I was thinking in terms of there is a lot of evaporation. Well I guess, as I thought through it more, I was thinking that it was not an indication of more evaporation, but it was just (let us say) the same evaporation. Immediately when you put it in anyway, it was more visible. Ahmm, as I think through it now, my belief is that it would evaporate less than the same pan left standing at room temperature and my reasoning there is that the air in the refrigerator is going to be relatively dense relative to the room temperature air, because at a colder temperature again its molecules are closer together and that in effect leaves less room to allow the molecules from the hot water to join the air. . . . Here, the subject first simulates what happens at the macroscopic level when you put a pan of hot water in the refrigerator. He imagines clouds

258 COOL AiR MASS

ALLAN COLLINS & DEDRE GENTNER WARM AIR MASS o « 0 o • / ° \ Q /a

/

Figure 10.9. RS's model of evaporation from a hot pan of water in the refrigerator of steam coming out of the pan and initially thinks there must be more evaporation. It is unlikely he would ever have seen what happens when you put a pan of hot water in a refrigerator. He probably constructed this process from some analogous situation(s), such as warm breath on a cold day, running hot water in a cold room, or mist rising off a lake, for example. He is correct about the visible clouds of vapor, but as he concluded later, these clouds do not represent more evaporation. In fact, they represent condensation of the evaporated moisture (which he did not know). This subject frequently considers different perspectives on a question. Thus, he did not stop with his macroscopic analysis; he also simulated what would happen at a microscopic level. Figure 10.9 shows his view of why cold air leads to less evaporation. It is a clear statement of what we have called the crowded room model. There is not enough space, so some of the water molecules bounce back into the water, as shown in the picture, whereas they do not when the air is warm. This is an incorrect model that RS probably constructed on his own. However, it leads to a correct prediction in this case. The last protocol segment shows RS following three different lines of reasoning: Q5: Does evaporation affect water temperature? If so, in what way, and why? RS: . . . My initial impression is that it doesn't. It does not affect the water temperature. The surfaces of that water, the exterior surfaces of that water are in contact with air or some other gaseous state, which allow molecules from the water to evaporate into that other gaseous substance. So I guess I don't see where the loss of water molecules would affect the water temperature. In a way, however, I guess those water molecules that do leave the surface of the water are those that have the highest amounts of energy. I mean, they can actually break free of the rest of the water molecules and go out into the air. Now if they have a, if they are the ones with the most energy, I guess generally heat is what will energize molecules, then that would lead me to believe that maybe, although it may not be measurable, maybe with sophisticated instruments it is, but maybe it would be measurable after your most energetic molecules have left the greater body of water. Those that remain are less energetic and therefore their temperature perhaps is less than when all of the molecules that have already left or were a part of the whole. So I guess that has led me into a circle to perhaps change my initial response - maybe now I could im-

HOW PEOPLE CONSTRUCT MENTAL MODELS

\2$9j.

agine where even if not measurable because of a lack of a sensitive enough InstfjiV ment, maybe my logic or noniogic (whichever the case may be) has alld\v6d me; to believe that maybe there is an effect on the water temperature. So I feetpretty good about i t . . . . I guess even the original model with the sun. striking the • surface of water, that I thought initially, threw me off. ;Maybe that ;is;accUr|te / . too, because the sun's rays certainly don't go down to the bpttom^fSo: i;;gUesf; > it depends how deep we are talking, of course. But it would penetrate Only so far into the surface of the lake and energize those.layers that areatJh^.top'.siir- ' • face of the lake or water, whatever the body of water is. So thos^top layers1 would then be more energized than deeper layers, and jt would obviously !beo more able to evaporate off. As they evaporate off, assuming conditions are constant and you have got a situation where the succeeding layers get struck by the same heat energy, in this case the sun, from the sun. And they become as the preceding layers were, they become energized and then go off. You are left with less and less water, where a surface is at a given temperature and the rest is at a lower temperature, but your volume is less. So therefore it'sounds like again, I am getting to the point where the average temperature would increase, ahmn, because you have less volume, but the surface is at a higher energy level then as a percentage of the whole. Initially, he cannot think of any way in which evaporation affects the temperature of the water and starts to conclude that there is no effect. This is a lack-of-knowledge inference, that we discuss elsewhere (Collins 1978; Gentner & Collins 1981). Then, based on the random speed model of water and either the rocketship or molecular escape model of escape (which make the same prediction here), he infers that the higher energy molecules are the ones that will escape, leaving behind molecules with less energy. This, he infers correctly, will cool the water. Said in another way, molecules headed up from the surface with less energy will not escape, whereas those with more energy will escape. This is a subtle inference that follows directly from the rocketship and molecular escape models. Notice he does not have the misconception that all molecules in the water are moving at the same speed (i.e., the equal speed model). His reasoning depends crucially on variation in speed (i.e., the random speed model). Later in his response, however, he comes up with an argument that leads to a conclusion opposite to the one above. He realizes that as water evaporates away, there is less water left for the sun to heat. Thus, given a fixed input of energy, the temperature of the water will rise. Why does he adopt this line of reasoning, given that nothing in the question supposed there was any further input of energy? RS seems here to be considering a canonical example of evaporation: the sun's warming a lake. Under these conditions, the temperature of the remaining water will indeed rise. Here, two generative models come into conflict. The first view is based on a microscopic model of molecules escaping. The second view is based on layers of water evaporating, leaving behind less water to be heated by

260

ALLAN COLLINS & DEDRE GENTNER

the sun. As in the previous protocol, the subject worked on the problem from different viewpoints; but in this case, two opposing conclusions seemed correct and he did not know how to choose between them.: We have tried to show how RS drew on a set of underlying component ,. "models to construct his molnmlar model of evaporation and how he used his model to find answers to novel questions. Based on this model, RS was able to deal quite successfully with the eight questions (he gave essentially correct answers on seven of eight questions). Nevertheless, his model was incorrect in the ways we have described. ANALYSIS OF SUBJECT PC'S MODEL OF EVAPORATION PROCESSES

The second subject, PC had many more difficulties in dealing with these questions. His view was much less coherent. He relied frequently on local analogies to phenomena he had observed or things he had heard about, shifting among them without checking their consistency. This shifting made it difficult to ascertain his models, especially his model of water itself. However, most of his responses suggest that he combined (1) the sandgrain model of water, (2) the heat-threshold model of escape, (3) the container model of the air, (4) the recapture model of return, and (5) the intrinsic state model of the liquid-vapor transition. What is most striking about his view is how he treats heat as an intrinsic property of individual molecules rather than as a property of aggregates of molecules, as experts do. The first protocol shows PC's response to question two: Q2: On a cold day you can see your breath; why? PC: The reason is because the air that you breathe, or rather the air that you should breathe out, comes from your body and is hot air. The air which surrounds your body, because it is a cold day, will be cold air. When the hot air that you breathe meets with the cold air of the atmosphere, it will tend to vaporize almost like steam from a kettle, which of course, can be seen. Thus unlike on a hot day, when there is hot air around you and the hot air that you breathe are the same temperature, roughly, you cannot see your breath because the steam will not be formed, but on a cold day because of the variation in the temperatures and the vaporization of your breath, you can see when you are breathing. This phenomenon would not occur on a hot day because of the similarity in temperature.

This response would be correct as far as it goes if the words vaporize and vaporization were replaced by the words condense and condensation, This could be a simple lexical confusion, but other answers suggest it is a conceptual confusion (i.e., he really does think that the visible cioud formed is vapor). One piece of evidence for this belief occurs in answer to question four, where he refers to "the water which has already condensed in the refrigerator" (see below), suggesting he does understand the

HOW PEOPLE CONSTRUCT MENTAL MODELS

^261

distinction between vaporizing and condensing. Furthermore, it is a common misconception to think of a cloud of steam as water vapor, gather than recondensed liquid water. His response suggests he does inothave a coterminus model of the liquid-vapor transition, as RShad, since Jie sees the transition occurring in the air. Rather, he appears to have ; con-: structed a version of the intrinsic state model, where the transition depends on a temperature difference between the air and the liquid \yater. :.:';:• His analysis relies on an analogy to a core observation he has seen many times - steam from a kettle. To solve the problem, he has. to find some factor in common between one's breath on a cold day and steam from the kettle that is different from one's breath on a warm day. The factor he comes up with is the temperature difference. Boiling water from a kettle is much warmer than room air. Likewise, warm breath is much warmer than air on a cold day. But warm breath is about the same temperature as air on a warm day. Since he believes that steam is vaporized water, he therefore concludes that the temperature difference must somehow cause vaporization of water. He does not push his analysis down to a microscopic level, as did RS, to account for how this might be. Instead, he stops with identifying the analogous factor in the two situations. PC's response to the fourth question shows some of the conceptual problems he has: Q4: Does a pan of hot water in the refrigerator evaporate more, less, or the same amount as the same pan left standing at room temperature? Why? PC: Hmmmm. Hot water, when it meets with - wait a second. Hot water will only evaporate if it is sufficiently heated. By placing the hot water in a refrigerator you are not heating it, you are cooling it, and in fact the reverse process will occur. That is, the water will decrease in temperature and the refrigerator's temperature itself will also decrease. Not only that, but also the water which has already condensed in the refrigerator, because of the heat from the hot water when you put it in the refrigerator, will go into water itself because of the difference in the temperatures. To sum up, the pan of water in the refrigerator will not evaporate at the same rate, but will evaporate at a lesser rate than the same pan left standing at room temperature for the above reason. This question came soon after the question about breath on a cold day, and PC starts out answering the same way, "Hot water, when it meets with - ." This line of reasoning would conclude "the cold air, vaporizes to produce clouds of steam, and so there is more evaporation" just as RS has at first imagined and PC had argued in response to question two. But PC aborted that line of reasoning, whereas RS pursued it and ruled it out as incorrect. The next sentence reveals PC's heat-threshold model of escape: that the temperature of the water has to be sufficiently high (presumably boiling) for evaporation to occur. Sometimes in his answers, PC invoked this heat-threshold notion, and other times he violated it, In answer to the question, PC concludes that evaporation will decrease

262

/

ALLAN COLLINS & DEDRE GENTNER

for two reasons. First, the water in the pan will cool, and hence evaporate less or not at all. Second, the water that has condensed in the refrigerator will tend to "return to the pan of water." Both ideas are essentially corr e c t , but there are several incorrect statements associated with the latter argument. One is that the refrigerator's temperature will decrease (just like the water's temperature) because a warm pah of water is put in it the opposite is in fact true. Then he seems to conclude that the warm r temperature of the water will cause condensation because of-the temperature difference with the air: a kind of inverse process from the one he argued for in response to question two. In fact, the warmth from the pan of water will tend to reduce condensation. Nowhere in this or other answers does he regard the air as anything other than a passive container for water and air molecules - he does not mention anything like the variable-room-size model that RS described, or the exchange-of-energy model that experts hold. In fact, he seems unaware that cooler air holds less moisture than warm air. However, he . does seem to have a notion in this and other answers that molecules from the air will return to the water. This notion seems to depend on what he calls "condensation" in this answer, but not so clearly in other statements he makes. We have characterized this as a recapture model of return, rather than an aggregation model. Even though he seems to say that recapture must be preceded by "condensation," it is not clear that his "condensation" involves aggregation of molecules, as in the expert model. Rather, in keeping with his intrinsic state model of the vapor-liquid transition, "condensation" may be a change in state that occurs individually to each molecule. PC's answer to the sixth question provides the clearest example of his view of heat as an intrinsic property: Q6: What would you do if you want to compress some water vapor into a smalUer space but keep the pressure constant? PC: Hot air rises. Vapor is air, ok. Therefore, if you have a greater amount of vapor and you want to compress it, all you do is you heat the vapor so that by heating it, one will be causing the molecules to react faster which would in, crease the temperature of initially some molecules of steam, which will then go to the top and which will eventually increase the temperature overall, which will all - all the molecules will want to go to the top of the container, and as a result one will have a level of steam at the very top of the container and a vacuum at the end of the container - now I have got to get that down. Initially, tthe temperature of some of the molecules of steam, which because hot air, in this case steam, will always rise above in the colder air, the hotter steam will rise to the top of the container. Eventually, all the molecules of steam will be all trying to get to the top of the container, which will cause a greater density -~ of the gas, i.e., steam, leaving a vacuum at the bottom of the container. * Figure 10.10 represents the microscopic model PC constructed to answer this question. He makes a classic error: He applies a correct macroscopic rule incorrectly at the microscopic level (Stevens & Collins 1980). He has

*.

HOW PEOPLE CONSTRUCT MENTAL MODELS MOLECULES DISTRIBUTED

a

a •

a a

a a

a .

a

.

•

• • a .

APPLY HEAT

e • ,

•

»

•

• a

•

APPLY^ HEAT*

a

. a

a

•

a .

' a

"a

•

* .

a

a

a a

• a

a

a

a

WHEN ALL MOLECULES HEATED

HEATED MOLECULES GO TO TOP

EVENLY

263

a

'' a

(VACUUM)

a

•

a

Figure 10.10. PC's model of how to compress water by heating it learned somewhere that hot air rises. The reason for this is that hot air particles have more energy, on the average, than cold air particles. This causes hot air to be less dense than cool air, and so it rises above cool air. But PC applies this aggregate property to individual molecules, where it does not apply. Thus, he imagines that each molecule, if it receives heat energy, will tend to rise to the top. Here, he reveals his view of temperature as intrinsic to individual molecules that underlies his heat-threshold model of escape and his intrinsic state model of the liquid-vapor transition. Eventually, in his view, each molecule will receive heat energy, and so they will all collect at the top, leaving a vacuum at the bottom. The correct answer is just the opposite. Cooling the vapor will cause each molecule to have less energy on the average, and so the same number of molecules will take up less space with no increase in pressure on the side of the container. PC's reasoning about evaporation is much less consistent than RS's. He shifts around among a variety of principles and models to reason about these questions. Not surprisingly, he is less accurate than RS: only three out of eight of his answers were correct, as compared to seven out of eight for RS. Conclusion This paper traces the way people construct models of evaporation processes by using analogies from other domains. Our thesis is that people construct generative models by using analogy to map the rules of transition and interaction from known domains into unfamiliar domains. Analogy is a major way in which people derive models of new domains (Gentner 1982; 1983; Gentner & Gentner 1983; Lakoff & Johnson 1980; Rumelhart & Norman 1981; Winston 1980). Analogy can serve to transfer knowledge from a familiar concrete domain to an abstract domain, such as emotions or marriage (cf. Lakoff & Kovecses this volume; Quinn this

264

ALLAN COLLINS & DEDRE GENTNER

volume). But even when there are many observable surface phenomena in the target domain, people still rely heavily on analogies in their .reasoning. Kempton's (this volume) work on people's models of home thermostats and our research on models of evaporation both show the importance of * analogical reasoning in physical domains. The existence of such analogical models raises two important questions for further research. First, how are different analogies combined? One goal of this paper is to examine how subjects combine component models constructed from different analogies. People vary substantially in the con-: sistency and stability with which they coordinate multiple analogies for a domain. We have shown how one subject, RS, combined his component analogies into a relatively consistent model of evaporation and used it to reason fairly correctly about novel questions relating to evaporation processes. His model contrasts with that of another subject, PC, with a less coherent understanding of evaporation. PC invokes different principles or models for every answer. He aborts one line of reasomng when it contradicts another line of reasoning, without trying to trace the reason for the inconsistency. He uses principles that apply at one level of analysis in reasoning at another level of analysis. A second, more basic question is, where do these models come from? Although they are partly idiosyncratic, they must be heavily influenced by cultural transmission. If these models were purely experiential, derived independently by each individual, it is unlikely that notions of molecules and temperature would figure so prominently. On the other hand, our subjects' models of such concepts as molecules differ in some rather striking ways from expert models. The interplay between scientific models and the idiosyncratic interpretations that individuals place on them is a topic ripe for serious study. Notes 1. This research was supported by the Personnel and Training Programs, Psychological Sciences Division, Office of Naval Research, under contract number N00014-79-C-0338, Contract Authority Identification Number NR 154-428. We thank Michael Williams for an engaging conversation on the analogy hypothesis of this paper and Ken Forbus for his comments on a draft of the paper. . 2. It is true that, in norma! outdoor conditions, warm air tends to be less dense than cool air, as in the variable-size-room model; but the difference in evaporation rate does not require this density difference. Even in a sealed container, warming the air will enable it to hold more water. References Collins, A. 1978. Fragments of a theory of plausible reasoning. In Proceedings of Conference on Theoretical Issues in Natural Language Processing, Vol. 2. Urbana: University of Illinois. Pp. 194-201.

HOW PEOPLE CONSTRUCT MENTAL MODELS

265

Collins, A. and D. Gentner 1982. Constructing runnable mental models. In Proceedings of the Fourth Annual Conference of the Cognitive Science Society. Ann Arbor: University of Michigan. Pp. 86-89. 1983. Multiple models of evaporation processes. In Proceedings of the FifthC. Annual Conference of the Cognitive Science Society. Rochester, N.Y.: Univer: sity of Rochester. de Kleer, J. 1977. Multiple representations of knowledge in a mechanics problem solver. In Proceedings of the Fifth International Joint Conference on Artificial Intelligence. Cambridge, Mass.: M.I.T. Pp. 299-304. Forbus, K. D. 1981. A study of qualitative and geometric knowledge in reasoning about motion. M.I.T. AI Technical Report 615. Cambridge, Mass.: M.I.T. Gentner, D. 1982. Are scientific analogies metaphors? In Metaphor: Problems and Perspectives, D. S. Miall, ed. Brighton, England: Harvester Press. Pp. 106-132. 1983. Structure-mapping: A theoretical framework for analogy. Cognitive Science 7(2):155-170. Gentner, D. and A. Collins 1981. Inference from lack of knowledge. Memory and Cognition 9:434-443. Gentner, D. and D. R. Gentner 1983. Flowing waters or teeming crowds: Mental models of electricity. In Mental Models, D. Gentner and A. L. Stevens, eds. Hillsdale, N.J.: Lawrence Erlbaum Associates. Pp. 99-129. Hayes, P. 1985. Ontology for liquids. In Formal Theories of the Commonsense World, J. Hobbs and R. Moore, eds. Norwood, N.J.: Ablex. Pp. 71-108. Kosslyn, S. 1980. Image and Mind. Cambridge, Mass.: Harvard University Press. Lakoff, G. and M. Johnson 1980. Metaphors We Live By. Chicago: University of Chicago Press. Reddy, M. 1979. The conduit metaphor. In Metaphor and Thought, A. Ortony, ed. Cambridge, England: Cambridge University Press. Pp. 284-324. Rumelhart, D. E. and D. A. Norman 1981. Analogical processes in learning. In Cognitive Skills and Their Acquisition, J. R. Anderson, ed. Hillside, N.J.: Lawrence Erlbaum Associates. Pp. 335-359. Stevens, A. L. and A. Collins 1980. Multiple conceptual models of a complex system. In Aptitude, Learning and Instruction, Vol. 2, R. E. Snow, P. Federico, and W. E. Montague, eds. Hillsdale, N.J.: Lawrence Erlbaum Associates. Pp. 177-197. Waltz, D. L. 1981. Toward a detailed model of processing for language describing the physical world. In Proceedings of the Seventh International Joint Conference on Artificial Intelligence, Ann Drinan, ed. Vancouver: University of British Columbia. Pp. 1-6. Winston, P. H. 1980. Learning and reasoning by analogy. Communications of the ACM 23:689-703.

-*

.PART IV -Negotiating social and psychological realities

•t

.

.

.

.

•i

Myth and experience in the Trobriand Islands1

Edwin Hutchim

This chapter examines how the knowledge of a myth is brought to bear on the interpretation of experience. This topic requires an eclectic approach since the issues it raises are simultaneously part of a venerable tradition in anthropology concerning the role of myth in society and part of a new research area in cognitive science concerning the nature of the processes by which people interpret and understand their world. The experiences to be interpreted are those surrounding an encounter between a Trobriand Island village and the spirit of one of its recently deceased members. I begin with a brief ethnographic sketch of the nature of spirits of the dead in the Trobriand Islands, which will enable the reader to make sense of the events reported. The next section describes an actual case of a spirit haunting a village. This haunting set the stage for the phenomenon we wish to understand - that is, an old woman's account of how she saw a cosmological question in these events and how a sacred myth provided her with an answer to her question. The heart of the paper is an examination of the relationship between the myth itself and the phenomena it is marshaled to explain. I argue that the myth has two kinds of connections to experience. One is an explicit link based on a belief in the power of mythic events as historical precedents. The events in the myth are seen as causes of important aspects of the experience. The second kind of connection is an implicit and unadmitted link based on a similarity of organization between the events of the myth and the experience of events of life. Although the historical link is emphasized by Trobrianders, an examination of the implicit link based on shared structure shows the myth to be a transformed description of repressed thoughts about contemporary relations between the living and spirits of the dead. That is, the myth is shown to be a cultural defence mechanism. I conclude with a discussion of how this role of myth bears on two otherwise anomalous properties of myths: the tenacity of native belief and the radical disjunction between historical time and mythic time.

V'M^',>VJ'
•.•.••

..•;-.

EDWIN HUTCH1NS

r

3AL0MA

TOTAtOi

KOSI

Figure 11.1. Classes of spirits of the dead

The ways of the spirits The class of beings that are spirits of the dead in the Trobriands all fall under the generic term baloma. In some contexts, including the recitation of myth, the archaic term yaluwa may be used instead. Within the category of spirits of the dead are several subcategories that partition the world of spirits on the basis of some of their salient attributes (see Figure 11.1). Each category has a specific term associated with it. It is important to note that baloma occurs both as a generic term for all spirits of the dead and as an unmarked specific term that can stand in contrast to other specific types of spirits. BALOMA

Before the marked terms can be discussed, more needs to be said about the nature of the unmarked specific category of baloma.1 When a per- ^ son dies, his or her baloma leaves the body and goes t o reside on the island of Tuma, a real island located about 10 miles northwest of the main island in the Trobriand group. Arriving at Tuma, the baloma meets a spirit named Topileta, who "acts as a kind of Cerebus or St. Peter in so far as he ad- - *** mits the spirit into the nether world, and is even supposed to be able t o refuse-admission" (Malinowski 1954:156). The baloma "live" normal lives on Tuma and return to their natal •; villages during the harvest season each year to partake of the spiritual goodness of harvested yams and valuables placed for them on special platforms out of doors. While they are in the village, the baloma sometimes indulge in mildly annoying pranks, such as making noises and moving things, but people d o not find them eerie nor do they fear them the way Europeans fear ghosts. At the end of the harvest feast, the baloma are unceremoniously driven from the village by gangs of children who shout and swing sticks. This driving-out is called yoba. The baloma then return „ to Tuma, where they remain until the next year's harvest feast. * At any time of the year, baloma may intercede invisibly in human affairs by bringing bad luck to people who have behaved badly. For exam\

MYTH AND EXPERIENCE IN THE TROBRIAND ISLANDS

;.2?i•

pie, a village that is constantly embroiled in petty squabbles may find a poor harvest because the baloma of the village feel that thepebpte -have not comported themselves with dignity; -;;; •••• ' ^ ^ Baloma frequently appear in dreams or even to people who are awake to announce to a woman that she will become pregnant or to suggest :a course of action to an Important person. In such cases, the baloma isSeen and recognized as a person who once lived. The only other circumstances in which baloma are visible to. the living are those near the boundary bf life and death. A person on a deathbed may see into the spirit world while ; speaking to those in the world of the living who are keeping the death vigil. Also, the spirits of the recently deceased may appear to the living. In the latter case, the terminology of spirits of the dead is elaborated in reference to the behavior of the spirits. TOTAL01

The spirits of the recently dead sometimes linger for a day or two after death to say good-bye to loved-ones. They may present themselves to persons either in dream or awake. These spirits are called totaloi, from the root -taloi, referring to the minor rituals of leave-taking performed by socially engaged persons when they part company. Such encounters are sad, but not at all frightening. For example, the spirit of a young girl appeared to her mother the day after her death to request that her favorite hymn be sung in church. KOSI

Other spirits called kosi stay around the village longer and frighten people. They are much more malevolent than baloma. When a kosi is in a village at night, villagers say that it "rounds the village." The verb "round" describes the movement of the spirit from house to house through the roughly circular layout of the village. The literalness of this usage was not apparent to me until, on the nights of the encounter described below, I could actually follow the progress of the kosi as the screams of those visited emanated first from one part of the village, then from another. The unpleasantness and the duration of visits from kosi are said to be related to the social character of the deceased while he or she was alive. If the deceased was a sorcerer, an adulterer, or a thief, for example, the kosi is expected to be malicious and will remain in the village for a long time. Some ofrayinformants speculated, in contradiction to Malinowski (1954:156), that the reason the kosi of evil people stay in the village for a long time is that Topileta refuses them entry at Turna and they then have no place to go.3 According to Malinowski (1954:154), Arosi are only encountered out of doors. In the days following the death of a villager, the sound of footfall in darkness on a jungle path at night or the otherwise unexplained rustling of vegetation in the deceased's garden enclosure may be taken as

272

EDWIN HUTCHINS

evidence of the presence of a -kosi. In the village, a kosi might call put a person's name or throw a stone or slap the thatched wall of a house much as a baloma might do during the harvest season. Summing up the reaction of Trobrianders in his times, Malinowski called kosi, "the frivolous and meek ghost of the deceased who vanishes after a few days of irrelevant existence" (1954:154). Modern-day Trobrianders agree that Malinowski's description is an accurate depiction of the behavior of kosi in the past, but times have changed. In more recent times, in addition to frighteri. ing people with their poltergeistlike activities out of doors, kosi have taken to confronting people in a visible form in their homes as they lie in what we would call hypnagogic sleep. The Trobrianders call this state kilisaia and describe it as resting with the eyes closed, but with the ears open and the mind awake. They are adamant that it is not sleep, and the events experienced in that state are real events of this world, not dreams or hallucinations. The assertion that these events are real is supported by items of evidence such as the fact that kosi knock things off of shelves and throw stones on the roof when everyone is wide awake. In those contexts though, the kosi is not seen. Encounter with a kosi In March 1976, Toigisasopa (a pseudonym), a prominent and feared citizen of a large Trobriand village, died. For more than a month following his death, the village was plagued by almost nightly visits from his kosi. Occasionally, the kosi appeared as the ghoulish figure of the rotting corpse of the deceased. As time went on, his reported appearance became increasingly revolting. • * The following is a translation of an excerpt of an interview with a woman who had twice been visited by this kosi. He was out walking about and he came to our place. I woke and saw him come .,.• in. His face was awful - completely black, his arms and legs were like those of an undernourished child and his belly was bloated. He just came in and grabbed my legs. I lay there. My body was completely numb and paralysed. Then I kicked and the old man [her husband] felt me kicking. He roused me and I grabbed him ' and cried out. He asked, "What was it?" "Toigisasopa's ghost!" By then my body was recovering from numbness a bit. The stench was awful. The first time he just looked black, but this second time, you know, it was as if there were holes in his body as well. It gave the house a terrible odor. Myth interprets life For nearly a month, I heard reports of the activities of the ghost. Many. villagers reported evidence of the kosi's presence, and the village as a whole * was quite disturbed by the length of the stay of the kosi.4 Near the end of Toigisasopa's stay in the village, an old woman came to me with the % following account.

MYTH AND EXPERIENCE IN THE TROBRIAND ISLANDS

273

One night,as she lay in her house, she heard banging on the side of her neighbor's house and heard her neighbors yelling to drive the kosi away. She later reported: \ • ••• . ••• .?"••:'•. i • I had gotten up and was sitting there! I thought, "perhaps that was' a kosi," This one will not soon disappear because when he went to Tuma, Topileta closed [the v way in]. Because he was a thief. He stole seed yams in the garden, and whatever he saw unattended in the village he would put in his basket as well. Everyone saw him do it. You see, this was his one bad habit. It was nothing of consequence; he seemed a good man except for this. Now he has died. Who knows, perhaps he was a sorcerer, perhaps . . . . Anyway, I thought to myself, "Why is it that the kosi and baloma see us, but we do not see them?" She said that in considering this question, she remembered that it had hot always been so. Long ago, the baloma would come into the village and sit with their kin, chewing betel nut and conversing in a pleasant way. They were generally amicable and even helpful. This old woman remembered the myth of Baroweni, and it, she claimed, provided the answers to her questions. Myths form one of several classes of oral tradition in the Trobriands. The major distinction of importance here is between sacred myths, called liliu or libogwa, and folk tales, kukwanebu. Both liliu and kukwanebu are sometimes performed in the story-telling sessions that occupy many idle hours in the rainy season. The events that are depicted in both genres are placed in time long, long ago beyond the stretch of ordinary historical time. Kukwanebu are told largely for their entertainment value, and tend to be a bit bawdy, whereas liliu are thought to have a message and the events they describe are often taken as historical precedents for the states of affairs in the contemporary world. Some myths establish sociocultural precedents concerning the ownership of land and the ranking of descent groups. Others concern fundamental states of existence, such as the inevitability of aging and death or, in the myth of Baroweni, the invisibility of the spirits of the dead. THE MYTH OF BAROWENI

A full translation of the telling of the myth as I recorded it is given below. I have included some additional commentary on a paragraph-by-paragraph basis where the telling of the myth assumes listener knowledge that is particular to Trobriand culture. By and large, these are things that from the Trobriand perspective go without saying. Hey, Baroweni's mother had died. She had died and Baroweni was already pregnant. When she was due to give birth, another woman in the village was dying. She was on the verge of death - they had already made the mortuary preparations for her. Baroweni went to her and said, "Go to my mother. Tell my mother, 'Hey, your child is pregnant and will soon give birth. Take food to her.'" This section establishes the relationships of the major actors in the myth to each other. The mother/daughter relationship cited is important because

274

EDWIN HUTCHINS

Trobriand mothers have a special obligation to care for their pregnant daughters. In fact, the female matrilineal kin of the pregnant woman are responsible for feeding and caring for both the new mother and the infant for several months after birth. Asking a dying person to take messages /' to Tuma is a common occurrence. Trobrianders make no effort to conceal their assessment of an ailing person's chances of survival from that . person, and those who are about to die provide an obvious communications link between the world of the living and the world of the dead. So the dying woman died and went [to Tuma]. She told her companion there (Baroweni's mother], "Your child is pregnant and close to giving birth, but what shall she eat?" She said, "She told me to come and tell you, 'You should take food to her.'" So the spirit [Baroweni's mother] got up and began cutting taro shoots. She raised the taro shoots onto her head and rose up [as a spirit rises when it leaves the corporal body]. She came. She continued on, what's it, at Tuma on the main beach at Kuruvitu, on their beach at Tuma. She came ashore and continued on there to Libutuma. There is an interesting mixing of attributes here. The spirit carries her load on her head as all Trobriand women do, yet the verb used to describe her rising is not one used for a woman, but one used to describe the rising of a spirit from a corporeal body. {directed to me] Shall I sing the song that you might hear it? "Baroweni, Baroweni, Barowentweni. I will set it down, I will set it down. Oh, my neck. Oh, my neck . . . Oh, my neck. We [inclusive dual] suppose a girl is standing with us whose name is yaluwa." As we shall see, variants of this ditty are sung at several points in the ;;i story. Baroweni's mother whines her daughter's name and complains of her desire to set down her load and the pain it causes in her weak spirit neck. The last phrase in the ditty suggests a dual persona for the mother. She supposes or assumes a girl is standing with her, but her name is just : "* yaluwa or spirit. So she walked and cried, and in that way she entered a village. She just cried, she was crying for her child. Because she was a spirit she couldn't touch anything or hold it. She had already died, and there was nothing she could hold. This makes clear the source of the mother's discomfort. She is carrying real taro, but as a spirit, she is known to be unable to carry anything. The explanation of this comes in the next sentence of the telling of the myth. However, this is a liiiu. This happened long long ago and for that reason [it is as it is]. \ Here is the first of several indications that inferences based on presentday understandings of the world are not always applicable to myth. That %

MYTH AND EXPERIENCE IN THE TROBRIAND ISLANDS

:275

it happened long, long ago is sufficient explanation of events that violate common-sense knowledge. I return to this point later. ••^•'^••r-\--^}^.;She just continued on to - what's its name - Yalaka . . .Buduwelaka. lYesat Buduwelaka she was just the same, crying, crying for her child.. '.r:w-sri'^' •• "Baroweni, Baroweni, Baroweniweni. I will setit down, I will set it do/Wp/jOh,•":; my neck Ya! Oh, my neck. She just scoops it up the girl that stands with us [in-: elusive dual]. Her name is yaiuwa. We are surely overloaded,"- ••;: -vriiv r r . This because she had already died. _ ; ; '"'T?^; ; The ditty Barowehi's mother sings is a magic spell she uses to;eause the taro to be carried. Since she is dead, she cannot carry things in a rior^ mal fashion. Notice that the mother's persona is completely confounded with that of the spirit girl whose presence she invokes. She says she herself would like to put the basket down, it is her neck that hurts, yet she refers to herself and the spirit girl using the inclusive dual form of the first person pronoun. She just brought it and continued on to Okupukopu. At Okupukopu she chanted just the same as she was crying, crying for her child. "Baroweni, Baroweni, Baroweniweni. I will set it down, I will set it down. Oh, my neck. Oh, my neck. Oh, my neck Ya! Oh, my neck. What is her name? What is she? Girl or person? But her name is yaiuwa." Because she was already dead there was nothing she could touch or carry on her head. And this: she^was of the time long ago, our [exclusive plural] ancestor's time, [then, directed to me] I shall take it to its conclusion that you should grasp it? The use of the exclusive plural possessive form here in referring to the ancestors makes it clear that these asides about the time of the ancestors and the special nature of events that happened then are directed at me as an outsider to the culture. I suspect she uses this form in order to say, "This is how it was with Trobriand ancestors. How it was with your European ancestors is irrelevant." She continued on from Okupukopu to Ilalima. At Ilalima it was just the same. She went then to Osapoula. It was night when she arrived and everyone was sleeping. She knocked on the coconut fronds [the wall of the house]. Her child woke up and said, "Who are you?" Having the spirit mother arrive in the village at night is plausible since spirits are most active at night. It also simplifies the story because even though the mother is at this point in time a visible spirit, it is possible for her to go unseen in the darkness while the village sleeps. She said, "I am your mother. I have brought your food. Open the house." Her child opened up her house. She went to her and saw her mother putting down the food basket she had been carrying on her head. In the night there she [mother] told her, she said, "Hey, go prepare the area behind the house as if, wa, as if you were to plant flowers there." She said, "Take these [the taro stalks] and bury them. They will be yours to

276

EDWIN HUTCHINS

eat with your child. When the seed root has sprouted, cut off the side roots. Cut off the side roots and replant the seed. The side roots alone you shall eat with your son. You prepare behind the house. I shall go beside the house and sit. I shall be watching you." This is interesting since a family metaphor is commonly applied to the form of taro propagation referred to here. The seed conn is called inala, mother, and the side roots are referred to as Utula, her children (Malinowski 1965, v.2:105-106). Given the traditional importance of yarns in the maintenance of descent group identity in the Trobriands, it is surprising that the mother brought taro rather than yams to feed her daughter. It is possible that the existence of this metaphor for taro propagation, which has no parallel in the cultivation of yams, made taro a more felicitous symbolic choice than yams. Her mother spoke well. She told her she would sit beside the house and watch, but Baroweni by herself was already going on with her things. She forgot. She boiled her food and ate. Her mother was sitting there watching her. She [Baroweni! picked up that container, a coconut shell bowl, a soup bowl. Like those containers that the Lukwasisiga clan drank from. She just picked it up, drank her fill and threw it out beside the house. She threw it out and it drenched her mother's body. Her mother felt it. She said, "Hey! Why did you dump that on me?" She said, "Oh my! Mother mine, chieftan's wife. Mother no . . . I just forgot. I forgot about you. [Don't be angry] because it can't be a killing since you are already a baloma. This event is outrageous in the Trobriand view of things. Remember that Baroweni's mother had gone to great effort and borne great pain to meet her responsibilities to her daughter. Trobriand children have a like set of responsibilities to their parents. Meeting these obligations is called velina, and failing to meet them has moral as well as jural consequences. When children are young and need support, parents supply what is needed\7 When parents grow old, the roles are reversed. Children must care for their parents in their old age. This role reversal is sometimes marked by a mother's addressing her grown daughter as inagu, "my mother." When • parents become infirm and require constant care, this job falls largely on the grown children. It is they who must feed their aged parents, bathe them, and, if necessary, carry them on their backs away from the village so that they can defecate in the privacy of the forest. These responsibilities,,. burdensome though they may be, are taken very seriously by all. It is understandable, however, that parents sometimes complain about not being well treated by their children, and children sometimes come to resent the imposition of a parent's needs on their lives. Here in the myth, Baroweni has failed to meet her obligations to her mother and she ha§ done so in a particularly offensive way. Her mother brought food, and Baroweni has thrown food on her mother. Worse yet, she has done so$> she says, because she has forgotten about her mother.

MYTH AND EXPERIENCE IN THE TROBRIAND ISLANDS

S277

WMv -,She said, "You have thrown out my soup. I shall return (to Tuma]. I shall return, llfec;•:--i-j shall go. You will stay here. I will split our [inclusive dual consumable] coconut. | I | ! S ; irhe lower half is yours. The half with,the eyes is mine. {Directed to me] When; | | f ; - i ; | s that coconut? [Iproduce a coconut and she demonstrates). "••:/;%r^ ||j;;:#;. ••; .^ ighe split the coconut. "This half is yours. The end with eyes we shall drink, |MJ ; ;;^land it -will be my coconut. I will go. I will go and then I will come back and See I^K^Cyou-You shall not see me." •.•?;-.>;• • .••- :• '-:'}-_r $$£$•- This ritual performance concludes the actual telling of the myth itself. Iils;;;^he symbolism of the eyes in the coconut is riot as transparent as it seems. S^V^JTbe.end of the coconut without eyes is called kwesibuna, which means liS:^;;!;iiterally the "cold" part. So the mother has not only taken the eyes'; she KS , : v &as taken the "hot" end of the coconut in a world where hot is.potent iU^:'",i"ahd powerful. ILK.'

Commentary on the myth

";.v The remainder of the text is the old woman's commentary on the myth J&',.!^;.-.'!Euid her attempts to show the connection of the myth to the world of ^'-.•" /experience. ;

{Commentary} Look, now-a-days people die and go [to Tuma] and their kosi come around, but we can't see them. He [Toigisasopa's kosi] sees us and wakes us up, but we do not see him, and this is the reason. Our ancestors have changed things. !£'.'-. They come and see us. Our fathers and mothers die and go away and then they ;cpine back and are watching us. We see them not. But this old woman in times long ago started this. This particular liliu here. Some think it is just a fairy tale {kukwanebu}, but it is real liliu. ;)r; However, if that woman, Baroweni had not done that, had not thrown out :,the soup, our mothers would be with us now. We would die. Later we would come back and stay and be seen. But here she made her mistake. She grabbed that soup cup, drank from it and threw it out, hitting her mother who was beside the house. The old woman cast an appropriate spell. She said, "Why did you throw soup #'& oh my body?" > .Notice the inference here: had Baroweni not made her mistake, our parents would be with us, that is, visible to us, now. This is a direct connection between the myth and the situation of life as it is experienced. ,;She said, "Oh, mother, I forgot about you." : She said, "You yourself have banished me. You have injured me. I will go back itoTuma]. .: Here, the terms of the mother's interpretation of Baroweni's act has escalated. The mother has no sympathy for Baroweni's excuse. She has |sp; equated Baroweni's mistake with a grievous form of social punishment, B|gv j;y°pG> banishment from one's village.

IP

P | g j -She got a coconut and told her child, she said, "I will split it in half you see. The fe;>? ^ ower e °d is yours. The end with eyes is mine. I will go away. When I come back, lg|.>;:; .1. will see you. You will not see me." So she went back and stayed.

278

EDWIN HUTCHINS

You see, the other day when Toigisasopa was going around the village as a spirit. 'We did not see him: Except those whosaw him all blackened. Look, [makes empty circles around her eyes with index fingers and thumbs] with his eye sockets empty. His whole body was black and when he went to someone, that's what they would see. This was just sleepers, in kilisald of course, who saw him thus. The other day I was listening to them screaming in the village. 1 thought to myself, "What is the source of this? Oh, this thing from long ago." My mind went to these words [this myth]. This is what the old woman had to say by way of explanation for the perceived invisibility of the spirits of the dead. By her account, she has used her knowledge of a myth to find an interpretation of ah important and puzzling aspect of life. That in itself is perhaps interesting, but not entirely surprising. It is a confirmation of Malinowski's claim that "myth is not an idle tale, but a hard worked active force" (1954:101). However, more remains to be said about how the force of myth is brought to bear on life. The historical connection In her commentary on the story, the old woman emphasizes the fact that the story of Baroweni is the reason for the invisibility of the spirits of the dead. Baroweni's mother declared that Baroweni would no longer see her. The relevance of this myth to the events surrounding the poltergeist behavior of the kosi in the village is based on the perception of the kosi as an (usually) invisible spirit of the dead. The connection is purported to be historical and causal, where the causality of the connection is based in the cosmological status of iitiu, sacred myths. The actions of Baroweni's mother (as a spirit) with respect to Baroweni (as a living person) set a precedent for all subsequent relations between spirits and living persons. This causal connection appears in discourse in the form of a pervasive Trobriand metaphor. The myth is the uula, root or cause of the current state of af-"' fairs, which is, in turn, the dogina, extremity or result of the myth. According to the old woman's account, it was this causal connection that led her from her question about the invisibleness of spirits of the dead : to the myth of Baroweni. On this view, the events described in the myth were identified as the cause of the salient aspect of the experience. This connection is also the one emphasized by Malinowski in his analysis of the role of myth in primitive psychology. In considering several related myths, including another version of myth of Barowem, which he had collected, he notes that such myths provide mundane precedents for some very unpleasant facts of life. What it actually does is to transform an emotionally overwhelming *. foreboding, behind which, even for a native, there lurks the idea of an inevitable and ruthless fatality. Myth presents, first of all, a clear realization ,; of this idea. In the second place, it brings down a vague, but great ap^

MYTH AND EXPERIENCE IN THE TROBRIAND ISLANDS

279

prehension to the compass of a trivial, domestic reality.;. -, , The separa-;: ;•>>;•. tion from the beloved ones after death is conceived as due to the careless':>-.' handling of a coconut cup and to a small altercation, (1954:137), "•• ^ - • • There are two loose ends to be attended to here. First, this tarjlaliang quote raises the question of the source of the great apprehension experienced in connection with the invisibility of the beloved ones after death. Malinowski is surely right that when Trobrianders consider the invisibility of the spirits of the dead they do so with apprehension. If this myth permits that invisibility to be conceived as due to a small altercation, what might it have been conceived as due to that provokes apprehension? That is, if it provides a substitute conception of the invisibility of the spirits, for what is it a substitute? What was the terrifying conception of the cause of the invisibility of the spirits that it replaces? Second, there is something paradoxical about this historical relation of myth to the nature of life as it is experienced in the present. Things happen in myths that all Trobrianders agree could not happen in life today. The carrying of the taro by the spirit woman in the myth of Baroweni is an example, and in the telling of the myth, the old woman repeatedly reminds us in that context that this is a liliu; that it happened long ago. Even though the events of the myth have direct causal relations to states of affairs in the present, the time of myth is not historical time. As Malinowski puts it, ". . . the distinction between the liliu and actual or historical reality is drawn firmly, and there is a definite cleavage between the two" (1922:303). Myth must be closely linked to life so that the events of myth can serve as precedents and causes for the events of life; yet myth must also be kept distant from life to protect it from mundane inferences about what might and what might not be possible by present standards.

A mythic schema When we look more closely at the myth itself and at how it is applied to the interpretation of the behavior of the spirits of the dead, we find other connections between the myth and experience. These connections have to do with similarities in organization between the concepts encountered in the myth and concepts encountered in life. The myth of Baroweni contains a structure of relationships among things and actors and actions. If we ignore the specific identities of the things and actors and actions and concentrate instead on the organization of the relationships among them, we see a structure I call a schema,5 When particular instances are plugged into the slots in the schema, we say the schema is instantiated, and the result is a proposition that is an assertion about the world. The schema underlying this myth encodes cultural knowledge about relations between the living and the dead. It is based on a set of implicit cultural theories about human motivation and

280

EDWIN HUTCHINS

human psychology. It implicitly attributes abilities, failings, behaviors, and reactions to deceased persons and to their survivors. "''/ • :;' The basic points of the myth can be summarized in the following simple three-term propositions. '':..' Ml. Baroweni

threw soup on

M2. Baroweni's mother.

became invisible to

Baroweni's mother Baroweni

These two propositions are related to each other in that the event describedin proposition Ml led to the event described in M2. Baroweni's mother's action in M2 somehow "caused" present-day spirits of the dead to be invisible to the living. M3. Baioma

are invisible to

the living

Since each event uniquely precipitated its successor, the negations of these propositions also form a plausible sequence. That is, if Baroweni had not thrown soup on her mother, then her mother would not have been angry at her and would not have made herself invisible and, so argues the old woman who told the myth, present-day spirits of the dead would still be visible to the living. It is clear how these propositions are connected to each other and how the old woman can use the connections to make inferences about how things might have been different than they are. It is less clear how this is connected to the world of contemporary experience.

The structural connection The question really is how one gets from the notion of contemporary spirits of the dead being invisible (proposition M3) to the idea of Baroweni's mother becoming invisible to Baroweni (proposition M2). No historical' * mechanism is offered in the telling of the myth or in its commentary, but both of these propositions are instantiations of the same structure, **x is invisible to^." This suggests that the relation might be analogical rather than historical. Support for the analogical interpretation requires a single schema that, when instantiated iiT different ways, generates the propositions of the myth as well as the propositions describing the conditions of life. Such a schema would be a structure composed of more general terms * than those found in the instances to be accounted for. To discover the underlying schema, we examine the terms in the propositions of the myth and of the description of the relevant bit of life and find for them the most specific category that is general enough to contain the set of terms. For example, Baroweni's mother, the kosi, and our dead parents are ail instances of deceased persons, whereas the living and Baroweni are instances of survivors. §

281

MYTH AND EXPERIENCE IN THE TROBRIAND ISLANDS INTERPRETING ACTIONS

'

• -"•:':•.'"•

Finding the appropriate general terms for the actions in the schema is more • difficult. Baroweni's mother leaves her and becomes invisible, the baloma are invisible to the living, yet the kosi is occasionally visible while frighten- '-••• ing people. Fortunately, there are other instances at hand-and;other..^ ethnographic evidence that can help determine the more general categories of action of which the throwing of soup and the coconut ritual aie instances. Within the myth itself, the throwing of the soup and the splitting : of the coconut are given metaphorical interpretations. Baroweni's mother says that her daughter has injured her and has banished her by throwing the soup on her. So Baroweni's act is an instance of injury and one of banishment as well. The mention of banishment here is interesting because in one version of the myth of Baroweni collected by Malinowski (1954:133-134), the same soup-throwing incident and subsequent coconutsplitting ritual were posed as being both the cause of the invisibility of the baloma and the origin of the banishment of the baloma to the island of Tuma. In that version of the myth, prior to Baroweni's mistake, the spirits of the dead were visible to the living and resided in their natal villages after their deaths. Before Baroweni's mistake, it seems, dying was not really much of an inconvenience. In fact, being dead looks quite a lot like being alive, with the possible exception that one gains considerable magical power by dying. Seen in this light, Baroweni's mistake is an enormously important event. It is as a result of her action that death acquires its most salient and distressing features: exile to Tuma and invisibility. Because of Baroweni's actions, death comes to mean removal of the spirit from the sphere of social intercourse. Now, in the telling of the myth and in providing commentary on it, my old informant gave several paraphrases of the conversation that took place between Baroweni and her mother. In one of those, she attributes the following words to Baroweni, "Gala bisimamatila, pela bogwa baloma""It is not a killing because (you are) already a baloma." Barowem protests to her mother that she could not have killed her because she is already a spirit. Where does this talk of killing come from? After all, Baroweni might have harmed her mother by throwing hot soup on her, but killing seems out of proportion to the nature of the mistake. It is, however, in perfect proportion to the consequences of that mistake. In spite of Baroweni's urgent denial, her act is a killing in the sense that before this act, there is no death as Trobrianders now know it. True, her mother had previously died, but her spirit was visible and interacted with her daughter as a living mother would. Death to Baroweni's mother did not mean what death means now. And, in fact, in terms of the Trobrianders' current understanding of death, Baroweni's mother did not "die" until this altercation had taken place. It is as a result of Baroweni's actions that death becomes the horrible state of separation from the living. In this light, the soup-throwing incident can be seen as a metaphor for killing, accidental

282

EDWIN HUTCH1NS

to be sure, but killing nevertheless. The mother's reactions to this are the ritual splitting of the coconut and the pronouncements about how things will be henceforth. The old woman telling the myth called this a just punishment for Baroweni, but it is also the mother's act of dying. With this act, Baroweni's mother declares she will leave, that is,-not. h^lp Baroweni with her child soon to be born, and when she returns she will be invisible. ,. .,., •.-•. -,-..--. ,• ';.., ,.»:,'...',;. Remember also that Baroweni's action occurs in the context of mother meeting her responsibilities to her daughter, but daughter, Baroweni^ reciprocates by negligently failing to meet her obligations to her mother. Thus, while the story is in one sense about a trivial mistake, it is also about a daughter's inadvertently ending her mother's life through negligence. This is a theme that is probably universal. Freud, writing about his European patients says, ' When a wife loses her husband, or a daughter her mother, it not infrequently happens that the survivor is afflicted with tormenting scruples, called "obsessive reproaches" which raise the question whether she herself has not been guilty through carelessness or neglect of the death of the beloved person. (1918:80) Freud describes the source of these thoughts as follows: Not that the mourner has really been guilty of the death or that she has really been careless, as the obsessive reproach asserts; but still there was something in her, a wish of which she herself was not aware, which was not displeased with the fact that death- came, and which would have brought it about sooner had it been strong enough. The reproach now reacts against this unconscious wish after the death of the beloved person.-; Such hostility, hidden in the unconscious behind tender love, exists in almost all cases of intensive emotional allegiance to a particular person, indeed it represents the classic case, the prototype of the ambivalence of human emotions. (1918:80) . ,.;.This ambivalence resonates perfectly with the Trobrianders' feelings about velina. We know that children often feel anxiety about meeting their obligations to their parents, and we know that in the Trobriands (as in our own * culture) elderly parents often complain that they are not properly treated by their children. If the historical causal link were indeed the only link between the myth and the experience it is marshaled to explain, then there are many scenarios that could arrange events in which spirits become in- . visible here ever after. Yet, this mythic structure mirrors the thoughts and fears that are experienced by Trobriand children (whether in childhood or as adults) on the death of a parent. In a world in which there are no natural deaths and in which elderly parents depend on their children for,, their very existence, a fleeting secret wish that one's burdensome parent;> were dead or doubts about having met one's filial obligations can easily lead to self-reproach on the death of one's parent. In the earlier discus-'%

MYTH AND EXPERIENCE IN THE TROBRIAND ISLANDS

;

283

sipn of Malinowski's account of the connection of myth to life, we asked, 'What could have been the original terrible conception of the cause of the invisibility of the spirits of the dead that must be replaced?'-The^answer for the survivors is that the dead are invisible because we killed them/This is the source of the Trdbrianders' great apprehension over the separation .v from the beloved ones to which Maliribwski referred. ^ • Reading the throwing of soup as a gloss for wishing death on and inadvertently killing the deceased makes the dynamics of the retaliatory nature of the myth clear. According to Freud,/the defence against the Hostility that was felt toward the deceased " : 7 !i .", . . is accomplished by displacement upon the object of hostility, .r ; r; namely, the dead. We call this defence process, projection. The survivor , will deny that he has ever entertained hostile impulses toward.the beloved dead; but now the soul of the deceased entertains them and will try to give vent to them. . . ." (1918:81) The anger of the deceased at the survivor, then, is a projection of the survivor's hostility toward the deceased. TRANSFORMING THE MYTHIC SCHEMA

The parallelism of these schemas is easy to show. The psychodynamic schema described in the paragraphs from Freud is as follows: PI. survivor P2. deceased

wished death to punishes

deceased survivor

Here, P2 arises from PI when the survivor projects to the deceased the hostility that the survivor felt toward the deceased. The myth mirrors this structure, although it is not a direct instantiation of it. The structure of the myth was: Ml. Baroweni

threw soup on

M2. Baroweni's mother

became invisible to

Baroweni's mother Baroweni

where the mother's act can be seen both as dying and as punishing Baroweni. The myth is reputed to explain the facts of life. The claimed connection is that Baroweni's mother's act somehow "caused" the observable phenomenon that the spirits of the dead are not visible to the living. In particular, it was claimed that dead parents are not visible to their living children. This is experienced by all surviving children. Furthermore, had Baroweni not made her mistake, the dead parents would be visible. Instantiating the psychodynamic schema in the first person (as experienced by the survivor) yields the propositions: 77. I T2. my dead parent

wished death to punishes

my dead parent me

284

EDWIN HUTCHINS

Obsessive scruples turn this active hostility, wishing death, into a passive animosity, negligent killing, and turn the punishment into a passive projected retaliation, becoming invisible and going into self-impOsed exile. LI, I L2. my dead parent

negligently killed is invisible to and leaves

my dead parent me

The first proposition here is one that elderly parents sometimes voice in their complaints about the way their children neglect them. The second is a statement of the phenomenon the myth is supposed to explain, this transformation of the mythic schema, then, provides a simple explanation of the invisibility of the spirits of the deceased, which is, after all, what is in need of explanation here, but it is a very painful explanation indeed. This is the schema that underlies the myth. It arises in the minds of the survivors following the death of a person with whom they have been involved in life. To this point, we have considered evidence about the myth and its telling as well as psychodynamic theory in order to discover the schema underlying the myth. Suppose this is the schema underlying the myth, how could the myth come to be structured the way it is? We have already seen that the defence process of projection operates in the composition of the internal structure of the schema. It is the source of the retributive nature of the myth. The defence process of projection and two others, intellectualization and displacement, also operate in the transformation of the propositions that represent the experienced relations between the living and the dead into the myth of Baroweni. Suppes ;; and Warren (1975) propose a scheme for the generation and classification of defence mechanisms that is directly applicable to sets of propositions such as those discussed. In their model, defence mechanisms are created by transforming propositions of the form, "self + action + object." The classification of the defence mechanism is based on the nature of the transformations applied. Among the transformations they describe are putting another in the place of self, projection; changing the nature '' of the act performed, intellectualization; and changing the identity of the object, displacement. The myth of Baroweni is produced by applying all three of these transformations to the original repressed propositions LI and L2 as follows. Let us begin with the underlying propositions: ;. LI. I L2. my dead parent

negligently killed is invisible to and leaves

my dead parent me

Projection changes the identity of the self, and displacement changes the identity of the object. In the case of the myth, Baroweni takes the place ^ of self and Baroweni's mother takes the place of one's own dead parent.

MYTH AND EXPERIENCE IN THE TROBRIAND ISLANDS

L Baroweni 2, Baroweni's mother

negligently killed punishes

285

BaroWeni's mother Baroweni "/ ; ^

Intellectualization changes the nature of the act performed. In the, case, of the myth, negligent killing becomes negligent handling of food, and, the punishment is the experienced invisibility of the dead. This brings us back to the propositions that summarize the myth. ; . :.-..^;;"i-:,;„;.;r Ml, Baroweni

threw soup on

M2. Baroweni's mother

"Baroweni's "mother'1 became invisible to , Baroweni

Having arrived at the schema underlying the myth, we are ready to return to the phenomenon that brought these issues to light in the first place - the terrifying visits of the kosi of Toigisasopa. By placing Toigisasopa in the role of decedent and those villagers who experienced the presence of his kosi in the role of survivors, substitution into the psychodynamic schema produces the following set of propositions: KL Villagers K2, Toigisasopa

wished death to punishes

Toigisasopa villagers

These propositions, like those involving the parents, are likely to be repressed. Given that this may be an important structure for the organization of ideas about the nature of relations between the living and the dead, it is easy to see the mechanism underlying Trobrianders' assertions that if a person is bad in life, his or her kosi will haunt the village for a long time and will be malevolent. Those who are sorcerers, adulterers, or thieves are likely to evoke hostility in their neighbors. Those who are none of these things are less likely to be hated and/or wished dead by their companions in life. Thus, the deaths of powerful and evil persons may evoke many reactions of this sort from the community, whereas the deaths of more sociable people are likely to evoke few such reactions. I have argued that the way the myth accounts for the invisibility of the spirits of the dead is not through the historical connection claimed by the Trobrianders, but through the fact that the myth is a disguised version of the inadmissible cognitive and affective structure experienced on the death of a loved one. We find also that a different instantiation of the same schema underlies the interpretation of the haunting of the village by the kosi. In each case, there is independent sociological evidence concerning the reality of the repressed propositions. What do they know? The application of schemas across sets of instances is a ubiquitous cognitive activity. Instantiation in conventional ways is involved in understanding, reasoning, and predicting (Hutchins 1980). Unexpected insights often

286

EDWIN HUTCHINS

seem to arise from unconventional instantiations. Metaphors and some types of humor are also based on the assignment of new instances to familiar schemas (cf. Lakoff this volume). This same process is also apparently at work in the creation and use of myths. This myth is both a charter or a precedent for an unpleasant fact of life and a cultural model of relationships between the living and the dead. The schema it embodies is as applicable to contemporary personal relationships as it is to those of the ancestors with each other. If what I have said is true, then there is an important problem for those of us interested in the role of cultural knowledge and belief in everyday cognition. When we turn to the complexities of cognition in real-life settings, distinctions between the realm of the cognitive and the realm of the affective begin to melt away.6 It is clear that a great deal of knowledge used in the interpretation of everyday events is never explicitly stated. The sort of knowledge that resides in these unpleasant and unstated instantiations of the mythic schema cannot be ignored. To the extent that they may influence memory, judgments, inferences, and other cognitive processes, they are things that are "known." Yet, in a sense, they are things that are too painful to be known. Trobrianders (or anyone, for that matter) need cultural knowledge to understand the myth, and they use the schema of the myth to understand, perhaps in a more profound sense than they can admit, the events of their everyday lives. Why sacred myths are sacred In the telling of the myth, the old woman went to some pains to assert the truth of the myth and to impress on me that the events in myth cannot always be made sense of in terms of what we know about the present-day world. Having examined the use of the liliu of Baroweni in the interpretation of these modern events, we can see why it is that the sacred myths, r are so adamantly defended. They are formulations that from the Trobri-' ander's perspective must be true. Were they not true, then experience could be exceedingly threatening. Remembering the myth must be a very rewarding experience since it allows the myth to perform its role as a defence ' mechanism. It allows the believer to confront the ugly subjective realities of deceased parents who can no longer be seen or a visit from a kosi with the sense that these are explicable phenomena. Not only can they be explained explicitly in terms of historical causality, but the process of re- ;. membering the mythic schema that explains the events also binds the dangerous, unstated, unconscious propositions to a conscious and innocuous isomorph. The situation is explained, and the disruptive propositions in the unconscious are transformed into acceptable elements of a descrip- \ tion of an event that happened to someone else, long, long ago. Malinowski documented the reasons that the Trobriand people gave ,. for the legitimacy of myth and interpreted their insistence on the truth "*

MYTH AND EXPERIENCE IN THE TROBRIAND ISLANDS

r

287

of myth as deriving from the necessity to maintain the historical connection to the precedents of the past. That is part of the reason the UHAQ sacred; as we have seen, however there is more to it than that; Theiiiyth as a defence mechanism must be both legitimized and protected from challenge. If the myth must be literally true to have its historical/causal effects and if, given what we all accept about how the world works frbw, the myth cannot be literally true, then what we all accept about how the world works now cannot be applicable to myth. The gulf between the present and the distant past ipmitibogwa), the larger-than-Hfe^qtiaiityfof the characters and their actions in myth, the unquestioned justice of their decisions, the insistence that in the past things were of a different sort than they are now, the denial that the inferences we would make today are applicable to the events in myth, in short, the whole collection of reasons people give for the legitimacy of myth, are a secondary defence structure erected to protect the primary defence of the myth.7 The sacred iiliu must be true because the putative historical/causal connection of myth to life depends on the myth's being literally true, and that connection is the only connection between myth and life that can be explicitly recognized. If the myth is to be recalled and used as an interpretive resource in understanding some troubling real-world event, there has to be some connection between it and the event other than the inadmissible fact that it shares a common schema with the unconscious propositions evoked by the event. The historico-causal link provides that connection. Conclusion We began with a description of an actual encounter between a Trobriand village and the spirit of one of its deceased members. For one villager, at least, this raised the question of why the spirits of the dead are nearly always invisible. We saw how myth was marshaled as an interpretive resource to provide an understanding of this troubling aspect of the experience. The story of Baroweni's clumsiness and her mother's retribution are taken as an historical precedent for all subsequent interactions between the living and the dead. An examination of the myth itself, and of the putative historical-causal connection of the myth to experience, has shown that there is another, more compelling connection. That is that the myth is a disguised representation of repressed thoughts and fears concerning relations between self as survivor and the deceased. We cannot say by which link the myth was retrieved from the old woman's memory. But we can say that the way it accounts for the invisibility of the spirits of the dead is via the structural connection rather than the historical connection. The schema instantiated by the myth is the same schema unconsciously instantiated by survivors following the death of someone important to them. It embodies the anxiety about possible responsibility for the death and the projection onto the deceased of the repressed hostility of the sur-

288

EDWIN HUTCHINS

vivor toward the deceased. But the mythic version is a safe version. It is not about self. It is about someone else - a special someone else whose actions long ago caused all spirits to be invisible. The myth,'as a transformed instantiation of-this schema, is a culturally constituted defence mechanism. Furthermore, even though the kosi is sometimes visible, the underlying schema also describes the relationship of living villagers to kosi. We expect the hostility that is projected onto the deceased, and therefore the severity of the deceased's punishment of survivors, to be all the more intense when the deceased was hated in life. This appears from the Trobriand perspective as the observation that the kosi of evil people haunt the village for a long time. So, the schema that underlies the myth of Baroweni and explains the invisibility of the spirits also appears to be the schema that causes villagers to experience the kosi as well. This chapter shows that there is a living connection between myth and experience in the Trobriand Islands. By its structural connection to life, the myth provides a way of thinking about things that are too painful or too threatening to address directly. By way of its historical connection to life, it is both a causal precedent for the current state of affairs and a story that exonerates the living from culpability in the disappearance of the dead. Notes 1. Earlier versions of this paper were presented at the Conference on Folk Models held in 1983 at the Institute for Advanced Study, Princeton, New Jersey, and in the symposium organized by Dorothy Holland and Naomi Quinn for the 80th Annual Meeting of the American Anthropological Association and en-.,,, titled Folk Theories in Everyday Cognition. Field research during which the data reported here were collected was supported by a grant from the Social Science Research Council. Text processing facilities were provided by the Navy Personnel Research and Development Center, San Diego. I am grateful to Roy D'Andrade and Laurie Price for reading and commenting on an earlier '* draft of this chapter. Whatever errors it contains are my own. My greatest debt is to Bomtavau, who told me the myth and provided her own rich commentary on it. 2. See Malinowski's paper, "Baloma: Spirits of the Dead in the Trobriand ' Islands," for a more detailed discussion of the nature of baloma and kosi and of their relationships to the living. 3. I also collected a supposedly historical story in which a woman encounters a more horrible fate. The woman converted to Christianity and married a native pastor. After a few years of marriage, she fatally poisoned her husband, moved * to a different village, and remarried. When her second husband died, she moved back to her own natal village, where she eventually died. Her kosi plagued the village for months, and it was finally determined that Topileta had closed the doors of Tuma to her, and God had closed the doors of heaven as well. \ 4. Unfortunately, I do not know just who (or even how many) among the villagers actually claimed to have seen the kosi, nor do I know what relationship those few I talked to bore to the deceased. ~*<

MYTH AND EXPERIENCE IN THE TROBRIAND ISLANDS

-289

5. Current approaches to implementing cultural knowledge representations iri~. elude schemas, frames, scripts, and more. For our purposes, it is not important which approach is used so long as it captures the structural relationships of the terms of the propositions. 6. D'Andrade (1981:190-193) argues the importance of cognitive scientists' looking at cognition and affect together as related parts of meaning systems, 7. This device is not unique to technologically primitive societies. 'Cohsidei^for example, the following testimony given by a creation scientist in a recent court hearing, '"'" --.--•-.--• -..,.,-. We cannot discover by scientific.investigation, anything about the. ._...!.. creative process used by the creator because He used processes which are not now operating anywhere in the natural universe. (Lewin 1982:144) In order to assert the literal truth of accounts which, by our present criteria of truth and falsehood, cannot be literally true, the claim of the special nature of that time must be made. References D'Andrade, R. G. 1981. The cultural part of cognition. Cognitive Science 5(3): 175-195. Freud, 5. 1918. Totem and Taboo. New York: Vintage Books. Hutchins, E. 1980. Culture and Inference: A Trobriand Case Study. Cambridge, Mass.: Harvard University Press. Lewin, R. 1982. Where is the science in creation science? Science 215(4529):142-I46. Malinowski, B. K. 1922. Argonauts of the Western Pacific. London: Routledge and Kegan Paul. 1954. Magic, Science, and Religion and Other Essays. New York: Doubleday and Company. (First published in 1948; contains Baloma; The spirits of the dead in the Trobriand Islands, first published in 1916; and "Myth in primitive psychology, first published in 1926.) 1965. Coral Gardens and Their Magic. Bloomington: Indiana University Press. (First published in 1935.) Suppes, P. and H. Warren 1975. On the generation and classification of defence mechanisms. International Journal of Psycho-Analysis 56:405-414.

12 Goals, eventss and understanding in Ifaluk emotion theory1 Catherine Lutz

I have three goals in this paper. The first is to represent formally the knowledge about emotions held by the Ifaluk people of Micronesia. That knowledge can be seen to be structured in two fundamental ways: The first is in terms of salient events in everyday life, and the second is in terms of the culturally constructed goals held by the Ifaluk. The second aim is to address the question of the actual status of ethnotheory in social interaction; I stress the idea that the emotional understanding this ethnotheory allows is, in actual practice, an understanding that is negotiated between individuals. Third, I reject the view that ethnotheoretic models of emotion are aptly characterized as involving "cognition about emotion" or "thinking about feeling." Introduction I would like to tell two stories here. To understand each story, it is *" necessary to present the underlying cultural and cognitive model that structures the understanding of the characters in them. The first story is a simple one because it merely involves narration, an assertion made by one character and left unanswered by others. The complexity of the second story arises because it tells a more fully social tale; the characters not only theorize, they also attempt to convince others that their theory is at least plausible, if not the only possible route to proper understanding of the events at hand. The first story occurs on the atoll of Ifaluk in the Western Pacific. I am sitting outside the house of a sick person whom I have come to visit. The illness does not appear to be very serious, and I am idly chatting with ^ the woman sitting next to me. We pause for a moment and a girl of about 4 years of age approaches us. Asshenears, she does a little dance, makes a silly face, and waits. "She's cute," I think, and smile at her antics. The woman sitting next to me has observed this, and she reprimands me, say- .. ing, "Don't smile at her - she'll think that you're not justifiably angry." v In that statement, the woman is telling me much more than she explicitly says. Some of the inferences she would have had me draw are presented -%

GOALS f AND UNDERSTANDING IN IFALUK EMOTION THEORY

-291;

below in the process of outlining a formal model of Ifaluk theories about :> emotion. 7:CiC- V: The study of ethnotheory involves the identification of the knowledge structures that underlie speech, and more generally, understanding/This, knowledge is largely below the level of explicit awareness and generally V 7: remains unverbalized. One special circumstance that permits the recognition of both one's own and others' tacit knowledge is the crossing of cultural boundaries. The (at least partial) nonsharing of knowledge across those boundaries encourages the identification and verbalization of takenfor-granted realities. Thus, one special methodology that anthropology provides for the study of enthnotheories,2 is the immersion in environments that maximize the possibilities for misunderstanding. In the process, the mechanisms of understanding become more apparent.3 In their concern with cultural belief systems, anthropologists share the desire of cognitive scientists to make explicit what it is a person needs to know to come to appropriate understandings of people and events. These two perspectives also share a concern with representing the knowledge people have in the most accurate way. The goal has been to construct representations that are both general, or able to describe knowledge in many domains, and efficient, or able to process information quickly, accurately, and with minimal moves (Winograd 1977). The inability of the anthropologist to go beyond a literal understanding of the discourse going on around him or her in the early days of the field experience finds its analogue in the failure of the computer to be able to process information with a program that is flawed or incomplete in particular ways (although there are obvious and important ways in which this analogy is limited).4 The formal modeling advocated by many cognitive scientists has been used successfully by Hutchins (1980) to represent Trobriand knowledge about land tenure and land transfer. This chapter uses his approach in modeling some aspects of the ethnotheory of emotion among the Ifaluk. Events and emotions in Ifaluk ethnotheory Ethnotheories of emotion describe a fundamental and ubiquitous aspect of psychosocial functioning. They are used to explain why, when, and how emotion occurs, and they are embedded in more general theories of the person, internal processes, and social life. As they play a central role in the organization of experience and behavior, an examination of the structure of emotion ethnotheories can contribute to both cultural and psychological models of emotion and social action. In addition, the existence of dense networks of connections between this domain and other knowledge systems among the Ifaluk gives this ethnotheory wide ramifications in their social life. Although ethnotheory may be an explicit and abstract body of knowledge, it is more often pragmatic, being used as implicit assumptions in

.

292

CATHERINE LUTZ

daily discourse and understanding. The investigation of Ifaluk theories of emotion reported here thus includes the collection of several thousand instances of the use of emotion words in everyday talk and of natural definitions (Boehm 1980), as well as interviews that elicited more generally stated propositions about emotions. The most fundamental unit of any theory, implicit or explicit, is the concept. The central elements of Ifaluk emotion theory are the concepts of emotion that are represented in linguistic form by (among other things) words for discrete emotions. There are almost 100 words in at least occasional use that represent these concepts; a core group of 10 to 15 words can be heard in daily conversations, where they are used in the descriptions of striking or salient events. The Ifaluk define, explain, and understand emotions primarily by reference to the events or situations in which they occur. This aspect of their ethnotheory of emotion contrasts with our own emphasis on the internal and private, rather than the social, nature of emotion. For example, definitions of emotion terms collected on Ifaluk relatively rarely contain reference to the physiological feeling tone associated with a particular emotion; American English emotion concepts, on the other hand, are often defined by reference to the physical and/or private mental state of the person experiencing the emotion (Davitz 1969; also see Averill 1974; Izard 1977). The kinds of situations that are related in theory to emotions are depicted by informants at several levels of generality. They include the more specific, such as "Someone gets drunk and comes to your house every night," or "Your children are adopted to another village and don't come to visit you for a long time," or 'The pig eats your food," as well as the more general - "Something happens that we want to happen," or "There is something we don't know [or understand]," or "Something bad happens that we don't expect to happen." Which level of generality is chosen by informants depends on, among other things, the kinds of contrasts and comparisons being drawn in any particular case. Thus, although the emotion concept can be considered the primary element in this domain of Ifaluk ethnotheory, it is also evident that the definitions of these or any words are in fact themselves propositions of a ' particular type (Casagrande & Hale 1967). It is useful to look at the underlying structure of those propositions in terms of the modeling used by Hutchins (1980), who succinctly outlines the relationship between the terms of such formal modeling; using the example of land use rights in the Tro-,, briand Islands, he states that a relation always links one instance from the range of concepts that are people to one instance from the range that are economically appropriatable units of land. When a relation is stated in terms of such variable ranges als these, it is a schema. When the ranges are replaced by concepts (a process called instantiation because it is the assignment of specific instances to the.

GOALS AND UNDERSTANDING IN IFALUK EMOTION THEORY

.^93?:

Ev [Illness] Ev [Travel from the island] Ev [Lack of food] ====-> Em [Fago (compassion/ Ev [Nurturance] love/sadness)] Ev [Gentle or calm behavior) Ev [Rule violation by self] Ev [Spirit present] ==»«=> Em {Metagu (fear/ Ev [Call to eat by someone anxiety)] unfamiliar] Some emotion words in the Ifaluk language share situational definitions, including both those that can be described as general and those that are more specific. Determining which emotion term will appropriately apply in any concrete situation involves a determination of some other characteristics of the individuals involved in the situation. These characteristics include particularly their social status ranking relative to each other, the history of their relationships with each other, and the possible influence of third parties on the actions of those involved.7 The process whereby these forms of knowledge are applied to a particular case can be examined both as a cognitive and as a social process. In the extended

294

CATHERINE LUTZ

example of emotional negotiation examined later, these other factors in understanding that are secondary to the general event-emotion schema are treated in more detail. ' "'/•• In summary, emotion concepts on Ifaluk are the fundamental units of an ethnotheoretical system that informs the understanding of salient events. These concepts are not conceptual primitives but rather are complex in meaning and schematic in form. To litter an emotion word on Ifaluk is not primarily to evoke an image of internal churnings or of particular ways of hotly thinking; rather, it is to evoke an image of a particular kind of event, a particular relationship between a person and the world. More generally, we can expect culturally variable ethnotheories relating to any topic to be evident in equally variable word meanings. Central concepts or key words, then, are not simple insertions into the schematic relations of an ethnotheory but are themselves ethnotheoretic and schematic. " The Ifaluk use at least three other central schemas in the process of emotional understanding. One schema states: (2) If we experience Emotion X, then we may perform Act Y. or Em [ 1 -====> A [ ] This schema is familiar to us as it is a framework for emotional understanding in American ethnotheories of emotion as well. From both the American and the Ifaluk ethnotheoretical perspectives, emotions and action are closely linked. It is important to note, however, that the causal link between action and emotion is much more probabalistic in American ethnotheory, where an emphasis on the control of emotions means that a situation of "unexpressed emotion" very commonly occurs. For the Ifaluk, on the other hand, control becomes an issue only in special circumstances (which particularly * include cases in which the potential action is physically aggressive). In fact, one action commonly Unked to virtually all emotions in instantiations of the above schema is "Telling someone about the emotion." It is expected, , as a sign of maturity and intelligence, that a person will declare an emotional stance in appropriate situations (of which there are many). Notwithstanding the similarity in the general form of this schema in both American and Ifaluk ethnotheories, examining some specific propositional examples reveals several culturally distinctive instantiations of the schema. Em [Ker (happiness/ excitement)]

A [laugh] =====> A [talk a lot] \ A [misbehave] A ["walk around," i.e., •-,, neglect work, show off] "*

GOALS AND UNDERSTANDING IN IFALUK EMOTION THEORY

295;

A [give food] Em [Fago (compassion/= = - - - > A [cry] love/sadness)] A [talk politely] A [don't speak],,/-{:~.,,:-.n •., A [talk impoUtHyj.^^;;^,.;; Em [Song (justifiable ======> A [reprimand] •; ,,.,.,. . .A [don't eat]. . ; ; ;;./.^,i4^ anger)] A [pout] ' .'.J;.. For example, one culturally standard expectation is that people who are justifiably angry might refuse to eat. Alternately or additionaUy,Hhey:might speak to others without common politeness markers, such as the use of the term "sweetheart" before a request. People thus use particular behaviors as signs of specific types of emotional understandings in others. All the listed instantiations of schema (2) are not equivalent, however, in terms of either their frequency of occurrence or cultural salience. A frequently occurring proposition is likely to be one that is particularly useful (for reasons that can be ethnographically described). In addition, each proposition evokes ethnotheoretical corollaries and semantic associations of differing degrees of richness; for example, the laughter that accompanies ker (happiness/excitement) is a relatively impoverished instantiation of schema (2) when compared with the idea that "happy/excited people walk around." The later notion is frequently evoked in condemnatory gossip about others; its links the emotion to amoral behavior; and it justifies the generally unspoken idea that people ought to stay close to their homes unless an errand calls them elsewhere. A rich and useful model of any cultural knowledge system specifies the density or weight of each proposition enumerated. Another schema the Ifaluk use in discussing and understanding emotions is of the form, (3) If we experience Emotion X, then another person should or might experience Emotion Y. or Em 1 [ 1 ==^==> Em 2 [ ]* Propositional instantiations of this schema include the following: Em 1 [Song (justifiable > Em 2 [Metagu (fear/ anger)] anxiety)! Em 1 [Ker (happiness/ - = = - = > Em 2 [Song (justifiable excitement)] anger)] Em 1 [Chegas (romantic =====> Em 2 [Lugumet pride/self confidence)] (discomfort/guilt)] T h e numbers are used to distinguish the different actors involved in an emotion event.

296

CATHERINE LUTZ

Em 1 [Tang (frustration/ =====> Em 2 [Fago (compassion/ grief)] love/sadness)] 7 Schema (3) can be termed a basic level schema on the basis'of the frequency of explicit statements and implicit, necessary inferences that would instantiate it in everyday Ifaluk discourse. On the other hand, propositions of this sort may be formed through a chain of inference linking together propositions of types (1) and (2). For example, the statement that song (justifiable anger) in one person leads to metagu (fear/anxiety) in another person could be formed in the following way (note that one per-r son's actions become an emotion-producing event for another person): (2) Em 1 [Song] =====> A [reprimand] (1) Ev [reprimand] > Em 2 (Metagu (fear/anxiety)] Propositions of type (3) probably are formed sometimes in this latter way and in other cases formed directly on the basis of a schematic structure of type (3). The more common and frequent dyadic links between emotions, such as that between song (justifiable anger) and metagu (fear/ anxiety) likely are learned by children in the direct form of schema (3). Knowledge of these dyadic links, in fact, is evident in children at a very early age and in children who seem simultaneously unaware of some of the particular propositions of types (1) and (2) that form part of the chain of inference necessary to link conceptually emotions in two different individuals. In other words, children who do not demonstrate understanding of the link between, for example, song (justifiable anger) and rule violation are yet able to predict that song (justifiable anger) in one person will produce metagu (fear/anxiety) in the other. On the other hand, many ** statements about the relation between emotions in self and other are created de novo on the basis of schemas (1) and (2) alone. Whether a chain of propositions of types (1) and (2) or an instantiation of schema (3) is involved in any particular case is an empirical question; however, it can be assumed that cognitive operations of the latter sort are more likely when the more common emotion pairs are involved. Other evidence for the schematic nature of the structure underlying i propositions of this type is also available. In talking about the emotions, the Ifaluk treat them as fundamentally social phenomena rather than, as in the case of American ethnotheory, as predominantly internal psychophysiological events that are simply correlated with social events. A proposition of type (3), if generated by an American, probably more likely would have been made on the basis of inferences about the likely correlations between environmental triggers of an emotion in the first person and an emotion in the second. Among the Ifaluk, on the other hand, people „. are conceptualized as more directly influencing one another. This means * that emotions in one person can lead in a more immediate way to emotions in another. People on Ifaluk believe that they hold important re- %

GOALS AND UNDERSTANDING IN IFALUK EMOTION THEORY

297

sponsibility for the emotions of others; they may be held responsible for causing an emotion in another (as in the statement "He is needy," which is literally, "He causes me to feel compassion") or for responding With the correct emotion to another's. The expectation of an important degree of emotional symbiosis between individuals is implicitly outlined in ifaluk ;' emotion theory and lends meaning and schematic, form to propositions of type (3). The dyadic linking of emotions in Ifaluk theory is also, informed by the social roles and situational positions of the two individuals involved. The role structure that White (unpublished data) has identified as an im-: portant aspect of emotion attribution among the A'ara of the Solomon Islands is also involved here; the speaker, agent, and affected within an event may necessarily have different stances and hence different emotional responses. Although propositions of type (3) do not specify the position of the actors involved, such information is efficiently stored in the definitional propositions that are nested within them. For example, the pragmatic information encoded in the term song (justifiable anger) includes the notion that judgments about whether a particular act constitutes a rule violation are more aptly made by people of higher social status, including the chiefs and older individuals. Someone who claims to be "justifiably angry," therefore, is simultaneously claiming to be in a relatively superior position vis-a-vis the person who has erred. Conversely, the individual who asserts "fear/anxiety" appeals to that term's definitional schema, whose instantiations include a correlation between that emotion and events in which the individual is in a weak (and, hence, often dangerous) position in relation to others or the environment. A fourth and final schema underlines the Ifaluk view that the situations associated with emotions are not static. This schema represents the recognition that these situations develop in ways that are often predictable, thereby creating expected sequences of emotion in an individual involved in that situation. (4) If we experience Emotion X, then we may later experience Emotion Y. or Em [ ]---=~> Em [ ] Sequences of events that are of high frequency underlie many instantiations of this schema. Some examples include: Em [Ma (shame/ embarrassment)] Em [Rus (panic/fright/ surprise)]

== = *==> Em [Song (justifiable anger)] == = = => Em [Song (justifiable anger)]

298

CATHERINE LUTZ

.,:.

Em [Pak (homesickness)] =====> Em [Nguch (sick and tired/bored)] . Em [Ker (happiness/ =====> Em [Metagu(fear/ .':"..•" excitement)] anxiety)], :• /• ;Em [Filengaw (incapa- "=====> Em [Ma (shame/ L bility/discomfort)] embarrassment)] . -As with propositions of type (3), it is not necessary to postulate a single schema that necessarily generates all of the propositions of type (4). Rather, some of the propositions, in fact, can be generated as composites of other schemas. The sequence of ker (happiness/excitement) and metagu (fear/ anxiety), for example, follows from these three propositions: (2) Em 2 [Ker (happiness/ =====> A [misbehavior] excitement)] (1) Ev [misbehavior by ========> Em {Song (justifiable other] anger)] (3) Em 1 [Song (justifiable > Em 2 [Metagu (fear/ anger)] anxiety)] or, in more expanded form: (2) Em 2 {Ker (happiness/ -=-=-> A [misbehavior] excitement)] (1) Ev [misbehavior by > Em [Song (justifiable other] anger)] (2) Em 1 [Song (justifiable ====«=> A [reprimand] anger)] (1) Ev [reprimand] ^^z==z=> Em 2 [Metagu (fear/ anxiety)]8 .>•* This returns us to the story of the prancing little girl and the witless anthropologist with which this chapter began. I had interpreted the girl's antics with an ethnotheoretical framework in which happiness is the ultimate good (as well as being "inevitable" in children), but our adult companion viewed the child's behavior in light of the link that, in her view, holds between ker (happiness/excitement) and misbehavior. In an effort to head off the disruption that "happiness/excitement" is expected to bring, • the woman asked me to stop smiling to open up the possibility of appearing song (justifiably angry). This justifiable anger was expected to produce "fear/anxiety" in the young girl. This fear would, in fact, constitute a recognition of the error of her ways; in Ifaluk ethnotheory, the expe-; rience would be a positive one that would help her mature into moral awareness.

GOALS AND UNDERSTANDING IN IFALUK EMOTION THEORY

.299

Goals and schematic range Schemas are, by definition, very open and abstract knowledge structures.: It is clear, however, that in practice only a limited number .of. instantiations of particular schemas are found and/or acceptable in .any.culture.: Some mechanism must be posited that operates to restrict the range of variables that can be inserted into the terms of schemas. One important factor contributing to the cultural construction of that range as well as to the overall coherence of the ethnotheoretical system is goals. ,--:/ Goals have a central place in Schank and Abelson's theory of knpwl-: edge structures (1977:101-130); in this, goals parallel the traditional position of values in many anthropological theories that attempt to account for the organization and motivation of behavior. According to Schank and Abelson, understanding of events is enabled by the knowledge people have about the goals typically held, both by people in general and by particular types of individuals. Both the prediction and the production of behavior is predicated on detailed knowledge of the plans that can be used for achieving particular goals (1977:70-71). It is necessary to add that both knowledge about goals and the goals themselves are culturally constructed. Thus, knowledge about the goals of others allows for prediction of behavior, and goals assist in the construction, in the first place, of ethnotheoretical knowledge systems. This relationship between goals and knowledge can be stated more abstractly as a dialectical one between direction (goals or values) and structure (knowledge) in human meaning systems. Not all goals are created equal. In the first instance, goals exist on several levels from (in Schank and Abelson's example) the most general, such as "Eat," to the more specific, such as "Eat steak at The Steak Pit" (1977:109). At all levels of generality, however, cultural factors work to produce many of the goals held. The question of whether these various types of goals are hierarchically arranged is important. In anthropological discourse, the related question has been raised as to whether or to what degree the general goals entailed in cultural values actually order more specific day-to-day goals and behavior. On the other hand, it may be more apt to characterize things in the converse way (i.e., it may be that the more specific, grounded goals are the dog to the tail of more general goals). Although this issue cannot be dealt with here, it is important to link emerging cognitive science frameworks on the relationships between various types of goal-related knowledge structures with the trail of this traditional anthropological debate. Each goal also has differential weight or salience. Description of an ethnotheoretical system requires a treatment of the degree of emphasis on a particular goal within a cultural system; thus, although both the Ifaluk and Americans may have the goal of avoiding violence, rates of physical

300

CATHERINE LUTZ

aggression in the two societies and beliefs about those rates are in dramatic contrast, in part due to cultural differences in the importance attached to that goal. / ' With the notions of goal generality and goal salience in mind, it is possible to discuss the kinds of goals that operate to constrain and construct the Ifaluk ethnotheory of emotion. At the highest level in any system'are culturally constituted goals that apply across many or all domains of knowledge. A general goal such as "Be first" can motivate behavior in; a supermarket line, in the classroom, or in discussions with one's spouse/ In the Ifaluk case, some of the most important higher-order goals include "Avoid confrontation," "Share food with others," "Comply with the demands of those more highly ranked than oneself," and "Avoid aggression."9 These goals help structure emotion ethnotheory as well as other theoretical domains. Compare these goals with such central middle-class American goals as "Assert yourself," "Control yourself," "Get ahead," "Know thyself," and "Stand on your own two feet." Are there not goals that are more general than these latter examples in ethnotheoretical systems? It may be only social scientists and philosophers whose theories attempt to model human goals on a level more abstract that those of the sort just mentioned.10 In social science theorizing, it is frequently the case that one master goal is posited as the structure behind all or most behavior. Thus, all cultural goals, including those found on Ifaluk and in the United States, might be portrayed as examples of the goal of "Being identified by others as a good person" (e.g., Cancian 1975) or as subserving the goal of acquiring protein or maximizing cash inflow. At another level, there are goals that are specific to, or of much greater importance in, particular domains of knowledge. In the area of emotion, some Ifaluk goals can be stated in the following terms: "Understand the event" (this in contrast to attempting to understand the internal feeling ^ state or the emotion's role in one's own psychohistory), "Avoid illness by communicating one?s emotional stance to others," and "Present oneself as a mature and intelligent person by disclosing many of one's emotional perspectives." Finally, there is the most specific level at which goals exist - the level of the element or emotion word, where the degree of detail expands greatly. What I am claiming here is that emotion words entail goals; that is, when a case of anger or guilt, fago (compassion/love/sadness) or ker (happiness/ * excitement) is identified in a particular person, relationship, or situation, a goal is simultaneously and necessarily identified. Quinn (1982) has argued that goals are also implicated in what she terms key words, and that such words "generate" goals. What remains to be done is specifying the pro-'*. cess whereby emotion words and key words are linked with goals. A preliminary suggestion is that emotion words do more than just signal the .f presence of a goal; they may also actively produce goal directionality. By ^

GOALS AND UNDERSTANDING IN IFALUK EMOTION THEORY

301

saying that I am angry, I in fact, may produce a motive, or more likely deepen and clarify an existing motive.11 This additionally suggests that \ all understanding (including the emotional) is enabled or enhanced by social: discourse; in hearing what we ourselves and others say about emotions/ we come to understand better (or create) our goals and other perceptions. • Two types of specific goals are encoded in Ifaluk emotion wbrdsi''including action goals and disclosure goals. An action goal is amotivei to act, an inclination, a directional impulse. By many definitions, Amotions are intrinsically motivational (e.g., Izard 1977), and hence goal-IadenV Action goals for some of the most commonly used emotion terms oh Ifaluk include the following. Song (justifiable anger): "Change the situation by altering the behavior of the offending party." Fago (compassion/love/sadness): "Change the situation by filling the need of the unfortunate party." Ker (happiness/excitement): "Make use of the resources in the situation. Maintain situation." Nguch (sick and tiredness/boredom): "Persevere, or, If no rules would be broken, change the situation by breaking off the pattern of repeated noxious stimuli." Waires (worry/conflict): "Seek further information. Seek assistance in decision making." The action goals for these emotions differ in varying degrees from the goals embedded in similar but nonidentical English emotion words. "Anger," for example, might entail more aggressive goals than does song (justifiable anger). Compare the goals embedded in any one of the three American English terms needed to translate fago. Disclosure (or attribution) goals are the second type of goal embedded in each emotion word, and they arise by virtue of the degree of social acceptability connoted by each word. The disclosure goals related to Ifaluk emotion words are affected first by the general domain goal of expressing emotion verbally. As noted, statements of one's emotional posture are generally considered a sign of maturity and intelligence, and people frequently make statements of the form "I am (emotion word)" or "We are (emotion word)." The disclosure value of each emotion is somewhat different, however. Fago (compassion/love/sadness) and song (justifiable anger), for example, follow the general domain goal very closely; that is, it is considered good to attribute these emotions to oneself and to others. Both emotions are considered mora! judgments about the world that people should make. Two emotions in particular entail nondisclosure goals; these are gasechaula (hate) and ker (happiness/excitement). The action schema (number 2 above) for gasechaula (hate) includes intentionally harmful acts; to disclose hate, therefore, is to disclose the intent to harm, which violates

302

CATHERINE LUTZ

the fundamental Ifaluk injunction against aggression. As the logic of the meaning of ker (happiness/excitement) outlined earlier indicates, an action schema for this emotion includes misbehavior, another act that by definition incurs negative public reaction. In addition, several emotion words, including baiu (romantic love/happiness) and gas; (hap^. piness/pride/confidence), are taboo in mixed gender conversations, primarily due to the terms' sexual connotations. _ . . . . ,,;;,,. ; These latter examples raise the more general problem of the multiplicity of disclosure goals attached to single emotion words. Although a single predominant or overall disclosure goal can be identified for each emotion word, at least as important is the proliferation of goals based on a large number of pragmatic considerations. Ethnotheories may identify which emotions are acceptably attributed in four different kinds of circumstances - when the attribution is (1) to the self, (2) to others, (3) to particular kinds of others (for example, chiefs vs. commoners), and (4) when it occurs in particular kinds of situations (for example, in "crisis" vs. normal situations or in the presence of diverse audiences, such as "mixed company")- Thus, a particular emotion word may entail the goal of nondisclosure vis-a-vis the self, disclosure vis-a-vis others, attribution to relatives and not to nonrelatives, and nonattribution in "mixed company" but disclosure in single-gender groups. In sum, it is impossible to talk about ethnotheory without talking about the goals it serves and articulates. This is particularly the case in relation to emotion ethnotheories, as emotions are motivational by definition. To separate cold cognition from hot motivations in delineating ethnotheories is to follow the lines of an American ethnotheory about the dichotomous relationship between the categories of "emotion" and "thought," an issue" to which I return in conclusion. It has been suggested that a relatively limited number of general goals constrain the creativity of ethnotheories. Goals place limits on the number and kinds of concepts that can be inserted into the framework of ethno-"' theoretic schemas. As knowledge about aspects of the physical and social world is developed in the context of attempts to attain particular culturally constituted goals, it is necessary to examine ethnotheoretical knowledge * in the context of those aims it was developed to satisfy. The negotiation of understanding: an example Up to this point, Ifaluk ethnotheory has been described as if it were a transparent and exhaustive model used to interpret and respond unambiguously to the statements of others. In actual fact, the model is ambiguous enough to allow for both confusion and negotiation of meaning, in actual interactions. The inherent ambiguity of all human messages is51 due not only to their telegraphic nature but also to the fact that ethnotheories are incomplete, not entirely internally consistent or coherent, in ^

GOALS AND UNDERSTANDING IN IFALUK EMOTION THEORY

Sick Sid Woman

I

f / /

iiii.il Pregnant Woman

303

/ / /

Letaupo *"• — ^ "" — — —

—• -""

O

A

Female

Male Adoptive tie

Jk

Deceased

Figure 12.1. Relations among principals involved in the event

process of construction by individuals, and not universally shared in all details among the members of a culture. This section presents an example of the use of Ifaluk emotion ethnotheory in interaction. The example illustrates that ethnotheories find their meaning in use and in the process of negotiation that occurs each time a situation is linked with an emotion word. This negotiation occurs primarily over whether the case at hand is prototypical of the emotion concept; what is negotiated is the meaning to be assigned to events. The principal individuals involved in this case are a 32-year-old woman in her 9th month of pregnancy, her father, an older and a younger sister, two younger brothers, and her husband (see Figure 12.1). Word reached this family that a relative, through adoptive lines, of the younger sister Letaupo (a pseudonym) had fallen seriously ill on another island. On hearing this, Letaupo told me and another young woman that she was waires (worried/conflicted). She explained that her worry and conflict occurred because she wanted both to be with her pregnant sister and to be with the ill woman, who stood in the relation of "mother" to her. She did not have to explain that she held the goal that is strongly and universally subscribed to on Ifaluk, which is to be with a relative at the precise moment of death (people have threatened suicide at the prospect of being absent at such a time), nor that she assumed it to be a possibility that her sister might die in childbirth. She also did not have to affirm explicitly her belief that it is very important to be with a female relative during labor and to provide assistance in the seclusion period that follows for mother and child. By asserting her waires (worry/conflict), she declared herself to simultaneously hold those two, now conflicting, goals.

304

CATHERINE LUTZ

Soon after her remark, Letaupo's father came to tell her that he and her older sister would be going to the other island to be with the sick woman, and that she, Letaupo, would be staying with her pregnant sister. Letaupo leapt up, grabbed her young son and work knife, and ran out with them towards the bush. Several people who had seen this all transpire agreed that she was song (Justifiably angry), and that she had probably run into the bush with the intention of killing herself. She was chased by several other relatives, who told her that her father would change his mind " and allow her to leave the island. Later that afternoon, Letaupo, her father, and older sister all left for the distant atoll residence of their sick relative. By his initial decision, Letaupo's father had attempted to shift the definition of the emotional situation from one of value conflict in Letaupo's position to one in which obedience (rather than any emotional stance) was required of her. By her dramatic behavior, Letaupo had attempted to negotiate the use of the schema that is instantiated as: Ev (Rule violation]

> Em [Song (justifiable anger)]

She did this symbolically by her behavior; running into the bush in that manner was interpreted by other people as an instantiation of schema (2), whereby justifiable anger may lead to suicide. Em {Song (justifiable anger)] =====> A [Suicide] Negotiation continued after Letaupo's departure, however, because not all parties concerned would agree that her father's telling her to stay constituted a rule violation. Negotiation would also continue over the extent to which the father's leaving the island itself constituted a rule violation, thereby justifying song (justifiable anger) in others. .* After news of the travel plans of their father and sisters reached them, the adult brothers of the pregnant woman arrived at her household to lend support and to express their views of the behavior of the departing family members. Each man stated several times in the prolonged stay at the sister's"" home that he had come to tell her about his feelings and thoughts so they would "go away" in the process of being verbalized. Each brother's description of his position was different, however; the older brother said he was i nguch (sick and tired/bored) while the younger (who was very drunk on coconut toddy) declared he was song (justifiably angry). At several points during the afternoon and evening, each man said he had cried over the incident earlier in the day. Afterward, several people told me and others of their concern that other men in the village would gama (cause shame and embarrassment to) the elder brother in the days to come. This series of events is only explicable on the basis of some of the goals and ethnotheoretical propositions held by the Ifaluk. Island men should. ideally play an active role in problems and decisions affecting their sisters.s A woman's husband should in fact defer in matters such as the one at hand to the wishes of her brother(s). Within the group of a woman's brothers, ™

GOALS AND UNDERSTANDING IN IFALUK EMOTION THEORY

305

as within most groups on Ifaluk, primary responsibility devolves upon the eldest. In the present case, the brothers should have attempted to force one sister to stay on the island to care for their pregnant sister. Thus,; the sisters' leaving constituted a "failure" (rather than a rule violation) for the elder brother. The fear that the brother would be ashamed is abased on the definitional schema for shame in which: i, :,- .-;••,Ev [being observed in failure] =====> Em [Ma (shame/ "-'' embarrassment)!. Given that it is also believed that excessive shame can lead to suicide, Em [Ma (shame/embarrassment)] ===-=> A [Suicide] people expressed sensible concern over community reaction to the elder brother. It is also interesting to note the difference between the two brothers' definitions of the situation. In general, the events and action goals associated with nguch (sick and tired/bored) are more appropriate to a relationship among relatives than to one among nonrelated individuals. Although song (justifiable anger) occurs among relatives, the goal of maintaining interpersonal harmony and avoiding confrontation is even more important for kin. In comparing the following two series of propositions associated withnguch (sick and tired/bored) and song (justifiable anger), it should be evident that the goals associated with kin versus nonkin relations determine which action is more condusive to achieving each of those two types of goals. Given the knowledge structures involved in the Ifaluk ethnotheory of emotion, it is often necessary for individuals, like the elder brother, to declare nguch (sick and tiredness/boredom), where others would declare song (justifiable anger). (A) Ev [noxious social obligations] ==-==> Em [Nguch sick and tired/ bored)] Em [Nguch (sick and tired/bored)] =====> A [tolerate, persevere] (B) Ev [rule violation] > Em [Song (justified anger)] Em [Song Gustified anger)] = = = = => A [reprimand] Finally, the brothers' statements that they had cried entailed implicit ethnotheoretical propositions and constituted an attempted negotiation of the definition of the situation. Of the several emotions whose action shemas may be instantiated by "crying," the most common is /ago (compassion/love/sadness). Given that several propositions relating to fago (corapassion/love/sadness) include the following, Ev [Neediness] --===> Em [Fago (compassion/love/sadness)] Era [Fago] =====> A [Nurturance]

306

CATHERINE LUTZ

and given that the concept also entails the notion of rank, with more highly positioned individuals being able to experience /ago (compassion/love/ sadness) by virtue of their "nonneediness," it is likely that the brothers were asserting both their compassion for their sister's situation (as well as for their own embarrassing position) and their intention to take cait of her needs through the birth, as well as their siiperordinate status relative to their sister. The story continues. Let me pass over the long and involved sequence of events in which the principals who remained on the island continued to make creative use of emotion ethnotheory to present themselves and their goals to each other. In particular, discussions about the incident centered around the issue of responsibility and song (justifiable anger), with the pregnant woman sometimes expressing the latter, but more often allowing others to declare that emotion in themselves on her behalf. Approximately two months later, the woman had had a healthy baby, and the child's grandfather had subsequently come back to the island. One evening several weeks after his return, the father arrived unexpectedly at the household of his offended daughter. It was obvious that his purpose was to clear the emotional air. He said he had been thinking about the incident continually and had felt "bad inside." Saying that he had come to ask his daughter to "speak her nguch (sick and tiredness)," he proceeded to explain his actions to her; he had gone because he had /ago (compassion/love/sadness) for Letaupo's adoptive mother, but, he said, "I came right back" (referring to the fact that he had returned to Ifaluk before the death of the ill woman). He told his daughter he had called on the shortwave radio to ascertain her condition, and he had, before he left, asked his sister to take care of her during the birth and postpar: turn periods. He asked if his sister had in fact done what he asked her to do. If she had, he said, "Iwe"(O.K., enough), meaning that there was then no reason for his daughter to be song (justifiably angry). If she did not take care of the daughter, "Ye toar mele i be ser" ("There is nothing' I will say"), meaning he would have no rejoinder to her condemnation of him. The situation is one that can be defined as creating either song • (justifiable anger) or nguch (sick and tiredness). The father came to his daughter compelled to negotiate a resolution to the emotional impasse at which they had arrived. Even though he might have liked to define the situation as one in which no emotion was even recognizable, his family's .. general displeasure continued over a long period and required a countermove or response on his part. By asking her to "speak her nguch," he placed the event into a category less serious than that of song (justifiable anger). He also left the door open for further negotiation by expressing his willingness to listen. Implicit in his argument is that leaving the island when he did was not

GOALS AND UNDERSTANDING IN IFALUK EMOTION THEORY

307^

a rule violation if the people he instructed to take over his responsibilities had carried out their duties. He also implicitly acknowledged, however! the possibility of a rule violation, and hence song (justifiable anger). The father's discourse constituted a "thinking out loud" about the problem of defining the appropriate emotional description of the situation^fhisdiscourse Occurred within the genera! constraints of the ethnotheory of ; emotion he shares with his daughter, but its specific content reveals the goals the father had - goals of maintaining, if possible, the definition of his behavior as moral while reaching a more harmonious state of affairs with his daughter. The evening ended with the father's visit having been taken by his daughter as an apology, and relations between them returned to normal. <-•"•*• ' , ; • "-1 In sum, negotiation within and by use of ethnotheory occurs on at least two levels. The first is semantic and involves motivated discussion over the extent to which an event is a prototypical or good example of a particular emotion. The questions people ask of each other here are of the sort, "Is this song (justifiable anger) or nguch (sick and tiredness/boredom)? Is this a rule violation or simply a noxious but unavoidable situation?" These semantic questions are framed in terms of the underlying schemas that organize knowledge in the particular domain at hand. Thus, whereas negotiation occurs in these Ifaluk examples by use of the event schema, in the American case it would be at least as likely that the physiological or internal feeling state schema would be used; the questions then might be of the form, "Is.this anger or am I just tired? Is he sad or is that just a frog in his throat?" The second level at which negotiation occurs is pragmatic. People discuss whether it is appropriate to use a certain emotion concept with particular other people in particular contexts. Underlying much of the preceding incident, for example, is the question (not explicitly, rigidly, or absolutely answered by Ifaluk ethnotheory) of whether it is appropriate to associate a set of circumstances with song (justifiable anger) if relatives (and particularly older ones) are involved. 12 Conclusion This short outline of Ifaluk ethnotheories of emotion only suggests the extent and coherence of this knowledge system. It demonstrates, however, that the role of the emotion word is central for the storage and structuring of ethnotheoretical knowledge in this domain. Emotion concepts have embedded in themselves crucial cultural propositions and in turn are nested in larger networks of knowledge about persons, roles, and goals. The various culturally constructed goals that many Ifaluk share are coded in their emotion words. These words frame particular groups of events as similar and meaningful, and they organize action and role taking within

308

CATHERINE LUTZ

that frame. This paper suggests that an understanding of how individuals interpret events and of how they plan their total response to them can be built on an understanding of emotion ethnotheories. / I conclude by rejecting one possible interpretation of the structure of this analysis. It is not meant to be a model of how the If aluk "really feel," nor is it intended as a model of how the Ifaluk "think about their feelings." I have argued elsewhere that the concern with "true, underlying feeling" is a local cultural preoccupation and that the dichotomous categories of "cognition" and "affect" are themselves Euroamerican cultural constructions, master symbols that participate in the fundamental organization of our ways of looking at ourselves and others (Lutz 1985a; 1986), both in and outside of social science. These categories and their ramifications are neither universal nor do they appear particularly useful in their current unexamined state for several reasons. Our ethnotheories of the nature of the person, of the social, and of the mind lead us to bifurcate our studies into the cognitive and the affective, to conceptualize ethnotheories as involving primarily or exclusively the former, and then to perplex over how emotion can be "attached" to ethnotheories so dispassionately defined. Thus, the questions that must continually arise (and remain unanswerable) include, "Does any particular ethnotheoretical model have an effect on the way people feel and behave (for we know that thought, being not-affect, is thereby not motivational)?" and, more generally, "How does the structure of ethnotheory acquire direction and force (the latter being, we know, provided by the definitionally excluded emotion and its energy)?" In talking about ethnotheories, we play with the relations between the elements of cognition and affect but maintain the elements themselves as conceptual primitives. The dichotomy of thought and emotion is all the more powerful and seemingly irresistable as it is integrated into a large number of other basic cultural themes and dichotomies; thus, we talk about the naturalness of emotion and the culturalness of cognition, the uncontrollability of emotion and the controllability of thought (or rather the need for control of emotion and the lack of such need for thought), hidden and dangerous emotion and more overt and safe thought. One of the most powerful' cultural distinctions in which the thought-emotion dichotomy is enmeshed is between facts and values. The alienation of emotion from thought in our explicit theories has its parallel in the perceived irreconcilability of facts and values and represents another disadvantage to maintaining the... former dichotomy. The current ideology and practice of science no doubt underline the importance of the distinction between mental activities (thought) and bodily activity (physical labor as well as emotion), between the judgment of matters of fact (primarily executed in the realm of thoughts) and the judgment of good and bad, or matters of value (often associated with the emotional). It would perhaps be surprising if we did not tend to model

GOALS AND UNDERSTANDING IN IFALUK EMOTION THEORY

309

the world on the basis of our own activity and values as social scientists; if we did not find our informants also maintaining a time and a place for rational, cognitive thought and a time and a place for.'.dr^pnd.d^^ieiice, ' with ethnotheory then clearly representing their efforts in the Ipfmerfield, If the concepts of emotion and thought were more integrated in-aftalyses , of ethnotheories, it might be less tenable to maintain the position that social science develops independently of its social and cultural milieu. As an overly rigid distinction between facts and values is predicated prf there being an objective distinction between how we think about something and how we feel about it, a breakdown of the latter would undermine the former and, with it, the view that it is possible to do social science independent of one's socially and culturally constructed interests.:-v This paper, then, represents my motivated representation of how the Ifaluk passionately model a selected aspect of their psychosocial world. In calling this the Ifaluk "ethnotheory of emotion," I do not mean to imply that they, like us, habitually separate cognitive from affective functioning (Lutz 1985b). Their ethnotheory of emotion is, in both their own and my views, a theory of the whole. Rather than an ill-fitting overlay of cognitive understanding on inchoate emotional experience, what is described here is a set of Ifaluk propositions infused with goals and meaning, interest and intentions. Notes 1. An earlier version of this paper was delivered at the Conference on Folk Models held in May, 1983, at the Institute for Advanced Study, Princeton, New Jersey. A still earlier version was presented at the 80th Annual Meeting of the American Anthropological Association in Los Angeles in 1981, in a symposium organized by Dorothy Holland and Naomi Quinn and entitled Folk Theories in Everyday Cognition. I would like to thank Edwin Hutchins, Bonnie Nardi, and John Kirkpatrick as well as Dorothy Holland and Naomi Quinn for their helpful comments on earlier drafts. The field research on which this paper is based was conducted in 1977 and 1978 with the kind assistance of the people of Ifaluk and with the aid of NIMH Training grant # MH 5-T-2-14088 to Beatrice Whiting. 2. Folk theory, cultural theory, and ethnotheory are often used synonymously. The latter term is used in this paper to avoid the connotations of the word folk (in the American Heritage Dictionary, it is defined by use of the terms unsophisticated, unrefined, and common), which implicitly and unnecessarily posit a hierarchic relationship between this type of theory and others. 3. It is interesting to compare two of the major directions taken by people pursuing the long-standing anthropological interest in belief systems. Cognitive anthropologists have been interested in examining belief systems as systems of elements or concepts and as a set of relations between those elements (e.g., Casagrande & Hale 1967; Colby 1975; D'Andrade 1976). Another group of individuals has been interested in the simultaneous examination of the belief systems of the anthropological observer and of the observed (e.g., Dumont 1978; Riesman 1977). Despite their interest in the process of understanding and their concern with questions of validity, cognitive anthropologists have

310

4.

5.

6.

7.

CATHERINE LUTZ

tended to be relatively uninterested in examining the anthropological research act itself as a process of constructed understandings. The positivist tradition within which cognitive anthropology has developed has meant that its methodological task has been construed as one of "eliminating" observer bias rather than illuminating it. ".'"";/"' The analogue between computer and human subject is one with obvious limits. To maintain a perspective on those limits, it seems useful to conceptualize this particular analogical exercise as a cultural task. As we model the computer in our own image (and as our image of ourselves is affected By the presence of that technology in our midst), and as that image is a cultiir4liy constructed and culturally specific one, it follows that the'computer's" "failure" to process information in the correct manner represents a cultural judgment that we make about the proper or dominant goals, nature, and content of mental activity. • •.'••.. This usage follows D'Andrade (1976). In some cases, the indeterminacy noted may be on the part of the ethnographer, but in most cases it represents indeterminacy in observed use of the emotion terms. This indeterminacy of use often reflects the existence of important and specifiable pragmatic considerations; for example, a severe illness in a close relative will definitely call for /ago (compassion/love/sadness), whereas a minor cold in a stranger would not (the latter marginal case of illness might call for a marginal case of fctgo 'compassibn/love/sadness"). Or, frequent requests for tobacco from a beloved friend would not cali for being nguch (sick and tired), whereas those same requests from a mentally retarded neighbor would. In other cases, the indeterminacy reflects the fact that people can choose from an array of ways of thinking or acting on the basis of their own tip- (will/desire/feeling), which may be unpredictable; for example, in a proposition of type (2) described below, a person may choose to "walk around" but not laugh in conjunction with feeling ker (happy/excited). Thus, neither of the latter two behaviors has a determinate relationship with the emotion.: 1;. Although there are many other potential sources of indeterminacy in this model (e.g., it is an aggregate model, which should reflect the fact that some individuals hold to particular propositions with more certainty than do some other individuals), these three are the major factors leading to the use of the probabalistic broken arrow. It would be important, in a more detailed model that space will not allow, to distinguish among them. ^ Preliminary analysis of transcripts from interviews on emotion with middle- " class adults in urban upstate New York reveals that many definitions of emotions are of the latter type. For example, for one man, "[Annoyance is] a sort of a buzz that you have which sort of prevents you from really experiencing . things and trying to understand them in some sort of perspective. It's like a narrowing of perspective." And, for another individual, "Boredom to me I guess is having energy and you don't know what to do with it." G. White (unpublished data) presents a framework for examining the emotional discourse of the A'ara people of the Solomon Islands that includes many of these factors. The assessments made by A'ara include (1) moral evaluation * of an action, (2) role structure, (3) social relations, (4) responsibility, and (5) intentionality. The notion of "sadness," then, for example, points to a rule violation with a focus on the "lack of intention or on social ties between Perpetrator and Affected." As seen below, when the Ifaluk stress social ties,. • in such a negative situation, they tend to minimize the moral aspects of their* emotional judgments and frequently use the term nguch (sick and tired/bored) to describe the situation. Comparative contrasts such as these point to the-(l usefulness of an analysis of what White characterizes as "backwards inferenc- ^

GOALS AND UNDERSTANDING IN IFALUK EMOTION THEORY

^ "311"

v-

v :

ing" in emotion-word use and to the importance of such understanding for • . dealing with the problem of translation. .^;'v«;j^:;':,-.y.7\ 8. Diverse chains of inference may be involved with some of the other pfbpb^i-' tional examples. For example, pak (homesickness) often leads, to inaciivity (schema 2) brought on by the desire to sit and think about missing ldyfe^rips" .1 (schema 2). A situation of inactivity often produces nguch %ii0ti^fc0fyf[.;•;.'•, bored) (schema 1). Thus, pak - = = = = = > nguch. As another :exam'pl^ ; rai (panic/fright/surprise) is often caused by sudden, unexpected nbis0,'including" particularly noises that are thought to be those of a spirit (schema f).ff}rW (panic/fright/surprise) is found to be caused by an intentional humah-ag^iit •-/''. (if, for example, a child has secretly made the noise in order •tdtfn^hten''••'• " another), then it is expected that song (justifiable anger) might fpllovVjias intentional frightening would be considered misbehavior (schema:;i);;Thus, rus-==--> song. The Ifaluk have come to expect that one sort of event (e.g., spirit encounters) may sometimes turn out to be an event of another sort (e.g., spirit imitations or misbehavior). In sum, there are a variety of links between and among events, actions, and emotions that may be evoked in understanding sequences of emotion in an individual. :;~: 9. For any ethnographic observation, it is almost always possible to object that the description offered could equally well apply to one's own or other contexts and thus cannot be evidence for cultural difference. These objections nearly always arise, it seems, when the question of differential salience or emphasis in goal structures across cultures is ignored. 10. This type of thinking can sometimes be observed in laypersons whose goal it is to characterize, as simply as possible, the behavior of people whose "otherness" is being emphasized. Ethnotheories about women generated by men in many cultures, for example, often appear to describe women as "singleminded," that is, as motivated by a single (and often unworthy) goal. In general, the singleness of the other's supposed purpose is often a sufficient critique in comparison with the more multifaceted goal structure with which the self is theoretically endowed. Thus, it is important to question the social contexts in which goal descriptions, both academic and lay, are generated. 11. See Hochschild (1979), whose notion of emotion work is useful for modeling this creative aspect of emotional response. 12. I am not suggesting here that the Ifaluk are working with an implicit semantic/ pragmatic split in their thinking about emotional meanings. In the absence of the referential emphasis that exists in the Euroamerican ethnotheory of language (Good & Good 1982; Rosaldo 1982), the Ifaluk are more likely to recognize implicitly that they are engaged in the process" of simultaneously using and constructing their language and their emotional meanings.

References Averili, J. 1974. An analysis of psychophysiological symbolism and its influence on theories of emotion. Journal for the Theory of Social Behavior 4:147-190. Boehm, C. 1980. Exposing to moral self in Montenegro: The use of natural definitions to keep ethnography descriptive. American Ethnologist 7(l):l-26. Cancian, F. 1975. What Are Norms? Cambridge, England: Cambridge University Press. Casagrande, J. B. and K. L. Hale 1967. Semantic relationships in Papago folk definitions. In Southwestern

312

CATHERINE LUTZ

Ethnolinguists, D. Hym.es and W. Bittle, eds. The Hague: Mouton Publishers. Pp. 165-193. . . , . . ' . ,.,..}•„ Colby, B. N. . . . 1975. Cultural grammars. Science 187:913-919. < D'Andrade, R. G. 1976. A propositional analysis of U.S. American beliefs about illness. In Meaning in Anthropology, K. Basso and H. Selby, eds. Albuquerque: University of New Mexico. Press. Pp. 155-180. . Davitz, J. R. ..= 1969. The Language of Emotion. New York: Academic Press. Dumont, J.-P. 1978. The Headman and I: Ambiguity and Ambivalence in the Fieldworking Experience. Austin: University of Texas Press. Good, B. J. and M.-J. D. Good' 1982. Toward a meaning-centered analysis of popular illness categories: 'Frightillness' and 'heart distress' in Iran. In Cultural Conceptions of Mental Health and Therapy, A. Marsella and G. White, eds. Dordrecht, Holland: £>. Reidel Publishing Company. Pp. 141-166. Hochschild, A. 1979. Emotion work, feeling rules, and social structure. American Journal of Sociology 85:551-595. Hutchins, E. 1980. Culture and Inference: A Trobriand Case Study. Cambridge, Mass.: Harvard University Press. Izard, C. E. 1977. Human Emotions. New York: Plenum Press. Lutz, C (1985a). Depression and the translation of emotional worlds. In Culture and Depression: Studies in the Anthropology and Cross-cultural Psychiatry of Affect and Disorder, A. Kleinman and B. Good, eds. Berkeley: University of California Press. Pp. 63-100. (1985b). Ethnopsychoiogy compared to what?: Explaining behavior and -, consciousness among the Ifaluk. In Person, Self, and Experience: Exploring Pacific Ethnopsychologies, G. White and J. Kirkpatrick, eds. Berkeley: University of California Press. Pp. 328-366. 1986. Emotion, thought, and estrangement: Emotion as a cultural category. Cultural Anthropology 1:287-309. Quinn, N. 1982. "Commitment" in American marriage: A cultural analysis. American Ethnologist 9(4):775-798. Riesman, P. ' 1977, Freedom in Fulani Social Life: An Introspective Ethnography. Chicago: University of Chicago Press. (First published in 1974 as Societe et liberie chez les Peul Djelgdbe de Haute-Volta.) Rosaldo, M. Z. 1982. The things we do with words: Ilongot speech acts and speech act theory * in philosophy. Language in Society 11:203-237. Schank, R. and R. Abelson 1977. Scripts, Plans, Goals, and Understanding: An Inquiry into Human Knowledge Structures. Hillsdale, N J . : Lawrence Erlbaum Associates. Winograd, T. \ 1977. Formalisms for knowledge. In Thinking: Readings in Cognitive Science, P. N. Johnson-Laird and P. C. Wason, eds. Cambridge, England: Cambridge „ University Press. Pp. 62-71. ?

13 Ecuadorian illness stones CULTURAL KNOWLEDGE IN NATURAL DISCOURSE 1

Laurie Price

The theme of illness occurs frequently in conversation the world over. Illness stories, set apart from the conversational flow by certain structural features, encode cultural models of causation, extensive situation knowledge (Holland 1985) about appropriate behavior when someone is sick, and a vast amount of cultural knowledge about types of treatments and health specialists. Because the tasks of coping with illness fall heavily on family members and friends of an afflicted individual, there is a general need for access to cultural knowledge about ways to respond to different illnesses. Through hearing illness stories, individuals expand the cultural models they use to think about and interpret illness. During the course of conversation in the highland city of Quito, Ecuadorians often tell one another about health problems that they or others have suffered. Many different cultural models can be discerned in this discourse, but it is situation knowledge of social roles that is most dramatically and exhaustively encoded. Ecuadorian illness stories often transcend their topical focus and express general models of the family, of neighbor and friend relationships, and of social hierarchy in Ecuadorian society. Before discussing these models, this paper briefly summarizes the ethnographic context of illness story-telling and the structure of illness stories. Ethnographic context This analysis is based on one year's field research in a marginal, largely Mestizo neighborhood of Quito, Ecuador. Residents of the barrio, here called Las Gradas, tell numerous stories that can be broadly labeled misfortune tales. In addition to illness narratives, some of these stories concern car accidents, domestic conflicts, and presidential assassination. This discussion refers only to accounts of illness. Data are "extensive stretches of naturally situated talk," to use Michael Agar's phrase (1982:83). Even though I listened to many stories, I cite examples here only from the recorded narratives of 14 individuals discussing 4 cases of illness. As described in Appendix 1, 3 of the afflicted individuals are children; the fourth

314

LAURIE PRICE

is a middle-aged woman. The 14 narrators are all people who played an active role in the illness situation they describe. In addition to hearing conversation about these cases, I observed some of the events and relationships pertinent to each case. Such ethnographic observation promotes more valid analysis of narrative material. For instance, observation reveals aspects of a situation that narrators downplay or omit from their accounts; such omissions often suggest underlying assumptions about shared cultural knowledge on the part of the narrator. A NATURAL FORM OF DISCOURSE

Evidence that illness stories are a natural form of discourse, rather than a task imposed by the investigator, comes from participant observation in the social life of the neighborhood, from the form of the recorded narratives, and from the many references to other stories that informants say they heard from other people. Friends in the barrio told me illness tales before I showed any special interest in them. Once begun, narrators typically continue their stories with little or no prompting. In some of the transcripts, the same information is presented two or three times in the same basic form and even in many of the same phrases. This suggests that certain narrative sections are remembered by the narrator in chunks, perhaps with particular phrasings or key terms attached. Probably, the same chunks had been related before to other listeners. Few of the storytellers refer to being in an interview frame in Tannen's sense (1979). Most compelling of all the evidence for natural occurrence of this discourse is that about half of the 14recorded narratives refer to other illness stories that narrators had previously been told and had remembered. These "stories within stories" demonstrate that not only do people com* municate vital information about illness in story form, but they also remember and apply stories they have heard to situations that arise in their own or others' lives at a later date. For instance, in the narratives about Isabel, the child's mother and grandmother retell stories they had heard" about two other individuals with similar disabilities and their treatments. ILLNESS STORIES IN SOCIAL CONTEXT

Illness stories occur most often in conversation among good friends and among barrio residents and relatives living elsewhere. Women tell the most illness stories, and they tell them mainly to other women in settings of semiprivacy. Some individuals are especially adept at telling this type of story, and such "specialists" tell them more often than others.3 Barrio residents derive a number of cognitive and social benefits from telling and listening to illness stories, although these benefits are generally not consciously sought or recognized. ^ First, conversation about illness has problem-solving value because it transmits useful technical information about such things as home remedies, disease symptoms, health care specialists.4 Even if this information is no%

ECUADORIAN ILLNESS STORIES

•-MS

immediately relevant to the listener's current prpbl^sylt^gii^sjbj^;.^ her fund of cultural knowledge with which to meet futureJUnes^s^i Second, through exposure to the'mwy.rau^..propq^ti^^^i|ji^ > narratives, listeners expand and refine their own theories iabpuV di^asej Such propositions constitute the raw material from which.pepple;M^yb "folk systems of interpretation" (Bohannan 1957). In actual Uinesss!ttia|-' tions, a viable cultural model helps individuals locate prbfel^matib'experiences in a framework of meaning and allays some of ihean^ety ; associated with the situations. .:•,..-:• i./^^-j^iiiia-i^'^^' Third, many illness stories focus attention on the caretaker role of the narrator. The narrator frequently asserts in an implicit way;. "Ijdid |he right thing." This public declaration constitutes a way of negotiating the meaning of the illness events and may be an important source of social validation for the narrator. Finally, it may be hypothesized that sharing illness stories reinforces bonds of mutual support among individuals and intensifies friendships, much as sharing life stories appears to deepen social relationships among middle-class North Americans (Linde this volume). Structure of illness stories The illness story follows rules for internal coherence and can be regarded as a speech act in the sense of a component of a speech genre. It is an unlabeled speech act, though, and Ecuadorians do not as clearly acknowledge it as a distinctive unit in the way that North Americans acknowledge, for example, a therapeutic interview, lesson, or court case as distinctive units (Hymes 1972; Linde this volume). Several features set the illness narrative apart from the rest of the conversation, The beginning of the narrative is anchored in a particular time and place, the orientation, to use Linde's term (this volume). Typically, the orientation describes how the illness started, its earliest symptomology. Sometimes, however, the narrator begins the story further back in time with description of what caused the illness. In a few cases, the narrator's particular role in the events may lead to his or, her anchoring the story at the point at which he or she entered the action. Finally, if the narrator has already told the audience at some other time about an on-going illness, he or she may begin subsequent segments of the story at a time and place appropriate to the new installment. From the beginning time and place, the narrative moves forward chronologically, describing various events along the way that are pertinent to the illness, such as attempts to treat the malady, struggles to get funds to pay for specialist care, and family interactions about the illness. For an illness story to be coherent, the narrator must at some point discuss cause. This usually occurs at or toward the beginning of the story but may recur throughout the narrative, as the unfolding events are associated with changing notions of cause. Like life story segments, many illness narratives

316

LAURIE PRICE

contain a coda or end phrase. Common ones are: "And, so, that is what happened," or "And now, she is as you can see." Appendix 2 points out the structural features identified in one illness narrative used in this .," analysis. Cultural models in illness stories Stories told among Las Gradans acquaint listeners with entire episodes of illness that they have not personally experienced. As evident in the retelling of such stories, the entire episode often gets stored in listeners' own memories of events, in streamlined form. Such discourse plays a dynamic role in the construction and refinement of shared cognitive models pertaining to illness. Second-hand episodes contribute to development of schemas, particularly schemas for situations that do not occur very often. Furthermore, natural discourse about illness transmits factual information about treatment and specialist alternatives, causal notions, situation knowledge about social roles in illness, and a picture of the narrator's feelings about the' events he or she describes - the affective propositions expressed in a narrative (Labov & Fanshel 1977:105). Although notions about the causes of health problems are usually more explicitly stated, situation knowledge (Holland 1985) about appropriate behavior when someone falls ill constitutes the pith of what these narratives communicate and is the primary focus of this analysis. Such situation knowledge is organized within more broadly applied cultural models of the family, of extrafamilial social support, and of social hierarchy and biomedicine in Ecuador. ANALYSIS OF ILLNESS STORIES

Just as a life story reveals the cultural knowledge by which an urban addict survives (Agar 1980), and interviews with North Americans reveal cultural models of marriage (Quinn 1982), so do Ecuadorian illness stories '* contain numerous "traces" of cognitive models that bear on interpretation of illness. Because tacit (understood) knowledge shapes natural discourse to such a large degree, important traces are found not only in what « is said, but also in what is left unsaid. In telling illness stories, narrators take for granted that listeners share many of their assumptions about how the world works. The missing "shared knowledge" must be filled in, if outsiders are to understand the ^ logical connections among utterances, and the cultural models that underEe them. For example, Passage A in Appendix 3 forms part of a narrative by Olivia about her little sister Susana's acute illness at the age of 1. Olivia describes the treatment and says: "We did that and it helped her." Hes account does not mention the fact that, at the time, Olivia was quite young^ (12 years of age) to be taking an active role in the treatment and care of her sibling. If a similar story were related in North America, the speaker*

ECUADORIAN ILLNESS STORIES

317

probably would have included her age as a salient fact. In contrast, among'- ' ~ < marginal Ecuadorians, young girls are typically expected to carry grave - [ ' responsibilities in the care of younger sisters and brothers.'The Ecuadorian narrator might lose the audieiice if he or she emphasized this'feature of the action. By not making a special point about such behavior-on the part of a 12-year-old, such a narrative both reflects and reinforces cultural expectations that such behavior is "normal and natural." ,,-~, What are, from the outsider's standpoint, omissions of information (due to narrators' assumptions about shared cultural knowledge) tell us a great deal about cultural differences between North America and Ecuador; We must also assess parts of the narratives that seem unusual to us as outsiders in the opposite way: that is, why some facets of a story are presented with what seems to us to be overelaborate detail. Often, elaborations indicate a deviation from the standard expectations for role behavior. For instance, one narrator delivers a lengthy, detailed description of events that prevented her father's presence when her brother died. Clearly, this section of her story constitutes an apology for deviation from the Ecuadorian schema for dying. Narrative passages may also be elaborated not because they represent deviation from expectations but because they depict events that are in themselves highly salient to speakers and listeners in highland Ecuador. An example of this is the description of wakes and funerals in Ecuadorian illness stories. The elaborate detail accorded these events no more indicates deviation from the situation schema than does the elaborate detail accorded the dress of the bride and bridal attendants in traditional accounts of weddings in the United States. To understand discourse, we must first determine what the culturally salient actions in a given domain are. Second, we must pay careful attention to narrative context in order to discover how a particular section of highly elaborated discourse relates to the cultural model in question. Illness stories contain many other traces of cultural knowledge beside the inclusion/exclusion of detail and elaboration patterns. Some other traces explicitly used to analyze discourse include repetitions, key words, generalizations, metaphors, false starts, evaluative statements, and hedges (Bohannan 1957; Labov & Fanshel 1977; Lakoff & Johnson 1980; Quinn 1982; Tannen 1979). Many other discourse analysts undoubtedly base their conclusions on some of these linguistic patterns but are less explicit about doing so. An important question to consider is how each of these traces relates to the underlying knowledge structures that are generating the utterances. E. Hutchins (personal communication) puts the question this . way: "By what connections can we claim that generalizations, repetitions, key words, counterexamples, and evaluative statements (and inferences) are traces of knowledge structures in discourse?" A related question concerns the choice of which particular indicators or traces to use in an analysis. With analysis of key words, for example, is it implied that the knowledge structures behind that discourse are primarily organized by way

'• ; j 1 j

318

LAURIE PRICE

of key words? The following analysis emphasizes patterns of omission and elaboration to show how illness stories transmit cultural information about social roles in illness situations. Evaluative statements and a .narrative device I call the counterexample are also shown to be important vehicles ," for ideas about what is inappropriate or unexpected behavior within the Ecuadorian cultural model of illness. .-:.• Situation knowledge, organized within particular cultural models, consists of generalizations about the parts people play, the usual activities these entail, and the role-identities associated with a particular kind of situation (Holland 1985). Without exception, illness stories encode significant cultural knowledge about the role the narrator played in the events of an illness. Whether husband, neighbor, mother, or sister, the storyteller always focuses considerable attention on action in which he or she took part. The following discussion of cultural models and schemas must necessarily be limited to consideration of these knowledge structures as they relate only to the interpretation of illness. Although it is suggested that the cultural models described here are more broadly applied to other kinds of interpretations and other kinds of tasks, this analysis does not encompass those broader applications. CULTURAL MODEL OF THE FAMILY

Responsibilities and activities of mothers occupy a distinct and central place in the Ecuadorian model of the family. The mother is the core of the family, as conceived by marginal Ecuadorians. Mother also constitutes an extremely powerful symbolic entity in cultures that, like Ecuador's, are shaped by Hispanic Catholicism. The symbolic power of the mother image derives from associations with the Virgin Mary, "Mother of God," and with numerous miracles attributed to her - miracles that have detailed histories, concrete locations, and tangible commemorative events associated with them. Since motherhood is modeled after Mary, it symbolically entails that peculiar mixture of ravishing but saintly beauty made somehow more beautiful through suffering the pain of love and death. Motherhood as an image embodies the ideal of love that is superhumanly strong but that nevertheless offers practical help and is gentle and approachable. Motherhood in illness narratives. A striking example of cultural assumptions about mothers' roles is found in the narrative Sra. Maria tells about her crippled daughter, Susana. The assumptions are revealed not only by what she says but also by what she does not say. In her story of the events, Sra. Maria never mentions that for months she daily carried her 6-yearold daughter (in a cast from waist to ankle) down a 200-step flight of public stairs and 4 blocks to the nearest bus stop so the girl could go to physical therapy. This constitutes an extreme example in that the father of the child drives a bus parked near the house every night. During the many months

ECUADORIAN ILLNESS STORIES

•

'•$$?$•':

of this grueling routine, the couple never made arrangements ; ^tof^o$felfSl::-PS of the responsibility for-Susana's transportation to. the fath^s"5?!^i^^iKSrfS''S^/1 pie talking about the child's condition take for granted t h i s | i e ^ y 0 | j ^ ^ ^ ^ l | : S ; ^ fort on the part of the mother; such efforts are. the u n w i a r H ^ ^ ^ ^ 1 ^ ^ ^ / ^ mothers. General statements about mothers and couriterefei^iilli^lfc^^fJS^®^ stories reveal cultural knowledge about the significance of the:mdlh^yi' ; #;^;;^ role. Narrators say, "Only God knows the mother's heart," and ^li^aihf: ;-^:>JK^ of a mother is like that." Explicit general propositions about husbands; ;::v:.y^;; or daughters or other roles or role-identities do not appear in these :illness.: narratives. -.•. .^'^-::'/;;rK'.. .''•'•'" Counterexamples, or narrative descriptions of individuals who are not fulfilling their roles adequately or in expected ways, throw into sharp relief the role expectations attached to mothers in illness situations. Passage B of Appendix 3 describes the behavior of mothers who leave their children in the public hospital and do not participate actively in their care during hospitalization. The emotionally charged tone of the passage should be evaluated in light of knowledge about hospitals in Ecuador, with their inadequate nursing attention and risk of death. Both the tone and the repetition of "She didn't come, she didn't come . . ." reveal not only that the narrator is making a statement about a deviation from her cultural schema for the mother's role but also that she has strong feelings about such a deviation. This passage and others like it reveal feeling and thought as "parallel systems of processing" (D'Andrade 1981; Zajonc 1980). The mother is saying, in effect, "Mothers are supposed to be near their children and take part in their physical care, especially if they're in the hospital, because they will probably die otherwise." The emotions attached to this counterexample can be paraphrased as, "I feel bad about these sick, abandoned children and good about myself for not being a mother like that." Affective propositions are so central to most of these illness narratives that it can be said that if cultural models of social roles drive the narratives, emotional propositions are the fuel that empower them, More counterexamples about motherhood occur in the narrative told by Elsa's neighbor, Mercedes. In fact, Mercedes's account of Elsa's condition is one long counterexample; it thoroughly dramatizes a number of instances in which Elsa failed to meet her responsibilities as mother of a family. Passage C exemplifies counterexample material in Mercedes's illness story. The same facts also appear in accounts of the illness given by other individuals but are not presented to imply personal fault. For instance, Elsa's daughter Rosa mentions the multiple miscarriages as a . possible cause of her mother's illness but does not imply that these could have been avoided. The point here is that counterexamples can easily be constructed out of facts pertaining to an illness, but narrators who present counterexamples usually have an affective point to make. A corollary proposition to the one that the mother is the vital core of family life is that marriages that produce no children are in trouble. In the

320

LAURIE PRICE

Las Gradan model of the family, children not only bring fulfillment to the mother, they are also what "glues" (pegar) a man to his wife arid conjugal home over the long run. Air of this logically fits given'the initial premise that a family unit must have a mother to flourish; children are J what turn a wife into a "real" mother. •' Gender differences, collective responsibility. The cultural model of family revealed in these illness stories shows clear gender differentiation in expectations about decision making, economic arrangements, nursing, and other role activities associated with health problems. Female family members (e.g., wives, mothers, sisters, daughters) are expected to bear f the main burdens of nursing the ill at home and making forays into the , 1 specialist health-care system on behalf of their families. Although the prbtotypic family role for a woman is that of mother, the cultural model prescribes that the burden of caring for the ill be borne collectively by related females within a household compound. (Ethnographic observa, tion confirms this as a general behavioral pattern.) Expectations of collective female responsibility for therapy management and caretaking of >t children are reflected in narratives about 3-year-old Isabel. The child's mother, grandmother, and aunt reside in the same compound. In talking about attempts to cure the child, each of them uses the plural pronoun "we" rather than "I." Context shows that the "we" refers to the three , women. Males make cameo appearances in the accounts but are always clearly identified when they play a part.

)}

Cultural knowledge about male roles. What do accounts of illnesses con'• vey about Ecuadorian folk theory concerning male roles in such situations? The male head of a household may be described as padre de la familia (father of the family) or just as frequently asjefe de la familia s (chief of the family). This is partly an administrative position but alsot . A >' involves financial support of the family unit. All adult males in the household are expected to earn money and contribute some of their earnings to the family. Apart from this expectation, male roles in family life at home are not as clearly defined as are female roles - there is no single * , prototyplc male role identity nor set of usual activities within the home as there are for females. The model accords males the freedom to come •*' and go, with their primary responsibility being to support the family financially. Even though pain is manifestly and prominently part of cultural .,„. expectations about the mother role, male roles within the family are not associated with suffering or pain. It appears that the Ecuadorian cultural model of the family allows, indeed encourages, women to feel fully the anguish of a calamity that befalls a family member but does not prepare ' men either psychologically or socially to acknowledge that kind of anguish."" Illness narratives indicate that men have, or know where to get, cultural^ knowledge about seeking therapy and about home treatments. The nar-^

ECUADORIAN ILLNESS STORIES

•'.321

ratives also indicate, however, that no one expects men to get/involved :; in those activities if women in the family are available to assume the respon. sibility. In Passage D of Appendix 3, Carmen (the mother,of the toddler • : who fell on the public stairs) recounts events at a moment when the child's 5 condition worsened. She mentions Ramiro's reluctance to wake upland ; :,7; help care for the child but does not evaluate his behavior negatively. •'..":;."• Rather, she excuses it by pointing out the hard work that led to the father's fatigue. His refusal to get up does not indicate lack of affection for the child according to his wife's story. She later describes him as desperately wanting to obtain good specialist treatment for Lucia. While at home, however, he sleeps, secure in the knowledge that the mother will do everything she can do to care for the child and in the knowledge that he is not necessarily expected to get actively involved at home. What happens, given the Ecuadorian model of family life, when the mother of a family becomes chronically ill? Clearly, this kind of situation would be likely to generate different interpretations than the same condition suffered by a child or a male member of the family. Among other things, this type of situation can produce serious role strains within the family, which will be reflected in illness narratives. In Case 4, which is summarized in Appendix 1, Fernando takes an active part in managing his wife's therapy. His 17-year-old daughter is too young to take on full responsibility for this, although she does do all the primary caretaking of her invalid mother at home. (Neither she nor her father point to her full-time devotion to her mother's care as anything worthy of praise; such a role for the eldest daughter at home is the unmarked case.) Themes of money and treatment costs pervade Fernando's account of his wife's affliction. He talks at length about the expense of various consultations and treatments, giving precise figures, and also about his personal efforts to obtain funds to finance specialist care. Overall, men's narratives show more concern with the economic dimension of illness situations. In addition, female narrators often make a special point of noting the participation of men in the economics of health care. An example of this is found in Passage E, in which Isabel's mother and grandmother give credit to the child's grandfather, her mother's father, for playing a vital role in financial decision-making about treatment. This and other passages reveal that the cultural model of illness situations prescribes an active role for males in matters of money, especially in negotiation with formal financial institutions. (The primary source of commercial loans for marginal Ecuadorians is at their place of employment. Men are more likely than women to hold structured employment. So, men's greater responsibility for this aspect of coping with illness derives from both economic contingencies and cultural ideals.) Sra. Maria also emphasizes male participation in certain decisions made about surgery for her daughter. She describes her early attempts to get help for the child. The third doctor she consulted told her the child needed

322

LAURIE PRICE

an operation immediately. Her story continues, as noted in Passage F, in which she shows that her husband has the final say about treatment. A few years later, Sra. Maria took her daughter to traditional curers, and * then, following her mother-in-law's advice, she consulted a physician. First, she says that she went to see the doctor. Then she amends this and specifies that she and her husband went to see him. Passage G, like many of the narratives, indicates the special recognition given to a man for playing even a minor part in the management of an illness in the family.vNarratives reinforce the cultural expectation that women automatically assume major responsibility for caretaking and seeking therapy, while the final authority lies with men to make decisions about treatment that involves more than incidental expense. Fernando's story about his wife's problems demonstrates abundant knowledge about certain aspects of the specialist health care systems; the majority of this account centers on his efforts to get effective treatment for his wife. The narrative also reveals a cultural bias in favor of including women, even young ones, in efforts to manage the therapy of a sick family member. On most of his forays to seek specialist care, Fernando took along one or more of his older daughters. His grown sons who live at home did not participate actively in the illness situation. Their behavior elicits no special criticism, just as the daughter's caretaking behavior elicits no particular praise; both the sons and the daughters are conforming to the cultural expectations in a case of this sort. Folk "physics" of health behavior. Narratives reveal the cultural notion that a special impetus is needed to "get moving" on health problems. A key concept in the "physics" of motivation is embodied in the phrase hacer empeiio. This phrase means "to get someone (or oneself) moving to accomplish something." Even though people take for granted tremendous ambivalence about acknowledging a health problem as serious and about; "moving" into action to seek specialist care, they also expect family members to monitor each other's health and push for specialist consultations. In Passage H, Sra. Maria credits her mother-in-law with pushing her (putting her in motion) to see another surgeon about her child's in- * firmity. Isabel's mother and aunt discuss their efforts to get a cousin "moving" toward treatment. Illness narratives suggest that the typical "motivator" is a member of one's family network, but not necessarily of one's immediate domestic group. In addition, narratives reveal that pushing «. for action on a health problem does not necessarily entail expectations of follow-up help with the mechanics of finding or financing specialist care. MODEL OF EXTRAFAM1LIAL SOCIAL SUPPORT

Ecuadorians hold distinctly different notions concerning roles and social " support within the family and roles and social support characteristic of extrafamilial relations. Many illness stories implicitly address this ques- **

ECUADORIAN ILLNESS STORIES

'•323,

tion: "How do you mobilize other people to help with an illness in the family when they have expertise or access to assistance that you doft't have?" A counterbalancing and equally important component of this model consists of propositions to the effect that help available outside the family circle is quite limited and should not be depended on except in certain rare circumstances. -••• , r:''^.:;!v';>;v ; ' The proposition that only family is expected to help constitutes tacit knowledge and is expressed when illness story narrators simply omit comment on help that is not forthcoming from friends and neighbors. However, when asked about social support, many Las Gradans do not hesitate to state the proposition directly. They say, "neighbors don't help each other here"; "the only ones you can really depend on are family"; and "nobody (neighbors) really knows each others' problems here." This proposition is further embodied in other cultural representations that are part of the Las Gradas verbal tradition, in particular, proverbs. Proverbs such as those cited below reinforce the cultural model of extrafamilial social support by reminding people to keep their distance from others, to think in terms of reciprocity, and to beware of giving or receiving too much kindness, especially outside of a standard reciprocal exchange context. "Too many visits make you less welcome." "Give the hand, get grabbed by the elbow." "Every burro with its own saddle." "Today for you, tomorrow for me." "Hands that give, receive." "If you give bread to a strange dog, you'll lose the bread and lose the dog." Explanations of these proverbs by Las Gradans indicate that people interpret them primarily as advice to use caution in conducting relationships outside the family. Ethnographic observation shows that neighbors and friends do, in fact, help one another considerably in coping with various problems. But this fact in no way detracts from the authenticity of the cultural propositions that they do not, or the admonition that one should not depend on them to do so. Another counterexample reveals the proposition that only family will help. A neighbor of the toddler's family describes how doctors regard her supportive behavior in that family's crisis as quite unusual. She describes events at two institutions where she accompanies the child's mother to seek treatment. At both locations, staff members ask her: "Who are you? Are you the child's mother? If you aren't, why are you so concerned and involved in this?" Here, the narrator presents her own behavior as a counterexample. By elaborating on the surprise of the medical personnel, the narrative shows that the neighbor interprets her own behavior as an exception to the standard cultural model. Even while the cultural model of extrafamilial social relations leads

324

LAURIE PRICE

marginal Ecuadorians to limit expectations of aid from nonrelatives, it also embodies knowledge about culturally appropriate strategies for developing the kind of extrafamilial relationships that most likely will result in instrumental and other social support. People obtain much of this knowledge from descriptions other people give of how they managed to build these connections. Isabel's family recount how they pursued the ideal strategy for marginal Ecuadorians faced with serious illness. They found benefactors. As described in the narrative, the primary benefactor was located on the advice of a friend and a salesclerk. Isabel's mother and grandmother credit this female doctor (ctoctora) with saving the child's life and curing her infirmity. CULTURAL MODEL OF SOCIAL HIERARCHY AND BIOMEDICINE

Many different terms exist by which people in highland Ecuador place one another within a distinct social hierarchy. As Stark (1981) demonstrates, how people use these terms in one highland town varies systematically according to the social category in which the speaker fits. However, social classifications are organized in a very clear-cut way on the basis of palanca (literally, "lever," or "handle"). Degrees of palanca ranging from "no access to goods and services" to "high access to goods and services" are interpreted as the distinctive features that differentiated people from one another within the social hierarchy. As such, they were used as implicit markers, rather than as explicit labels, of social categories (Stark 1981:391). Las Gradans also conceptualize the social world as a place in which various categories of people have greater or lesser amounts of palanca. This term comes up in a number of the illness stories in connection with biomedical institutions. For instance, a friend of Isabel's family talks about their attempts to get assistance with medical expenses: "[The Patronato] is an organization that helps poor children, sick children, and like that. But you know the bureaucracy, no? If we have no 'leverage' (palancasj, they don't help us," The cultural model of social hierarchy found in Las Gradas conceptualizes people as possessing different amounts of palanca. It also divides them in a dichotomous way according to wealth. This divi-v sion is represented in the terms pobres (poor people) and ricos (rich people). Residence in certain barrios of Quito almost automatically certifies one as a pobre or a rico. It is tacitly understood that biomedical institutions belong to and are staffed primarily by people belonging to the latter,, category. Models of wealth and palanca together form the cognitive landscape that pervades schemas of biomedical health care. Physicians in illness stories. In their illness tales, Isabel's family describes * a greedy, unethical doctor who lied to the family during the early stages of their attempts to get treatment for the child. Their descriptions of this physician were constrained and understated, partly because the narrators^

ECUADORIAN ILLNESS STORIES

325

were uncertain about where I stood with relation to doctors and other professionals, but more because lack of ethics is nothing remarkable in a doctor from their standpoint. A narrative about a different case shows the ramifications of the proposition that physicians are, as a rule, hot to be trusted. The mother of the toddler who suffered fatal head injuries says, "I imagine, frankly, that the doctors gave her a poisonous injection so she would die right away." In effect, this narrative proposes, that it is not out of the question to suspect doctors of hastening the death of an ill child. • ••• "•'"-v?^v^^:i^fe.In contrast, narrators from Isabel's family describe the female doctora who befriended them in glowing and elaborate detail. She is transformed into another counterexample. By emphasizing and elaborating the doctora as a character in their stories, the narrators again reveal that expected behavior on the part of physicians does not include real affection for patients, genuine sympathy, "vows of poverty," or being willing to wait for payment. Their effusively grateful language communicates knowledge about how they expect the typical doctor to act. Because the doctora's behavior constitutes a surprising deviation from the cultural model of social interactions with professionals, Isabel's family attempt to explain this exception to the rule. They do this through references to person theory (Holland 1985). Passage I expresses their proposition that the doctora is kind and compassionate toward poor children because she is pretty and an unmarried woman with no children of her own. Many other passages in the accounts of Isabel's illness also reveal person theory at work, including direct quotes of the doctora's explaining her own unusual behavior as due to "conscience" and other personal factors. How do counterexamples relate to knowledge structures concerning social roles in illness? First, narrators who present counterexamples are talking about "exceptions to the rule" that are usually highly salient to them. The exception matters because the rule matters. For instance, counterexamples concerning mothers and doctors indicate that these roles are pivotal ones in the domain of illness behavior. Second, this material throws into sharp relief the cultural model of social roles by capturing the essence of what a mother or doctor does not act like, in the usual course of events. By definition, counterexamples pinpoint the "difference that makes a difference" (Bateson 1972; Hutchins, personal communication) from the emic point of view. Thus, from the two counterexamples discussed here, we find that a standard entailment of the mother role is that, "A mother tends her children, whatever happens." A standard entailment of doctor can be embodied in this proposition: "A doctor expects his fee immediately, whatever happens." Nurses in illness stories. The Ecuadorian model of biomedical institutions clearly distinguishes doctors from nurses in terms of their interac-

326

LAURIE PRICE

tional style with pobres. Before specific cultural propositions about nurses are identified, it should be noted that enfermeras (nurses) is a category that marginal Ecuadorians use to refer to practically ail the nonphysician practitioner roles women play in biomedical treatment systems; in other words, Las Gradans do not generally distinguish between the Ecuadorian equivalents of registered nurses, licensed practical nurses, and nurses'aides. The differences in licensing and education between these practitioner.types are not salient to most Las Gradans. • u;;, •;,•;, Medical doctors are thought of as typically greedy and not compassionate, but at least they can be approached by pobres. As Sra. Maria says, "Oh the doctors, yeah, they're nice enough. They answer when they are greeted." While you may not get good treatment from a doctor, at least they are 'civil.'" Marginal Ecuadorians put a great deal of emphasis on respectful interaction styles, particularly on greeting and leave-taking rituals. In contrast to physicians, nurses (particularly nurses who work in hospitals) are widely described in natural discourse as groseras. This term can be roughly defined as "rude, uncivil, brutal." One narrator says: There are children who can't move [in the hospital]. They can't move and the nurses come and throw their food on the bedside table, [saying] "If you can eat, eat; if not, then don't eat." That's how it is. That's how this type of Seiiorita is - they have no heart, we could say a word. There was one child that couldn't eat, bedridden like this, they had him with a weight attached to his little leg. I fed him when I came. But, the nurse said to him: "There it is. You have to eat it. What do you want me to do, put it in your mouth? If you can eat, go ahead and take it, go on, eat," she said. Two things should be pointed out about the proposition that nurses are groseras. First, not all illness story narrators communicate that proposition, although they would certainly all recognize it as a common idea in their neighborhood. Second, the cognitive model of social hierarchy'"'' as applied to biomedical systems in Ecuador dramatically differentiates hospitals (and hospital nurses) from private clinics (and clinic nurses, by extension^ Thus, cultural propositions about nurses may not have as much = to do with categorizing the type of person who works as a nurse as with categorizing the typical behavior of nurses who work within a certain type of institution. Clinics versus hospitals. Private clinics in Ecuador are like small-scale hospitals, with their own surgical facilities and patient accommodations. Climes are expensive and are associated with ricos (rich people). Hospitals, on the other hand, are public and are supposed to charge nothing, by a. recent act of national law. Hospitals are associated with pobres and with* inadequate, uncaring treatment. Passage J is profoundly shaped by cultural knowledge about how to maneuver as a "marginal" in urban Ecuador.^ It reveals a proposition that doctors and nurses, with few exceptions,

ECUADORIAN ILLNESS STORIES

327

should not be trusted to give adequate or compassionate care if pneis poor. A concrete expression of this proposition is found.in references to "tipping" hospital personnel. The reference to "tips" in Passage J reveals, the assumptions on the part of barrio residents that hospital staff, expect to be paid directly for giving good service to patients and, that inadequate care will be given to patients who fail to observe this custom, v,A :;;:*;. Schemas of therapy management in getting biomedical treatment are.. profoundly shaped by the distinction people make between elirdcs.and hospitals and encompass a behavioral routine that guides. Las, Gr.adan.jn-r: terpretation of efforts to get biomedical care. It is considered important that therapy managers at least attempt to get private clinic treatment before resorting to hospitalization. Only after inquiring about the expense of clinic treatment and finding it too expensive do they turn to hospitals, Not only is this a behavioral routine, but it also constitutes a highly salient element in illness narrative. Many of the 14 narrators gave attention to the clinicfirst, hospital-second pattern. As explained in a later section, when narrators describe their efforts to obtain treatment at private clinics first, they are making a social and affective statement that they "did the right thing." A cultural model of social hierarchy informs schemas concerning biomedical treatment among marginal Ecuadorians, These 14 illness narratives show that biomedical practitioners are associated with higher social status; a distinction is made between doctors and nurses based on interpersonal interaction styles; private clinic and public hospital treatment facilities and personnel are regarded as dramatically different, partly on the basis of social class factors; and, on the whole, their cultural model of biomedicine guides marginal Ecuadorians to be wary and mistrustful, unless they manage to develop a personal relationship with someone who has palanca in the biomedical system. As the analysis presented in this section shows, illness tales communicate extensive cultural knowledge about social roles in illness situations, including expectations of male and female kin, neighbors and friends, and physicians and nurses. Communication of such information through conversation and narrative both augments and reinforces shared cognitive models used to interpret illness situations. These models constitute a subset of cultural models of social relations in general, as held by marginal urbanites in the highlands of Ecuador. However, illness narratives do more than encode situation knowledge. The following sections show how narrative communication also acts as a dynamic vehicle for subtle negotiations about illness causation and about feelings associated with illness situations. Theories of causation in illness stories Illness narratives reveal much greater fascination with the causes of an affliction than with its symptoms or diagnosis; stories present numerous theories about what caused this or that health problem. A preference for

328

LAURIE PRICE

problems of cause has been documented in other cultures as well. For instance, North Americans have been shown to think more about "the consequences and preconditions of illnesses" than the attributes or features that define disease (D'Andrade 1976). Ecuadorians take seriously the task of identifying causes of a sickness, especially in cases of serious or chronic affliction. During informal discussions, people evaluate the truth value" of the causal propositions of others. Cultural consensus emerges as participants in a conversation work out the relationships between different beliefs and so refine their explanatory models. Narratives put forth three kinds of instrumental, or immediate cause. These include physical factors (e.g., temperature), emotional factors (e.g.;, anger), and behavioral factors (e.g., "She was playing on the stairs.'^ (Glick 1967). Some narratives also suggest "ultimate" causes, notions that involve some degree of moral judgment of the ill person or family. MULTIPLE NOTIONS OF CAUSE

It is striking that many narrators put forward a number of different causes for any given illness. Sra. Elsa's daughter, for example, attributes her mother's disease to at least six distinctly different factors: too many miscarriages, anger (iras) at children and husband, anger held inside, her son's death, drinking hot medicinal teas prescribed by a curer but not staving in bed afterward, and doing laundry too soon after giving birth to her last child. The daughter's narrative also cites other causes for the disease that have been identified by various health-care specialists and other family members. Sra. Elsa and her husbancTrefer to many of the same causal notions independently of their daughter and each other. The family has obviously discussed the problem of causation, and family members are aware of each others' theories, although no consensus has been reached. Intrafamiiial verbal communication about illness is another way in which individuals refine cultural models of illness. Some narratives demonstrate not only multiple notions of cause, but' also an effort to judge the truth of various causal propositions, particularly those promoted by other people. Susana's sister first states that the child's hip malformation was caused by her mother's work as a weaver when preg- nant with Susana. Later in her narrative, she returns to the question of cause, presenting a causal notion she has heard from her mother and rejecting it. In Passage K, we see an individual striving for logical consistency between various body notions and propositions about illness causation. „ Several narrators take such a logical and "constructive" approach to causal theories bearing on illness. In telling illness stories, some individuals are also striving to clarify relationships between different causal notions. Moral cause and situated social purposes. Relatively few of the 14 recorded narratives express the proposition that an illness or accident was due to someone's immoral behavior. When these propositions are expressed^

ECUADORIAN ILLNESS STORIES

329.

it is usually by a speaker who is not related to the sufferer. A neighbor of Carmen and Ramiro's attributes their child's accident and subsequent; death to moral shortcomings of family members. The father drinks and abandoned his wife temporarily; the mother does not pray enough; The narrator notes that the child's grandmother had actual caretaking respori-; sibility for the toddler when she fell and yet does not place any blame; on this person for negligence. What might seem to us as North Americans, to be guilt is not evaluated that way by the neighbor. She is operating with a different set of assumptions about family role responsibilities and about what kinds of actions bring punishment from God and whatjkmds do not.

••"'""."•'":•">',

A neighbor of Sra. Elsa's traces her illness to many different causes; a number of which consist of irresponsible or immoral behavior. The neighbor, Mercedes, frames her most extreme proposition about moral shortcomings in a way that allows her to get the idea across but dissociate herself from proposing it: They say that when an illness can't be cured by doctors, it's a thing caused by the devil. . . . Dice que, or "they say," is an oft-occurring linguistic device in these stories; it functions to allow the speaker to say something without taking a stand on the truth value of the statement. When propositions about moral causation are expressed in stories, they reveal a great deal about Ecuadorian models of proper role behavior. The extent to which these notions are openly asserted depends on the narrator's social relationship to the person who is ill and on the narrator's purposes in conversation in a given social context. However, whether narrators openly express such notions, they probably often have these in mind as interpretational possibilities.5 Thus, social context factors should be taken into account in assessing illness beliefs as a distinctive body of cultural knowledge. How would the system of beliefs about illness that can be distilled from analysis of illness stories compare with the sets of interrelated propositions about illness resulting from a more formal elicitation approach in the same population? This question cannot be answered here since a formal elicitation of illness beliefs was not carried out in Ecuador. However, it seems likely that many illness narratives would reveal more causal propositions of the moral type than would an elicitation study. People tend to think of and talk about moral causes in the context of particular, on-going social relationships. Moral causation probably most often comes to mind in atypical cases of illness - very serious or chronic afflictions. Formal elicitation procedures, on the other hand, may access beliefs only on a general and rather abstract level: Respondents focus on the unmarked case that is largely detached from knowledge of specific people and is irrelevant to the normal social purposes of the respondent in talking about illness.

330

LAURIE PRICE

The point here is that cultural knowledge can best be conceptualized as a system of ideas that people use according to their needs and purposes, rather than as a deterministic grid that directs their thinking. The next, and final, section considers natural discourse from the standpoint of people's social and emotional needs in conversation about illness.. Not only do narratives communicate situation knowledge and causal notions, but they also dramatically communicate the narrators' feelings and purposes in the situations they describe. Communication of feeling in illness narrative

'-

Affect powerfully influences how people apply cultural knowledge to cognitive tasks, as Hutchins shows (this volume). His analysis identifies the affective (in this case, subconscious) connections that lead a Trobriand villager to use a particular myth to explain a seemingly unrelated phenomenon of malevolent spirit visits. It is not surprising that affect also influences the task of formulating interpretations of illness and that many illness narratives make important statements at the affective level. Most of the recorded narratives analyzed here express feeling in a powerful way, while ostensibly only describing events. The pattern of using narrative to communicate strong feelings indirectly has been documented in other cultures and conversational genres, most dramatically by Labov and Fanshel (1977) in a study of North American psychotherapeutic discourse. In a classic example, they show how a client expresses angry rejection of a therapist's request by telling a story about her aunt. The expanded meaning of the client's story was inferred partly from the authors' knowledge of the client's life (revealed in other therapy sessions) and from such physical factors as the speaker's intonation, pauses, repetitions, and movements. The affective propositions in these Ecuadorian illness narratives have also been inferred partially through contextual information and nonverbal cues. "1 DID THE RIGHT T H I N G . "

Many of these illness stories dramatically encode the affective proposi- * tion that "I did the right thing" (i.e., "I am a good mother/daughter/ neighbor/husband"). The more caretaking responsibility a narrator has for the afflicted person, the more pronounced this affective message tends to be. The more distant the teller is, socially, from the events and the in-. dividuals involved in them, the less pronounced this affective message is. Logically, the "I did therightthing" proposition appears in its most distilled and dramatic form in stories mothers tell about their children's illnesses. The Ecuadorian schema associated with "mother" very clearly incorporates an affective component. Just as the term hot fuses together a represents^ tion of the "factual external condition" and how we feel about that (D'Andrade 1981), so the term mother among marginal Ecuadorians^.

ECUADORIAN ILLNESS STORIES

331

automatically entails a large potential for pain. People are aware of-this component of the role and even point it out with general statements.^Part of this cultural knowledge is directly experiential.'The personal-experiences',' however, are shaped and reinforced by cultural representations, particularly religious representations. :.;..'.•••' '.n'::::c:yiM^Sj<:f( ; A mother describes her second round of struggle to get treatment for her child (Case 1). The public hospital to which Sra. Carmen wants the child admitted tries to get her to take the girl to another hospital across town. In Passage L, the woman asserts implicitly that she .did all she possibly could to ensure the child's recovery by struggling to keep her in what she thought was the safest hospital available. The samesmother presents a second variation of the proposition "I did the right thing" when she describes her resumption of normal life after a long period of incapacitating grief. First, she recounts the events in which a female acquaintance advises her not to cry any more. She describes how friends urge her to moderate her grief, reminding her that the child is in heaven and is watching over and praying for her mother and the rest of the family. Even more central to Sra. Carmen's narrative is description of a dream she had in which her deceased daughter appeared as an angel. In the dream, the child asks her mother not to grieve any more, because the mother's tears are keeping the child from entering a beautiful paradise. Both sections of narrative implicitly assert that the narrator is "doing the right thing" by overcoming depression about her child's death and going on with life. . In his story about his wife's disease (Case 4), Fernando communicates two major affective propositions. The familiar "I'm doing the right thing" assertion is embodied in his repeated and detailed accounts of the many hospitals, clinics, doctors, and religious healing specialists he has taken his wife to see for diagnosis and treatment. The second proposition is the less common, "This illness (ill person) makes me angry and frustrated!" Fernando communicates this proposition through detailed description of how his wife accuses him of being unfaithful, makes unreasonable demands on his time, and does not "really try" to get well (see Passage M). Another account of Elsa's illness, by her caretaker daughter Rosa, contains the complete and spontaneously included narrative of how Rosa's brother died from cancer. As part of her narrative, she recounts her brother's dying words: My aunt told me, "Your brother said this to me. He told me to tell you that he is going to take care of you, to keep watch over you, in order that you go on caring for her." Because I am the only one [of us in the family] that takes care of my mother, that takes care of her, that has the strength to take care of her. The affective point of this segment of narrative can be paraphrased: "I am struggling, and afraid of this burden, but I am proud of the strength I have shown so far." Of all the illness situations, that of a disabling chronic

332

LAURIE PRICE

disease for which many treatments have been tried and failed provokes the most thought and feeling. Just as the situation is complex and emotionally charged, so are the stories told about it. / WHY SAY IT WITH NARRATIVE?

•

-r r

"- • •'- •• -• -"'- ••-• .= .-',.

People often express affective propositions implicitly, rather than explicitly, precisely because making these assertions directly in conversations violates rules of social interaction. Because illness experiences that are worth talking about tend to be highly emotionally charged, feelings may be what narrators most want to communicate to their listeners. Affective propositions that are prominent in many Ecuadorian illness stories can be paraphrased as follows: "I did the right thing"; "My part in these events was important"; "I care deeply about this (ill) person and am upset by his or her suffering"; and "This situation makes me very angry and frus- . trated." One other affective proposition that many illness narratives clearly express concerns specialist health care, particularly biomedical health care. This emotional theme can be paraphrased along the lines of; "We resent being deceived and treated disrespectfully by medical personnel without compassion; we feel bad being so powerless." Since Ecuador offers no recourse to victims of medical malpractice, illness stories constitute one of the few ways to air grievances and warn others of especially hazardous health-care facilities. Although documentation of this pattern awaits further research, illness stories suggest that Ecuadorians prefer to exchange information about emotions by means other than emotion words. The recorded narratives analyzed here contain very few direct statements about feelings (statements of the sort: "I was very sad about that"; or "They have been very anxious and worried"). We can imagine that in the United States someone telling a story like that of Elsa's daughter Rosa would include direct statements about emotion states, such as "I feel so frustrated, but determined, in this situation." In Ecuador, however, the narrative description of events im-* plies the emotions of the narrator, and native listeners are enculturated to be sensitive to this dimension of the communication. In addition, some social negotiation about the interpretation of events may occur when the * story is told. Often, this negotiation is itself implicit rather than explicit, as, for instance, when a listener responds to a narrative about a child's accident with the comment, "those public stairs are certainly a menace to children," or with a personal story about medical mistreatment. Discussion As this analysis of 14 recorded narratives shows, Ecuadorian illness storiesencode a great deal of cultural knowledge; some of this knowledge is ex^ pressed directly, but much of it is expressed implicitly. In the task of interpreting and talking about illness, Las Gradans draw on cultural modelsi

ECUADORIAN ILLNESS STORIES

3,33

of the family and of extrafamilial social support, a culturaTmodel of.social'\^-^--'li hierarchy (with associated schemas of biomedical institution's) ^ahii -:a'.set;; / • of interrelated notions about illness causation. Narratives show^ of the cultural knowledge Las Gradans bring to'bw'ih;^fei^riB^4^^,^"'-::^:^;J^ ness episodes is situation knowledge. References to perfon';kn£wi«^^ values and attitudes, general character,'person'"typ^s'^are''^are^ elaborate personality trait model to whichNorth Americansi often re§drV: when interpreting behavior does not appear to be part of the explanatofy tool kit Ecuadorians use in talking about illness.6 •\-\\\i*l^£i!0]/J;-$-J' Two kinds of roles are thrown into sharp relief in illness narratives; roles played by close family members and roles played out in interactions between the poor and biomedical personnel. Some family roles, such as that of mother, are seen in these stories to entail many obligatory activities in illness situations; other roles allow for more leeway and personal choice, Illness sets in motion superhuman efforts by mothers that are, typically, taken for granted. It may also set in motion superhuman efforts by men; but, if so, these efforts will be highlighted in the narratives. The cultural model of family also specifies the feelings that are predictably and acceptably attached to each particular role. For instance, it is accepted that male relatives may feel and express resentment about the demands of being a caretaker for someone who is ill, whereas women, typically, may not. Expectations about family roles in illness situations, as reflected in illness discourse, can be regarded as a subset of a broader model of roles in everyday family interactions. Likewise, much of the cultural knowledge encoded concerning interactions with medical personnel and institutions is also used to understand encounters with other nonmedical professionals and institutions. Much of the social role knowledge discussed here is not restricted to interpretation of only illness situations but should also show up in narratives about other kinds of situations. On the other hand, the same cultural models may be used somewhat differently in accomplishing other "tasks," for instance advice-giving or actual treatment-choice. Hearing illness stories can influence knowledge structures in several ways. Individuals gain information about entire illness episodes without having personally experienced those events. Such secondhand episodes are integrated with those the person has directly experienced and become part of his or her current scripts concerning illness. Since illness situations are quite variable, no one is likely to experience enough episodes firsthand to develop robust scripts. Verbal transmission of the outlines of other episodes also affects development of general cultural models. Because illness stories occur in the context of conversational exchange, they contribute to "socially generated knowledge" (Clement 1982). Cultural models are carried by individuals, but they are partly constructed and refined through conversation and reflection. Nowhere is this more clearly seen than in the discussions of causation in illness narratives. As demonstrated here, there is a fascination with causal propositions, and narrators not

AM

a T?«^

334

LAURIE PRICE

infrequently turn their attention (and that of listeners) to the truth value of various propositions. , :/ ,:-, ... Finally, it is argued here that many illness narratives have an important affective dimension that must be taken into account in analysis of this discourse. Propositions such as "I did the right thing" are often what people most emphatically communicate in telling an illness story. Ecuadorians rely on narrative communication more than do North Americans to express feelings they are not allowed to express directly in conversation. Furthermore, feelings as well as thoughts and accounts of particular events are socially negotiated through exchanging stories and assessing responses to them. To sum up, the illness story as told among this population of marginal urban Ecuadorians may be regarded as a culturally identified (although unlabeled) way of talking about illness, with certain structural features that set it apart from ordinary conversation. This type of discourse encodes multiple cultural models, most importantly, situation knowledge about social roles in illness. The discourse also implicitly encodes affective propositions made by narrators about the events and provides a vehicle for social negotiation about thoughts and feelings. Verbal communication of this type supplies both raw material and a forum for the refinement of cultural knowledge bearing on illness. For all of these reasons, the verbal traditions concerning illness that exist in Las Gradas are important in the transmission and generation of cultural models.

ECUADORIAN ILLNESS STORIES

335

Appendix 1 ECUADORIAN ILLNESS STORIES - CASES AND .NARRATORS rr^fp.V4:

1. Lucia - toddler who suffered head injuries in falling on the public stairs in front of her house; taken to hospital, released; condition deteriorated; taken back to different hospital, died there. Narrators - mother, female neighbor, another female neighbor. ' --. 2. Isabel - 3-year-old born with foot deformity; treated at private clinicforlong • period of time with no improvement; ultimately received operation in a different private clinic thanks to intervention of a doctor who befriended the family; considerable community involvement in helping this family. Narrators - mother, aunt, grandmother, male neighbor, female community leader. 3. Stasaiia - 6-year-old who suffered severe hip malformation, starting at the age of about 1 year (possibly due to polio); many different health care specialists consulted and therapies tried to bring about cure; ultimately received operations on both legs in the public children's hospital, then physical therapy, then spiritual curing after physical therapy proved less than totally effective. Narrators - mother, child's older sister. 4. Sra. Elsa - middle-aged woman with degenerative nervous system disorder of about 5 years' duration; mother of nine children, youngest age 2; many different health care specialists consulted, diagnoses made, and therapies tried; family under great strain in trying to cope with the illness and the recent death of the oldest son in the family. Narrators - husband, caretaker daughter, female neighbor, and Sra. Elsa.

336

LAURIE PRICE

Appendix 2 STRUCTURAL FEATURES OF A TYPICAL ILLNESS NARRATIVE

(Case I, Mother Speaking) Initial events of illness [Orientation] Causation Chronology of events [Some of these segments may be "explanations," in Linda's sense (this volume), but they occur in chronological order within the narrative.]

Coda

Susana walked normally at 1 year of age. Then she got an infection. We took her to a clinic. At the clinic, they gave her IV and four injections. We believe the injections created little "pellets" or balls in her hips. From then on (after the injections), she began to limp, sway from side to side like a duck. When she was 2, I took her to a doctor in " Ambato; he said we had to operate immediately. My husband raised objections and there was no money. Then, I consulted a bone specialist in Latacunga. He said there was no reason to operate, that she would be an invalid all her life. That caused me so much pain, I didn't have her examined by anyone else. Then we came here to Quito to live. My mother-inlaw was the one who "got me going" to go consult Dr. G. in the hospital to see if he could operate. Before going to see him, I took her to Conocoto for treatment from a "bone-setter"; but that took too much time and I couldn't keep doing it; besides she wasn't getting better. At this time, she wasn't in pain. Then I [correction! we, my husband and I, went to see the doctor; he gave us hope and : >; said "Bring her in." The doctor operated on the first leg. In physical therapy, the doctor said it wasn't so good. I massaged the bone here at home with a remedy I know, and the bone began to move bit by bit into place. Then they operated on the other one. The outcome still isn't known. After 6 weeks, they will look and see how if is. For the first one, they called me after 4 weeks to see how it was; it was 3 months before they took the cast off. ._ I believe it will be 3 months for this one, 'i_ too. 1 am praying to God that it will all come out OK, whatever happens after that. '"^

;

ECUADORIAN ILLNESS STORIES

•5337.'y

V:.-]

[Questions by investigator lead to further n/arraty&Mgfi&ts^ •'Interview w ith'sbcjaf^IcCT^Ii^^^S^:^'^^ < '•'/•:r.,-•••-. • • concerning possibility^-ofjeduc^^ '..:. •.,•*;.•:.-;-:-•. -'*-•,-. - •'^ledies. "sed. and >care p f ^ u s ^ ^ •

- . --•

• past.';;

.-A.,-- -•-.;_

^•.•••':)^:^M^^-^'^'••^^^

: Stories of two. children in the hpspjtal,;Jhe jj'-r' •'•'• ::';"':••'; . .nurses' interactions with them, and;her:iQ#n:;:>''7'V. ':^:u^-': Final coda

I hope that God's will is to cure her. There is no {other] help, nothing to do but bear the ;, pain of a mother.

338

Appendix 3

LAURIE PRICE v

PASSAGES FROM ECUADORIAN ILLNESS STORIES

A. Narrator: Olivia, older sister of Susana, ehild with hip malformation (Case 3). "The neighbors said she had eaten something, maybe, something that stayed in her intestines and couldn't get out. So they said we had to give her an internal bath with, camomile, lemon [and glycerine] in order to clean everything out of there. We did that and it helped her. Then we put grape leaves on her stomach with almond oil, you see, curing her on the outside. We put that on her and she got better. . . ." B. Narrator: Maria, mother of Susana, child with hip malformation (Case 3). "I have seen mothers go to leave their children in the hospital there. They [the children] are forgotten there and the mothers don't go to see them again. One Senora left a little boy about 2-years-old. He was there, he was there 15 days. They had already operated on his little leg, and his mother didn't come, she didn't come. I don't understand that. She didn't come. I don't know how it turned out, since my daughter was discharged. I don't understand it. {Pause.} The poor little thing just stayed right there. I didn't know if she would come get him or not. Another Senora from Santo Domingo did the same thing." C. Narrator: Mercedes, neighbor of Sra. Elsa (Case 4). "She went to events like soccer games, outside [al aire] almost unclothed. To dances, in that air, can you believe it? And, on the second or third day after miscarrying [the same thing]. . . . One should take care of oneself, and the Senora, note that in one month it's said she had three miscarriages, in one month. That is, neither she nor her husband were taking care [of her health] to conceive again right away. . . . [Elsa does not take her medicine] because she doesn't want to. She's irresponsible. Doesn't it seem so to you? If she were a responsible person, she would try to get better, strive to overcome it. The way I see it, overcoming things in order to keep the home together is the essence of the "mother." And she doesn't like any of that, no. So there it is. And that's why her family doesn't love her." D. Narrator: Carmen, mother of toddler who suffered head injuries in a fall (Case 1). "My daughter started to have convulsions, to shake. I saw it and tried to wake my husband up because he was sleeping. I told him that the child was getting worse. He had gotten up at dawn and worked until late at night. So I told him, 'Look, the child's getting worse.' He didn't answer me. Then he said, 'Let me sleep.'" E. Narrators: mother and grandmother of Isabel, child with foot deformity (Case 4). Mother: Then we came to our senses about my child, who was crippled in the same way. . . . She wanted to talk and she couldn't and she cried. So in view of that, we took the risk of trying to find someone who would"''~ operate. Grandmother: But, on the other hand, my husband said: 'My dear daughter,' he said, 'and the money, the money? Where are we going to get it? Where "%

ECUADORIAN ILLNESS STORIES

•y^y^:-.'-

are we going to get it from?' And he applied [for a Ibanj ' l ^ ^ ^ ^ f ^ ^ ^ V r - ; ' able to do it, because he works at the utility company^lBufct^li^ refused.lt looked like everything had gotten messed tip/'Hndjf^^^ obtain the money, we just couldn't. We had no place i $ g # i t ^ ! ^ F. Narrator: Maria, mother of Susana, child with hip malfe-rmation^Ca^ • "But, my husband told me that it couldn't be that, all of a suddenrða^-;^;^^.^ she had to be treated, to be operated on. And, at that time, there was no money.::;:;;; T 4 ^ •;::So she stayed like that. . . ." '••''-'' •
340

LAURIE PRICE

put her in the hospital. . . . She will for sure get better if we put her in the clinic, keep her in there. We will avoid a lot of frustration and hassle that way, for me and my daughters. And for me, too, in order to ,be able to get out again. Right now I have no way to go anywhere. I have to be here doing whatever [make-work], just so she can see that I'm here. It irritates her so much if I go out anywhere. . . ." Notes 1. Thanks go to the National Science Foundation and R. J. Reynolds, for financial support of this research project. Much appreciation also to Dorothy Holland, Ed Hutchins, and Wendy Weiss for their valuable contributions to the development of this paper. Eariier versions of the paper were delivered at the "Conference on Folk Models" and at the 80th Annual Meeting of the American Anthropological Assocation in Los Angeles in 1981, in a symposium organized by Dorothy Holland and Naomi Quinn and entitled "Folk Theories in Everyday Cognition." 2. There are two primary types of natural discourse about illness between nonspecialists in Ecuador: illness stories and therapeutic recommendations. Therapeutic recommendations pertain to ongoing health problems and may be solicited from, or offered spontaneously by, family members, friends, or neighbors. Even though the structure of this discourse is conversational, it does not follow the pattern of chronological narrative that characterizes illness stories. Therapeutic recommendations lead rather naturally into narratives, however, if a story exists to be told, if time permits, and if the social context is appropriate. 3. For Ecuadorian women, the genre of illness storytelling constitutes what we might call "shop talk" because one of their special gender roie responsibilities is management of family illnesses. Such conversations have immense probiemsoiving value, and women tend to launch into conversation about illness much more readily than men. One exception occurs when men are known to be carrying out direct therapy-managing roles in an illness situation; these men will be the recipients of therapeutic recommendations from friends and family and will be told pertinent illness stories, just as female caretakers are recipients of such communication. >••• 4. Although illness stories do transmit information about specific remedies and home-treatment principles, there is less of it than might be expected given the widespread use of home therapies in Las Gradas. Why do cognitive heuristics such as home treatment sequences (Mathews 1983) not figure more promi- ? nently in this discourse? A plausible explanation is that home treatments are less valued than those prescribed by specialists and so receive less attention. Another probable explanation is that narrators deemphasize or omit these details from their accounts because they assume their audience already possesses that information or will ask for it specifically if they need to know. Just as ,£, a storyteller in the United States would assume that listeners know the heuristic of drinking plenty of fluids and taking aspirin for a cold, so do Ecuadorian narrators take for granted that everyone knows the more common herbal remedies and specialist treatment options. This cultural pattern also affects ethnographic data collection. To obtain detailed information about such-' treatments, the ethnographer must usually ask about them directly. 5. Reasons for not openly acknowledging propositions about moral causes for •%

ECUADORIAN ILLNESS STORIES

,3,41

a particular illness include family loyalty, self-implication by such ah interpre-. tation, and inappropriate audience. 6. Whether North American narrators instantiate propositions about personality traits in telling illness stories is a question that requires further research. Only with documentation of the illness storytelling genre cross-culturally can analysis be made of differences between the models that shape the discoursed arid ::. of the social or historical reasons for such differences.

References Agar, M. 1980. Stories, background knowledge and themes: Problems in the analysis of life history narrative. American Ethnologist 7(2):223-240. 1982. Whatever happened to cognitive anthropology: A partial review. Human Organization 41(1):82~86. Bateson, G. 1972. Pathologies of epistemology. In Steps to an Ecology of Mind. New York: Ballantine Books. Pp. 478-487. Bohannan, P. 1957. Justice and Judgement among the Tiv. London: Oxford University Press, Clement, D. 1982. Samoan folk knowledge of mental disorders. In Cultural Conceptions of Mental Health and Therapy, A. J. Marsella and G. M. White, eds. Dordrecht, Holland: D. Reidel Publishing Company. Pp. 193-215. D'Andrade, R. G. 1976. A propositiona! analysis of U.S. American beliefs about illness. In Meanings in Anthropology, K. H. Basso and H. A. Selby, eds. Albuquerque: University of New Mexico Press. Pp. 155-180. 1981. The cultural part of cognition. Cognitive Science 5(3): 179-195. Glick, L. B. 1967. Medicine as an ethnographic category: The Gimi of the New Guinea Highlands. Ethnology 6:31-56. Holland, D. 1985. From situation to impression: How Americans use cultural knowledge to get to know themselves and one another. In Directions in Cognitive Anthropology, J. Dougherty, ed. Urbana: University of Illinois Press. Pp. 389-411. Hymes, D. 1972. Models of the interaction of language and social life. In Directions in the Ethnography of Communication, J. J. Gumperz and D. Hymes, eds. New York: Holt, Rinehart and Winston. Pp. 35-71. Labov, W. and D. Fanshel 1977. Therapeutic Discourse: Psychotherapy as Conversation. New York: Academic Press. Lakoff, G. and M. Johnson 1980. Metaphors We Live By. Chicago: University of Chicago Press. Mathews, H. 1983. Context-specific variation in humoral classification. American Anthropologist 85{4):826-843. Quinn, N. 1982. "Commitment" in American marriage: A cultural analysis. American Ethnologist 9(4):775-798.

342 Stark,L.

LAURIE PRICE ...'•.:. , • / ' , . . ^ / ; , - : : ' , . . . , . . ' , , ' ,

"

-"••'"•

1981. Folk models of stratification and ethnicity in the highlands of Northern "'• Ecuador. In Cultural Transformations and Ethnicity iti Modern Ecuador, N. Whkten/ed. Ufbaria: University of Illinois Press. Pp: 387-401. Tanrien, D.: • i":^'-- -;•• v-";-'r ;;-^...:>-.- \--\]^-- ••/•' ""••'•' - 1 9 7 9 . What's in a frame? Surface evidence for underlying expectations. /« New Directions in Discourse Processing, Vol. 2, Advances in Discourse Processes, R. O. Freedle, ed. Norwood, N.J.: Ablex Publishing Company. Pp. 137-181. Zajonc, R. B. 1980. Feeling and Thinking. American Psychologist 35:151-175. '*•-*•

14 Explanatory systems in oral Ufe stories1

Charlotte Linde

This chapter defines the explanatory system using oral life stories of middle-class American speakers as a source of data. An explanatory system is a system of beliefs and relations among beliefs that provide the environment in which one statement may or may not be taken as a cause for another statement. More specifically, an explanatory system of the type discussed here is a system of beliefs that occupies a position midway between common sense, the beliefs and relations among beliefs that any person in the culture may be assumed to know, if not to share, and expert systems, which are beliefs and relations among beliefs held, understood, and used by experts in a particular domain. An explanatory system is a system of beliefs derived from some expert system but used by someone with no corresponding expertise or credentials. Note that the term common sense as defined here closely corresponds to the notion of cultural models or cultural theories as used throughout this volume. The term expert system or some close equivalent is similar to the usage of Kempton, and Collins and Gentner (this volume). The term explanatory system represents an intermediate level that, to my knowledge, has received relatively little attention. To clarify the definition of explanatory system, let us turn to two (constructed) example pairs. The first (1) gives a reason that relies on commonsense beliefs to account for professional choice. (la) How did you come to be an accountant? (lb) Well, I guess I have a precise mind, and i enjoy getting all the little details right. In contrast, in order to understand (2b) as a relevant answer to (2a), the hearer must know if not share the popular Freudian explanatory system which locates the real causes of events in early childhood. (2a) How did you come to be an accountant? (2b) Well, my mother started toilet training me when I was six months old. Using this notion of explanatory system, this paper first discusses the methodological background and data for the study of explanatory systems,

344

CHARLOTTE LINDE

then examines in detail the explanatory systems that have been found, arid finally discusses the historical relation between common sense, explanatory systems, and expert systems. / The life story This section first defines life story as a technical term and then discusses the types of data and methodology necessary for its study. This discussion is necessarily brief, as the present study forms part of a larger study of the creation of coherence in oral life stories (Linde n.d.). THE LIFE STORY

The definition of life story used in this study is an attempt to render precise and accesible to analysis a notion in common use by speakers of English. Intuitively, we believe that we "have" a life story and that any normally competent adult has one. This nontechnical sense of life story is something like, "What events have made me what I am"; "What you must know to know me." Technically, the definition of a life story is: All the individual stories and the relations drawn between them told by an individual during his or her entire lifetime that satisfy the following criteria: 1. The stories contained in the life story make a point about the speaker, not about the way the world is. 2. The stories have extended reportability. That is, they are tellable over the course of a long period of time. The first point in this definition is that an entire life story is constantly added to and changed in the course of a lifetime, so that to have a record of an entire life story would require all of the talk ever produced by a given speaker. This is in principle possible, in practice impossible. Nonetheless, it is possible to study the life story by using a small sample of the stories that comprise it, since the interest of the present study is not the specific form of the entire story but rather the principles by which it is constructed. The next point is that not all stories told by a speaker form a part of the life story. One criterion for the inclusion of a story as part of the life story is that its evaluative point is directly about the speaker or some event framed as relevant to the speaker, as opposed to a story that makes some .„ evaluative point about the world in general. That is, I would want to consider as part of my life story a story about what happened to me when I was in the hospital, but not a story about the same events told to show what's wrong with hospitals. &,:A further criterion for inclusion is that the story have extended report- ~ ability. This means that the story can be retold over a long period of time. Thus, the story of a career decision, or a divorce, or a major illness are •'&

EXPLANATORY SYSTEMS IN ORAL LIFE STORIES

•345'

relevant and reportable over a major portion of one's life:3ri icbntrast^ a story about the funny thing the guy in the health food storesmdisVe^prJti able for a day or two at most as a story about what;ha£pened ;$6'!;m^ (although it may retain extended reportability as a storyaboat^hatpep-;; : pie in hearth food stores are like). * •••'':'• :'0§^i$$M^ A final point is that the life story is not defined simply as a particular'': subset of stories but as the relations the speaker draws among thern as well. Thus, as new stories enter the life story, old ones may have-to •be. dropped or revised to maintain coherence. It is this form of relation among stories that permits the life story to express our entire sense of what our life has been about without ever necessarily forming a single narrative organizing our entire lives. Note that life story defined here is related but not identical to autobiography and life history. Both the life story and the autobiography are produced by a person presenting his or her own life in a form that will make sense to other people. They differ, first, in their medium - spoken versus written language. The life story also differs from autobiography in being a temporally discontinuous, ongoing account, whereas an autobiography is a crystallization at a given time of what has happened to the author and what it means. Finally, autobiographies are subject to the economic process of publishing, and so published autobiographies tend to be by people either socially defined as important or interesting, or by undistinguished people what can be taken as typical members of a class of interest - cancer victims who survived, or housewives who have found God, for example. In contrast, every adult is entitled to and indeed required to have a life story. The notion of life story also contrasts with life history, a term mainly used in anthropology, although also somewhat current in psychology. The life story in anthropology is usually presented as a form of autobiography of someone in another culture, a presentation of his or her experience and sense of self that is the collaborative product of the subject and the anthropologist. The major difference is that the presentation of self in this form may not be at all a natural discourse form for the subject; indeed, the whole notion of self present in the subject's culture may be quite different from that of the anthropologist. (For an extended review of the notion of life history, see Langness & Frank (1981:ch.4) THE DATA OF THIS STUDY

The data of this study come from oral interviews on choice of profession. Fourteen people were interviewed; they were chosen because they all had professions that were important to them and that could be expected to form a major part of their life stories. The interview began with the question, "What is your profession" and went on to ask "How did you arrive at that," or "How did you become a ." When the speaker had

346

CHARLOTTE LINDE

apparently finished, I probed for more, either by a further question or by an extended silence. . : / The interviews focused on the choice of profession since,.'at least for middle-class professional speakers, it is a necessary part of one's account of oneself. One need only recall how often one is asked, "What do you do?" to verify this. Another demonstration of the centrality of profession is to consider the case in which one is acquainted with someone but does not know that person's occupation; this is anomalous, and after some time, sinister. One's occupation is a publicly available piece of information, from which many inferences may be drawn as to what sort of •per-. son one is. ~ Note that profession is public knowledge in just that social world in which it plays a major role in the definition of the self. This is, of course, not true for all societies, nor even for all classes of our own society. However, with that restriction, we may note that professional choice works excellently as an interview topic because it is so relevant to self-presentation and because it forms a relatively public portion of one's biography. The data are then analyzed using the methods of discourse analysis. For a full discussion of discourse analysis, see Linde {1981; n.d.). In brief, the component discourse units are identified - in the case of the life story, they are narrative, chronicle, and explanation. The internal structures of these units are then analyzed. Of particular relevance for the study of explanatory systems is the analysis of the evaluative structure of these units, that is, the portions of text that comment on what has been said, informing the addressee of how this is to be understood. The construction of coherence An important property of the life story, both linguistically and psychologically, is that it must be coherent. Its coherence is not a property of the life, but rather an achievement of the speaker in constructing the story. Coherence in this sense is a property of texts; it is one set of relations by which we may understand a text. Specifically, the coherence of a text consists of the relations that the parts of the text bear to one another. A text may be described as coherent if its parts, whether on the word level, phrase level, sentence level, semantic level, or level of larger units, can be seen as being in a proper relation to one another and to the text as a whole. Thus, a narrative will be heard as coherent if it is clear for the hearer what relation its parts bear to one another - that is, given that A and B were said, why B follows A (Becker 1977). We can recognize at least three different levels of coherence in these texts. The first level is the basic linguistic level of text structuring devices.:: The devices of this level are widely used across an entire language and are dependent on the obligatory morphological and syntactic categories of that language. An example of a device at this level is this narrative'"

EXPLANATORY SYSTEMS IN ORAL LIFE STORIES

347

presupposition: the rule of interpretation stating that the order of ;p'ast-,: tense main-clause verbs shall be taken as the order of events! •^h^^e<;bE|'.^: level of coherence is the level of implicit philosophical categories^^uch'-.'-'.:'. as, in English, causality, accident, continuity, and .discontinuity^ These } categories depend on the coherence established by the previous j^yei.aruj.j-^ in some sense represent an elaboration of it. The third level of coherence; •"•! is the level of semiexplicit explanatory systems, the level of interestjfQr, this paper. The systems at this level all presuppose the categories oftthe previous level. That is, they all assume the existence of causality and jthen. give different answers to the question of what a possible cause might be.:: We turn now to a more detailed discussion of these levels. ; THE LEVEL OF MORPHOLOGY: THE NARRATIVE PRESUPPOSITION

As mentioned in the previous section, the narrative depends on the fact that tense is an obligatory category in English - all sentences have a main verb, and this main verb has some tense. Further, speakers of English take tense to be iconic; that is, we assume there really is such a thing as time, and that it naturally consists of past, present, and future. This belief permits the narrative presupposition: the assumption that the order of a sequence of sentences or main clauses matches the order of events. The following examples illustrate the automatic character of this assumption. (3) I got flustered and 1 backed the car into a tree. (4) I backed the car into a tree and I got flustered. In (3), getting flustered is assumed to precede backing into the tree; in (4) it is assumed to follow backing into the tree. These examples further illustrate the relation of tense and the beginings of the assumption of causality. The natural logic of English is post hoc, ergo propter hoc; in (3) we assume that getting flustered caused the speaker to back into a tree, in (4) we assume that backing into a tree caused the speaker to get flustered. More complex forms of causality are built on this foundation. COHERENCE PRINCIPLES: CAUSALITY AND CONTINUITY

We have seen that at the morphological level, the temporal order of clauses, or narrative order, represents the major resource for the creation of the implication of causality, since any temporal order is likely to be understood as a causal order as well. We turn now to the level of the major coherence principles used in organizing life stories, which are primarily causality and continuity, both of which are based on the previous establishment of order. In telling a life story, one major task for the speaker is to establish that the causal sequence of events is an adequate one. We define adequate causality as a chain of causality that the hearers can accept as a good reason for some particular event or sequence of events. For the data of this study, establishing adequate causality for the choice of profession means estab-

348

CHARLOTTE LINDE

iishing that there are obvious good reasons for the speaker's choice of profession, or showing that even if the reasons do not seem at first glance to be acceptable, actually there are reasons for accepting these reasons. Establishing adequate causality for a given professional choiceyor for any sequence of events, is both a social and an individual achievement.' Part of the social aspect is the fact that there are cultural and subcultural beliefs about proper lives, proper sequences, and proper reasons for professional choice. For example, one socially sanctioned and very strong form of adequate reason for a career choice is ability or character If a speaker can claim, "I like that sort of thing," or I'm good at it," a standard form of adequate causality has been established and no further reasons need be given. Similarly, we have beliefs about what would be inadequate or frivolous reasons for choosing a career; for example, "I was in love with a girl who was enrolled in premed, so I decided to become a doctor."2 In addition to this store of common-sense beliefs about adequate causes, there is also a strong element of individual creativity in the creation of adequate causality. The individual speaker's adroitness or ineptness in framing his or her story can also determine whether a given sequence will be heard as adequate or inadequate. To establish adequate causality, not only must the reasons be adequate, but the chain of reasons must also be neither too thin nor too thick. Too thin an account suggests that one's life has proceeded at random, discontinuously. Too thick an account suggests that the speaker, implicitly accepts a deterministic or fatalistic theory of causality. Neither of these extremes is generally acceptable, and both are subject to correction. The correction may be made by the speaker, following a deterministic account with an accidental one, or vice versa. That is, the speakers often perform a sort of philosophical wobble around a socially determined equilibrium point, which carefully avoids taking any position to its logical extreme. If the speaker does not maintain this equilibrium, the addressee may correct the story in the direction of equilibrium. In studying adequate causality in the interview situation, we may recognize an account as adequate by the fact that its speaker treats it as sufficient; that is, the speaker does not follow it with another account and holds to the same account even if challenged by the interviewer. In this set of data, I found two forms of adequate accounts of the reason for a professional choice: those based on character traits and those based on multiple noncontradictory accounts. One of the most powerful accounts possible for choice of profession is character. Speakers appear to take character traits as a primitive, referring to them as obvious causes for career decisions, with no need to account for these traits. We thus find examples like (5) given as an account for a career as an English professor, and (6) as an account for a career as an editor.

EXPLANATORY SYSTEMS IN ORAL LIFE STORIES

w%

(5) I seemed to be very good in reading and analyzing 'books '•and^rih^stf4^ became an English major and from then on since I knew! ^oqld ^o ^tfaife^ in whatever I chose as I could possibly go, getting a Ph.D. becaihe'^necessity}'$ (6) I always was nitpicky, I was always good at grammar,urni;Ilikeio^orrec't^' things rather than create them. And I'm interested enough:jn"Tea^mg;ind'';l' I like to spend the time reading, that at least now that I'm on journals thtt\ are even vaguely interesting, it can be a lot of fun. _• •:};:\:?-;:i:-.?:{;^-i-i'.':. Both of these are given as adequate and final reasons fpr the choice pf profession. Another very strong form of adequate causality is richness of account. An account may be rich because it covers a long period of time; that is, the reason for the choice of profession is located far back in the speaker's past. Of course, one form of temporally well rooted account is the explanation by character trait. But citing ambition, family interest in some sphere of activity, schooling, or childhood hobbies may equally permit a professional choice to appear well-rooted in time. Another way an account may be rich is by containing many noncontradictory reasons for a choice. In one such example, a speaker who is a professor of English studying science and literature gives the following accounts for her profession: She was at a university that was trying to bring together academic disciplines and medicine; she had good verbal skills because of her family's interests; she had an interest in science but not the aptitude for it; in her undergraduate career, she had encountered one noted scholar with a similar blend of interests; her quarrelsome family led her to be a peacemaker, which led to some attempt to reconcile science and literature. Such multiple noncontradictory accounts provide adequate causality because they show that a choice was not random or insufficiently motivated. In addition to adequate causality, there are accounts that must handle the fact that the speaker perceives the causality or the ordering of events to be in some way insufficient or problematic,These can be divided into accounts structured in terms of accident and accounts containing a socially recognizable discontinuity. In the context of a life story, accident means the absence of sufficient, socially recognized causality. When an event has been presented as an accident, a common pattern is to narrate that event as an accident and then follow it by a demonstration of how it was not really an accident, since there were multiple routes to the same point in life. Thus, although the particular route was an accident, the goal was not. In telling a story, the narrator may perceive and present the events as discontinuous. A (constructed) example of such discontinuity is, "I was a banker until I was 35, and then I dropped it all and became a potter." As members of the culture, we recognize that this is a conventionally discontinuous sequence of professions. One of the clearest patterns in these data is that this form of discontinuity must be managed in some way; some

350

CHARLOTTE LINDE

evaluation or explanation of the discontinuity must be given. Among the forms of management of discontinuity found in these data are;,.. < i. Discontinuity as Apparent Break. Discontinuity is seen as only apparent, not real. That is, a later state appears to be very different from an earlier one, but they can be seen as having characteristics that make them continuous. 2. Discontinuity as Temporary. An apparent break is shown to be due to a temporary discontinuity; for example, an early interest was abafr; doned and then resumed. 3. Discontinuity of Actor. The speaker may distance himself or herself from the protagonist of the story. Such distancing focuses the blame of discontinuity, so to speak, on some other person than the present speaker. 4. Discontinuity as Continuity. In this logically more complex strategy, the speaker uses a series of discontinuous events to establish that discontinuity forms a continuity for him or her. An example of this strategy is, "I think I could be a dilettante all my life as long as I could do it with some intensity." EXPLANATORY SYSTEMS

Moving up from the level of the management of causality, we turn to the level of the explanatory system. We have defined the explanatory system as providing the conceptual environment in which something may or may not a cause of something else. Thus, within a Freudian explanatory system, the course of development of childhood sexuality may be a cause of adult character, whereas in other systems, one would look elsewhere for the cause of character, or might not seek it at all. In later sections, we see that the explanatory systems present in these data are popular versions of expert systems. However, not all explanatory systems are expert systems. We may view the ordinary system of common sense as the most pervasive explanatory system and the most unnoticed. As defined, common sense is the system of beliefs assumed to be shared by everyone and requiring no special circumstances or standing for its use. Thus, a speaker who said, "I quit banking because I didn't like it" would be invoking an explanatory principle from common sense. In contrast, a speaker who said "I quit banking because my father was a banker, and I've always had a love-hate thing with my father" would be invoking a psychological explanatory principle. In later sections, we explore some of the explanatory systems present in these data. We do not, however, consider the elements of the system of common sense, as this is a task beyond the scope of the present study.

EXPLANATORY SYSTEMS IN ORAL LIFE STORIES

:351

Definition and function of explanatory systems This section gives a fuller definition of explanatory system and attempts to sketch its function for speakers. One view of the explanatory system is that it not only guides the construction of a story but also provides an evaluation of what the story means and what its value is. When a life story is analyzed by an analyst other than the speaker, a number of evaluation principles might be used. The most obvious is factuality: Do the events narrated correspond to reality? If the analyst is a biographer or a district attorney, this is an appropriate criterion. However, since my interest in this entire research program is how speakers construct life stories as coherent, it is possible to work entirely with the internal relations of the text, bracketing the question of relation to some postulated real world. However, there also is a more sophisticated form of evaluation by an analyst. The analyst may subscribe to some system that gives a principled basis for deciding whether the speaker's account is one that could possibly be valid, using the organization of the account, its coherence, and its choice of topics. Such a system may be a political theory, a psychological theory, or a religious theory, and so on. For example, for a Marxist analyst, none of the accounts of career choice given by these speakers could possibly be valid since they are not based on, and indeed give no account of, social class or economic circumstances. Within this explanatory system, real explanation lies within the domain of economics, and so all other forms of explanation represent surface phenomena or false consciousness. Similarly, a Freudian analyst can claim to have resources to determine whether an explanation given by a speaker is a correct product of genuine insight or whether it is the result of one of the deceptive complicating processes the mind uses to hide from itself. (See Frank 1981 for a discussion of the basis of such validation.) As these examples show, an explanatory system is a system that claims to give a means for understanding, evaluating, and interpreting experience or accounts of experience and usually, as a consequence of that understanding, also gives, either explicitly or implicitly, a guide for future behavior. Presumably, this definition would include such local explanatory systems as theories of the right kind of food to maintain optimal health, or the best way to win at roulette. However, the interesting examples of explanatory systems are those with some reasonable claim to completeness, those purporting to explain most or all realms of experience, not merely local areas of life. The most obvious current examples are Freudian psychology, or more generally, psychoanalytic theory, Marxism, and most religious systems. An understanding of much of experience as the product of a worldwide conspiracy of the Bavarian illuminati or the unrecognized influence of higher plane entities would equally qualify as explanatory systems, albeit less widely held ones.

352

/

CHARLOTTE LINDE

This definition of explanatory systems is not equivalent to any concept in general use. It is related to the term belief system or cultural system as used, for example, in Geertz (1973) but differs in that the explanatory system is specifically a semiexpert system, related but not equivalent to either belief systems shared by an entire culture or belief systems exclusive to some class of experts. We now turn to the question of how these explanatory systems function for their users. If we look at the social conditions under which expert systems are used to evaluate experience, we find that such a system is often used by a person with special status, trained in its terms and argument forms, to criticize some nonexpert account of experience. One obvious example is the use of case history material by a psychologist. The subject produces a life story, or this life story is summarized, and then the psychologist can use expert knowledge to demonstrate what is really going on. Depending on theoretical persuasion, the psychologist may use the array of psychoanalytic techniques and arguments, or the armamentarium of diagnostic psychology, including personality tests, and projective tests. In either case, these implements of expertise have primacy; they confirm or invalidate the correctness of the speaker's own account of his or her life. It is not experts alone who have theories that offer the "real" reasons for people's behavior and hence can be used to evaluate the correctness of any explanation given. Any adult has a wide variety of theories available that can be used to make sense of his or her life. These theories may be explicit systems requiring formal subscription, such as a particular religion or political philosophy. Or, they may be extremely inexplicit belief systems implicit in the culture, which require no formal recognition or allegiance and which, without careful analysis, pass unnoticed as theories, appearing to be just one further example of how we speak about these things. When we examine how explanatory systems are used, we find that because they are all popularized versions of expert theories, they permit the speaker an extra level of distance, allowing him or her to be an expert, to step back from the personal account to give a deeper, or apparently more objective, or truer account than can be conveyed in a common-sense narrative. To clarify the nature of the distance created by the use of an explanatory system, note that all first-person narration has an implicit distance between the narrator and the protagonist. This distance permits us to tell a story about some bad or ill-judged or embarrassing action we took, since even though the protagonist acted in this way, the narrator knows better and so is able to tell the story in a way that indicates present allegiance to the norm that was in the past broken. However, the use of an expert explanatory system introduces an additional layer of distance beyond that of narrator and protagonist. Although the narrator may know better than the protagonist, they are still the same order of being, operating by the same rules, with the same type of knowledge. In contrast, the expert is different from the protagonist, possessing different and superior knowledge of why the protagonist acts and how he or she should act.

EXPLANATORY SYSTEMS IN ORAL LIFE STORIES

353

THE SET OF EXPLANATORY SYSTEMS IN THE DATA

The theories I have found thus far in these data are versions of Freudian psychology, behaviorism, astrology, Zeitgeist theory, and Catholic the6iy of sin and confession. Each theory has its associated expert ~ a practicing psychoanalyst, an academic psychologist, an astrologer, a sociologist, or a priest. This paper explicates the first three theories, since they are most clearly represented in my data. .;! This list of explanatory systems raises a number of questions! Is the list exhaustive? How many more explanatory systems would be discovered by doing more interviews, or interviews on another topic? How many such explanatory systems can be present in a culture at a given time? It would take a different research program and a rather extensive one to answer these questions. However, my estimation is that this list is not at all exhaustive; other interviews and particularly interviews on other topics would undoubtedly produce further examples. However, the number of explanatory systems would not be very great. There is a necessary limit on the number of such explanatory systems that will be present in a given culture at a given time, since one's addressee must at least recognize.if not share any explanatory system one choses to use. Certain absences in the list of explanatory systems in these data are particularly striking. For example, no one cites economic opportunities or class limitations as factors in professional choice; the speakers all appear to assume that professional choice is dictated by personal abilities and by degree of psychological adjustment, which either permits or prevents one from seeing and talking advantage of opportunities. This view of a world without politics and without social class appears to be particularly American (Lasch 1979; Schur 1976; Sennett 1977).3 Another striking absence is the lack of invocation of race or ethnicity to explain available opportunity. Since all the speakers are white, it is perhaps understandable that race is not dominant in their understanding of the factors that shaped their professional lives. However, the sample does include people from a variety of ethnic groups and religious backgrounds, only one of whom mentions this as a factor influencing his personal psychology. This is in strong contrast to a previous generation of speakers, for whom ethnic origin, and more specifically, generation of immigration were crucial in understanding anyone's life story.

Explanatory systems in these data Of the explanatory systems in these data, the most common appears to be that of Freudian psychology. I thus begin with it, give the fullest discussion of it, and use it as a detailed example of the relation between the expert and the popular versions of an explanatory system.

354

CHARLOTTE L1N.DE

POPULAR FREUDIAN PSYCHOLOGY

To get the flavor of this explanatory system, consider the following example, which comes as the conlusion to a story about how the speaker took a banking job that he did not like and that did not last very long. (7) . . . so I didn't really make much of a decision there. I think that's one way of looking at it. I made a decision. It was the decision that I didn't tike at the time, so that's why I have the sense that I was forced into it, but there are ail kinds of psychological things that make you do things at various moments in your life. From a common-sense point of view, this seems to be a rather weak justification of a bad decision. That is, common sense suggests that one should be in control of one's action, or if one is not, that one should be forced into making decisions by strong external forces. However, further analysis suggests that this text implies an explanatory system that gives a way of making sense of the phenomenon of doing things that one does not like or approve of. The speaker himself gives the clue to the system he is invoking by the phrase "psychological things." This system is a psychological model, more specifically, as we see, a popular version of Freudian psychology. A number of components to this popular version show up in the present data. These are: 1. The splitting of the self into component parts, which are in disagreement. 2. The notion that real causes are to be found in childhood and childhood experiences. 3. The notion of levels of personality, some of which are deeper than others. Let us begin with the split of the self into component parts in disagreement. The speaker of (7) feeis that he was "forced" into a decision and attributes this to "psychological things that make you do things." This fragmentation of experience and the identification of the speaker with the fragment that is not in control is a strong component in popular Freudian psychology. Another related notion at work in this example is the idea , that one can act for reasons of which one is unaware or does not understand. Thus, we hear statements like, "1 was in Paris at the time and I thought I was happy, but really I wasn't." A more extended example of this phenomenon is (8). (8) I married someone who was older than I was, who was a bit of a self-centered tyrant, and for that 1 got, uh, to whom I was not very, even sexually attracted, %

EXPLANATORY SYSTEMS IN ORAL LIFE STORIES

§55.:ki-

and was very temperamentally different than 1 am, and iih althoughTw)isV.•:':<•'. in a certain way attracted to him sexually I was not; really, ''^ <>'*-•:>z^Z&y .;.'•' Here, again, we have an example of someone telling about acting because, of feelings which she felt, or thought she felt, but which were nptfe^i.- :• • Both of these assumptions about the nature of action are based':pn"g>^\ ;['•'.':• complex form of distancing. The first is the distance the speaker:rndiK tains from the protagonist of the story. As discussed, this is 'p'a^';6f |iny:/ form of narration. However, in stories containing this additional element of a Freudian explanatory system, the speakers, in achieving'this^ distalcei,-: are able to place themselves in the position of an expert, commenting on . their own life. .;.;;.A further form of distancing, buried more deeply within the story, is the distancing between the component parts of the person. The part that feels or believes it feels is not the part that causes action. As the speaker of (7) indicates, this permits a curious paradox in which one is simultaneously active and passive, forced into action by oneself. Another, perhaps more familiar, tenet of popular Freudian psychology is the notion that real causes are to be found in childhood and childhood experiences. We see a version of this belief in (9), combined with the metaphor of splitting oneself into component parts. The speaker, who is also the speaker of (8), is explaining her combination of literary and scientific interests. She has already told a number of stories about how she arrived at these interests, which have to do with the opportunities were available to her in the course of her education. She then steps back, as it were, from this historical presentation, to give what the language interestingly allows us to call a deeper reason. (9) Another idea that, well, in my more - uh I often wonder why I have this need to make, to write, to kind of split myself off in two directions or to try and take on many, to take on what might seem to be warring, 1 see what I do as, as reconciling things and why do I need to reconcile things? I came from a family that was very conflict-prone. And I was the peacemaker in the family. And I don't like conflict. And I, or rather, I, I am really very upset by it. The assumption of this explanation, which permits the speaker to offer it as an explanation, is that one's childhood emotions and actions are crucially relevant in providing an explanation of any further developments. It is important to note that I am not claiming that only Freudian psychology admits childhood experience as important and that no other explanatory system admits it. Rather, the claim is that Freudian psychology gives it a primacy, which means that any adequate explanation within this system is likely to include it. A further theme of this explanatory system, which is related to the splitting of the individual into component parts of the personality, is the

356

CHARLOTTE U N D E

metaphor of levels, the notion that some of these parts are deeper, or further down than others. We get an oblique reference to this notion in (10): (10) And 1 think that the source of one's power professionally comes from some deep s-, deep thing. And if you don't tap that, you're sunk. The idea here seems to be that the components of the personality are ordered vertically, with the higher ones being more accessible to consciousness and the lower ones being out of the reach of ordinary selfinspection. A further part of this metaphor is that the deep components have a certain amount of autonomy and can, in unseen ways, motivate actions, which may or may not be in agreement with the desires and plans of the more conscious surface components. The foregoing discussion has presented a sketch of a semiexpert theory of the mind, which contrasts both with a common-sense theory of the mind and with the professional Freudian model. D'Andrade (this volume) provides an explication of the common-sense or cultural theory of the mind, which is clearly quite different from the Freudian model in both its expert and semipopular form. The most interesting locus of difference lies in causality - the cultural model treats conscious mental states as having central causal powers, whereas the Freudian model treats unconscious mental states as the causal center. The popular Freudian explanatory system also differs from a professional or expert Freudian model in a number of ways. These differences are discussed in detail in the section on.the synchronic relation between popular and expert systems. In brief, like all cultural models, the popular Freudian system represents a considerable simplification of the original model, reducing both the number of themes and concepts it uses and the complexity of the connections between them.

BEHAVIORIST PSYCHOLOGY

The Freudian psychological explanatory system is quite familiar because of its pervasiveness in the culture. In contrast, let us now turn to behaviorist psychology, a rare system, represented in these data by only one speaker. More specifically, one speaker presents her life in behaviorist terms, taking the role not of the conditioner but of the subject of conditioning, not Skinner but one of Skinner's pigeons. Three themes in her account of her life lead one to an analysis of her explanatory system as behaviorism. These themes are: 1. The need for reinforcement, and reinforcement as the cause of action. 2. The separation of the self from emotion. 3. Nonagency - the self described in such a way as never to be an active agent in causing events,

EXPLANATORY SYSTEMS IN ORAL LIFE STORIES

'"::•'M5J\

Reinforcement is'.themost obvious component^ofilhe?beh^i^itis3>^^ planatory system, and itislthe ^ctbt^Ms :isp^ak:er.:^se> ^xplicitly:^ 5 !!! plaining her historyiAftefcbllegeishV^rked ^ ; ^ ^ i ^ ^ i r k ^ | i ^ ' i t | ^ i ^ went back to school to'geta^aster^ide|ree^ time of narration, she Worked as a compMer p r p g r a m ^ r ; ; ^ h e i : w o ! ^ ^ focuses in her account of her careerhistory .are^'chai^e'';fr^^^cikn work to programming and the ".fact that -'She: has held .t)flujribe¥^^gl;; gramming jobs at different companies/She"feels.ltiia^.JiotH.feiFi-^eMl^a^ff-'need accounting for, and she accounts for both of themas : arlslnf^m1 her need for reinforcement. :.-'.\^.i >^ftfc: (11) I like concrete results. 1, I need to be able to see what I've done. I need that kind of reinforcement. If you get me out of that mode, I'm just, I have little tolerance for it. I, I need the strokes of seeing what I'm doing as I'm doing it. (12) So when I was doing social work, and I did that for five years, I need to have, I need results. I'm a result person. OK. I need to, I like to manipulate symbols, I like those kind of machinations in your head. And I like to see what I've done. OK? If I, I can not go extended periods without that, I don't have the tolerance for it and it was, I came to that realization while doing social work, that I had to find something that was, that gave me that kind of concrete kind of [unclear]. (13) I guess my chief feeling about XYZ Co. [where she formerly worked as a programmer] was that it was very, there was no point in working. Because it didn't matter whether you did or whether you didn't. There were no strokes for working, there were no punishments for not working. This notion of reinforcement as the motive for behavior is identical to Skinner's views, with the additional specification of her personal need for a reinforcement schedule with a very short time delay. Although the terms reinforcement and strokes have become part of the common vocabulary, this account of one's life is not coherent within the commonsense explanatory system but only within the expert system of behaviorism. Another similarity to Skinner's thought lies in this speaker's treatment of her emotions and mental states. In her stories about her professional history, the speaker frequently mentions that she must have had certain emotions, but does not claim them directly. For example: (14) And it began to be very demoralizing to me. Uh so I left and whatever I, arid I had at that time decided that I really wanted to get to, I guess somehow I had decided I wanted to get serious again about working. The correction from "I had at that time decided . . ." to "I guess somehow I had decided . . ." is a perfect example. Most psychologists would regard this as a small but pathological form of distancing oneself from one's emotions. However, from a behaviorist point of view, the belief in autonomous mental states that act as determinants of behavior is an

358

CHARLOTTE LINDE

error, indeed the fundamental error in understanding human psychology. However, it is an error that nontechnical language almost forces one into. The speaker here denies any actual belief in or sensation of decision as an autonomous mental entity, while indicating some loyalty to the ordinary forms of discourse and understanding, which require them. ;. Closely related to the nonexistence of mental states is the third theme: nonagency. What this term is intended to convey is that this speaker does not use personal agency in the way most speakers do. Normally* speakers describe their experiences in such a way that they are active agents in,their stories: "I did so and so"; "I decided such and such"; "I solved the problem facing me." This speaker does not construct her stories in this way: The active agents are either abstract motivations or other persons, not herself. The following story is an example of both kinds of agency. The situation is that the speaker was working as a social worker, which she did not like, and was investigating the possibility of becoming a computer programmer by speaking to a professor of computer science. (15) I spoke to Dr. K a long time. He decided, oh yeah, I had really gone over there, wasn't asking to go to graduate school. I was asking to take a course in the evening. And 1 figured I could qualify because I had graduated from V and it wouldn't, what big deal would it be for him to let me take a course. I, by the time I was out of there, he had me enrolled in this whole program, and he says, "Oh you have to do it this way." And it must have appealed to me, because i went ahead and did it. In this story, Dr. K is the agent, not the speaker. It is his somewhat surprising actions that move events along. The speaker had intended to take a single course, but as a result of Dr. K's actions, finds herself enrolled in a master's program, and a simultaneous BS in mathematics. Even in the smaller level details of sentential construction, we find that she is not the agent. For example, note the nonagency of the phrase "By the time I was out of there" as compared with the possible "By the time I left there." The theme of nonagency is extremely strong, so much so that even when it is directly challenged, it is still maintained. As part of the interview procedure, after I had obtained the speakers' own accounts of their professional history, I attempted to reframe the questions, using explanatory systems different from those the speakers had used. I did this in an attempt to see to what extent the speakers were willing to adapt their explanatory systems to my apparent beliefs and to what extent they would maintain them in the face of a proposed alternative conceptualization. Answer (16) represents a response to such a challenge. I had asked the speaker about choice points in her life, a formulation that implies choices as relevant entities. Although the speaker does not explicitly deny the existence of choice points, or the possibility of choice, the story she tells, apparently using this vocabulary, is completely consistent with her previous .; system of nonagency.

EXPLANATORY SYSTEMS IN ORAL LIFE STORIES

359.

(16) Speaker: I chose to flunk out of a masters program in psychology, Uhhuli;^ That was a choice point. . /•••^\-i^ru^${^ Interviewer: What do you mean, you chose to flunk out? • • - ].,'*;,:,^-> • Speaker: I got one too many C's. ..-....,, :•.>—» ^st-^^^''-'--• Interviewer.'Why? , , •^•^•^:-,K-hy>M^'--./':' " Speaker: Because I just didn't know that I w~. I wasn't i n ^ e ^ t ^ ^ | 3 u ^ ; : i ^ : ;; couldn't bring, 1 had this big struggle with myself as.to'' why;'l,]was''dpmg';';; what I was doing and I decided well I, I just have to do thisbeckuse/l"^^. ",';•;• supposed to do this, and it just didnt, it was, well, there wasn't ahy^ rewards" in it for me. And uh it's an interesting choice because I could haVepiclcecl^ myself up and left, which I didn't do, I just sat there and waited for "it!: V to all come crashing down on me until "Oh jolly!" So, but that,-but'. definitely a choice that I made somewhere in that, in that interim. ;

To speak of choosing to flunk out of a program by doing nothing until one is thrown out is clearly a subversion of the ordinary sense of choice. For a speaker operating within a psychoanalytic belief system, such a formulation might represent a recognition of personal responsibility for events that might ordinarily be taken as beyond one's control. In this case, however, I believe the speaker is denying my formulation of life as motivated by choices that occur at choice points. It would be extremely difficult, both conceptually and in terms of etiquette, to make explicit and then deny my presupposition that choices exist and are relevant; rather, she has implicitly redefined the sense of choice until it can be used in a story that does not conflict with her system. Comparing this speaker's explanatory system to the popular writings of Skinner (1948; 1979), the two are in agreement. The one striking difference is that Skinner's model provides a place for the agent, the force that arranges the contingencies of reinforcement. In the laboratory, of course, this is the experimenter. In the ideal community, it is a planner, or a board of planners, who experiment with various social and physical arrangements that will produce happy and good people. In the ordinary state of affairs, the environment rather haphazardly supplies'positive and negative reinforcement, which, in Skinner's view, would be better supplied by a more conscious agent. In the examples we have seen, and in the entire life story given by this speaker, there is no sense that reinforcements are provided by a particular agent. She speaks of the reinforcement process as something sought by the person to be reinforced, but in a rather passive way, something like Skinner's notion of shaping a behavior. (17)

. . . and the whole process of finding what it is that gives you your strokes OK? is not an easy one.

(18) So that's what you're asking about, the process of finding out what is positively reinforcing to you OK? is not an easy thing to do and I don't know how people hit upon . . . .

360

CHARLOTTE LINDE

This is related to Skinner's notion of the environment as agent, except that she has not formed any notion of the environment as an entity that acts on her. A number of questions remain. One concerns the rarity of this ex>" planatory system. Since only 1 speaker out of 14 in the sample uses this behaviorist explanatory system, is it justifiable to posit it as an explanatory system present in the culture, or should it rather be considered an idiosyncracy of this speaker alone? This system is unique in the data of this study, and I believe that even in a considerably larger sample, it would be quite rare. However, although this explanatory system sounds somewhat odd, its relation to a known expert system is recognizable. Furthermore, it is comprehensible; we may be surprised by the sorts of things the speaker cites as reasons, but we understand why they count as reasons. This is in strong contrast to fully idiosyncratic explanatory systems, whose connections are incomprehensible. We may also ask how this speaker came to use such a rare system. This is a tantalizing question, but if it is answerable at all, it requires psychological and biographical information that lie beyond the scope of this study. ASTROLOGY

Another explanatory system present in these data is astrology. This system is used only Jokingly and only in the relaxed later part of the interview as it modulated into a flirtation. This is not surprising. Since astrology is a less respectable explanatory system than the others we have examined, it is less likely to turn up at all in an interview situation. Therefore, only traces of the system are present; it is impossible to give a listing of all the themes present in the system. The most striking use of astrology is to give an explanation of character and character differences. It is used, both in the examples below and in casual uses I have heard, to provide an understanding of why an incomprehensible or obnoxious character trait cannot be changed, or why two otherwise reasonable people cannot get along with one another. Unlike psychological explanations of character types, astrological types are not presented as pathological and, therefore, there is no demand or expectation that the person of a particular type could or should change that type by becoming better adjusted or more mature. The following examples of use of an astrological explanatory system are taken from an interview with a single speaker. (19) Speaker: Yeah. The one thing that I don't miss from the East Coast are the arguments. Interviewer: (Laughs) Yeah. Speaker: OK. (Laughs) Talking, yeah, uh but the arguments, and almost the creation. It's sort of like, my mother does that too. She's really beautiful. She called me up one night and she said "How are things?"

EXPLANATORY SYSTEMS IN ORAL LIFE STORIES

361;

And I said "Fine, you know, kind of beautiful, right in the middle.^'^ch:>; :: is where I like things, being a Libra. And she says, "Oh, I can^t^s^dV;.;;>••: them when they're that way." She says, "I gotta have some excite^enjy':::;<[:':\. in my life." She says, "I noodge your father a little bit to,^£^"^^ , , ] J!•..^^> , : . , : creates an argument just to get a little stimulation. Yeah. I, IhaVs •not;/. ; for me. That's for my m-, who's a Scorpio, my mother is a Scorpio, so^: : ; -: (20) Speaker: What's your sign? '-V '!:'•";;%:.• Interviewer: Uh, Capricorn. •• •} '1 '-Speaker: Perfect. Methodical. They like it to go right. Intelligent. YeahTInterviewer: I'm not methodical. '•'••:•!?'• Speaker: You're not? Your mind works that way. Interviewer: Hm. (Laughs) You should see my drawers. Speaker: The hell with the drawers. (Laughs) Your mind works very methodical. I have a very close friend that's a Capricorn. And he uh, you know, he, everything is, he does it right the first time. He can't do a harassed job. It'd bug the shit out of him. He figures, you know, and I agree with him, too, that the easiest way is the right way. Interviewer: Capricorn has always struck me as one of the more boring signs. Speaker: Hmmmm. Not to me. Well, I dig minds, OK. So minds have uh, uh, they turn me on. I dig when I I'm talking to people, somebody intelligent. These examples represent a popular form or popular use of astrology as an explanatory system. In the more serious use, it presents, as we have seen, an explanation and reconciliation of character types. Its use as a flirtation device functions both by rekeying the talk on a less serious level than that of interview and by providing a legitimate way of talking about the other person's character, Like the other explanatory systems discussed in this study, astrology has in addition to its popular form an expert version. Like other experts, the astrologer may provide not only an account of character but also advice on how to live.

The relation of popular to expert explanatory systems Thus far, this study indicates the existence of a number of popular explanatory systems and claims a relation to corresponding expert systems. It is now necessary to consider the nature of this relation. The discussion first examines the synchronic relation between a popular explanatory system and its corresponding expert system and then considers the historical development of popular explanatory systems from expert systems. The discussion uses popular Freudian psychology as a model, since this is the explanatory system most fully represented in the data and about which there is the most historical evidence. It should be emphasized that I do not claim to have given a full picture of the Freudian explanatory system as it is represented in popular thought. Rather, I am attempting

362

CHARLOTTE LINDE

to demonstrate the existence of popular explanatory systems and to untangle some of the relations between popular and expert systems. SYNCHRONIC RELATION BETWEEN POPULAR AND EXPERT EXPLANATORY SYSTEMS

Let us begin by considering the current relation between the popular and expert Freudian systems. The first point is that the popular version uses a very small number of the concepts present in expert Freudian psychology. Most notably lacking are any references to theories of sexual development and sexual functioning. These are, in fact, the concepts that first come to mind in considering the influence of Freud on popular culture. However, the topic of the interview - professional history - does not encourage discussion of sexuality, although a number of speakers do mention it obliquely, as we have seen. There are also many other Freudian concepts not present: the Oedipus complex; the tripartite structure of ego, superego, and id; and the stages of psychosexual development, for example. This lack appears to be characteristic of all relations between popular and expert explanatory systems. A professional astrologer looking at the popular use of astrology as an explanatory system would perceive a similar impoverishment of the system in including only birth sign and failing to consider rising sign, moon, and so on. A second point is that isolated concepts have been borrowed, but not the entire system as system. The concepts of popular Freudian psychology do not require the dense interconnections of Freud's argumentation. Similarly, we may say that the difference between a popular use of astrology and a serious professional use comes in the drawing up of an astrological chart, which permits the expert to investigate patterns, rather than relying on isolated facts about individual signs. Finally, and perhaps most important, those concepts of the expert explanatory system are included that do not contradict other popular theories of the mind and the reasons for human behavior. That is, Freudian psychology, as it filters into popular thought, is an expert explanatory system that supplements other more widespread theories, rather than supplanting all of them. THE HISTORICAL DEVELOPMENT OF EXPLANATORY SYSTEMS

This section considers the process by which an expert explanatory system can develop, give rise to popular explanatory systems, and finally become part of common sense. To understand this process for any given explanatory system would require extensive research in intellectual history, research that has, for the most part, not yet been done. Therefore, the current discussion is restricted to a schematic model of how such historical development might occur. Figure 14.1 shows such a model. The first part of the process is the move from common sense to expert theory. This is the point at which the originator of some expert model

EXPLANATORY SYSTEMS IN ORAL LIFE STORIES EXPERT THEORY

363

PRIOR EXPERT THEORY .

EXPERT THEORY'

^EXPLANAfoRYS^STEiV]

•cr&w-x-

COMMON SENSE THEORY

Figure 14.1. Historical development of an explanatory system of the nature of mind, human behavior, and so on begins the development of the model. The new expert model arises within a context of currently held ideas, both those of common sense and of already existing expert systems. Much of the new expert system will be in opposition to common-sense theories. At the same time, there must be common-sense constraints on the new theory; too entirely radical a theory of mind could not be recognized as a theory of mind at all. Other relevant expert theories extant at the time will also have an effect on the type of models that can be constructed. We now turn to the second leg of the diagram in Figure 14.1, the move from expert system to explanatory system. This is the process by which this diffusion of ideas takes place from expert into culture. As mentioned, the full expert theory does not become part of this explanatory system; only a limited subset of ideas is chosen. How this selection takes place is a fascinating question. It seems likely that the common sense of the period affects the selection so that the most radical and startling portions of the theory are not chosen. Finally, we turn to the last leg of the diagram, the move from explanatory system to common sense. As a given explanatory system becomes better known and more widely held, it begins to move closer to common sense and may eventually come to form a part of common sense. As an example, we take the notion of the Freudian slip, which seems to be a part of the general, common-sense body of accepted notions and which does not require the support of the Freudian explanatory system to be comprehensible. To document the historical process suggested in Figure 14.1, extensive research in the history of popular culture would be required. Such research

364

CHARLOTTE LINDE

is not currently available for any of the explanatory systems discussed in this paper, although there is some'related research on the history of the development of Freud's ideas.4 / * In the absence of such detailed reseach, this section has attempted to indicate a model for such historical studies and the value they would have in understanding both explanatory systems and common sense. Issues in linguistics and anthropology This section discusses some of the questions about explanatory systems that may arise in linguistics and anthropology and sketches directions for further research in this area. LINGUISTIC ISSUES

From the point of view of linguistics, this work attempts to show that one component of oral life stories is the explanatory system, which furnishes a system other than that of common sense by which the actions narrated can be understood and justified. A possible criticism arising from this point of view is whether this analysis makes too much of the available evidence. That is, we have seen speakers using such phrases as "strokes" or "reinforcement" or "deeper reasons" or "Libra." Since these are common in ordinary conversation, what justification is there to use them to assert the existence of as complex and hypothetical an entity as the explanatory system? One argument for the postulation of* the explanatory system is that speakers do not use metaphors from several systems. That is,we do not find a single speaker using both Freudian and behaviorist metaphors. Since the mixing of metaphors is a common phenomenon in speech, when we find metaphors kept separate in this way, it argues that they come from separate systems, which, in some sense, are known to the speakers. A second argument is that in narratives exhibiting the effects of a particular explanatory system, the details of construction at the level of the sentence and the entire narrative are consistent with the content of the explanatory system claimed for the life story. This effect is particularly evident in the behaviorist life story examined earlier; however, it is present to some extent in all life stories that use an explanatory system. These two arguments together suggest that however common the metaphors of these explanatory systems may be, they do indicate the existence of a number of systems, rather than an unstructured collection of concepts that can be used promiscuously and without regard to coherence. CULTURAL ISSUES

From the point of view of anthropology, or of social history, a number of questions are raised that cannot be answered within the scope of this

EXPLANATORY SYSTEMS IN ORAL LIFE STORIES

':'•'•/Mi'-.':'•

paper. One issue is how many explanatory systems are present in.-a^^^S^culture, arid what they are. This question could &$^0$ffffi0ii$?\%;'/'l/\; sive study of a number of types of discourse on;&punjl^$f'diffej&n^B^;^./. topics, as well as an analysis of the most widely disseminated i$^pfw^pi-^Q'•'' ular culture. A second issue is the relation between ^ Q p M ^ : « ^ . ^ ^ , ; f v ^ ^ ; ' forms of an explanatory system. As already discussed, the 'irebduHT^utd^ be investigated using the usual techniques of social )^g^.^t)^^^wi';..^r. "': is more psychological: How serious are these explanatory systems fprtheir^^^; .: users and how are they related to their behavior? This is a more difficult-; question, but it could perhaps be approached using a mixture of psychological and participant observation techniques. .; ,' These questions, I believe, indicate interesting and valuable directions for research. The present research indicates the existence of a level of conceptual organization that is necessary to understand the structure of life stories. This is already an extremely important finding since it allows us to understand one mechanism by which people make sense of their own lives. Notes 1. I am grateful to Eleanor Rosch and Dan Slobin for introducing me to autobiography as an area of investigation. My interest in discourse analysis and my understanding of it are due to William Labov. I have also been greatly aided by discussions with Joseph Goguen, A. L. Becker, Veronika Ehrich, Gelya Frank, Nathan Hale, George Lakoff, Willem Levelt, Livia Polanyi, Naomi Quinn, Larry Selinker, Sandra Silberstein, and James Weiner. My primary debt is to the people who told me something of their lives and whose stories are here treated dispassionately as objects for analysis. This work was partially supported by Office of Naval Research, Contract Number N00014-82-K-07U. 2. The examples of this paragraph have been constructed by the author. 3. The data discussed were gathered from 1977 through 1979.1 have noted that starting in 1980, as economic conditions worsened, stories of professional choice have begun to include at least a mention of economic circumstances, although they still do not mention social class. 4. For example, Elienberger (1970) discusses the social history of the development and spread of Freud's thought. Sulloway (1979) discusses the relation of Freud's theories to the theories of biology extant at thattime.Bakan (1958) discusses Freud's debt to the Jewish mystical tradition of linguistic interpretation. Hale (1971) gives a history of the acceptance and modification of Freud's thought in the American professional psychological circles. AH of these works, though, concentrate on Freud's relation to other expert systems and to his acceptance by other experts. There is no systematic history of the effect of Freud's thought on the general public. References Bakan, D. 1958. Sigmund Freud and the Jewish Mystical Tradition. New York: D. Van Nostrand Company.

366

CHARLOTTE LINDE

Becker, A. 1977. Text building, epistemology, and aesthetics in Javanese shadow theater. In The Imagination of Reality, A. Becker and A. Yengoyan, eds. Norwood, N.J.: Ablex Publishing Corporation. Pp. 211-243. ' .-.,'. ~\ t* Ellenberger, H. 1970, The Discovery of the Unconscious; The History and Evolution of Dynamic Psychology. New York: Basic Books. i Frank,G. 1981. Venus on wheels: The life history of a congenital amputee. Unpublished Ph.D. Dissertation. University of California, Los Angeles. Geertz, C. •• •-••• • • -•• 1973. The Interpretation of Cultures. New York: Basic Books. Hale, N . J . 1971. Freud and the Americans, Vol. 1, The Beginnings of Psychoanalysis in the United States. New York: Oxford University Press. Langness, L. L. and G. Frank 1981. Lives; An Anthropological Approach to Biography. Novato, Calif.: Chandler and Sharp Publishers. Lasch, C. 1979. The Culture of Narcissism. New York: W. W. Norton and Company. . Linde, C. 1981. The organization of discourse. In Style and Variables in English, T. Shopen and J. M. Williams, eds. Cambridge, Mass.: Winthrop Publishers. Pp. 84-114. n.d. The Creation of Coherence in Life Stories. Unpublished manuscript in preparation. Structural Semantics, Palo Alto. Schur, E. M. 1976. The Awareness Trap; Self-Absorption Instead of Social Change. New York: Quadrangle Books. Sennett, R. 1977. The Fall of Public Man. New York: Alfred A. Knopf. Skinner, B. F. 1948. Walden Two. New York: MacmtHan Company. 1979. The Shaping of a Behaviorist. New York: -Alfred A. Knopf. Sulloway, F. J. 1979. Freud: Biologist of the Mind. New York: Basic Books.

PARTY An

15 Models, "folk" and "cultural" PARADIGMS REGAINED? 1

Roger M, Keesing

Some ten years ago, I reflected in a series of papers (1972a; 1972b; 1972c; 1974) on the limitations of cognitive anthropology as then practiced: "ethnoscience," "ethnographic semantics," "the new ethnography." I argued that in its preoccupation with inductive rigor and cultural uniqueness, cognitive anthropology had followed the premises of American descriptive linguistics, even as the Chomskyan revolution had made these premises untenable in the realm of language. Analyzing the semantics of folk classification on the promise that the methods and models so developed could be extended so as eventually to produce a "cultural grammar" was, I thought, nai've. The focus on artificially simple and simplified problems (especially the semantics of kin terms in their genealogical senses) had deflected our attention from the deep complexities of meaning and context and deep questions about the rule-govemedness of social behavior. I pointed to the incredible complexity of the organization of mind and brain, as then beginning to emerge in the neurosciences and artificial intelligence research. In this light, anthropologists' attempts to describe cultural knowledge seemed curiously unrealistic: Cognitive anthropology has so far been an Alice in Wonderland combination of sweepingly broad aspirations and ludicrously inadequate means. We have been cheerfully and optimistically using high school algebra to explore the most profound mysteries of the natural world. (1972c) Yet, my message in these critiques was not entirely pessimistic. In "Paradigms Lost," I noted that pioneers of the linguistic frontier such as Charles Fillmore and George and Robin Lakoff had made impressive incursions into the territory claimed by cognitive anthropology, and I suggested that paths connecting the two disciplines were opening up once more. In my 1972 University of Illinois lecture on "Toward a New Cognitive Anthropology," I noted that while proponents of "The new ethnography" were trying to characterize formally the cultural knowledge of native actors, artificial intelligence (AI) researchers were trying to write "cultures" for automata. There and in an assessment of kinship studies (1972b), I pointed toward a potential convergence between AI research

370

ROGER M. KEESING

and cognitive anthropology: Both were making parallel simplifying assumptions, and both were facing the same intractable problems of pattern and relationship; each had something to learn from the other. I further suggested that a nascent understanding of categorization in terms of prototypes, contexts, and fuzzy edges in anthropology (Berlin & Kay), cognitive psychology (Rosch), linguistics (Lakoff), and AI patternrecognition research could revolutionize the simplistic models of thenprevailing ethnographic semantics. That in 1983 a surviving little band of battered pioneers of the early cognitive anthropology could sit around a conference table with George Lakof f, with students of Rosen and Kay pursuing complexities of semantics in terms of prototypes and fuzzy edges; with Jaime Carbonell, who is trying to teach automata to reason analogically as human beings do; and with a new generation of cognitive anthropologists collaborating with AI researchers, suggests that my guarded optimism and glimpses of a new cognitive anthropology were not so far off the mark. But is all this more than collective optimism, fraught with the same old naivety? Do we simply have new catchwords for the 80s - "folk models," "prototypes," "schemas," "instantiations," "conventional metaphors" - that disguise the same old difficulties with new labels? Is the interdisciplinary venture genuine, or have we cognitive anthropologists just enticed outsiders into our wild goose chase? Or joined theirs? A detailed summary and retrospective of the past decade of cognitive anthropology would be tedious. However, some assessment of the new developments in relation to the past, tolhe future, and to the deep problematic of characterizing what native actors know and how that shapes what they do, may be worthwhile. An ambiguity in this task needs to be made explicit at the outset. My assigned task, at the conclusion of the Conference on Folk Models, was to assess retrospectively the papers presented in the conference and the theoretical concern with folk models that had been their focus. After an initial version of my paper had been circulated - indeed, my critique may have been one of the stimuli to the change - the concept of folk models was recast as cultural models by the editors. Moreover, the set of papers initially given and discussed at the conference was considerably reduced, and some key contributors to conference discussion are not represented in this volume. Was 1, then, to present a retrospective critique of a conference on folk models? Or, was I to summarize a volume addressed to essentially the same problematic, differently labeled, and with a substantially reduced set of papers? I have chosen a middle course, addressing myself mainly to papers included in the volume, but alluding where worthwhile to other participants in the conference and points raised there. I also look both at folk models and their mutated version, cultural models.

MODELS, "FOLK" AND "CULTURAL"

371

Cultural knowledge and cultural theory

;

I have argued (1974; 1981) that an ideational theory of culture does not commit us either to a monolithic view of "a culture" as a shared system of symbols or to a deterministic view of culture as directly generating behavior. An ideational theory of culture can look at cultural knowledge as distributed within a social system, can take into account the variation between individuals' knowledge of and vantage points on the cultural heritage of their people. It can also view cultural knowledge as shaping and constraining, but not directly generating, social behavior.:i'"••""' In their pursuit of cultural grammars, however, pioneers of the new ethnography had seldom dealt with variation and the distribution of knowledge. The cultural code was an idealized set of rules assumed (heuristically) to be shared; and despite the protestations about accounting for appropriate, not actual, behavior, a view of human beings as rulefollowing and appropriateness-maximizing had pervaded early cognitive anthropology. This orientation left little room for the processes of cocreation, negotiation, and contextual shifting on which ethnomethodologists and other social interactiomsts have focused. The assumption that behavior is pervasively rule-governed led in many cases to a strategy of inference whereby unconscious rules were posited to account for observed behavior, even where there was no evidence for the "rules" other than the behavior they were assumed to be generating. Has an emerging new cognitive anthropology transcended the limitations of a view of an idealized native actor mechanically enacting implicit rules? The past decade has seen an increased concern with problems of cognitive variability, which had been foreshadowed by Sankoff (1971), Wexler and Romney (1972), and others. This has been most clear in study of folk classification, as in Gardner's (1976:464) examination of cognitive variability and sharing and "the model of the omniscient informant." Have the frontiers been carried still further, toward an understanding of how the distribution of knowledge and perspective articulates with the structure of social systems? I see in the conference papers and volume some grounds for optimism, some for pessimism. In place of the idealized native actor, we are presented with "folk" and "experts." The folk, in turn, may have different models from one another (e.g., Kempton's alternative models of how thermostats work, the alternative metaphors of marriage explicated by Quinn, and the folk models of electrical current as water or lemmings discussed by Dedre Gentner at the conference). There thus is room here for variable versions of a common culture - and that I take to be an advance. But the folk/expert dichotomy, if we take it seriously, would seem itself to represent a folk model of society: and one which, I argue below, is itself fraught with dif-

372

ROGER M. KEESING

ficulties, Cultural models may steer us clear of some of these traps; but the reconceptualization raises further problems. The conception of human beings as pervasively rule-following is being transformed into a more sophisticated framework for looking at culture and behavior. The cognitive models explored in these papers constitute paradigms through which one places constructions on everyday experience; conventional metaphors represent simplified, transformed, culturally defined realities. But these are frameworks of interpretation, not sets of rules and routines native actors follow to maximize the appropriateness of behavior. There is room here for choice, for alternative constructions, for creativity. There is also an emerging concern with the creative social processes of co-construction of situationaliy shared and negotiated realities. Quinn's paper (this volume) and her other recent work vividly show how an understanding of a marriage is negotiated, fashioned together in (as well as from) ongoing interaction by husband and wife; Price (this volume) shows how cultural interpretations of disease are fashioned in the process of interaction. We are even learning that lexical semantic analysis may require reference to the situational context of speech, as Eve Sweetser's elegant analysis of the semantics of lie in English (this volume) illustrates. These developments and continuing advances by students of pragmatics in linguistics and speech act theory in philosophy make it clear that the static and deterministic models of early cognitive anthropology were sterile not because they were (or aspired to be) cognitive. They were sterile because, as preliminary mappings of what human beings know about their world, they were not placed within an adequate, multisided paradigm of how what we know articulates with the social worlds we create together (and which, dialectically, create us and what we know). This is a problem to which I return. In my review of cultural theory (1974) and more recent assessments of culture (1981; 1982) I have noted that a cognitive view of culture, while potentially allowing us to interpret the distribution and variability of knowledge and the situational co-construction of shared worlds, renders it difficult to capture the publicness and collectiveness of culture as symbol systems (Geertz 1972). As Varenne (1984:291) has perceptively written in critically assessing "individualist" theories of culture, the collective tradition of a people is in an important sense external and transcendant in relation to any individual; such a cultural tradition is always already there. It is in this sense that ideology, or culture, is an external social fact that is part of the environment of individuals. To the extent that it is part of the environment, it is something to which individuals will adapt and against which they may react. . . . This sense of externality and transcendance, which is easily lost if we either reduce a culture to the "head" of an idealized omniscient native actor or distribute versions of "it" through the "heads" of a population of

• -07?

MODELS, "FOLK" AND "CULTURAL"

native actors, is potentially captured more clearly in the xuytionM^Qlk models, cultural models, or folk knowledge (Clement 1982). These; modejs are at once cultural and public, as the historically cumulated knowledge • of a people and the embodiments of a language, and.;.cognjtiye,^as V•'[ paradigms for construing the world. • :• ^U- -Y^rn'M^f^P-----^ • Models, "folk"and "cultural"

^

^ #g:

The central concern with folk models at the Princeton conference and its Los Angeles predecessor represents a.confluence of several streams/of thought. One is the continuing quest in cognitive anthropology to c&pture the uniqueness of a culturally constructed world, now tempered with realism: If one cannot go from folk taxonomies boldly out to write a cultural grammar, one can usefully seek a partial but systematic model of a single target domain. Another stream of thought comes from Lakoff and Johnson's (1980) exploration of how conventional metaphors build on paradigmatic, experientially based models. The models embedded in and constitutive of language itself offer hope of systematic methods of inference; we can use as data materials less elusive than evanescent behavior, less directly constituted by ethnographic interviewing, perhaps, than the spuriously "hard" data of early ethnoscience. Another stream comes from semantic theory, particularly the prototype semantic theory pioneered by Eleanor Rosch (1973; 1975; 1977; 1978) and Brent Berlin and Paul Kay (Berlin & Kay 1969; Coleman & Kay 1981). These models of the cognized world build on prototype relationships in a number of ways that promise to make prototypy, along with conventional metaphor, a pervasive organizing principle. We find in these papers not only prototypic or canonical exemplars of categories (the chairiest chair), but also the prototypic idealized interactional situation (the purely informative speech act) and the prototypic sequence or scenario (an idealized family structure or life history). Finally, the concern with models of everyday reality in cognitive anthropology parallels, and is closely tied to, an emerging concern with naive theory or mental models in cognitive psychology (see, e.g., Gentner & Stevens 1982; Johnson-Laird 1981; McCloskey, Caramazza, & Green 1980), a tradition represented here by Collins and Gentner (this volume). The focus on folk or cultural models by cognitive anthropologists and fellow cognitivists in kindred disciplines raises a number of theoretical and methodological issues. I look at most of them in the sections to follow; some bear examination at this stage. A first question, which I raise here and touch on again in the next section but do not attempt to answer systematically, is how we are to define folk or cultural models so they usefully delimit some sectors, but not all, of the cultural knowledge of individuals. What are such models? And what

::

374

ROGER M. KEESING

do human beings know that does not comprise these models? There are at least two important dimensions here: One concerns their sharednessy as folk or cultural constructions; the other concerns their "mo&el-ness." The original characterization of these models as "folk" did not reflect their adherence among the untutored masses,2 but their common-sense nature. Such models comprise the realms of (culturally constructed) common sense. They serve pragmatic purposes; they explain the tangible,-the experiential (hence perspectivally egocentric), the probable; they assume a superficial geology of causation; they hold sway in a realm in which exceptions prove rules and contradictions live happily together. I return to the experientially based, culturally conventional commonsensicality of such models. What, then, makes them "models"? Presumably, it is their paradigmatic, world-proposing nature. These cultural constructions of the everyday world do not consist of disconnected bits of cultural wisdom, expressed in precepts, parables, proverbs, or pragmatic* probabilistic operating strategies, but of the world-proposing (however simplified and internally or externally contradictory) models embodied or expressed in these bits. Such models, then, are not presented to us in what everyday people say and do in their everyday lives, or in the stuff of metaphoric talk; they are represented, in fragmentary surface facets. We must infer the more coherent, if unarticulated, models that lie beneath (as we infer native actors must, in learning them). How such models of everyday reality relate to other modes and realms of cultural knowledge then becomes a central question. How diverse are such models in different culturally constructed worlds? What constrains their diversity? This second question is raised by D'Andrade (this volume) at the end of his superb paper on American models of the mind. In many realms, folk conceptualizations - of emotions, time, mental processes, and so on - are turmng out to be more similar than might have been expected (especially in the light of anthropological dogmas about cultural uniqueness and the relatively arbitrary nature of cultural constructions). Conventions for talking about mental processes are different on Ifaluk and in the United States in ways congruent with differences in cultural precept and practice, but they are not very different - more shadings of value and emphasis than unique conceptualizations. If that turned out to be true of Balinese and Tikopia and Eskimo, as well, their varied ways of saying the same thing could not necessarily be taken as expressing deep truths about the cognitive organization of information. It would, I think, tell us more clearly about the experiential nature of consciousness, the way members of our species experience and perceive the operation of mind/brain. Loosely speaking (with apologies to Paul Kay), it tells us not about the nature of human heads, but about how it feels to live inside them. A third question, implied by D'Andrade (this volume) and raised more squarely by Linde (this volume), is the degree to which scientific discourse builds on and with such common-sense models. The physical sciences have

MODELS, "FOLK" AND "CULTURAL"

•375

progressively penetrated through and beyond everyday, commonTsense models of experienced "reality." Yet the physicist leaves subatomic^par^ tides and relativity behind, the mathematician non-Euclidean geometry* as they enter the parking lot and drive home through a world of seemingV ly solid objects, flat surfaces, and straight lines. Moreover,'conventional/\ :7 metaphors and common sense cannot be expunged from the natural languages to which the most precise scientists must have recourse {Kuhri .: 1979). Disengaging what passes for behavioral science from "folk" models and conventional metaphors is hopeless: Most of psychology (and,!we could add, much of sociology and anthropology) reflects common-rsense cultural models of the mind and reified metaphors. r Are "folk" or "cultural" models cognitive? Granted that conventional models of everyday reality are a worthwhile focus for anthropological concern, why should they be analyzed cognitively? Or, put another way, anthropologists have been characterizing such models for decades as (elements of) other peoples' cultures; nowadays, our symbolist colleagues interpret them as public systems of shared meaning. What is gained by interpreting models of everyday reality as cognitive codes rather than as cultural texts? Let me exemplify with the work of Clifford Geertz, who watched with some wonder at the Princeton conference as cognitive anthropologists explored the realm of culturally constructed common sense as if it had just been discovered. Much of Geertz's work has explored this realm in cultural terras. Consider, for example, his brilliant essay in which the cockfight as cultural text is illumined by Balinese models of society and sociality (1972). Consider the depiction of the Moroccan scenario of Cohen and the sheep (1973) as articulation and negotiation of cultural models. The comparative interpretation of conceptions of self and person in Bali and Morocco (1966; 1974) depicts culturally constructed models of the most fundamental takens-for-granted about human-beings-in-the-world. Finally, Geertz's essay on "Common Sense as a Cultural System" (1975) explores conceptions of personhood, sociality, sexuality, and other realms of the everyday. There is a double twist to cognitive anthropology's emergent concern with models of everyday cognition - with folk models of the mind, of emotion, of mundane experience. A good deal of Geertz's work, and, of course, much of phenomenology (Husserl, Schutz, Heidegger, MerleauPonty), is precisely about models of everyday cognition: about what thinking, remembering, understanding, seeing, communicating, getting angry, the passage of time are, experientially, to human beings; and, more or less explicitly, how they are culturally constructed, how they are shaped by language and metaphor. What, then, is specially to be gained by characterizing models of everyday experience in cognitive terms? Are putatively cognitive accounts of models of the mind more perceptive,

376

ROGER M. KEESING

systematic, or powerful than, say, a psychoanalytically informed culturalsymbolist account such as Paul's (1976) analysis of the Sherpa temple as a "model of the psyche"? Do we gain from our cognitive assumptions and , = commitments? Or, do we achieve, perhaps more pretentiously and less. gracefully, what is achieved through the word-magic of a Geertz depicting personhood in Bali (1966) or a Schutz (1962; 1967) or Natanson (1967; 1970) depicting the phenomenology of self and role? ---v" •=• ~v*-v--The papers in this volume provide ample evidence that there are wide differences, theoretical and methodological, between symbolists and contemporary cognitivists; and that the cognitivists are achieving important new insights. Where symbolists mainly view their task as interpretive, cognitivists mainly view theirs as scientific. For the latter, the primary data - transcribed tapes of interviews or other discourse or other "hard" records of the behavior of individuals - are crucial, and the inferences drawn from them systematic and explicit. The naive inductivism of much early ethnoscience is tempered now with realism and appreciation for interpretive insight as well as with rigor; however, concern for methodological precision emerges clearly in these papers. It is misleading that Geertz's metaphor of culture as text to be read suggests close convergence: The texts of the cognitivist exploring models of everyday reality mainly comprise what individual subjects and informants said and did. Inevitably, the perspectives these approaches yield are quite different. Where symbolists find coherence and systematic structures, cognitivists are likely to find in their primary data contradictions, alternative constructions, small and partial coherences, individual variations. This difference in perspective does not rule out our finding global structures or systematic models underlying apparent diversity - witness D'Andrade's model of American models of the mind or Lakoff and Kovecses's model of models of anger embodied in American English (both in this volume). It is perhaps no coincidence that the most coherent and global models of everyday reality described in these chapters are the schemas built into a language. Even here, however, as Kay's paper reminds us, there are alternative if not quite contradictory models expressed in conventional usage. Cognitivists' engagement with such models does take them into domains already extensively explored by symbolists and others; but the insights this exploration is beginning to yield are complementary to, and in some important ways corrective of, those discovered along different paths. I am concerned that this exploration takes place with adequate appreciation of how extensively mapped this territory already is and of the insights to be gleaned from earlier work. Cognitive anthropology grew up curiously innocent of social theory; it need not remain so in its maturity. Past explorations of everyday cognition and common-sense constructions of reality include not only those of the phenomenological tradition and its offshoots (Husserl, Schutz, Heidegger, Merleau-Ponty, Foucault, Ricoeur) but also those of the Marxist tradition and its offshoots (Marx, Gramsci,

MODELS, "FOLK" AND "CULTURAL"

flV-\

'

Althusser, Bourdieu). In this Marxist tradition, the realm of comnwh&enp-'^:' .is viewed as refracting as well as reflecting, disguising as well : ^ ^ ^ ( f ; ; iuminating, shaped by as well as shaping the realities of a social wbfld^::;.s I am also concerned about what I see as a problematic theoretical:Sfs*it: junction between folk models as cognitive structures M d ^ ^ ^ ^ ^ a ^ l ^ ^ - ^ v V , ; ' / ; ' cultural patterns. Early cognitive anthropologists radically oversimpIifiecUf: matters by reducing "a culture" to an idealized cognitive systemv'However^l.:;- •••' the solution proposed by Clement (1982) - that we see folk knowledge' : as having its locus "in the group" and not in the individual - seemsItoTJJ::: . pose another set of dangers. Clement contrasts her concern with folk knowledge with the earlier assumptions of ethnoscience in these terms: In . . . ethnoscience, attention is focused on cultural competence . . . mentally held principles and recipes that the individual has learned which allow him or her to behave in a culturally appropriate manner. . . . In contrast. . . , folk knowledge is viewed as an aspect of the group. Folk representations, the means through which folk knowledge is expressed, are . . . products of the institutionalized patterns of information processing and knowledge distribution within the group. (1982:194) If what we study is collective, however, if its locus is "in the group" and not in "mentally held principles and recipes," can we usefully regard ourselves as cognitivists and effectively and legitimately borrow models and methods from the cognitive sciences? What we want to study, I think, ultimately is how individuals cognize and use models they (partly) share with others, models that are common coin in the community. That will require that we see folk models both as collective and social and as "mentally held principles and recipes." Perhaps, then, we need to distinguish between models (/), the pool of common-sense knowledge and understandings of the community, and encoded in its language; and models (//), the (alternative, partial) versions of these models cognized and invoked by individuals in everyday perception, thought, and interaction. If early ethnoscience erred in imagining that we could adequately characterize level (/) as an idealized version of level ill), we equally err, I think, if we try to characterize level (/) using theoretical frameworks derived from and appropriate to cognitive science (chunking, instantiations, etc.). It is true that the pools of ideas in communities can include only ideas that are thinkable and knowable by individual humans. As systems, however, these pools are not subject to many other constraints that structure the cognition of individuals, and their dynamics are quite different. We can be wiser cognitive anthropologists if we distinguish these two loci of conceptual models and levels of analysis and engage ourselves with the nature of each and the interplay between them. Replacing folk with cultural in designating the models we are concerned with avoids some of the problems raised by the term folk (and the folk model of society it itself embodies) to which I pointed in earlier ver-

378

ROGER M. KEESING

sions of this commentary. In locating the models as collective and shared, however, it perhaps renders more problematic than folk did attempts to analyze such models as if they represented the cognition of individuals. Perhaps an example, to which I later return, will serve to clarify. Consider the folk models constructed around the concept of luck in American culture as a set of public, largely shared conventions for talking, for explaining outcomes, for characterizing good or bad fortune. Linguistic conventions ("some people have all the luck/' "thank my lucky stars") and cultural icons (rabbits' feet) are cultural coin of the community. Yet, at a level of individual cognition, these cultural models of luck (folk models /) are drawn on in quite different ways. When I talk about having all the luck on the tennis court on a particular day - the day when all my letcord shots dropped over and the occasional desperate lunge yielded a winning drop volley off the racket frame - I do not surmise that my "luck" was anything but random chance. Other people, on the tennis court or in poker games or reflecting on the trajectory of life, invoke (I think) luck with a conviction that the outcome of events is ultimately determinate, or at least skewed in favor of some, against others. Some people carry rabbits' feet or other amulets, wear "lucky" headbands or use "lucky" putters, convinced that this will affect the outcome of events. Individuals perceiving contradictions between their "rational" and magical beliefs may invoke the folk concept of superstition to rationalize their conjunction. In the cognition of individuals, conceptions of luck may be central (e.g., to a gambler) or peripheral, may be invoked with metaphysical implications or without, may be "believed" or not. The same conventional linguistic expressions and metaphors regarding luck may be used by different speakers of American English with radically different implications: The folk models (I) embodied in language structure our talk whatever our personal folk models (II) of luck. In the cognition of individuals, folk models (II) articulate with, exist side by side with, evoke and are evoked by other systems of folk models drawn from the pool of the community in diverse ways. To take another example, my color vision differs from that of the majority in the community in that I am neurophysiologically incapable of making color distinctions (greens vs. reds, in some perceptual environments) made by normal members of the population. I and a substantial minority are constrained to use a conceptual system we lack the perceptual equipment to use properly, or at least conventionally. Like other people in the population, however, we turn green with jealousy, red with anger; we use folk models (/) of the community as best we can. If cognitive anthropologists are to pursue more deeply and more formally the place of folk models in the mind, not simply o/the mind, then some further questions are in order. One characteristic of such models, or of some of them, is their probabilistic and partial nature. They are, as elements of culturally constructed common sense, models that work

MODELS, "FOLK" AND "CULTURAL"

379

well enough for everyday use. But human beings, operating in a universe of unique constellations of events, must deal with the atypical,: the improbable, the unexpected - not simply with ideal, types, canonical circumstances, the probable, and the normal. As Haviland (1977) andJRah-.•;.'•' dall (1976) observe, we must have ways of engaging, construirig,;arid acting on circumstances our rules of thumb do not aptly comprehend - despite our high tolerances for living with contradiction, for not only holding. mutually incompatible models but also for maintaining models in the face of their demonstrated inadequacy. ("The exception proves the rulej";as Geoff White's proverb [the volume] might remind us.) 1 But do we have, in addition to the models of culturally constructed common sense, deeper grammars to which we have recourse in the face of the atypical and complex? Do common-sense models, enabling us to deal with the world probabilistically on the basis of typical scenarios, "canned" or formulaic routines, and ideal types, operate in conjunction with cognitive processes whereby we decompose situations into their deeper constituents and interpret them with an underlying grammar when we have to? Or, do we simply muddle through by extrapolation, approximation, and ad hoc invention? I raised these questions more than 15 years ago (1970) in looking at what I took to be an underlying grammar on which folk sociological models of roles were constructed; and I reflected again on the adequacy of such an inferred grammar when, five years later, I was assessing my exploration of roles in relation to the state of play in cognitive anthropology: •-,v.. .-.. ..,••. One facet of the miraculous and still mysterious process of everyday, mundane social life is the way we act, and perceive others to be acting, in culturally standardized capacities. We enact roles: or so social science theory, drawing on the theatrical metaphors of folk sociology, would have it. (1975:5) But since no adequate description of roles had ever, to my knowledge, been formulated - and certainly not in a non-Western culture - I could not assume that the metaphor of 'role', derived from our folk sociology, would turn out to fit the way the Kwaio [of the Solomon Islands] organized their knowledge of their social world. 'Social identity' and 'role' might not turn out to reflect the deeper and unconscious levels at which cultural algorithms were coded. (1975:9) I had provisionally inferred, using the method described in my 1970 paper, that the labeled more-or-Iess standardized capacities in which Kwaio acted were often composites of underlying elements. This conception of social identities as fractionated elements from which we formulate social personae has . . . advantages analytically and theoretically. The elements, if they have any cognitive salience, are ways of organizing knowledge at an unconscious level. . . . In this view, social identities serve as "building blocks" or elements that, by "grammatical" rules of combination, are combined into composite social personae with

380

ROGER M. KEESING

predictable role entailments. We are conscious of, and often label, typical high-probability combinations. They are the *pre~packaged' units of a culture. Navigating probabilistically in our social world, we expect physig cians or attorneys or gas-station attendants to be male, and expect fathers to give their daughters away as brides; yet we can deal in culturally ap-:;:: propriate ways with less familiar and less typical combinations. (Keesing > 1975:16) _M. .._,•_, Some are improbable, some rare, and a few probably unique. The point is that instead of taking a small set of ready-made social identities as the basic units of a cultural inventory, we would view our actors as continually creating 'social personae'... out of the elements provided by the culture. (Keesing 1970:434) My point is that a cognitive theory of folk models, as culturally constructed common sense, would perhaps in the long run not want to take these models as representing cognitive organization but as representing a set of operating strategies for using cultural knowledge in the world; they comprise sets of shortcuts, idealizations, and simplifying paradigms that work just well enough yet need not fit together without contradiction into global systems of coherent knowledge. A cognitive theory of such models of everyday reality would, I think, ultimately want to take them not as the constituents of cultural knowledge but as their surface manifestations. (I have similarly argued that, contra the early ethnoscience, an adequate cognitive theory would not assume that lexemes or taxonomies are the fundamental constituents of cultural knowledge but rather would see labeling and taxonomic classification as emergent phenomena to be explained in terms of more fundamental constituents and processes [Keesing 1970:444-45; see also Randall 1976].) Against this line of speculation can be counterposed an argument, articulated by Quinn and Holland in their introduction, that it is the simplifying power of the models of culturally constructed common sense, in codifying the prototypical, that allows human beings to cognize such vastly complicated natural and social worlds. The problem of how we human beings deal - whether by extrapolation, analytic decomposition, feature-weighting, flexible modes of pattern matching, or whatever - with the atypical, the marginal, the fuzzy, or categorically liminal - has received little attention in cognitive anthropology to this point. It squarely confronts AI researchers, notably in pattern recognition; and it faces us in various forms when we work with prototype analyses, whether in semantics or other realms. (See, for example, Kempton 1981 and Kronenfeld, Armstrong, & Wilmoth 1985.) Cognitivists in anthropology and kindred fields need to confront these problems, at once formal and substantive, more directly. It may be, as Lakoff (1984) has recently argued, that explorations of the logics and patterns of prototype categorization will in the long run not only revolutionize

MODELS, "FOLK" AND "CULTURAL"

'381 ;

our understanding of cognition but also demand refonriulatioribtihe basic Aristotelian premises of western logics. My point is that if we afegbjtig to be cognitivists, in relation to models conceptualizing everyday reality, we need to be good ones, to explore deeply the way cultural ;fcnqwledge. is organized and applied to the world. . ''':/:•,^'r^;'^^^5S,^•,;;"::, If our models of other peoples' models are to fit into the emerging conceptualizations of cognitive science, and at the same time are to fit into; the wider anthropological enterprise, we need to be asking strategic questions about the structure of cultural knowledge and the way it is used in ongoing social life. •Culture, memory, and behavior I noted a decade ago that despite a common boundary beween cognitive anthropology and cognitive psychology, a curious gap had arisen between them. Pioneers of cognitive anthropology such as Frake and Conklin had read and learned cognitive psychologists such as Miller, Galanter, and Pribram (1960). But especially as psychologists probed memory experimentally, memory as conceptualized psychologically and culture as conceived anthropologically had become strangely separated. Short-term and longterm memory represented "real" cognitive processing capacities; but what was remembered was artificial, an artifact of experimental procedures. Human beings, of course, remember particularities -people, places, experiences; but they also abstract from particularities to derive general rules, models of meaning, expectations and strategies and routines. These abstractions from experience, rules for deciding what is . . . , what can be . . . how one feels about it . . . what to do about it . . . and . . . how to go about doing it. (Goodenough 1961:552)

comprise cultures, as cognitive systems. Early attempts in artificial intelligence and cognitive psychology to create models of memory dealt mainly with memory as a psychological construct - that is, with remembered particularities (see Norman & Rumelhart 1975). But as AI models were pushed further, the need to simulate the "internal models of reality" (Gregory 1969) on the basis of which complex organisms perceive and act became clear. Automata had to be programmed with "cultures"; and much progress in AI in the last decade has been in this direction. I return later to the growing convergence between cognitive anthropology and artificial intelligence. My concern at this stage is to look more closely at the interplay between memory of particularities and generalized "rules" in ongoing behavior. In 1974,1 had raised the question of the degree to which the knowledge human beings use to act in the world is generalized. Could it be that "the

382

ROGER M. KEESING

cultural code" is as much an artifact of our assumptions and methods as "memory" was for experimental cognitive psychology? It remains an open question to what degree human action is guided by a general conceptual code, a theory of the world and the game of social life that can be disentangled from the particularities and immediacies of each individual's unique experience and life space. (Keesing 1974:93) Do we human beings interpret the unique situations of life by extrapolating from unique past situations as well as by applying generalized models of the possible and appropriate? John Haviiand warns that: it is difficult to distinguish an "ideational" component, which involves knowledge of the general rules of the culture, from knowledge of a wide set of contingencies which are in no sense common to a cultural tradition. We ordinarily have thought of one's cultural competence as composed of codes: conceptual schemata for, say, plants and animals, kinship systems, political structures, and so on. The conceptual schemata have, we assume, an independent existence prior to any particular configuration of animals, any set of actual kin, any actual political operation. But in gossip the nonparticular is irrelevant before the actual; the contingencies determine the general principles - for they are all there is. In gossip, the world becomes more than ideal schemata and codes; it rests on the Who's Who, much expanded, on history, on reputations, on idiosyncrasies, on exceptions and accidents. Gossip exalts the particular. Much of an actor's cultural competence rests on a vast knowledge of contingent fact, raw unconnected trivia. . . . (Haviiand 1977:181) Similar questions about the relationship between cultural models, individual experience, and situational factors are explored by Brown (1985). Could it be that the "cultural code" itself is partly an artifact of the elicitation process? Do we and our informants enter into a process of cocreation, with normative statements as our collective work? Noting "the processes of collective creation whereby situations are defined, social identities assigned, and rules and relationships negotiated in everyday life," I have recently suggested that normative assertions are very often contextually created to suit the purposes of the moment, not simply cited. The 'rules' we ethnographers have decontextualized, in turn, may often be constructions fabricated as coEective 'work' by the ethnographer and his or her subjects. The ethnographer fashions, with informants, situations in which the construction of normative statements is their joint 'work', their co-creation. Normative statements that emerge in this process of co-creation are then objectified and decontextualized in the production of ethnographies. (Keesing 1982:43) Randall (1976:546) has asked whether cultural "rules" are in part situationally created and co-created even in the realm of folk categorization:

MODELS, "FOLK" AND "CULTURAL"

383

. . . folk taxonomies as we know them from ethnographic tradition [may ' be] . . . constructed by applying normally unexploited principles of native logic to various scraps of knowledge lying around in the mind. Randall suggests that folk taxonomies are in large measure artifacts of our elicitation procedures and suggests that cognitively stored complexes or associations of characteristics and prototype images are more likely to be the bases of actual categorization than are taxonomic trees. This, he suggests, may be because instead of consciously systematizing, most people tackle a different task . -. . Ii.e.,1 to operate adequately in a physically demanding, complex, and often dangerous socioecological environment. Doing this does not involve constructing taxonomic trees, but rather, in a particular situation, selecting a contrast set of characteristics which is both sufficiently specific to achieve a practical and safe result and sufficiently general to accomplish one's purposes efficiently. (1976:552) Similar concerns about the situational structuring of supposed cultural schemata are expressed by Frake (1977) and Agar and Hobbs (1985). We must also be wary, in our pursuit of folk models, of doing what we seemingly have done with folk taxonomies: creating more global and coherent models than our subjects in fact cognize. Folk models may by their nature have a partial and situational ad hoc quality, a lack of global systematicity. I am concerned that some of what we take to be folk or cultural models may not exist until our strategies of questioning lead informants to create them; or worse yet, until their responses provide fragments out of which we create them. Perhaps one does not need a generalized model of how thermostats operate, or what electrical current is, until an ethnographer induces us to formulate one, or elicits situational responses suggesting that we are implicitly using one. Perhaps all we need are operating strategies that can be internally contradictory precisely because they are not motivated by any systemic model. That, then, leads to further speculations about the cognitive processes involved. To what degree do we extrapolate from knowledge of the particular - the canonical case, the prototype example, the closest experience as well as strategically formulating and choosing abstract models that allow high enough probabilities of adequacy, models that work well enough? Explorers of the cognitive frontier, in anthropology, AI, and other fields, do not yet know. Here, Price's observations of disease diagnosis and interpretation in Ecuador (this volume) are particularly interesting. The "experts" in the process of folk diagnosis she describes have recourse to particular actual case scenarios, from which they derive interpretations of a new case, its possible causes, and best treatments. It would seem that cognitive interpretation of new situations entails a close interplay between extrapolation from the particular and application of the general, as Haviland (1977) had glimpsed in the gossip of

384

ROGER M. KEESING

Zinacantan. Are there then separate cognitive compartments, where memories of the particular and models of the general, rules and codes, are stored and organized? Or, are.the general "rules" hung onto, illuminated with, and organized in terms of knowledge of prototypical particularities? The latter intuitively seems more plausible and potentially powerful a system; and I am cheered by Jaime CarbbnelPs observations (personal communication) that recent research in artificial intelligence is exploring just these possibilities. Here, again, prototype conceptualizations that take as background prototypical contexts (that can then be modeled on or remembered in terms of actually experienced events or exemplars) may provide a crucial link between knowledge of the general and memory of the particular extrapolation from past experience to interpret new experience. Cognitive anthropology and artificial intelligence This brings me to the convergence, which I had anticipated a decade ago (see also Colby & Knaus 1974), between cognitive anthropology and artificial intelligence. It had seemed curious to me that AI researchers were beginning to write simple "cultures" for robots with enormous difficulty while the "new ethnographers" were aspiring singlehandedly to write cultural grammars of everything the "natives" knew. Given that the most mathematically sophisticated researchers in the world, working in teams, were hard-pressed to program "knowledge that" allowed automata to recognize and manipulate a few objects in artificially simplified environments, it was hardly surprising that a lone ethnographer in a jungle, virtually illiterate in formal languages, was hard-pressed to write more than token segments of a cultural grammar of Subanun or Kwaio. Had anthropologists anything to contribute to the development of what I foresaw as an emerging formal science of communications, a general theory of information - "knowledge" - in biological systems and its simulation in automata? Could anthropology's cognitive theories of culture begin to be realized in some more substantial and less evanescent and illusory sense through collaboration with AI researchers and others exploring the formal organization of biological information systems? A decade ago, I had foreseen the prospect of positive collaboration: . . . If cognitively oriented anthropology is to survive we will have to add our strengths, our knowledge of cultural variability, to the strengths of interdisciplinary teams working on the frontiers of artificial intelligence research and related fields. The problems of how humans organize their knowledge . . . are too vastly complicated for any single researcher or any single discipline to tackle them alone . . . without the anthropologist's help the cybernetician may well create robots that find their Chinese counterparts inscrutable. (Keesing 1975:20)

MODELS, "FOLK" AND "CULTURAL"

385

[Anthropologists] have, in our studies of other cultural worlds, gained a grasp - mainly intuitive, I think - of how those worlds vary, and how real human beings think, perceive, and choose. . . . The challenge is to make [this knowledge] available to colleagues with the mathematical powers to incorporate our implicit knowledge into their explicit models; and to maintain a continuing dialogue with them that keeps their models anchored in human realities. (Keesing 1972) I had been concerned that in the early 1970s, AI researchers (especially in pattern recognition) seemed to be making simplifying assumptions parallel to those that were creating a false sense of accomplishment in ethnographic semantics (especially analyses of kin terms). The mysteries whereby human beings recognize patterns across contexts lie at the, heart of semantic analysis; cognitive anthropologists studying kinship semantics in narrow genealogical frames were spuriously simplifying the task. AI researchers, by artificially holding frames constant and seeking to capture relational patterns by feature analyses, seemed to be following a similar strategy. I had belatedly realized, as I put it in 1975, that in my own attempts at cognitive analysis, I had been "hunting elephants with a fly swatter." But I suggested that my hunt "posed rather clearly some of the complexities of the phenomena" and hence served "as a corrective against the simplifying assumptions so often made in artificial intelligence and related fields to make dead flies look like elephants" (1975:21). Both cognitive anthropology and AI still seem prone to celebrate more or less successful fly hunts. At least serious collaboration has begun, however, as witness the work of Ed Hutchins on one side of our conference table and Jaime Carbonell on the other. More formally sophisticated models, both in semantic analysis (prototypes, feature weighting) and pattern recognition, the growing dialogue and common vocabulary emerging from the work of such scholars as Schank (Schank 1980; Schank & Abelson 1977), and engagement with anthropological problems, cognitively framed, by AI researchers such as Klein and his colleagues (Klein 1983; Klein et al., 1981) give cause for measured optimism. Conventional metaphor Lakoff and Johnson's provocative 1980 book on conventional metaphor has kindled interest in cognitive anthropology and acquired a central place in the growing concern with folk or cultural models. Lakoff and Johnson (and others less widely read by anthropologists, such as Ortony 1979 and Sacks 1979) have been developing a view of language as pervaded by metaphors that are neither "creative" nor "dead" but conventional and fundamentally constitutive of our ways of everyday talk. Lakoff and Johnson, in particular, have developed a view of metaphor as paradigmatic and experientially based. ^j&m***

386

ROGER M. KEESING

A metaphoric schema, establishing a universe of discourse in terms of another universe of discourse (LOVE IS A JOURNEY, TIME IS MONEY), in effect defines the kind of paradigm that has been.
MODELS, " F O L K " AND "CULTURAL"

387

language. And in my conference paper (Keesing1985) and discussions, I warned of the danger of imputing to our subjects metaphysical theories that seem to be implied by conventional Ways of talk:' Out of our" conv^ri-, tional ways of talking about luck, I suggested, a nonnative observer could be led to infer a metaphysical theory of an invisible substance of which people have more or less. We anthropologists, with our bent for cultural exotica and our propensity for viewing cultures as radically unique and diverse (not to mention our imperfect command of field languages), may often have imputed metaphysical notions to our subjects that in fact represent no more than conventional tropes: "Some people have all the luck/'3 That leads to a further epistemological concern about conventional metaphors and about the models, folk or cultural, on which conference and volume have been focused. Perhaps, working here on a terrain much more slippery than that of folk classification (though that was more slippery than we realized), we really need to draw on our own intuitions as native speakers, need a deeper command of semantic shadings and hidden connections than we acquire in learning fieldwork languages. At least that could be one implication of the fact that so many of the ethnographic papers of the volume (10 of 13) deal with our native language and culture. "Folk" or "cultural" models in society I argue in the second section that early cognitive anthropology was naively reductionists in its tacit premise that cultural rules generate behavior that social interaction is an epiphenomena! outcome of a shared code, a product of rule-following and maximizing of appropriateness. Early cognitive anthropology was, I think, equally naive in its tacit assumption that the institutional structures of society are themselves epiphenomenal - that cultural rules generate social systems as well as behavior. Such a view was perhaps intended only as a heuristic corrective to the alternative determinisms - ecological, economic, functionalist - then prevailing in anthropology. However, we nonetheless implied the Subanun, Hanunoo, Trukese, or Kwaio society was a more-or-less direct outcome of the rules of a cultural grammar; cognitive anthropology's task was to uncover these mainly implicit rules. As heuristic corrective, this view doubtless had some value, especially when applied to a small-scale classless society in the Philippines or Solomon Islands. But cultural "rules" are themselves historically situated, shaped (although not determined) by economic and ecological constraints and social processes. Cultural knowledge does not simply vary from individual to individual. What actors know, what perspectives they take, depend, even in the least complex classless societies, on who they are - whether they are male or female, young or old, leader or led. Early cognitive anthropology, taking its models from linguistics, grew up curiously innocent of social theory. It was not simply its concern with

388

ROGER M. KEESING

trivia, its aura of scientists objectivity, that brought "the new ethnography" into collision with the radical anthropology catalyzed by the blood of Vietnam. An anthropology that reified the cultural status quo into tie determinant of institutions as well as behavior, that ignored class and history "and economy, that ignored the ideological force of cultural rules, could scarcely be the needed "new" anthropology, revolutionary in both senses. Both cognitive anthropology and the various modes of radical anthropology have mellowed, become less strident in their critiques, less sweepingly millenarian in their claims and aspirations. Yet cognitive anthropology remains, I think, curiously innocent of social theory. . In the conference focused on folk models, I became something of an , ideological gadfly about this issue. The model of society assumed there seemed, from the perspective of Marxist social theory, to be itself a folk model manifesting itself as science. But like other folk models, this one - of society as comprising "experts" who properly understand and properly control knowledge and society and "folk" - has strong ideological force. Like other folk models, it disguises and mystifies real structures of class and power. Models are created for the "folk" as well as by them. As instruments of ideological hegemony, such models - whether we call them "folk" or "cultural" - may legitimate and perpetuate the status quo: whether they be models of Hinduism that reinforce and perpetuate the position of Brahmin priests or the models of illness among the Ecuadorian poor that allow the masses to contend with grim struggles of life and death while the country's oligarchs go to the Mayo Clinic. Folk models in the form of ideologies of patriotism and empire have led hundreds of thousands of Europeans to "glorious" deaths in service of their rulers. Not all folk models so clearly serve ideological ends. (Such an argument would be hard to make of models of electricity or anger.) But cognitive anthropology's emergent concern with conceptual paradigms of this kind, whether it labels them as folk or cultural models, needs surely to be tempered by sophistication about the sociology of knowledge and the uses of ideology. The concept of culture as shared and societal embodies, as social theory, deeply conservative premises (see, e.g., Abercrombie 1980). We need to keep in view (at least in our peripheral vision) the production, control, distribution, and ideological force of cultural knowledge. Anthropologists, like other social scientists, are perforce specialists as well as generalists. Specialized work in cognitive anthropology may leave no room for long excursions into the class structure of society and the production and control of the knowledge we seek to map.4 But we need to be much more cautious and wise than were the pioneers of "the new ethnography" about embedding the cognitive systems we purport to explore and map within social systems and about our place in the muitidisciplinary advances of social/behavioral science. An emergent social/behavioral theory must bridge between the institutional and abstract and the essentialities of humanness. Social systems are

MODELS, "FOLK" AND "CULTURAL"

389

constructed out of, as well as constrained by, what human beings .are; and how. human beings cognize their worlds constrains and shapes how , humans-in-societies reproduce them. Exploration of conceptual mddels, of metaphors, of the construction and co-construction of meaning, of how what we see is constituted by what we know, can contribute, impprtantly to an emerging composite understanding of humans-in-societies. ^ We need a specialized concern with cognition. There is challenge enough . in trying to keep up with rapidly moving frontiers in AI, psychplingui^iics, linguistics, and other subfields of cognitive science. Perhaps it is too much to expect cognitive anthropologists and other explorers of conceptual models of everyday reality to be social theorists as well. At the same time, however, I believe we must also step back regularly to assess how what we are exploring articulates with a wider concern with humans-in-societies. Otherwise, we run the danger that beset "the new ethnography" - of radically misconstruing the place and power of the provisional models. Cognitive anthropology, I think, must be pursued in the long run as part of a multidisciplinary, multisided, and mutually informed exploration.

Notes 1. I was asked by Dorothy Holland and Naomi Quinn to assess the Conference on Folk Models, and developments in cognitive anthropology against the background of my earlier critiques of the field. I have interpreted the invitation as mandate for the self-indulgence reflected in these pages, where I refer to my earlier critiques as if they were divinely inspired. I ask the reader's indulgence for this egocentric perspective on a field in which I have for 20 years been a marginal participant. 2. Clement suggests that cognitive anthropologists are using folk "in a broad sense to refer to any group of people participating in a cultural tradition," and that - with no pejorative intent - they "use 'folk' in this broad sense to legitimate the study of indigenous classification systems in their own terms" (1982:211). 3. See Keesing 1985 and Lutz 1985 for discussions of the problem of "false exoticism." 4. For a recent attempt of mine to articulate Marxist perspectives on ideology with recent work on prototype categorization and conventional metaphor, in examining concepts of race and ethnicity, see Keesing (in press).

References Abercrombie, N. 1980. Class, Structure and Knowledge: Problems in the Sociology of Knowledge. Oxford: Basil Blackwell. Agar, M. and J. Hobbs 1985. How to grow schemas out of interviews. In Directions in Cognitive Anthropology, J. Dougherty, ed. Urbana: University of Illinois Press. Pp. 413-431. Berlin, B. and P. Kay 1969. Basic Color Terms: Their Universality and Evolution. Berkeley: University of California Press.

390

ROGER M. KEESING

Brown, M. •••/' 1985. Individual experience, dreams and the identification of magical: stones in an Amazonian society. In Directions in Cognitive Anthropology, J, Dougherty, ed. Urbana: University of Illinois Press. Pp. 373-387. ^ Clement, D. H. 1982. Samoan folk knowledge of mental disorders. In Cultural Conceptions of Mental Health and Therapy, A. J. Marsella and G. M. White, eds. Dordrecht, Holland: D. Reidel Publishing Company. Pp. 193-213. Colby, B. N. and R. Knaus 1974. Men, grammars, and machines: A new direction for the study of man. In On Language, Culture, and Religion: In'Honor of Eugene A. Nida, M. Black and W. A. Smalley, eds. The Hague: Mouton Publishers. Pp. 187-197. Coleman, L. and P. Kay 1981. Prototype semantics: The English word 'lie.' Language 57(l):26-44. Frake, C O . 1977. Plying fram.es can be dangerous: Some reflections on methodology in cultural anthropology. Quarterly Newsletter of the Institute for Comparative Human Development l(3):l-7. Gardner, P. M. 1976. Birds, words, and a requiem for the omniscient informant. American Ethnologist 3(3):446-468. Geertz, C. 1966. Person, time and conduct in Bali: An essay in cultural analysis. Yale Southeast Asia Program, Cultural Report Series. (Reprinted in C. Geertz, The Interpretation of Cultures. New York: Basic Books, 1973. Pp. 360-411.) 1972. Deep play: Notes on the Balinese cockfight. Daedalus 101:1-37. (Reprinted in C. Geertz, The Interpretation of Cultures. New York: Basic Books, 1973. Pp. 412-453.) 1973. Thick description: Toward an interpretive theory of culture. In C. Geertz, The Interpretation of Cultures. New York: Basic Books. Pp. 3-30. 1974. 'From the natives' point of view': On the nature of anthropological understanding. Bulletin of the American Academy of Arts and Sciences 28, No. I. (Reprinted in Meaning in Anthropology, K. H. Basso and H. A. Selby, eds. Albuquerque: University of New Mexico Press, 1976. Pp. 221238.) 1975. Common sense as a cultural system. Antioch Review 33(1):5~26. (Reprinted in C. Geertz, Local Knowledge: Further Essays in Interpretive Anthropology. New York: Basic Books, 1983. Pp. 73-93.) Gentner, D. and A. L. Stevens, eds. 1982. Mental Models. Hillsdale, N.J.: Lawrence Erlbaum Associates. Goodenough, W. H. 1961. Comment on cultural evolution. Daedalus 90:521-528. Gregory. R. L. 1969. On how so little information controls so much behavior. In Towards a Theoretical Biology, Vol. 2, C. H. Waddington, ed. Chicago: Aldine Publishing Company. Pp. 236-246. Haviland, J. 1977. Gossip, Reputation and Knowledge in Zinacantan. Chicago: University of Chicago Press. Johnson-Laird, P. N. 1981. The form and function of mental models. In Proceedings of the Third Annual Conference of the Cognitive Science Society. Berkeley: University of California. Pp. 103-105.

MODELS, "FOLK" AND "CULTURAL"

391

Keesing, R. M. 1970. Toward a model of role analysis. In A Handbook of Method in Cultural Anthropology, R. Cohen and R. NarroU, eds. Garden City, N.Y.: Natural History Press. Pp. 423-453. 1972a. Paradigms lost: The new ethnography and the new linguistics. Southwestern Journal of Anthropology 28(4):299-332. , l * 1972b. Simple models of complexity: The lure of kinship. In Kinship Studies in the Morgan Centennial Year, P. Reining, ed. Washington, D. C : Anthropological Society of Washington; Pp. 17-31. . 1972c. Toward a New Cognitive Anthropology. Address presented to Institute for Communications Research, May 10, University^ of Illinois, Urbana. 1974. Theories of culture. Annual Review,of .Anthropology 3:73-97. 1975. Explorations in role analysis. In Linguistics and Anthropology: In Honor of C. F. Voegelin, M. D. Kinkade, K. L. Hale, and O. Werner, eds. Lisse, Netherlands: Peter de Ridder Press. Pp. 385-403. 1981. Cultural Anthropology: A Contemporary Perspective. 2nd ed. New York: Holt, Rinehart and Winston. 1982. 'Cultural rules': Methodological doubts and epistemological paradoxes. Canberra Anthropology 5(l):37-46. 1985. Conventional metaphors and anthropological metaphysics: The problematic of cultural translation. Journal of Anthropological Research 41(2):201-217. (in press). Racial and ethnic categories in colonial and postcolonial States: Sociological arid linguistic perspectives on ideology. In Studies on the Adequacy of Theories, Paradigms and Assumptions in the Social and Human Sciences, M. O'Callaghan, ed. Paris: U.N.E.S.C.O. Kempton, W. 1981. The Folk Classification of Ceramics: A Study of Cognitive Prototypes. New York: Academic Press. Klein, S. 1983. Analogy and mysticism and the structure of culture. Current Anthropology 24(2): 151-180. V Klein, S., D. A. Ross, M. S. Manasse, J. Danos, M. S. Bickford, and K. L. Jenson. 1981. Surrealistic imagery and the calculation of behavior. In Proceedings of the Third Annual Meeting of the Cognitive Science Society. Berkeley: University of California. Pp. 307-309. Kronenfeld, D. B., J. D. Armstrong, and S. Wilmoth 1985. Exploring the internal structure of linguistic categories: An extensionist semantic view. In Directions in Cognitive Anthropology, J. Dougherty, ed. Urbana: University of Illinois Press. Pp. 91-110. Kuhn, T. 1979. Metaphor in science. In Metaphor and Thought, A. Ortony, ed. Cambridge, England: Cambridge University Press. Pp. 409-419. Lakoff, G. 1984. Classifiers as a reflection of mind: A cognitive model approach to prototype theory. Berkeley Cognitive Science Report No. 19. Berkeley: University of California Institute of Human Learning. Lakoff, G. and M. Johnson 1980. Metaphors We Live By. Chicago: University of Chicago Press. Lutz, C. 1985. Depression and the translation of emotional worlds. In Culture and Depression: Studies in the Anthropology and Cross-cultural Psychiatry of Affect and Disorder, A. Kleinman and B. Good, eds. Berkeley: University of California Press. Pp. 63-100.

392

ROGER M. KEESING

McCloskey, M., A. Caramazza, and B. Green . ; 1980. Curvilinear motion in the absence of external forces: Naive betiefs'about the motion of objects. Science 210:1139-1141. ., ; / , Miller, G. A., E. Galanter, and K. Pribram. "•''/'' ^ „" I960. Plans and the Structure of Behavior. New York: Holt/Rinehart and Winston. .'.-'--^ :•-••— ,.,......,..,.. .^ Natanson, M. 1967. Alienation and social role. In Phenomenology in America, J, M. Edie, ed. Chicago: Quadrangle Books. Pp. 255-268. 1970. The Journeying Self: A Study in Philosophy and Social Role. Reading, Mass.: Addison-Wesley. Norman, D. and D. E. Rurhelhart 1975. Memory and knowledge. In Explorations in Cognition, D. Norman and D. E. Rumelhart, eds. San Francisco: W. H. Freeman and Company. Pp. 3-32. Ortony, A., ed. 1979. Metaphor and Thought. Cambridge, England: Cambridge University Press. Paul, R. A. 1976. The Sherpa temple as a model of the psyche. American Ethnologist 3(1): 131-146. Quinn, N. 1982. "Commitment" in American marriage: A cultural analysis. American Ethnologist 9(4):775-798. Randall, R. A. 1976. How tall is a taxonomic tree? Some evidence for dwarfism. American Ethnologist 3(3):543-553. Rosaldo, M. Z. 1980. Knowledge and Passion: liongot Notions of Self and Social Life. Cambridge, England: Cambridge University Press. Rosch, E. 1973. On the internal structure of perceptual and semantic categories. In Cognitive Development and the Acquisition of Language, T. M. Moore, ed. New York: Academic Press. Pp. 111-144. 1975. Universals and cultural specifics in human categorization. In Cross-Cultural Perspectives on Learning, R. Breslin, S. Boucher, and W. Lonner, eds. New York: Halsted Press. Pp. 177-206. 1977. Human categorization. In Studies in Cross-Cultural Psychology, Vol. 1, N. Warren, ed. London: Academic Press. Pp. 1-49. 1978. Principles of categorization. In Cognition and Categorization, E. Rosch andB. Lloyd, eds. Hillsdale, N.J.: Lawrence Eribaum Associates. Pp. 27-48. Sacks, S., ed. 1979. On Metaphors. Chicago: University of Chicago Press. Salmond, A. 1982. Theoretical landscapes: On cross-cultural conceptions of knowledge. In Semantic Anthropology, D. Parkin, ed. New York, London: Academic Press. Pp. 65-S7. Sankoff, G. 1971. Quantitative analysis of sharing and variability in a cognitive model. Ethnology 10:389-408. Schank, R. 1980. Language and memory. Cognitive Science 4:243-284. Schank, R. and R. Abelson 1977. Scripts, Plans, Goals and Understanding: An Inquiry into Human Knowledge Structures. Hillsdale, N.J.: Lawrence Eribaum Associates.

MODELS, "FOLK" AND "CULTURAL"

393

Schutz, A. 1962. Collected Papers, Vol. I, The Problem of Social Reality, M. Natanson, ed. The Hague: Martinus Nijhoff. 1967. The Phenomenology of the Social World. G. Walsh and F. Lehnert, trans. Evanston, III.: Northwestern University Press. Varenne, H. 1984. Collective representation in American anthropological conversations about culture: Individual and culture. Current Anthropology 25(3) :281-299. Wexler, K. N. and A. K. Romney 1972. Individual variations in cognitive structures. In Multidimensional Scaling: Theory and Applications in the Behavioral Sciences, Vol. 2, Applications, A. K. Romney, R.N. Shepard, and S. B. Nerlove, eds. New York: Seminar Press. Pp. 73-92. White, G. 1980. Conceptual universals in interpersonal language. American Anthropologist 83(4):759-781.

Topics marked with an asterick (*) are the focus of one or more whole chapters; see table of contents. Abelson, Robert, 19-22, 35; see also Schank and Abelson Abercrombie, Nicholas, 388 •affect, in cultural models, 15, 118-19, 122-7, 142-3, 195-221, 286, 290-312, 330-2, 334; see also cross-cultural comparisons Agar, Michael, 222, 313, 316, 383 Althusser, Louis, 377 analogy, as basis of cultural models, 10, 29, 154, 243, 263, 280; see also metaphors analogy hypothesis, 247-8 Anscombe, G. E. M., 114, 118, 120 Arewa, E. Ojo, and Alan Dundes, 152 artificial intelligence research, 5, 19, 310, 369-70, 380, 381, 384-5 astrology, 360-1 Austen, Jane, 137-8 authority, and use of cultural models, 9-10, 46, 55, 59, 64, 73, 75, 92, 171; see also language and action Bakan, David, 365 Bateson, Gregory, 325 Becker, Alton, 63 behaviorism, 356-60 Berlin, Brent, and Paul Kay, 145, 370, 373 Boehra, Christopher, 292 Bohannan, Paul, 315 Bok, Sissela, 64-5 Boltanski, Luc, and Laurent Thevenot, 104 Bourdieu, Pierre, 377 Brown, Michael F., 382 Burrows, Edwin G., and Melford Spiro, 141

Burton, Michael, 85; see also Burton and Romney Burton, Michael, and A. Kimball Romney, 80, 108 Cancian, Frank, 300 Carboneli, Jaime, 154, 370, 384, 385; see also Carboneli and Minton Carboneli, Jaime, and Steven Minton, 154 Casagrande, Joseph, and Kenneth L. Hale, 292, 309 causality: in cultural models, 117-18, 121-7, 188-90, 212, 294, 297; myth and historical causality, 278, 285; in narrative, 327-30, 333-4, 340-1, 347-9; in schemas of chained propositions, 33, 162-3, 181-8,311 Caws, Peter, 5-6, 7-8, 32 change, in cultural models: as a result of experience, 236-7; as result of interview, 10, 234-6; see also historical change Chomsky, Noam, 369 Clement, Dorothy, 16, 333, 372-3, 377, 389; see also Eisenhart and Holland; Harding and Clement; Holland; Holland and Skinner; Quinn and Holland cognitive anthropology: and artificial intelligence research, 384-5; criticisms of, 23, 80, 376-7, 387-9; definitions and history of, 4, 5, 13-18, 222, 309, 369-70; and symbolic anthropology, 375-7; see also ethnoscience, formal semantic analysis cognitive psychology, 222-3, 370, 373 cognitive science, 18-19

396 cognitive-structure analysis, 85, 103-5 cognitive tasks: generalization, 182, 317; inference, 15, 152, 159, 161-2, 168, 243, 244, 310-11; memory storage, 33-4, 112, 157, 170, 316; problem solving, 169, 254-63, 314; reasoning, 29, 180-8, 246-7, 249, 260; similarity judgments, 164-8; word definition, 51, 79, 170, 292, 300-2, 310; word use, 23, 24, 292 Colby, Benjamin N., 309; see also Colby and Knaus Colby, Benjamin N., and Rodger Knaus, 384 Cole, Michael, and Sylvia Scribner, 222 Coleman, James, 106 Coleman, Linda, and Paul Kay, 17, 43-5, 48, 50,51, 54-5,373 Collins, Allan, and Dedre Gentner, 10, 17, 29, 34, 154, 192, 248, 373 common sense, 350, 375; see also cultural models, folk models component models, definition of, 17, 26, 33-4, 112, 189-90, 249-54, 307, 333 componential analysis, see formal semantic analysis Conklin, Harold, 381 conscious models or theories, 68-9, 76; see also expert systems under cultural models and folk .models consciousness, 75, 140 counterexample, as narrative device, 318, 319, 323, 325 creativity, of cultural models, 109, 214-16, 243, 302, 372, 379-84 cross-cultural comparisons: emotions, 220, 292, 294, 296-8, 308, 310, 330; death, 282; energy use, 232; illness, 316-17, 328; methods and purposes of comparisons, 108, 311, 317, 374, 386-7; models of the mind, 141-6; proverbs, 8, 151-5, 161, 323; social roles, 80, 108, 333, 341; speech genres, 60-2, 152, 170,315 cross-cultural universals, see cross-cultural comparisons cultural knowledge, see cultural models cultural models: and artificial intelligence, 18-19; coherence and organization of, 7-8, 10-U, 20, 76, 239, 264, 302, 315, 376-8; evaluation of concept, 370, 372-4, 377-8, 380-4; and expert systems, 309, 343, 350, 352, 361-5, 371-2, 374-5; general-purpose, 34; methodological issues, 5-6, 16-17, 156,

INDEX 222, 329, 373, 376, 382-3; and se'man- ' tic theories, 23-4, 43-5, 62-3, 75-6, 88; see also folk models D'Andrade, Roy, 16-17, 25, 33-4, 80, 289, 319, 328, 330, 374; see also D'Andrade and Wish D'Andrade, Roy, and Myron Wish, 114 Davidson, Debra, 106; see also Holland and Skinner decision making, studies of, 4 de Kieer, Johan, 244-6; see also de Kieer and Brown de Kieer, Johan, and John Seely Brown, 222-3 DeSousa, Ronald, 211 directive force, 7-8, 9-13, 161, 166-7; see also authority, language and action discourse analysis, as methodology, 16, 17-18, 173, 317-18, 346 discourse types, 17, 45, 46, 52-4, 106; see also speech acts, speech genres DiSessa, Andrea, 222, 239 distribution of cultural models: 106-7, 113, 118, 353, 372-4, 377; see also variation Dougherty, Janet, 222 Dumont, Jean-Paul, 309 Dyer, Michael G., 152 Eisenhart, Margaret, and Dorothy Holland, 106 Ekman, Paul, 145; see also Ekman, Levenson, and Friesen Ekman, Paul, Robert W. Levenson, and Wallace V. Friesen, 219-20 Ellenberger, Henri, 365 "emotion, see affect, cross-cultural comparisons Ervin-Tripp, Susan, 64 ethnoscience, 14, 49-50, 80, 371, 373, 377, 380; see also cognitive anthropology, formal semantic analysis ethnotheory, definition of, 291-2, 309, 330 expert models, see cultural models, folk models experts, communication with nonexperts, 53,70 "explanatory systems, 343, 350, 351-3; see also cultural models Fillmore, Charles, 23, 31, 43, 44, 88, 112, 369 Fjeliman, Stephen, 4

INDEX folk models: and cultural models, 4, 309, 370, 377-8, 389; definitions of, 64, : •' 68-9, 76, 113-14, 373-4; ana dcpert ; systems, 32, 76, 139-41, 192, 223, 234, 237; see also cultural models folk theory, see folk models Forbus, Kenneth D., 244-6, 248 formal semantic analysis, 43, 63, 6 7 / 164-8, 370; see also cognitive anthropology, ethndscience """' •;-<•-Forman, Don, 64 Foucault, Michel, 376 Frake, Charles, 19, 381, 383 frame semantics, 23, 44, 45, 307 Frank, Gelya, 351 Frege, Gottlob, 75 Freud, Sigmiind, 282 Freudian psychology, 140, 350, 353-6, 361-4 Friedrich, Paul, 138 Gardner, Peter, 371 Geertz, Clifford, 5, 352, 372, 375-6 *gender roles, 318-22 Centner, Dedre, 248, 263, 371, see also Collins and Gentner; Gentner and Stevens Gentner, Dedre, and Albert L. Stevens, 223, 373 Geoghegan, William, 4 Gerber, Eleanor, 141 Gilsenan, Michael, 61-2 Gladwin, Hugh, and Christina Gladwin, 4 goals, derived from cultural models, 22, 299-302 Goffman, Erving, 44 Good, Byron, and Mary-Jo Delvecchio Good, 31! Goodenough, Ward H., 4 Gordon, David P., 47 Gramsci, Antonio, 376 Gregory, Richard L., 381 Grice, H. Paul, 44, 45, 58, 60, 70, 120 Hale, Nathan, 365 Harding, Joe, and Dorothy Clement, 80, 108 Harris, Marvin, 5 Haviland, John, 379, 382, 383, 386 Hayes, Patrick, 244-6 •hedges, 7, 56-7, 70, 72-3, 153, 317 Heidegger, Martin, 375, 376 Heider, Fritz, 127, 159, 162, 169 historical change: in cultural or folk

397 models, 11, 105-8, 146, 231, 362-3; and myth, 269, 282, 286-7 '' ' >•'• "•••••" Hobbs, Jerry R., 383 •' .-'.••':-^ Hochschild, Artie/311 ' ~ ''"'/'" •;"'"•''";;-"'-" ' Holland, Dorothy, 222,; 313,-316,'325; 5ee: ; - : also Clement; Eiserihart and Holland; '" Harding and Clement; Holland and Skinner; Quinn arid Holland Holland, Dorothy, and Debra Skinner, 9, ' • 15,20 • - • ;'-i'-='--i- .•--v.?.? Holy, Ladislav, 6 / ':' Husseri, Edmund, 375, 376'"" -'•"••'x Hutchins, Edwin, 8, 14, 17, 25, 179, 291-3, 317, 385; see also Hutchins and Levin Hutchins, Edwin, and James Levin, 169 Hymes, Deil, 315 ideology, 13, 55-9, 308, 388 indexicality, in language, 68, 72 interviewing, as methodology, 79-81, 88-9, 109, 156-9, 173, 254-6, 345-6 Izard, Carroll, 301 Johnson, Alien, 4 Johnson, Mark, 386; see also Lakoff and Johnson Johnson-Laird, Philip, 223, 373; see also Johnson-Laird and Steedman Johnson-Laird, Philip, and Mark Steedman, 27 Kay, Paul, 10, 16-17, 24, 32, 44, 64, 192, 222, 376; see also Berlin and Kay; Coleman and Kay; Kay and McDaniel Kay, Paul, and Chad McDaniel, 43-4 Keesing, Roger M., 6, 13, 80 Kempton, Wiliett, 10, 17, 26, 192, 371, 380; see also Kempton and Lave Kempton, Wiliett, and Jean Lave, 232 key words, 300, 317 Klein, Sheldon, 385 Kripke, Saul, 75 Kronenfeld, David B,, James D. Armstrong, and Stanley Wilmoth, 380 Kuhn, Thomas, 375

Labov, William, and David Fanshel, 316, 330 Lakoff, George, 24, 26, 30-1, 43, 56, 64, 70, 179, 286, 380-1; see also Lakoff and Johnson, Lakoff and Kovecses Lakoff, George, and Mark Johsnon, 153-4, 211,218, 221-2, 373, 385

INDEX ;.

398 Lakoff, George, and Zoltan Kovecses, 16, 20,26,30, 154, 263, 376 '•• Lakoff, Robin, 44, 51, 55, 369 Langacker, Ronald, 23 Langness, Lewis L., and Gelya Frank, 345 ,. ..,; ...,.,;.-, language and action, 6-9, 68, 107, 159, 222, 225, 330; see also authority, directive force . Larkin, Jill, 223 Lasch, Christopher, 353 Lave, Jean, and Barbara Rogoff, 222 learning, of cultural models, 12, 21-2, 50, 105, 144-5, 264, 296 Levi-Strauss, Claude, 5 Lewontin, Richard C , Steven Rose, and Leon J. Kamin, 13 lexicography and semantics, 68 life story, definition of genre, 344-5 Linde, Charlotte, 9-10, 17, 32, 192, 315, 346, 374 linguistics, 23-4, 114, 364, 369-70, 372; see also semantics, speech acts, speech genres literature, 50, 137-8, 376 Lutz, Catherine, 6, 9, 25, 141-4, 161, 386 McCloskey, Michael, 223, 239; see also McCloskey, Caramazza, and Green . McCloskey, Michael, Alfonso Caramazza, and Bert F. Green, Jr., 222, 373 Malinowski, Bronislaw, 270-2, 278, 283, 286, 288 Mandler, George, 140 Marsella, Anthony, and Geoffrey White, 222 Marx, Karl, 376 Marxism, 376-7, 388-9 Mathews, Holly, 4, 340 meaning, theories of: cognitive-structure analysis, 80, 85, 103-5; literal meaning, 70-1, 196; materialist and idealist, 5, 9; referential meaning, 75, 311; see also cognitive anthropology, symbolic anthropology, prototypes, semantics memory, 381-2; see also memory storage under cognitive tasks *mentai models, 222-3, 373 Merleau-Ponty, Maurice, 375, 376 Mervis, Carolyn, and Eleanor Rosch, 44 Messenger, John C , Jr., 151 metaphors: arbitrariness of, 220; conceptual, 198, 221; constitutive, 218, 385; conventional, 385-7; in cultural models,

30, 211, 218, 286, 364, 386-7; lexical' productivity, 198; ontologicai, 218; source and target domains in, 28, 30, 154, 170, 199, 201, 247-8; see also Lakoff and Johnson metonymy, 30-2, 80, 196-7, 208 Miller, George, Eugene Galanter, and Karl Pribram, 381 Montague, Richard, 75 morality, invoked in cultural models, 54-9, 328-30 motivation, 45, 279, 301, 322 multidimensional scaling, 15, 81, 84, 103-5, 109, 170-1 multivariate analyses, 16 Murtaugh, Michael, Kathenne Faust, and Olivia de la Rocha, 222 naive theory, see mental models narrative, as type of discourse, 9, 17, 152, 314-16, 332, 346-50, 352 Natanson, Maurice M., 376 native speaker's intuitions, as data source, 16-17, 156-7 Natsoulas, Thomas, 140 •natural discourse, as data source, 16, 313 negotiation, of meanings and cultural models, 304, 306-7, 328/371-2 Neisser, Ulric, 85 new ethnography, see ethnoscience Norman, Donald, and David Rumelhart, 381

Ochs Keenan, Eleanor, 61 Ortony, Andrew, 385 participant observation, as methodology, 88-9, 272-3, 288, 291, 292, 313-14, 323, 329 Paul, Robert, 376 person, cultural models of, 158-9, 161-4, 212, 333 phenomenology, 375, 376 philosophy, 114,300,372, 375 Piaget, Jean, 50 politeness, 54, 58-9 popular models or systems, 354-6, 358-9, 361-5; see also cultural models, folk models power, see authority Price, Laurie, 6, 9, 17, 372, 383 propositions: in cultural models, 25, 33, 72-3, 167, 222, 289, 295; in illness nar-

INDEX

399j?y

ratives, 323; in myth, 280, 284; see also Schur, Edwin, 353 ,;; ...... causality, schemas ..:•,;•?.;,':'.>'• ..Schutz, Heinrich, 375-6 ,; prototypes, in cultural models, 22^3, Schwartz, Gary, .106 4 3 - 4 , 55, 80, 87, 217, 372-3, 383 science: as a cultural model, 46, 289; see , •proverbs, see cross-cultural comparisons also expert systems under cultural ;.;.<•., psychoanalysis, see Freudian psychology - ;• models and folk models ; psychological reality of cognitive models, scripts, concept used by Schank and 14, 2 2 0 - 1 , 3 6 5 , 3 8 3 , .-^.. K -=.-;«.* ~.^.- ••••< Abelson, 19, 170 Searle, J o h n , 44, 114, 118 P u t n a m , Hilary, 75 ; self, cultural models of, 12-13, 118-19, Quinn, Naomi, 17, 20, 2 3 , 25, 222, 300, , 121, 131, 190, 288, 296, 346 316, 371-2, 386; see also Quinn and semantics: and cultural models, 23-4, , ' Holland 4 3 - 5 , 6 2 - 3 , 75-6, 88, 307; comQuinn, N a o m i , and Dorothy Holland, binatorial, 6 7 - 8 ; lexical domains, 14, 380 15, 4 3 , 103-5, 170, 369, 380, 382-3; and pragmatics 6 7 - 8 ; truth-conditional, race, 106, 353 45-8; see also ethnoscience, formal Randall, Robert, 14, 379, 380, 382-3 semantic analysis, meaning Rappaport, Roy A., 46, 232 semiexpert systems, see explanatory repression, in interpretation of myth, 282, systems 284-5 Sennett, Richard, 353 responsibility, of speakers, 56-7, 209, simplified worlds, 22, 44, 45, 63, 89, 282, 297 101-5; see also scenarios Rice, G. Elizabeth, 85, 222 Skinner, B. F . , 356, 359 Ricoeur, Paul, 376 social class, 98, 106-7, 324-7, 3 4 6 / 3 5 3 , ' Riesman, P a u l , 309 365, 388 Rosaldo, Michelle, 311, 386 social theory and cognitive anthropology, Rosch, Eleanor, 23, 44, 7 3 , 370, 373; see 376-7, 387-9 also Mervis and Rosch specialist systems, see expert systems Rumelhart, David, 112; see also Norman under cultural models and folk models and Rumelhart; Rumelhart and speech acts, types and definitions of, 53, Norman 56, 114, 315 Rumelhart, David, and Donald Norman, speech genres, types and definitions of, 46, 49, 62, 273, 315-16, 330, 340, 263 344-5; see also cross-cultural Ryle, Gilbert, 5, 114 comparisons Sperber, Dan, 15 Sacks, Sheldon, 385 Sperry, Roger, 140 Salamone, Frank A . , 151, 152 Spiro, Melford, 11, 144; see also Burrows Salrnond, Anne, 386 and Spiro Sankoff, Gillian, 371 Spradley, James, 80 scenarios, 87, 103, 105, 170, 243-4 Stark, Louisa, 324 Schachter, Stanley, and Jerry Singer, Sternberg, Robert, et at., 127 219-20 Stevens, Albert L., and Allan Collins, Schank, Roger, 20-2; see also Schank 248 and Abelson Strathem, Marilyn, 109 Schank, Roger, and Robert Abelson, 20, Sulloway, Frank, 365 299, 385 Suppes, Patrick, and Hermine Warren, schemas: affective, 294, 330; definitions 284 of, 22, 25, 44, 85, 112-13, 179, 285-8; mythic, 279-80, 283-5; organization of, Sweetser, Eve E . , 16, 20, 88, 372 symbolic anthropology, 375-7 12-13, 112; proposition-schemas and image-schemas, 18, 24-30, 179, 244 Tannen, Deborah, 314 Schmidt, Charles, and J o h n D'Addamio, Tarski, Alfred, 75 169

400

INDEX<

Vendler, Zeno,U4, 145 themes in cultural models, 20-1, 35 Thomason, Neil R., 51, 55 •Verschueren, Jef,'57-8 •• transmission of cultural models, see •••learning •'' : .truth value: accuracy of folk theories, • 233, 237-8; in cultural models, 46,' "' •'Waltz,'David,'243-" -. ' • • Ward, Barbara, 5 •. 52-4, 56-7,-70-1, 286, 328, 351; literal • Wexler, Kenneth, and A.- Kimball ----truth, 59, 287, 289; see also cultural • : models, folk models ' ••"•'•• • . ' Romney, 371 : • White, Geoffrey, 6, 8, 15,-25, 80,115, Tyler, Stephen^ 80' : : : ~'- • 156, 297, 310, 386; see also Marsella - " a n d W h i t e •"-''• — • • Varenne, Herve, 372' "• Whorf, Benjamin Lee, 189 variability of cultural models: in a single Wimmer, Heinz, and Josef Perner, 50 person or account, 232, 258, 263, 264, Winograd, Terry, 291 328, 364; within a culture, 220, 239, Winston, Patrick Henry, 263 303, 310, 371-2, '387; See also distribuWiser, Marianne, and Susan Carey, 223 tion, cultural models

**££«->