Art's Agency and Art History (New Interventions in Art History)

Osborne / Art’s Agency and Art History 1405135379_1_pretoc

Final Proof

page i 23.1.2007 3:05pm Compositor Name: PDjeapradaban

Art’s Agency and Art History

Osborne / Art’s Agency and Art History

1405135379_1_pretoc

Final Proof

page ii

23.1.2007 3:05pm Compositor Name: PDjeapradaban

New Interventions in Art History Series editor : Dana Arnold, University of Southampton New Interventions in Art History is a series of textbook mini-companions – published in connection with the Association of Art Historians – that aims to provide innovative approaches to, and new perspectives on, the study of art history. Each volume focuses on a specific area of the discipline of art history – here used in the broadest sense to include painting, sculpture, architecture, graphic arts, and film – and aims to identify the key factors that have shaped the artistic phenomenon under scrutiny. Particular attention is paid to the social and political context and the historiography of the artistic cultures or movements under review. In this way, the essays that comprise each volume cohere around the central theme while providing insights into the broader problematics of a given historical moment. Art and Thought edited by Dana Arnold and Margaret Iversen (published) Art and its Publics: Museum Studies at the Millennium edited by Andrew McClellan (published) Architectures: Modernism and After edited by Andrew Ballantyne (published) After Criticism: New Responses to Art and Performance edited by Gavin Butt (published) Envisioning the Past: Archaeology and the Image edited by Sam Smiles and Stephanie Moser (published) Edges of Empire: Orientalism and Visual Culture edited by Jocelyn Hackforth-Jones and Mary Roberts (published) Psychoanalysis and the Image edited by Griselda Pollock (published) Museums After Modernism edited by Griselda Pollock and Joyce Zemans (published) Material Identities edited by Joanna Sofaer (published) Art’s Agency and Art History edited by Robin Osborne and Jeremy Tanner (published) Exhibition Experiments edited by Sharon Macdonald and Paul Basu (forthcoming) Pop Culture and Post-war American Taste edited by Patricia Morton (forthcoming)

Osborne / Art’s Agency and Art History

1405135379_1_pretoc Final Proof

page iii

23.1.2007 3:05pm Compositor Name: PDjeapradaban

Art’s Agency and Art History

Edited by

Robin Osborne and Jeremy Tanner

Osborne / Art’s Agency and Art History 1405135379_1_pretoc

Final Proof

page iv 23.1.2007 3:05pm Compositor Name: PDjeapradaban

ß 2007 by Blackwell Publishing Ltd except for editorial material and organization ß 2007 Robin Osborne and Jeremy Tanner BLACKWELL PUBLISHING

350 Main Street, Malden, MA 02148-5020, USA 9600 Garsington Road, Oxford OX4 2DQ, UK 550 Swanston Street, Carlton, Victoria 3053, Australia The right of Robin Osborne and Jeremy Tanner to be identified as the Authors of the Editorial Material in this Work has been asserted in accordance with the UK Copyright, Designs, and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs, and Patents Act 1988, without the prior permission of the publisher. First published 2007 by Blackwell Publishing Ltd 1

2007

Library of Congress Cataloging-in-Publication Data Art’s agency and art history / edited by Robin Osborne and Jeremy Tanner. p. cm. — (New interventions in art history) Includes index. ISBN-13: 978-1-4051-3537-5 (hardcover : alk. paper) ISBN-10: 1-4051-3537-9 (hardcover : alk. paper) ISBN-13: 978-1-4051-3538-2 (pbk. : alk. paper) ISBN-10: 1-4051-3538-7 (pbk. : alk. paper) 1. Art and anthropology. 2. Art–Historiography. 3. Art and society. I. Osborne, Robin. II. Tanner, Jeremy, 1963N72.A56A78 2007 701’.0309—dc22

2006020375

A catalogue record for this title is available from the British Library. Set in 10.5/13pt Minion by SPi Publisher Services, Pondicherry, India Printed and bound in Singapore by COS Printers Pte Ltd The publisher’s policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufactured from pulp processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover board used have met acceptable environmental accreditation standards. For further information on Blackwell Publishing, visit our website: www.blackwellpublishing.com

Osborne / Art’s Agency and Art History

1405135379_2_toc Final Proof

page v 12.1.2007 2:14pm Compositor Name: SJoearun

Contents

Series Editor’s Preface Preface List of Illustrations Notes on Contributors Introduction: Art and Agency and Art History Jeremy Tanner and Robin Osborne 1 Enchantment and Sacrifice in Early Egypt David Wengrow 2 Agency Marked, Agency Ascribed: The Affective Object in Ancient Mesopotamia Irene J. Winter

vii ix x xiii 1 28

42

3 Portraits and Agency: A Comparative View Jeremy Tanner

70

4 The Agency of, and the Agency for, the Wanli Emperor Jessica Rawson

95

5 The Material Efficacy of the Elizabethan Jeweled Miniature: a Gellian Experiment Jessen Kelly

114

6 Representational Art in Ancient Peru and the Work of Alfred Gell Jeffrey Quilter

135

7 Gell’s Idols and Roman Cult Peter Stewart

158

Osborne / Art’s Agency and Art History

1405135379_2_toc Final Proof

page vi 12.1.2007 2:14pm Compositor Name: SJoearun

vi Contents 8 Sex, Agency, and History: The Case of Athenian Painted Pottery Robin Osborne 9 Abducting the Agency of Art Whitney Davis Index

179 199

220

Osborne / Art’s Agency and Art History 1405135379_3_posttoc Final Proof

page vii 12.1.2007 6:02pm Compositor Name: SJoearun

Series Editor’s Preface

New Interventions in Art History was established to provide a forum for innovative approaches to and perspectives on the study of art history in all its complexities. Art’s Agency and Art History takes Alfred Gell’s posthumously published work Art and Agency: an Anthropological Theory (1998) and provides a series of critical interventions which carry the significance of the subject well beyond Gell’s immediate anthropological readership. The volume brings together essays from leading academics working across a wide geographical and chronological span to offer an authoritative and innovative consideration of the ways in which arguably one of the most important pieces of ‘‘theory’’ published in recent decades relates to art history in its broadest constituency. Indeed, the chapters combine to take up the challenge to traditional disciplinary boundaries between the anthropology of art and art history that is implicitly offered by Gell. The volume offers ways of thinking through the complex and sometimes difficult theoretical and methodological issues Gell’s work raises that will be particularly useful for students working in art history and related fields such as archaeology and classics. Gell aims to replace the emphasis on aesthetics and the communication of meaning with a concentration on the ‘‘material agency’’ of art which is evident in processes of social interaction. In order to do this, Gell proposes a set of concepts which touch on almost every aspect of art production and reception, from issues of representation to artistic creativity, artist–patron relationships, and the social effectiveness of art. In the introductory chapter of Art’s Agency and Art History, the editors position Gell’s theory in relationship to key methodological and theoretical approaches in the critical tradition of art history, as well as to more recent sociological and semiotic approaches. The remaining chapters offer not only case-studies of how these ideas

Osborne / Art’s Agency and Art History

viii

1405135379_3_posttoc

Final Proof

page viii 12.1.2007 6:02pm Compositor Name: SJoearun

Series Editor’s Preface

work to inform a fresh understanding of art, but also significant new interpretations in the fields of anthropology, archaeology, art history, classics, Egyptology, Near Eastern studies, oriental studies, and preColumbian studies. The volume thus makes a welcome addition to a series that seeks to offer a theoretically informed transdisciplinary analysis of issues that are important for our understanding of the visual world. Dana Arnold London 2006

Osborne / Art’s Agency and Art History 1405135379_3_posttoc Final Proof page ix 12.1.2007 6:02pm Compositor Name: SJoearun

Preface

This volume originated in a panel organized by Robin Osborne at the Oxford meeting of the Theoretical Archaeology Group in December 2000. The eventual transformation of that panel into this book was dependent upon the enthusiasm of Dana Arnold and Jeremy Tanner. An informal day conference, involving most of the participants here, held in King’s College, Cambridge in summer 2003 usefully moved the project toward this fruition. The premise of the original panel was that Alfred Gell’s Art and Agency deserved to be better known among archaeologists since, whether or not one liked the particular approach taken by Gell, the questions which he posed more sharply than anyone before – questions about art’s agency but also about what makes an artwork distinct – were ones which archaeologists ignored at their peril. The premise of this book is that those same questions are urgent also for the art historian, and that disciplinary division between anthropology, archaeology, art history, classics, Egyptology, Near Eastern Studies, Oriental Studies, Pre-Columbian Studies, etc. has impeded dialogue over questions of great mutual interest – and questions which are most usefully illuminated by the very different sorts of materials which scholars in those disciplines have variously at their disposal. Robin Osborne and Jeremy Tanner June 2006

Osborne / Art’s Agency and Art History

1405135379_3_posttoc Final Proof

page x 12.1.2007 6:02pm Compositor Name: SJoearun

Illustrations

0.1 Trobriand canoe prow-board

4

0.2 Gell, Art and Agency, table 1 ‘‘The art nexus’’

13

0.3 Sir Joshua Reynolds, Portrait of Dr. Johnson

15

0.4 Gell, Art and Agency, fig. 4.3/1

16

0.5 Nkisi figure (nail fetish), British Museum

17

0.6 Gell, Art and Agency, fig. 4.4/2

18

0.7 Gell, Art and Agency, fig. 4.4/4

20

1.1 Mastaba tomb and surrounding burials, Saqqara, First Dynasty

33

1.2 Transformation of a prototype: ‘palace fac¸ade’ forming part of the Horus name of King Djet, on his funerary stele from Abydos

35

1.3 Painted decoration of Mastaba tomb fac¸ade

36

2.1 Plaque of Ur-Nans˘e of Lagash, found at Tello, c.2600

BC

46

2.2 Detail, registers 2–3 of the Stele of Ur-Namma of Ur, found at Ur, c.2110 BC

47

2.3 Statue ‘‘B’’ of Gudea of Lagash, found at Tello, c.2115 BC

48

2.4 Clay Tablet with seal of Ur-Lisi, ensi of Umma, citing ruler Amar-Su’en of Ur, Ur III period, c.2045 BC

49

Osborne / Art’s Agency and Art History 1405135379_3_posttoc Final Proof page xi 12.1.2007 6:02pm Compositor Name: SJoearun

Illustrations xi 2.5 Detail, obverse of the Stele of Eannatum of Lagash, found at Tello, c.2560 BC

51

2.6 Reverse of the Stele of Eannatum of Lagash

52

3.1 Portrait of Eizon, 1280 AD. Saidaiji, head temple of the Shingon Risshu, Nara Prefecture

75

3.2 Objects discovered within the portrait of Eizon

76

3.3 Portrait of Epicurus. Cast of Roman copy after Greek original of c.270 BC

79

3.4 Icon of Virgin and Child with saints. Sixth century AD

86

4.1 Anonymous, ritual portrait of the Wanli Emperor

96

4.2 Yellow robe embroidered with dragons from the tomb of the Wanli Emperor (1621)

105

5.1 Nicholas Hilliard, the Heneage Jewel (exterior)

118

5.2 Nicholas Hilliard, the Heneage Jewel (interior)

119

6.1 A Moche portrait head vessel

137

6.2 The Maritime Frieze on the interior wall of a patio on the summit of Huaca Cao Viejo

145

6.3 Possible ‘‘basic’’ units of the motif of the Maritime Frieze

146

6.4 The patterns of the Maritime Frieze drawn beyond its borders

147

6.5 The Maritime Frieze

149

6.6 Color patterns in the Maritime Frieze at Huaca Cao Viejo

150

6.7 The Prisoner Frieze at Huaca Cao Viejo

153

6.8 Roll-out photograph of the Prisoner Frieze

153

7.1 Roman imperial, marble copy of the Aphrodite of Cnidus by Praxiteles, original c.340s BC

164

7.2 Reverse of bronze coin from Roman Cyprus (reign of Septimius Severus, c.AD 193–211)

167

Osborne / Art’s Agency and Art History 1405135379_3_posttoc Final Proof

page xii

12.1.2007 6:02pm Compositor Name: SJoearun

xii Illustrations 7.3 Marble funerary statue of a faceless goddess from Cyrene, fifth century BC

174

8.1 Reworking of Gell’s fig. 4.4/4

183

8.2 Nearchos’s aryballos; side view

186

8.3 Handle plate of Nearchos’s aryballos

187

8.4 Hydria ascribed to Kleophrades

192

8.5 Raymond Coxon, The Intruder

193

Osborne / Art’s Agency and Art History

1405135379_3_posttoc

Final Proof

page xiii 12.1.2007 6:02pm Compositor Name: SJoearun

Notes on Contributors

Whitney Davis is Professor of History and Theory of Ancient and Modern Art and Chair of the Department of History of Art at the University of California at Berkeley. He is the author of several books on prehistoric, ancient Egyptian, and modern art, archaeology, and aesthetics and on the history of psychoanalytic theory, as well as many articles on art theory and cultural history, most recently ‘‘Visuality and Pictoriality’’ in Res: Anthropology and Aesthetics 46, 2004, 9–31, and ‘‘Decadence and the Organic Metaphor’’ in Representations 90, 2005, 131–49. Jessen Kelly is a Ph.D. Candidate in the Department of History of Art at the University of California at Berkeley, where her research has focused primarily on Netherlandish painting of the fifteenth and sixteenth centuries. She is currently working on a dissertation on issues of chance and futurity in Northern Renaissance art and material culture. Robin Osborne is Professor of Ancient History at the University of Cambridge. He writes widely across the range of Greek history, Greek archaeology and classical art history. He is the author of Archaic and Classical Greek Art (Oxford, 1998) and of Greek History (London, 2004). Jeffrey Quilter is Deputy Director for Curatorial Affairs, Peabody Museum, Harvard, and was previously Director of Pre-Columbian Studies, Dumbarton Oaks. His books include Life and Death at Paloma: Society and Mortuary Practices in a Preceramic Peruvian Village (Iowa City, 1989) and Narrative Threads: Accounting and Recounting in Andean Khipu (edited with Gary Urton) (Austin, 2002). Jessica Rawson is Professor of Chinese Art and Archaeology at the University of Oxford and Warden of Merton College; she was previously

Osborne / Art’s Agency and Art History 1405135379_3_posttoc Final Proof

page xiv 12.1.2007 6:02pm Compositor Name: SJoearun

xiv Notes on Contributors head of the Department of Oriental Antiquities at the British Museum. Her areas of study include Chinese ancestral rituals and burial practices of the period 1500–700 BC and the unification of China in the third century BC. Among her publications are Chinese Ornament: the Lotus and the Dragon (London, 1984), Western Zhou Bronzes from the Arthur M. Sackler Collections (Washington, DC, 1990, Chinese Jade from the Neolithic to the Qing (London, 1995), Mysteries of Ancient China (London, 1996), China: The Three Emperors, 1662–1795 (edited with Evelyn Rawski) (London, 2005). Peter Stewart is a lecturer in Classical Art at the Courtauld Institute of Art in London. His publications include Statues in Roman Society: Representation and Response (Cambridge, 2003), Roman Art (Oxford, 2004), and articles on various aspects of Roman art and culture. Jeremy Tanner is a lecturer at the Institute of Archaeology, University College London, where he teaches classical archaeology and comparative art. His The Invention of Art History in Ancient Greece has just been published by Cambridge University Press. David Wengrow is a lecturer at the Institute of Archaeology, University College London, where he teaches comparative archaeology of the Eastern Mediterranean and Middle East. His research addresses the nature of early social transformations in Egypt and Iraq, and the role of the remote past in the formation of modern political identities. Recent publications include The Archaeology of Early Egypt, 10,000 to 2,650 BC: Social Transformations in North East Africa (Cambridge, 2006). Irene Winter is William Dorr Boardman Professor of Fine Arts at Harvard University. A former MacArthur prizewinner and Slade Lecturer at the University of Cambridge, she worked and traveled extensively in the Middle East and South Asia. She is currently working on two books, Visual Affect: Aesthetics and Ancient Mesopotamia, and The Stele and the State: Victory Monuments of Naram-Sıˆn of Agade.

Osborne / Art’s Agency and Art History

1405135379_4_000 Final Proof page 1 12.1.2007 2:15pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History Jeremy Tanner and Robin Osborne Alfred Gell’s Art and Agency: an Anthropological Theory is a strikingly original intervention in the anthropology of art, and also a controversial one. It has been celebrated, elaborated, and fiercely criticized.1 Gell’s treatise is of immediate interest to art historians. He breaks out of the restricted interest in ‘‘tribal’’ or ‘‘primitive’’ art characteristic of the preponderance of work in the anthropology of art, and offers a theory of art which is anthropological in the broader sense, namely one which can equally effectively analyze either a West African nail fetish or Marcel Duchamp’s Large Glass. For Gell all these are objects which act as agents, through which we are able to grasp ‘‘mind’’ as an external disposition transcending the individual. Gell’s theory is as provocative as it is wideranging and stimulating: what kind of understanding of art is possible when one rejects ‘‘meaning’’ and ‘‘aesthetics’’ as proper objects of analysis? But it is one which should not be ignored: Gell’s theory of art not only offers a new way of looking at art, but also provides a genealogy of art history, suggesting that it is just a minor variant transformation of underlying cognitive dispositions shared with fetishists and idolaters from the Congo to Kathmandhu.2 In this introduction, we first outline the key themes of Gell’s program (‘‘Art and Agency: The Program’’). We explore their background in recent debates in anthropology and their resonances with strands of critical theory in art history (‘‘Art and Agency between Art History and Anthropology’’). We then describe in more detail the core of Gell’s analytical apparatus, his ‘‘art nexus,’’ which embodies the most important elements of Gell’s theoretical contribution to the history and anthropology of art

Osborne / Art’s Agency and Art History

1405135379_4_000 Final Proof page 2 12.1.2007 2:15pm Compositor Name: SJoearun

2 Jeremy Tanner and Robin Osborne (‘‘From the Art Nexus to the Gellogram’’). This provides essential background to the varied appropriations, elaborations and critiques of Gell offered in the chapters which form the substance of this volume, ranging from ancient Egypt to early-modern England, from imperial China to preColumbian Peru, which we briefly introduce (‘‘Art and Agency in Art History’’).

Art and Agency: The Program Art and Agency is a polemical work. Gell defines his intellectual program in part by attacking conventional anthropological (and art-historical) approaches. In an earlier paper, Gell had already advocated ‘‘a complete break with aesthetics.’’3 In Art and Agency he further argues that conventional art-historical approaches cater primarily to the aesthetic sensitivities of a Western art public, and assimilate the kinds of experiences entailed in the encounter with African or Oceanic art to that of the museum-going art lover in the West – all at the expense of meaningful sociological understanding of how the artwork circulates, and produces effects, in its original setting. He also criticizes iconographic approaches to art, for their exclusively cultural focus on meaning and symbolic communication, an approach derived from Panofsky,4 but also highly influential in the anthropology of art, in particular through the work of Anthony Forge.5 Culture, Gell suggests, is not the proper frame within which to contextualize art, since culture is merely an ‘‘abstraction’’ in contrast to ‘‘the dynamics of social interaction . . . a real process . . . unfolding in time.’’6 Instead of the traditional focus on aesthetics and meaning, Gell chooses to ‘‘place all the emphasis on agency, causation, result and transformation,’’ viewing ‘‘art as a system of action intended to change the world, rather than encode symbolic propositions about it.’’7 All social agency, Gell argues, is realized through the medium of objects. Humans realize their intentions, and thus exercise agency, through the medium of artifacts as ‘‘secondary agents’’ which distribute their agency in the causal milieu.8 A soldier is only a soldier by virtue of the weapons which make possible his capacity for violence: Gell cites the example of Pol Pot’s soldiers scattering landmines in Cambodia, objects which were constitutive of their agency, their capacity to inflict violence on a spatially and temporally expanded scale. Artworks differ from other artifacts, such as landmines, in the manner of their agency. Artworks are ‘‘indexes’’ which distribute agency

Osborne / Art’s Agency and Art History

1405135379_4_000 Final Proof page 3 12.1.2007 2:15pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 3 by calling forth on the part of the viewer or ‘‘recipient’’ an ‘‘abduction’’ or inference of their origins in an act of manufacture. The visual properties of the index, in particular their visual complexity or technical virtuosity, realize the agency of their producer (whether the artist or the patron who commands his services) through the impact they have on the viewer or ‘‘patient,’’ inferring and experiencing the primary agent’s potency at one and the same time. The canoe prow-boards of Trobriand Islanders (figure 0.1), richly decorated with circles painted in bright colors with strong tonal contrasts, were weapons in the psychological warfare which was one component of kula-exchange. Overcome with awe at the magical powers inferred to be at the disposal of the canoe-owner by virtue of the captivating power of the prow-decoration, the viewers of such imagery were expected to lose hold of their normal wits, and trade their kula-shells at less than their true value. Likewise, the decoration on the Asmat shield, which forms the frontispiece to Art and Agency, was designed not to elicit aesthetic interest, except in the most elliptical understanding of the term, but rather to instill fear on the part of the opposing warrior, similarly to the terrifying gorgoneion, which was a popular device on the shields of ancient Greek hoplites.9 In these works of art, ‘‘beauty’’ and the decoding of ‘‘symbolic meaning’’ are not much at issue.10

Art and Agency between Art History and Anthropology Gell’s statement of his program needs to be seen against the background of dominant trends in anthropological theory and the anthropology of art against which he was reacting. Once placed in this context, it can be shown to have significant features in common with some of the more significant recent critical interventions in mainstream art history. Gell offers such strands of art-historical theory a potentially robust foundation in contemporary cognitive psychology, as well as a rigorous formalization in models designed to facilitate comparative historical and cross-cultural analysis. Gell’s polemics are strongly shaped by developments interior to the anthropology of art. He is particularly concerned to distance himself from ‘‘the slightest imputation that art is ‘like language’,’’ communicating symbolic meanings on the basis of language-like components.11 This concern is founded in a reaction against the structuralist anthropology of the 1970s, in which art was analyzed on the basis of linguistic

Osborne / Art’s Agency and Art History

1405135379_4_000 Final Proof page 4 12.1.2007 2:15pm Compositor Name: SJoearun

Figure 0.1 Trobriand canoe prow-board, courtesy of Dr. Shirley Campbell.

Osborne / Art’s Agency and Art History

1405135379_4_000 Final Proof page 5 12.1.2007 2:16pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 5 models. Scholars like Faris made extensive use of linguistic analogy in decomposing systems of artistic design, such as Nuba body art, into basic elements (visual forms such as circles, zigzags etc) which could be combined according to phase-structure rules, to produce ‘‘well-formed’’ utterances or representations, such as ‘‘poisonous snake.’’12 Gell, who worked extensively on body-art,13 objects to these ‘‘visual grammars’’ on the ground that such sets of visual forms do not share the structural properties of languages – for example arbitrary and diacritical relations between their base components, and a structured hierarchy of levels from phonemes through morphemes to syntax.14 Such approaches, Gell argues, elide both the social features of artworks as components in networks of social relations, and their specifically visual characteristics as presenting objects, characteristics grounded in highly specific material technologies of production.15 These criticisms are relevant also to the semiotic turn in art history. Gell makes no allusion to the work of such art historians as Norman Bryson or Mieke Bal, whose use of the analogy between language and art is rather differently conceived from that of the anthropologists with whose work Gell is concerned, and who apply the modes of analysis of the literary critic as well as those of Jakobsonian linguistics.16 Such work itself raises issues relating to its elision of the technologies and the social relations of production similar to those raised by anthropological work based in the language model. Bryson himself noted that applying the Saussurean conception of the sign uncritically left the critic in danger of ending up with ‘‘a perspective in which the meaning of the sign is defined entirely by formal means, as the product of oppositions among signs within an enclosed system.’’17 He acknowledged that his own Tradition and Desire: from David to Delacroix was ‘‘an analysis of what painting can become in the hands of those who both fear and desire that the meaning of a painting is, always, another painting.’’18 And yet Gell’s claims, although expressed in distinctly different language, run remarkably parallel to the case Bryson makes that ‘‘A virtue of considering the visual image as sign is that having relocated painting within the social domain, inherently and not only as a result of the instrumental placing there by some other agency, it becomes possible to think of the image as discursive work which returns into society.’’19 We should not too readily, therefore, assume that Gell’s critique of linguistic models should be taken to distance him from all aspects of the ‘‘linguistic turn’’ that has so marked post-structuralist and post-processualist work.

Osborne / Art’s Agency and Art History

1405135379_4_000 Final Proof page 6 12.1.2007 2:16pm Compositor Name: SJoearun

6 Jeremy Tanner and Robin Osborne Gell’s strictures concerning aestheticism also have a very specific anthropological target, namely the contemporary program in anthropological aesthetics developed by Howard Morphy, Jeremy Coote, and Anthony Shelton, but raise issues that reach well beyond anthropology. Coote and his colleagues specifically de-emphasize the art object as a focus of interest in favor of a more abstract focus on ‘‘valued formal properties of perception’’ characteristic of cultures, and specifically criticize attempts in the early anthropology of art to ‘‘relate aesthetics to social organisation or social structure.’’20 Coote argues that although the Dinka in southern Sudan do not have any tradition of manufacturing art objects, they do have a rich aesthetic experience, largely based in making metaphorical connections between their visual experience of cattle – their color configurations and other bodily forms, such as horn-shapes, patterns of movement – and wider aspects of experience. The anthropology of art, Coote suggests, should be concerned with identifying the set of formal qualities recognized and valued within a society, and the repertoire of visual skills through which they are instantiated. Art historians will recognize the influence of the early work of Michael Baxandall devoted to understanding fourteenth- and fifteenth-century Italian painting in fourteenth- and fifteenth-century terms.21 Gell objects that in seeking to abstract such a cultural aesthetic, Coote unwittingly assimilates Dinka aesthetic orientations to modern conceptions of aesthetic disinterestedness, developed in the philosophy of Kant.22 This ignores the social basis which gives life to the visual preoccupations identified by Coote. To understand why Dinka have these ‘‘aesthetic’’ interests in oxen, Gell argues, one must explore the role of oxen as ‘‘mediating elements in social praxis.’’23 Far from being disinterested, the Dinka appreciation and celebration of oxen is linked to the status ambitions of the youths who care for them and celebrate them. ‘‘What makes oxen ‘aesthetic’ is the role they play in locally dominant forms of competition.’’24 For Gell the aesthetic is of interest only when its social agency is considered. Notwithstanding his strictures, Gell does invoke both aesthetics and symbolic communication in the course of his analysis, but in a way which at least in principle is integrated with his emphasis on the primacy of social agency and perceptual cognition, rather than culture and symbolic meaning. While he is not interested in the analytics of beauty – ‘‘pure aesthetics’’ in the sense of Kant’s third Critique – Gell is very much concerned with ‘‘transcendental aesthetics’’ on the model of Kant’s first Critique, which explores ‘‘how the human sensory capacity construes and

Osborne / Art’s Agency and Art History

1405135379_4_000 Final Proof page 7 12.1.2007 2:16pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 7 gives form to stimuli.’’25 The exclusion of purely symbolic bases for artistic effects puts the spotlight upon exactly how those effects are achieved. In Gell the alternative to the purely symbolic sometimes seems to be some kind of transcendental aesthetics, universal perceptualcognitive bases for visual response, by virtue of which the abductions from indexes with which he is concerned have the ‘‘causal’’ character his theory asserts. Decorative motifs, which do not refer to or represent anything beyond themselves, provide a particularly favorable ground for developing his line of argument. In the sixth chapter of Art and Agency, ‘‘The critique of the index,’’ Gell argues that the patterns of decorative art have an intrinsic liveliness by virtue of the interrelationships between neighboring motifs. Four planar transformations – reflection, translation, rotation and glide-reflection – operating on constituent motifs form the basis of all patterns, and generate relationships of such complexity that they are indecipherable, thus slowing perceptual scanning of the entranced viewer, trapped in an interminable exchange with the decorated object. Apotropaic patterns, like Celtic knotwork and labyrinths, exemplify this ‘‘sticky’’ character of decorative patterns to a particularly marked degree. They thus lend themselves to protecting thresholds, both of buildings and, as tattoos, of orifices of the body: demons fascinated by the patterns get stuck, like insects on fly paper, and are thus diverted from acts of malevolence. Further, it is probably these perceptual qualities which explain the use of representations of one type of apotropaic pattern, the labyrinth, in contexts of transition between the worlds of the living and the dead as far apart in time and place as the passage grave at New Grange in Ireland and the tattooed bodies of the dead in the New Hebrides.26 Of course, these motifs may also permit abductions of symbolic themes, specific to particular cultures, but these bear a family relationship to each other by virtue of the fundamental cognitive indecipherability which is the basis on which such labyrinthine patterns exercise their characteristic agency. In practice, Gell recognizes that symbolic meanings and modes of communication are also crucial to art, and he is indeed an accomplished exponent of iconographic method. The eighth chapter of Art and Agency is an extended analysis of style in the art of the Marquesan Islands in Polynesia, particularly tattooing. Gell focuses on one particular motif, the etua, and shows how it can be transformed into numerous other motifs according to principles of production homologous with certain structural principles which also inform social interaction in the

Osborne / Art’s Agency and Art History

1405135379_4_000 Final Proof page 8 12.1.2007 2:16pm Compositor Name: SJoearun

8 Jeremy Tanner and Robin Osborne Marquesas. In doing so, he shows how the etua, a kind of generic godling, can be transformed into the hope Vehine or ‘‘buttocks women’’ and the Vai o Kena, ‘‘the bath of Kena,’’ a hero who plays a crucial role in the charter mythology of tattooing.27 Like other motifs in the corpus of Polynesian tattooing practices, which Gell analyzed in Wrapping in Images, ‘‘their nomenclature and meaning . . . is hardly intelligible except in the light of manifold mythic motifs and ritual symbolic devices,’’28 which Gell recounts in his fuller exposition of the material, and shows to be connected to the agentive function of tattooing in a way parallel to the style of Marquesan art, albeit on a different cultural level, a more specifically symbolic one. Thus, one design, the Malu, Gell glosses as ‘‘perfectly sealed vagina,’’ which he is able then to refer back to a mythological archetype of vaginaless woman who plays a central role in the charter mythology of tattooing.29 The symbolic meanings of these motifs – for example the ‘‘flying fox’’ which wraps itself in the second skin of its wings when at rest – are internally related to the agentive function of tattooing which Gell identifies, namely the creation of a second skin, which both ensures the retention of stored vital essence and protects the bearer from invasion by external agencies by sealing bodily orifices. Tattooing thus was ‘‘a means of magically fortifying young men and preparing them for military exploits.’’30 Iconography, symbolic meaning, and aesthetics do have a role to play in Gellian art analysis, but only properly framed within a robust sociology and cognitive psychology. It is not difficult to find parallels between Gell’s project and certain strands of art-historical work. Following the example of David Freedberg in The Power of Images (1989), a number of art historians have been concerned to explore the performative aspects in particular of religious and ritual arts.31 In seeking to distance himself from aestheticism and iconography, Gell also resonates with the concerns of some of the more important recent work critical of ‘‘new’’ and post-structuralist art historians. Donald Preziosi sees both as elements of a ‘‘eucharistic’’ model of art, in which the role of the interpreter is simply to reveal the truth immanent within the work itself.32 He, and other scholars advocating semiotic approaches to art, have criticized as logocentric the traditional focus on meaning immanent in the work and the language-based model of signification according to which such meanings are decoded.33 Paralleling Gell’s interest in the presenting image, such critics of traditional art-historical interpretation have sought to open out analysis in a number of directions. In particular, they focus on the ‘‘figurative,’’ the

Osborne / Art’s Agency and Art History

1405135379_4_000 Final Proof page 9 12.1.2007 2:16pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 9 specifically visual ways in which, for example, stories are narrated through pictures, rather than shortcutting analysis by seeing the image as secondary illustration of some primary text.34 Like Gell’s, such approaches see meaning as a transaction in time, produced in the context of a relationship between work and recipient. Production aesthetics is increasingly supplemented by reception aesthetics, exploring ‘‘the lines of signification opening out from the work’’35 as it is appropriated by individual viewers engaging the work in both its original context36 and the multiplicity of other settings in which a work might find itself through the vicissitudes of history.37 It is increasingly recognized that adequate analysis of the engagement between work and recipient requires categories additional to linguistic ones. W. J. T. Mitchell, asking ‘‘what do pictures really want?,’’ advocates an approach to pictures which goes beyond ‘‘looking at them as vehicles of meaning or instruments of power’’ and seeks to understand ‘‘the uncanny personhood of pictures,’’ and the kinds of physical, social, and erotic engagement to which that gives rise.38 He argues that a focus on the role of vision in ‘‘mediating social relations’’ in ways not reducible to language might provide theoretical resources to elaborate Michael Fried’s account of the ways paintings ‘‘arrest’’ and ‘‘enthrall’’ beholders as the necessary precondition of any symbolic communication.39 Semiotic approaches seek to supplement the ‘‘semantic’’ analysis characteristic of iconography, with attention to issues of rhetoric, which would analyze the ‘‘affective efficacy’’ of works of art.40 Affect is a dimension of experience not easily amenable to analysis in terms internal to linguistics or iconography. Correspondingly, scholars have sought extra-cultural bases to the agency of visual art. Bal and Bryson allude to the implications which Freudian psychology might have for the limits to the conventionalism in the functioning of visual art.41 Drawing on the phenomenology of perception, Effimova (1997) argues that it was the sensory effects of the paintings like Alexander Laktionov’s Letter to the Front that constituted the ‘‘realism’’ of Soviet Socialist Realism, not the banal party dogmas encoded in the iconography.42 Powerful manipulation of the play of light and shadow, suggesting dazzling sunlight and parching heat, exercised a magical affective charge by literally enlivening the viewers, making them feel more real, by the force of the paintings’ appeal to the human sensorium. With relatively few exceptions, however, art historians still show ‘‘minimal disciplinary awareness of [the] perceptual and cognitive psychology’’ which would be the necessary theoretical foundation to make good on

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 10 12.1.2007 2:16pm Compositor Name: SJoearun

10 Jeremy Tanner and Robin Osborne such promises.43 Consequently they tend to revert to language-like models of cultural signification. Thus Bal and Bryson describe reception in language-like terms, as involving viewers bringing ‘‘discourses of spectatorship’’ to bear on works of art.44 Similarly, the development of the concept of ‘‘visuality,’’ intended to focus analysis on the specifically figural aspects of art as opposed to language-like iconographic meanings, has served to bracket vision, emphasizing history and cultural convention at the expense of the physiology and psychology of vision.45 Where art history has started from semiotics and symbolism, and made partial openings toward psychology and sociology, Gell founds his program in psychology and sociology, leaving openings (notwithstanding the rhetoric) to semiotics and symbolism. In this respect, Gell offers a potentially fruitful complement to the kinds of moves beyond traditional theoretical paradigms which some art historians are trying to make. Furthermore, he develops an analytic apparatus to formalize these general orientations in the history and anthropology of art in order to permit systematic comparison across time and space. This he accomplishes through his account of the ‘‘art nexus,’’ to which we will now turn.

From the Art Nexus to the Gellogram Gell’s development of the ‘‘art nexus’’ in Art and Agency provides the means by which he seeks to give strong psychological and sociological foundations to the broad orientation to art he had developed in his paper ‘‘Technologies of Enchantment.’’ Chapters 2–5 of Art and Agency develop the art nexus concept in some detail, and represent the core of Gell’s new approach to art analysis. Gell’s conceptual panoply is in fact remarkably economical. He works from just four base concepts – artist, index, prototype, recipient – related to each other as ‘‘agents’’ or ‘‘patients’’ primarily through a cognitive operation Gell calls ‘‘abduction.’’ His whole framework is elaborated through the systematic exploration of the possible relationships between these conceptual elements. According to Gell, ‘‘art like situations’’ are those in which a material ‘‘index’’ permits ‘‘the abduction of agency’’ by an observer.46 Indexes are visible, physical objects. Abduction is a particular kind of cognitive operation, a probabilistic mode of inference, for example that a smile suggests a friendly person. It thus differs from strictly causal inference, or from purely logical deduction. The interpretation of the smile involves an

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 11 12.1.2007 2:16pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 11 abduction specifically of ‘‘agency,’’ implying some social other, with will and intentions, and the capacity to act in relation to, and thus affect as ‘‘patient,’’ the observer who has made the abduction, and reciprocally to abduct the intentions of and be affected as ‘‘patient’’ in turn by the actions of the observer performing the role of agent in their turn. The concept of agent implies that of patient and vice versa. Artists as agent act on indexes as patients, shaping through their skills the form works of art assume. But the material of the index can also act on the artist, as patient, setting limits, by virtue of its material properties, to the final form the work of art takes. Although the concept of abduction is ultimately drawn from Peirce’s semiotics, Gell’s more proximate source is Boyer’s work in cognitive psychology. Boyer’s work has been concerned to explain the extraordinarily wide occurrence of certain cultural phenomena, in particular in religion, such as a belief in some kind of spiritual beings or supernatural agents.47 Boyer argues that the propensity of human beings to believe in supernatural agencies, and the widely shared characteristics of what constitutes the supernatural, can be explained by constraints on the acquisition and representation of concepts given in certain intuitive mental modules, or information-processing devices, with which all human beings are equipped by nature. These include ‘‘naive’’ theories of physical objects, biology and mental processes, which account for the capacity of very young children to make distinctions between such categories as living things and objects, and then between plants, animals, and persons and natural objects and artifacts. Boyer goes on to argue that the great variety of supernatural concepts can be understood in terms of variations on a very limited number of templates, combining these basic ontological categories with a single breach or transfer, such as a person who has no body (violating physical expectations) and lives for ever (violating intuitive biology). The combination of basic category with striking but limited violation makes such concepts particularly powerful and highly memorable. This facilitates their distribution in the population of cultural representations, and explains the family resemblances of such concepts to each other. Amongst these basic templates Boyer identifies is an intuitive psychology, based on the assumption that others people’s behavior ‘‘is consistently directed by implicit principles concerning motivation, intentions, memory, reasoning and so on.’’48 These basic assumptions are not learned inductively from experience, or transmitted through culture, but are inbuilt orienting principles. Drawing on Boyer, Gell assumes that a powerful

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 12 12.1.2007 2:16pm Compositor Name: SJoearun

12 Jeremy Tanner and Robin Osborne interest in and capacity for the abduction of agency is built into the human mind. Further, such abductions are always in the nature of things materially mediated, whether through the body as a material medium of communication (smiles), or in artifactual form. Indeed, it is only on this basis that other more abstract modes of linguistic and symbolic communication could be erected. Further, it is by virtue of this intuitive social intelligence, and the interpretative procedures which constitute it, that we can abduct agency from objects, and that objects in particular can thus have an agency which is to all intents and purpose equivalent to human agency.49 The kind of agency specifically relevant to art-like situations is the agency which can originate an index through an act of manufacture.50 All manufactured objects can in principle motivate the abduction of the agency of their producer. Art objects are distinctive in so far as they exercise their social effectiveness, their agency, by means of the technical processes embodied in them by their producers, and their agency is realized in the process of abduction and by the way the index shapes that process.51 It is this which makes an artist and a work of art different from a Pol Pot minelayer and a mine. Thus, in the case of the Trobriand canoeprow, the dazzling effects of the colors and the eye-spots induce the abduction on the part of the viewer or ‘‘recipient’’ of the magical skills of the artist who produced the prow, and demoralization in relation to the power of their potential kula-exchange partner who has such forces at his disposal. Similarly, the technical skill of Ingres in depicting silk is part of what captivates us in his portrait of Mme Moitessier. Our admiration of the skill of the artist, transcending our capacity fully to grasp its operations, and respect for or envy of the person who commissioned the portrait and is thus able to command such skill, are internally constitutive of the agency of the work of art.52 The final concept in Gell’s repertoire is that of ‘‘prototype.’’53 This he uses to designate the model or entity which may be represented in an image – Mme Moitessier herself in Ingres’s portrait. As with his other concepts, Gell sees the relationship between prototype and index as being causal, rather than purely conventional. The physical facts of her appearance causally constrained Ingres’s choices in producing the index which is the portrait of Mme Moitessier. Artists, indexes, recipients, and prototypes can all be related to each other as agents and patients, and all by virtue of processes which Gell would see as being nearer to ‘‘causal’’ ones than to the calculus of symbolic

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 13 12.1.2007 2:16pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 13 meanings. Their interrelationships comprise the ‘‘art nexus.’’ Gell seeks to formalize implications of his theory of how art works by systematically exploring the logically possible relationships between his key terms. He tabulates (figure 0.2) the logically possible relationships of agency and patienthood between the four terms artist, index, prototype, and recipient. This allows him to map the space of possible art-like situations. ‘‘Artist as agent, prototype as patient’’ describes the situation of the creative artist characteristic of Western art history in which the very form of the work is something first conjured up in the imagination of the artist (artist agent, index patient). ‘‘Prototype agent, artist patient’’ is what would commonly be regarded as the typical formula for a portrait. It also describes the

AGENT Artist Artist as source of creative act

Artist

P A T I E N T

Index

Artist as witness to act of creation

Material stuff shaped by artist's agency and intention

Appearance of prototype dictated by artist. Prototype Imaginative art

Recipient

Recipient's response dictated by artist's skill, wit, magical powers, etc. Recipient captivated.

Index

Prototype

Recipient

Prototype Material Recipient controls artist's inherently cause of dictates to artist action. artist's action appearance of the form it (as patron) prototype assumes imitated by artist. Realistic art. Index as Recipient the cause of Prototype cause of the itself: dictates the origination and ‘self- form taken by form taken made’ index by the index Index as a ‘made thing’ Image or Prototype Recipient has actions of as cause of power over prototype index the prototype. controlled by Volt sorcery. means of Prototype index, a locus affected by of power over index prototype Index source of Prototype has Recipient as power over the power over patron recipient. Image recipient. of prototype used Recipient as to control actions ‘spectator’ or recipient. submits to Recipient as Idolatry. index. spectator

Figure 0.2 Gell, Art and Agency, table 1 (p. 29) ‘‘The art nexus.’’ ß Oxford University Press, 1998.

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 14 12.1.2007 2:16pm Compositor Name: SJoearun

14 Jeremy Tanner and Robin Osborne popular view of realistic art or photography in general. ‘‘Index agent, artist patient’’ corresponds to the oriental tradition of using found objects – objects in nature which captivate the artist – as works of art, like the feature rocks of Zen gardens. ‘‘Index agent, recipient patient’’ describes the kind of passive spectatorship assumed where art serves the state as propaganda.54 But it also describes ways in which an index might act on a patient in ritual contexts – the holy icon of the Virgin which heals the believer who touches it. ‘‘Recipient agent, index patient’’ is the implicit model of Elsner’s Art and The Roman Viewer, which explains the changed style of late antique art as the response to the changed needs of the viewer. Of course, all actual art situations are somewhat more complex than these binary relations, and Gell’s account of abduction duly presupposes that in practice an index should always be part of any art-like situation. More realistic modeling of the agent–patient relationships in Gell’s table involves combining certain types, or at least adding an index as one of the elements. Indeed the index functions as the pivot of most art situations – mediating the agency which is realized through it, and indeed the patienthood experienced in the abduction of agency from it.55 To accomplish this Gell develops two techniques, a kind of logical algebra of agent–patient relationships which allows him to express the complex hierarchy between elements in a simple linear form, and the slightly more complex treestructures, felicitously termed ‘‘Gellograms’’ by Whitney Davis, which allow the most complete modeling of art-situations, and perhaps the most perspicuous. Gell develops a simple algebra which produces expressions such as:56

[[[Prototype-A] ! Artist-A] ! Index A]
! Recipient-P The primary agent is always to the extreme left of the formula – here the prototype; and each of the secondary agents (the artist and the index) is patient with respect to the agents to their left. The long arrow is used to indicate the core of the situation, pointing toward the primary patient, affected by the index as agent. The patient does not have a direct relationship with the prototype, but nevertheless stands in a relationship of patienthood to the prototype by the abduction of the agency of the artist who produced the index, and of the agency of the prototype which somehow causally constrained the artist’s shaping of the material which was to form the index. The formula describes a passive spectator being ‘‘causally affected by the appearance (or other attributes) of a prototype of

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 15 12.1.2007 2:16pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 15 an artwork (the index), when this attribute is seen as itself causal of this response.’’57 Gell exemplifies this situation by the portrait of Dr. Johnson (figure 0.3). The posture of Johnson and his dominant orientation toward the viewer engenders a feeling of awe toward this English national figure, an awe which doubtless also affected the painter, and determined his particular portrayal of Johnson. Where the fame of the painter outweighs that of the sitter, as in contemporary viewings of Leonardo da Vinci’s Mona Lisa, the relationship between artist and prototype is reversed:

[[[Artist-A] ! Prototype-A] ! Index-A]
!Recipient-P This indicates that the compelling power of the portrait of Mona Lisa lies more in Leonardo’s artistry than in the sitter’s appearance.58 The underlying logic of these formulae can be represented more fully by means of tree-diagrams which cash out the recurrent relationships of agency/patienthood somewhat compressed in the formulae. Thus the formulae for the Johnson portrait (top) and the Mona Lisa (bottom) can be represented as follows (figure 0.4). The base-man represents the foundation of the nexus in question, the primary agent. Thus in the case of the portrait of Johnson, Johnson is the base-man acting on Reynolds as patient, while Reynolds under the impact of Johnson acts on the index,

Figure 0.3 Sir Joshua Reynolds, Portrait of Dr. Johnson, courtesy of Tate, London 2005.

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 16 12.1.2007 2:16pm Compositor Name: SJoearun

16 Jeremy Tanner and Robin Osborne Index Agent

Agent Agent

Patient

Patient

Patient

‘base man’

Index Agent

Patient Patient

Agent Agent

Patient

‘base man’

Figure 0.4 Gell, Art and Agency, fig. 4.3/1. ß Oxford University Press, 1998.

which in turn (by virtue of having been affected by Reynolds having been affected by Johnson) affects the patient-recipient. The importance of such diagrams is that they show how indexes contain within themselves hierarchically subordinate sets of agent and patient relationships. It is this which accounts for the complexity of the abductions of agency which may be made simultaneously from a single object on a number of different levels.59 In the case of the portrait of Johnson, and the Mona Lisa, the reader may not feel all that much more enlightened by the Gellograms, since the relationships are ones which are sufficiently simple, or sufficiently culturally familiar, that we can grasp them intuitively. Such formal modeling can, however, play a very valuable role in explicating the practical logic of more culturally distant or more complex situations. Gell takes the example of ‘‘nail fetishes’’ (figure 0.5) from the Congo in West Africa to demonstrate the analytic value of tree-diagrams.60 Aesthetic approaches to these anthropomorphic figures pierced with nails have not gone much beyond rather trivial analogies with their (fortuitous) parallels with some Western representations of suffering (St. Sebastian, Christ).

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 17 12.1.2007 2:16pm Compositor Name: SJoearun

Figure 0.5 Nkisi figure (nail fetish from the Congo region of Africa), British Museum 1905, 0525.3. ß British Museum.

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 18 12.1.2007 2:16pm Compositor Name: SJoearun

18 Jeremy Tanner and Robin Osborne Gell quotes the description by Dennett of the rituals which informed the making and use of such images: A palaver [meeting] is held, and it is decided whose Kulu [soul] it is that is to enter into the Muamba tree and to preside over the fetish to be made. A boy of great spirit, or else, above all, a great and daring hunter is chosen. Then, they go into the bush and call his name. The Nganga [priest] cuts down the tree, and blood is said to gush forth. A fowl is killed, and its blood mingled with the blood that they say comes from the tree. The named one then dies, certainly within ten days. His life has been sacrificed for what the Zinganga consider the welfare of the people. They say that the named one never fails to die. . . . People pass before these fetishes (Zinkici Mbowu) calling on them to kill them if they do, or have done, such and such a thing. Others go to them and insist on their killing so and so, who has done, or is about to do, some fearful injury. And as they swear and make their demand, a nail is driven into the fetish, and the palaver [business] is settled so far as they are concerned. The Kulu of the man whose life was sacrificed upon the cutting of the tree sees to the rest.61

Translating Dennett’s description into a tree-diagram (figure 0.6) brings out the practical logic of what, on a first reading, seems merely bizarre. This particular diagram has some unusual properties which make sense of

Recipient-P2 Victim

Index-A Fetish exercises judicial authority over

Fetish Supplicant Recipient-A1 invokes (by driving Index-P in nail)

Muamba tree Index-A

Priest Artist-A cuts down

causes death of

Brave hunter Index-P

Muamba tree Brave hunter Wild animals Index-P causes death of

Figure 0.6 Gell, Art and Agency, fig. 4.4/2. ß Oxford University Press, 1998.

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 19 12.1.2007 2:16pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 19 the unusual ways in which nail fetishes operate. First, the index has two base-men, both the priest, who chops down the tree which forms the index, and the brave hunter whose death is caused by the blood spilling occasioned by the cutting down of the Muamba tree. Second, the index is characterized by a dynamically oscillating cycle of passivity and activity, which constitutes its agency as a judicial authority. The passive tree is cut down, but then actively causes the death of the hunter, who is incorporated into the index through the mingling of spilt blood. In turn when the fetish, comprising hunter and tree, is acted on as patient by the supplicant, driving in the nail, the lethal power of the brave hunter, now hierarchically embedded within the fetish, is released against the victim. It is by virtue of this complex hierarchically embedded network of relations embodied in the nail fetish that it is able to perform its judicial role. The consultant, and the victims of such images, see not just a figure pierced with nails, but the whole complex network of relationships reaching backwards in time, and potentially able to reach forwards in time, indexed by the nail fetish figure. Each time it is acted upon by a consultant who drives a nail into its flesh, the agency of both priest and hunter, the two base-men, is abducted and mobilized in the fetish’s exercise of judicial authority over its victim. Gell develops his analysis of images to which supernatural power is ascribed in a subsequent chapter concerned with idolatry. Although created in very different circumstances to the Muamba tree image, the wax model, written name, portrait, or other object which is made to stand for a person so that what is done to the object may happen to the person, can be seen to rely on a process of abduction of agency which is essentially parallel to that involved in the case of the nail fetish. Gell goes on to show that this process is parallel also to that involved in the worship of idols, and indeed to all religious worship: ‘‘Since, in this world, God’s presence is inherent in . . . works of human agency . . . [h]is agency is enmeshed in ours, by virtue of our capacity to make (and be) his simulacrum.’’62 Gell’s insistence that human agents may be, as well as make, the image of God takes him, as he acknowledges, well away from the discussion of aesthetic or artistic form. In insisting on mapping the agency of the distributed person Gell necessarily emphasizes the psychological, sociological and, where that person is divine, theological. The strengths and the drawbacks of such an emphasis emerge particularly clearly from Gell’s analysis of Vela´zquez’s Rokeby Venus. Gell concerns himself with the slashed Rokeby Venus that was the product of an attack by the suffragette Mary Richardson in 1914. Gell is

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 20 12.1.2007 2:16pm Compositor Name: SJoearun

20 Jeremy Tanner and Robin Osborne attracted to this painting by Mary Richardson’s own statement: ‘‘I have tried to destroy the picture of the most beautiful woman in mythological history as a protest against the government for destroying Mrs. Pankhurst, the most beautiful character in modern history.’’ Gell brings out the parallelism in a Gellogram (figure 0.7) in which the violent relationship of Mary Richardson to the Rokeby Venus is parallel to the violent relationship of the prison warders (NB, not the government) to Mrs. Pankhurst, and entitles the slashed painting ‘‘Mrs. Pankhurst’’: Mrs Pankhurst becomes for Gell an inspiration for Mary Richardson in the same way that Venus was a prototype for the artist Vela´zquez. This is a provocative analysis, but how helpful is it? The Gellogram cannot express the full set of relationships implied in Mary Richardson’s short statement: nothing in the Gellogram brings out the special status of Venus not simply for Vela´zquez but in the common perception of mythological history; by placing Mrs. Pankhurst’s inspiration of Mary Richardson above the prison warders’ violence to Mrs. Pankhurst in the tree, the Gellogram implies that the violence was prior to, rather than simply parallel to, the inspiration, and hence we lose the sense that it is the beauty of Mrs. Pankhurst that is a crucial factor for Mary Richardson. More significantly, like Mary Richardson’s own statement, the Gellogram offers no explanation for why Mary Richardson might choose this Venus. When faced with this situation, which is culturally rather close to us, the Gellogram seems thin and impoverished.

The‘Slashed’ Rokeby Venus

sensation, outrage

Index 2-A Mary Richardson

Artist 2-A Mrs Pankhurst Prototype 1-A

Prison warders Recipient 1-A violence

The Rokeby Venus Index 1-P

violence

Mary Richardson Recipient 2-P

inspires

Mrs Pankhurst

Recipient1-P (the art public, the government, etc.)

Velázquez Artist 1-A

(=)

Venus Prototype 2-P

depicts

Venus

Artist -P inspires

Prototype 1-P

Prototype 2-A

Figure 0.7 Gell, Art and Agency, fig. 4.4/4. ß Oxford University Press, 1998.

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 21 12.1.2007 2:16pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 21 The epistemological status of Gellograms is open to debate. Are they objective mappings of the logical structure of the art-like situations they represent? Do they have a strongly subjective component by virtue of the role played by the abduction of agency? Different cultures or contexts of viewing may ascribe agency differently. One might switch the Johnson formula to the Mona Lisa one, simply by placing the Johnson portrait in the context of some kind of historical survey of art history, or a one-man show on Reynolds. Part of the value of this kind of formal modeling is that it opens up such issues in an analytically clear way. Who – both empirical ‘‘patients’’ and analysts – goes along certain pathways, and why? Why are some lines of abduction unraveled, others not? In addition to being a new tool to explain what is going on in the context of individual art situations, Gell’s formulae and diagrams offer what is potentially an extremely valuable tool for historical and comparative analysis. They allow one to focus on fundamental underlying relational structures. To what extent do similar kinds of social structure give rise to similar structures of artistic agency? How are the characteristic ways in which the agency of art is exercised in particular artistic traditions transformed by structural changes in cosmology? Which of the logically possible structures regularly occur, and which rarely if ever achieve historical instantiation and why? The last Gellogram in Art and Agency occurs in the discussion of idolatry in the chapter on ‘‘The Distributed Person,’’ and it maps ‘‘Ocular exchange as the medium for intersubjectivities.’’63 Gell himself finds no use for diagrams in the penultimate chapter, ‘‘Style and Culture.’’64 This abandonment of the diagram corresponds to a move in focus from the individual, particular work, to ‘‘larger unities.’’ Gell insists that ‘‘Artworks do not do their cognitive work in isolation; they function because they cooperate synergically with one another, and the basis of their synergic action is style.’’65 What follows is ‘‘minute analyses of specific Marquesan artworks and their transformations,’’ at the conclusion of which Gell insists that ‘‘visual style is an autonomous domain’’ and denies that ‘‘ ‘culture[,]’ in some general sense, is responsible for the visual style of artefacts,’’ while continuing to maintain that artworks ‘‘are the outcome of social initiatives which reflect a specific, socially inculcated sensibility.’’66 Just as earlier Gell has presented the artwork in relation to the distributed person, so now he suggests that ‘‘Artworks are shares or portions of a

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 22 12.1.2007 2:16pm Compositor Name: SJoearun

22 Jeremy Tanner and Robin Osborne distributed object corresponding to all of the artworks in the Marquesan system, distributed in time and space.’’ The distributed person was mapped in a Gellogram, but no similar attempt is made to map the distributed object. In the final chapter Gell fills out this notion of the distributed object. Each separate work in an artist’s oeuvre ‘‘is a modification, a recension, of previous ones, the leftovers of a particular cycle in a career-long generate-and-test sequence.’’67 Gell pursues this through an investigation of Duchamp’s The Network of Stoppages, a palimpsest of three works and hence a very particular demonstration of the way in which each work by a particular artist ‘‘invites us to adopt a particular perspective on’’ all other works by that artist.68 He reinforces this point with reference to the Maori meeting-house, which again ‘‘is an object which we are able to trace as a movement of thought, a movement of memory reaching down into the past and a movement of aspiration, probing towards an unrealized, and perhaps unrealizable futurity.’’69 By making this final move Gell makes it clear that his claims about the distributed object do not depend upon any particular mind: a society as well as an individual can constitute the appropriate level at which to understand unities of style. Why is it that Gell gives up the formal analyses via the Gellogram? Arguably Gell altered his aim in the final chapters of the book. The analytic mode was designed to shake off the traditional terms in which works of art are described and to enable the structures of the artwork’s power to be uncovered. But the structures of power, although analyzed for the individual work of art, do not reveal what is individual to that work of art. In particular they do not reveal what that work of art knows.70 For Gell the attraction of the ‘‘generate and test’’ model of creative agency is that it captures the way in which ‘‘ ‘‘thinking’’ takes place outside us as well as inside us,’’71 and that process of modification demands that we respect the artist’s oeuvre as a temporal object. Gellograms give us no idea of the changing perspective in which each stage of the sequence is viewed, no indication of the problem of events and tenses. To see both X as a precursor of Y and Y as a recapitulation of X involves shifting our vantage point,72 yet Gellograms present both (and neither) in a single scheme. By the last chapter Gell is insisting that ‘‘the ensemble of an artist’s works, strung out in time, constitutes a dynamic, unstable, entity; not a mere accumulation of datable artefacts. We can only appreciate it by participating in its unfolding life.’’73

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 23 12.1.2007 2:16pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 23

Art and Agency in Art History By no means all who have an interest in Gell’s ideas will wish to take on board the full analytical apparatus of the Gellogram. Indeed, both the extent to which one uses such devices and the elaboration of any particular diagram (potentially endless) depend on interpretive tact and the analytic payoff that such diagramming might have for a particular research problem. Consequently in the papers which follow, some have chosen just to use Gell’s core conceptual framework – emphasizing the agency of objects as an alternative to traditional focus on ‘‘meaning’’ – where others have chosen to use Gellograms to clarify the situations which they are interested in. David Wengrow shows how the sort of analysis which Gell applies to the Maori house can usefully be applied to Egyptian Mastaba tombs, both promoting an understanding of their combination of physical impenetrability with conceptual penetrability and revealing their social role in distinguishing between those who do and those who do not understand. Irene Winter employs analysis of the grammar of Sumerian and other languages to bring out the difference between agency which is explicitly asserted within a culture and agency ascribed by abduction by analysts from outside the culture. Winter points out that although the image may be potentially more ambiguous than a text, the way in which the language community from which the image originates understands agency has to be fed into the abduction of agency in the case of the image. The positive advantages of Gell’s ‘‘art nexus’’ as a helpful framework for the comparative analysis of art are stressed by Jeremy Tanner. He shows how indigenous conceptions of portraiture in ancient Greece, medieval East Asia, and Byzantium all correspond to Gell’s formula for portraits. Culturally variant conceptions of how the prototype causally determines the index, however, refract the basic portrait formula in quite distinctive ways specific to each case, and explain the rather different artistic technologies through which the artists of each culture sought to bind the prototype to the index and thus mediate its agency to recipients. The importance of understanding the wider cultural background of the image is further stressed by Jessica Rawson, in her paper on the Wanli emperor and his portrait. Having emphasized the usefulness of Gell’s formalization of relationships between artist, index, prototype, and recipient for mapping different effects, Rawson questions whether the agency of

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 24 12.1.2007 2:16pm Compositor Name: SJoearun

24 Jeremy Tanner and Robin Osborne artifacts can be understood as self-contained and independent of the broader intellectual framework. In examining another royal portrait painted more or less contemporaneously with that of the Wanli emperor, but in a very different cultural background, Jessen Kelly puts emphasis on the usefulness of the idea of the technology of enchantment as she extends the use of Gell to works firmly in the Western artistic tradition. She finds Gell’s notions of captivation and enchantment very directly applicable to Elizabethan miniature portraits, but emphasizes how symbolic decoding was an essential part of the social agency of these works. Technologies of enchantment play a major part also in Jeffrey Quilter’s examination of the art objects of a culture about which we know very little except what can be deduced from the artifacts themselves, the Moche. Taking two different products of Moche art, Quilter asks how far Gell’s notions of the art nexus, the technology of enchantment, and the distributed person can be applied to each. Quilter argues that the ideas of the technology of enchantment and the distributed person prove particularly good to think with in regard to Moche wall art, but that ignorance of how other social agents worked makes it unclear how to apply Gell’s idea of the art nexus when dealing with truly prehistoric societies. Peter Stewart turns his attention to Gell’s work on religious images. He acknowledges that the power of ancient images was not bound up with their precise form, and finds Gell’s notion of the distributed person helpful, but argues at the same time that iconography does matter. He also notes that Gell’s analysis is entirely consonant with ancient theories of causation, but that awareness of the contradictions of anthropomorphic representation of supernatural powers is something present within societies which embody divinities in this way, not something imposed from outside. The final two papers look particularly closely at how Gell’s arguments work. Robin Osborne explores at some length the value of the Gellogram, drawing attention in particular to the question of how exactly we gloss the arrows which indicate the relationships between the parties involved. Taking Greek painted pottery as his test case, he explores how we might map differently reactions to art and to pornography. At the same time Osborne urges that Gellograms share the limitations of much essentially structuralist analysis in accounting for change over time. Whitney Davis concludes the volume with a close critique of the way Gell develops his argument. He discusses the ways in which Gell’s work shares

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 25 12.1.2007 2:16pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 25 with much traditional art history the redescription of the artwork in other terms, and proceeds to set Gell both in his anthropological context and against wider theoretical currents in the 1980s and 1990s. He explores the necessary opacity of abduction of agency in the case of art, and argues that, for all Gell’s resistance to the aesthetic, it is the very aesthetic qualities of the work of art that resist abducting agency in an unlimited regress and which ensure that agency coheres in the art object itself.

Notes 1

2 3

4 5

6 7 8 9 10 11 12 13 14 15 16 17 18 19

Celebration: C. Pinney and N. Thomas, eds., Beyond Aesthetics: Art and the Technologies of Enchantment (Oxford, 2001); criticism: R. Layton, ‘‘Art and Agency: a reassessment,’’ Journal of the Royal Anthropological Institute 9 (2003), pp. 447–64. We are grateful to James Whitley for comments on an earlier draft. A. Gell, ‘‘The technology of enchantment and the enchantment of technology,’’ in J. Coote and A. Shelton, eds., Anthropology, Art and Aesthetics (Oxford, 1992), pp. 40–63 at p. 43. Criticized in Gell, ‘‘Technology,’’ p. 43. A. Forge, ‘‘Art and environment in the Sepik,’’ Proceedings of the Royal Anthropological Institute for 1965 (London, 1966), pp. 23–31; ‘‘The Abelam artist,’’ in M. Freedman, ed., Social Organization: Essays Presented to Raymond Firth (London, 1967), pp. 65–84. A. Gell, Art and Agency: an Anthropological Theory (Oxford, 1998), p. 10. Ibid., p. 6. Ibid., p. 20. Ibid., p. 6. Ibid., pp. 6, 71; Gell, ‘‘Technology,’’ pp. 40–6. Gell, Art and Agency, p. 14. J. Faris, Nuba Personal Art (London, 1971). Gell, Wrapping in Images: Tattooing in Polynesia (Oxford, 1993). Gell, Art and Agency, pp. 163–5. Gell, ‘‘Technology,’’ p. 43. M. Bal and N. Bryson, ‘‘Semiotics and art history,’’ Art Bulletin 73 (1991), pp. 175–208. N. Bryson, Vision and Painting: the Logic of the Gaze (London: 1983), p. xii. N. Bryson, Tradition and Desire: from David to Delacroix (Cambridge, 1984), p. xix. N. Bryson, ‘‘Semiology and visual interpretation,’’ in N. Bryson, M. A. Holly and K. Moxey, eds., Visual Theory (Cambridge, 1991), pp. 61–73 at p. 70.

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 26 12.1.2007 2:16pm Compositor Name: SJoearun

26 Jeremy Tanner and Robin Osborne 20 J. Coote, ‘‘ ‘Marvels of everyday vision’: the anthropology of aesthetics and cattle-keeping Nilotes,’’ in Coote and Shelton, eds., Anthropology, Art and Aesthetics, pp. 245–73. 21 Painting and Experience in Fifteenth Century Italy (Oxford, 1972). 22 Gell, Art and Agency, pp. 1–4; ‘‘On Coote’s ‘Marvels of everyday vision’,’’ Social Analysis 38 (1995), pp. 18–31. 23 Gell, ‘‘On Coote’s ‘Marvels’,’’ p. 24. 24 Ibid., p. 29. 25 H. Morphy in J. Weiner, ‘‘Aesthetics is a cross-cultural category,’’ in T. Ingold, ed., Key Debates in Anthropology (London, 1996), pp. 251–4 at p. 252. 26 Gell, Art and Agency, pp. 81–94. 27 Ibid., pp. 174–5. 28 Gell, Wrapping in Images, p. 95. 29 Ibid., p. 86. 30 Ibid., p. 57. 31 Most notably H. Belting, Likeness and Presence: a History of the Image before the Era of Art (Chicago, 1994). 32 D. Preziosi, Rethinking Art History: Meditations on a Coy Science (New Haven, 1989), pp. 15–17. 33 Ibid., p. 16; Bal and Bryson, ‘‘Semiotics,’’ pp. 175–6. 34 For examples of such work see Bal and Bryson, ‘‘Semiotics,’’ p. 175, and M. Bal, Reading ‘‘Rembrandt’’: Beyond the Word–Image Opposition (Cambridge, 1991). 35 Bal and Bryson, ‘‘Semiotics,’’ p. 179. 36 J. Elsner, Art and the Roman Viewer: the Transformation of Art from the Pagan World to Christianity (Cambridge, 1995). 37 R. H. Davis, Lives of Indian Images (Princeton, NJ, 1997). 38 W. J. T. Mitchell, ‘‘What do pictures really want?’’ October 77 (1996), pp. 71–82. 39 M. Fried, Absorption and Theatricality: Painting and the Beholder in the Age of Diderot (Berkeley, CA, 1980). 40 Bal and Bryson, ‘‘Semiotics,’’ p. 193. 41 Ibid., p. 199. 42 A. Effimova, ‘‘ ‘To touch on the raw’: the aesthetic affections of socialist realism,’’ Art Journal, Spring (1997), pp. 72–80. 43 Preziosi, Rethinking, p. 49. 44 Bal and Bryson, ‘‘Semiotics,’’ p. 179. 45 C. Wood, ‘‘Introduction,’’ in idem, ed., The Vienna School Reader: Politics and Art Historical Method in the 1930’s (New York, 2000), pp. 9–72 esp. p. 26; cf. R. Nelson, ed., Visuality Before and Beyond the Renaissance: Seeing as Others Saw (Cambridge, 2000). 46 Gell, Art and Agency, p. 13.

Osborne / Art’s Agency and Art History 1405135379_4_000 Final Proof

page 27 12.1.2007 2:16pm Compositor Name: SJoearun

Introduction: Art and Agency and Art History 27 47 P. Boyer, ‘‘Cognitive constraints on cultural representations,’’ in L. A. Hirschfeld and S. A. Gelman, eds., Mapping the Mind: Domain Specificity in Cognition and Culture (Cambridge, 1994), pp. 391–411; ‘‘What makes anthropomorphism natural: intuitive ontology and cultural representations,’’ Journal of the Royal Anthropological Institute 2 (1996), pp. 1–15. 48 Boyer, ‘‘Cognitive Constraints,’’ p. 403. 49 Gell, Art and Agency, p. 15. 50 Ibid., p. 23. 51 Gell, ‘‘Technology,’’ p. 44, Art and Agency, p. 68. 52 Cf. Gell, Art and Agency, pp. 24–5 on recipient, pp. 68–72 on skill and captivation. 53 Ibid., p. 25. 54 P. Zanker, The Power of Images in the Age of Augustus (Ann Arbor, 1988). 55 Gell, Art and Agency, p. 53. 56 Gell’s examples: ibid., pp. 51–9. 57 Ibid., p. 53. 58 Ibid., p. 33. 59 Ibid., p. 54. 60 Ibid., pp. 58–62. 61 R. E. Dennett, At the Back of the Black Man’s Mind; or Notes on the Kingly Office in West Africa (London, 1906; repr. London, 1968), p. 93. 62 Gell, Art and Agency, p. 114. 63 Ibid., p. 120. 64 Ibid., p. 154. 65 Ibid., p. 163. 66 Ibid., pp. 214, 216, 220. 67 Ibid., p. 237. 68 Ibid., p. 250. 69 Ibid., p. 258, the last paragraph of the book. 70 Compare Peter de Bolla, Art Matters (Cambridge, MA, 2001), p. 55 for concern with ‘‘what a response to the question ‘What does this artwork know?’ might be.’’ 71 Gell, Art and Agency 236. 72 Ibid., p. 241. 73 Ibid., p. 242, our emphasis.

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 28 12.1.2007 2:19pm Compositor Name: SJoearun

1

Enchantment and Sacrifice in Early Egypt David Wengrow

Decorative patterns applied to artefacts attach people to things, and to the social projects those things entail. Alfred Gell, Art and Agency, p. 74

Art or Enchantment? In his autobiographical novel Waiting for Snow in Havana: Confessions of a Cuban Boyhood, the historian Carlos Eire recalls an early childhood visit to a church.1 His account begins with memories of frightening religious icons and the unfamiliar smell of incense. It then turns, quite unexpectedly, to the ornamented surfaces of the church pews. The author recalls how his mind was drawn to the endless rows of wooden slats. He remembers trying at first to make sense of the hollow patterns they formed, which reminded him of an optical illusion. Were they the outlines of chalices or opposed human faces? What matters, he concludes retrospectively, was the paradox – the sensation created by images that were absorbing, precisely because their meanings and grammar escaped definition. More than anything else it was his desire to merge with and inhabit the paradox that came to dominate Eire’s earliest recollection of a church service. Writing a century or so earlier, Emile Durkheim chose the image of a child entering a church congregation to elucidate the subject matter of the (then nascent) discipline of sociology. The church-member, he observed,

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 29 12.1.2007 2:19pm Compositor Name: SJoearun

Enchantment and Sacrifice in Early Egypt 29 finds the ‘‘beliefs and practices of his religious life ready-made at birth; if they existed before him, they existed externally to him.’’2 Social facts, as he famously argued, are defined by this quality of exteriority, which they share with things. They cannot be observed through introspection, because they are not present in the mind of any one individual. Rather they are constituted in the flow of interaction between individuals. The observer of social facts, wrote Durkheim, must put aside the ‘‘confused, fleeting, subjective impressions’’ that make up personal experience, and adopt instead the critical and detached perspective proper to the scientific observation of objects.3 What then is the observer of social facts to make of Eire’s recollection, confused, fleeting, subjective – and yet instantly recognizable and durable in memory – as it is? What are the links between experiences of the kind he describes and the longer-term processes that bind individuals into larger social projects? Alfred Gell’s Art and Agency carves out a theoretical space within which questions of this kind become central to understanding human social conduct. He contends that objects which act, even momentarily, to enlarge or deflate the personal agency of human actors – that is, their capacity to form decisions and influence others – play a key role in the mechanics of social encounters. The church, from this perspective, does not symbolize a community of believers. Nor does it constitute a ready-made system of social relations. Rather it forms a complex set of material strategies for keeping children (and adults) riveted to pews, so that they can be exposed to a range of spiritual beliefs and holy rites. Eire’s experience also points toward another central argument of Art and Agency : that the visual trickery of the carved church pew is no less significant in securing this result than a depiction of Christ or the story of the Passion. It forms an integral part of the building’s psychological arsenal, comparable (at least in its effect on the young Eire) to the dazzling ornamentation of a Maori meeting-house, within which ‘‘all merely private, independent, thought is overwhelmed and only those cognitions which actually emanate from the house, those cognitions which are part of the house’s very structure, are attainable.’’4 I will return to the lessons of the Maori meeting-house in due course. Gell’s emphasis on the use of objects to manipulate the senses for social ends – what he calls ‘‘the technology of enchantment’’ – differentiates his work from many other approaches to art and architecture. In terms of our own received categories of experience it almost seems more useful, or at least intuitive, to think of Art and Agency as a social critique of advertising

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 30 12.1.2007 2:19pm Compositor Name: SJoearun

30 David Wengrow strategies rather than art. This is just one of a number of dichotomies that Gell’s work brings into question, another being that between art and magic. He ascribes particular importance to objects and images that exploit ‘‘characteristic biases of human visual perception so as to ensnare us into unwitting reactions.’’5 Familiar examples include sharp color contrasts, eye-spots, complex geometrical patterns, knots and bindings, things that seem unnaturally small or large, or that are partially concealed, and the layering or nesting of one thing inside another. Objects exhibiting these features by no means exhaust Gell’s subject matter, although they do serve to identify a ‘‘primordial kind of artistic agency.’’6 Nor does he go quite so far as to argue that the social properties of objects are rooted in hard-wired mental responses. What he is, in my view, concerned to show is how, in the course of encounters between people, sensory effects are deployed both routinely and ceremonially as cues to initiate certain chains of thought and action, and to suppress others, like cognitive charges injected into the social formation. In what Ku¨chler has termed the ‘‘fluid mechanics of persons and things,’’ objects and images merge with human actors, extending and diverting their social agency beyond the physical confines of the body.7 In Art and Agency Gell also reserves particular scorn for a basically Western art world that habitually excludes such effects from its evaluation of art objects. Somewhat inexplicably, he is scarcely more generous to those who seek to reunite the cognitive effects of art objects with indigenous frameworks of cultural perception and symbolic meaning. The weaknesses and paradoxes of this latter critique have already been widely pointed out. The former critique is, however, an important one. Serious intellectual engagement and emotional profundity have long been considered proper responses to art. When we see people exhibiting these kinds of behavior in the vicinity of objects that they are not touching or using, other than in religious contexts, it is usually a sign that art is believed to be present. By contrast, sensory tricks (the very term denotes cheapness and deception) are marginalized in the high cultural discourse of art appreciation. From the connoisseur’s point of view, the quick thrill they afford is – like fast food to the gourmet – a low and easy form of consumption, best confined to the circus, the children’s entertainer, or the public billboard. Until quite recently, genuine shock, laughter, confusion, and revulsion were considered inappropriate responses to art, while professional advertisers have turned their stimulation into a commercial form of artistry. It seems significant, given Gell’s rejection of the notion of art as symbolic

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 31 12.1.2007 2:19pm Compositor Name: SJoearun

Enchantment and Sacrifice in Early Egypt 31 communication, that these same responses are visceral and often prevent the use of language, leaving us literally lost for words and prone to action, or to being acted upon. When a nineteenth-century critic wrote of ancient Egyptian sculpture that it is ‘‘more calculated to astonish the eye than to enlighten the mind’’ he was, by implication, consigning it to a category of things (‘‘curiosities’’) that are not properly considered as art.8 The contrast here was with GrecoRoman sculpture, which at the time elicited entire literary odes on the nature of truth and beauty. That which was found to be spontaneously entertaining, dazzling, grotesque, or amusing was also deemed superficial, vulgar, and plebeian. Many modern art-works may be understood as attempts to rescue these same types of response from trivialization and exploitation, by installing objects that evoke them within the quasi-sacred spaces of galleries and museums. One irony of this process has been the colonization of these spaces by shops selling cheap reproductions of modern art-works, so that objects intended as satirical commentaries on the world of commodities are themselves worked back into the consumer market as mass-produced images. For Gell, the task of the anthropology of art clearly is not to enfranchise previously excluded forms of expression by learning how to respond to them ‘‘as art’’; rather it is to rediscover the importance of enchantment in the conduct of all social life and, by extension, in the historical formation of past societies. The salience of ancient Egyptian art in considering such broad matters has traditionally resided in its status as an enduring alternative to Western principles of representation, notably through its unfamiliar treatment of visual form and perspective.9 Here, however, I wish to focus upon a period of early Egyptian history from which very few pictorial representations have survived. The First Dynasty, covering the opening centuries of the third millennium BC (c.2950–2775 BC), witnessed the consolidation of a territorial state extending from the Mediterranean coast to the First Cataract of the Nile at Aswan, under the centralized rule of a sacred monarch. My particular interest will be in a type of funerary architecture known in the literature as the ‘‘mastaba tomb,’’ which first appeared during this period and – in this, its earliest form – seems to exemplify an extreme case of the agency of art objects; that is, their ability to ensnare individuals into acts of submission.10 Not only were mastaba tombs striking works of artistry, with elaborately patterned surfaces, but during the First Dynasty their completion also involved the ritual sacrifice of many individuals, whose bodies were incorporated into the architectural

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 32 12.1.2007 2:19pm Compositor Name: SJoearun

32 David Wengrow plan of the tomb complex. Such practices, which were not confined to mastaba tombs, reached a pinnacle during this period and then were largely expunged from elite funerary rites in Egypt. Accordingly they constitute a relatively brief phase in the cultural formation of ‘‘the state,’’ best viewed as a continuum rather than as a once-and-for-all evolutionary threshold. I will not be concerned here with the vexed, and probably unanswerable, question of who was buried in mastaba tombs (kings, perhaps, or high courtiers), but with defining the agent-like properties of the tombs themselves as indexes of social power, and their role within a wider politics of inclusion and exclusion, centered upon relationships between the living and the dead.11

Mastaba Tombs of the First Dynasty A high escarpment borders the west bank of the Nile at North Saqqara. Over the course of the First Dynasty this prominent strip of land was colonized by a series of mastaba tombs, their fac¸ades aligned along the escarpment edge, overlooking the meeting-place of the valley and delta below. The Arabic word mastaba means simply ‘‘bench,’’ and describes (albeit a little disparagingly) the niched and painted superstructures of these tombs, which would have been highly visible features within the early dynastic landscape (figure 1.1).12 All mastaba tombs, with a single exception at Naqada in southern Egypt, comprised two distinct structural elements: a series of subterranean chambers, housing the main burial in a central pit, and a platform or ‘‘superstructure’’ that remained visible, but impenetrable, above ground. The subterranean chambers were hewn deep into bedrock and lined with brick, plaster, timber, and reeds. Traces of painted matting have sometimes been detected on the walls of the burial pit, and at least one such chamber at Saqqara was faced with gold foil, embossed to resemble similar reedwork. Bodies, mostly removed or destroyed long ago by looters, were interred in large wooden coffins, of which only the base usually remains. Slightly later examples suggest that they may have been decorated with external niches, replicating on a smaller scale the outer appearance of the tomb superstructure.13 Throughout the First Dynasty, new structural features were added, heightening the theatrical qualities of the tomb as a setting for ritual action. They included the construction of imposing staircases, descending into the burial chamber along brick-lined ramparts

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 33 12.1.2007 2:19pm Compositor Name: SJoearun

Enchantment and Sacrifice in Early Egypt 33

Figure 1.1 Mastaba tomb and surrounding burials, Saqqara, First Dynasty (reconstruction after Emery, Great Tombs, II, pl. 1).

and thus enhancing movement into and out of the tomb as a choreographed social spectacle. The completion of funerary rites was marked by sealing this entranceway with a series of massive stone portcullises, lowered on ropes: a dramatic act of closure, fixing in collective memory the relationships formed during the course of the ritual.

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 34 12.1.2007 2:19pm Compositor Name: SJoearun

34 David Wengrow A further phase of construction involved the establishment of the large mud-brick superstructure with its foundations at ground level, encompassing the chambers below. In the middle-to-later part of the First Dynasty (reigns of Den and Anejib) an intermediate stage of building was sometimes introduced, resulting in the creation of a low tumulus or stepped platform encased inside the superstructure. The form of the superstructure itself was always rectangular, with the largest examples extending over 50 m in length and standing more than 3 m high. Its hollow interior was divided into a series of magazines, the purpose of which was to be filled with objects, including vast quantities of food, wine, and luxury oils (presented in ceramic and stone containers), serving-vessels, cosmetic equipment, and finely crafted furniture, but also agricultural implements and craft tools. In filling the tomb with objects, participants in elite funerary rites also filled it with the names attached to those objects in the form of incised marks, seal impressions, and inscribed tags. Many scholars have attempted to distill the identity of the tombowner or patron from these names. Typically, however, they refer, not to any one individual, but to a cluster of individuals: a collectivity, comprising a king, his successor, and between one and four high officials.

The Death of an Artist Considerably more can be said about the external appearance of mastaba superstructures than about their internal de´cor. During the First Dynasty their outer surfaces conformed to an ornamental design known as the ‘‘palace fac¸ade,’’ comprising a continuous series of large and small niches sunk into the walls at close, regular intervals. The basic constituent of the palace fac¸ade design, a panel ornamented with a series of vertical recesses, also provided the prototype for a motif, known as the serekh, which enclosed the writing of royal names. The serekh was usually surmounted by an animal figure, typically the Horus falcon, referring to that aspect of the royal person that embodied a deity. The combination of serekh and Horus name (not a birth name but one of a series of epithets acquired by the king upon accession) appears on a wide variety of prestigious funerary objects ranging from stone stelae (figure 1.2), known mainly from early royal burials at Abydos, to the vessels and implements deposited en masse within tombs. The serekh was also considered an appropriate form for the ornamentation of elite, and perhaps only royal, bodies in death, as

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 35 12.1.2007 2:19pm Compositor Name: SJoearun

Enchantment and Sacrifice in Early Egypt 35

Figure 1.2 Transformation of a prototype: ‘palace fac¸ade’ forming part of the Horus name of King Djet, on his funerary stele from Abydos. After J. de Morgan, Recherches sur les origines de L’Egypte (Paris: Ernest Leroux, 1897), fig. 797.

exemplified by jewelry recovered from high-status tombs at Abydos and Giza.14 In the making of high-status burials, the ‘‘palace fac¸ade’’ was therefore deployed at a variety of scales, from the miniature to the

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 36 12.1.2007 2:19pm Compositor Name: SJoearun

36 David Wengrow monumental. This mediating role was most fully realized in the case of mastaba tombs, where the image of the palace fac¸ade forms a unifying element, linking both the static and mobile parts of the tomb (i.e. the structure and its moveable contents) and its inner- and outermost surfaces, emanating from the body and its accoutrements via the coffin and decorated burial chamber to the exterior walls of the superstructure. Mastaba superstructures, which provided no physical point of access to the tomb’s interior, were brightly painted. Surviving traces of bold, multicolored designs have been detected on a number of tomb fac¸ades, occasionally in good states of preservation (figure 1.3). They replicate in precise detail the appearance of decorated reed mats lashed to a framework of wooden poles, and have a kaleidoscopic quality that is at once mesmerizing and disorienting.15 Positioned below them, riveted with wooden pegs to a low bench surrounding the superstructure, were row upon row of large animal heads modeled in clay, into each of which were inserted the horns of a bull. Some three hundred of these heads were discovered, still in place, around one mastaba at Saqqara. Each was carefully located in front of a niche, emerging from the tomb to confront the viewer at waist height, as if charging toward him. As I have already mentioned, human as well as animal bodies were incorporated into the architectural scheme of mastaba tombs, forming a protective wall of subsidiary burials around the superstructure. Over sixty individuals were interred around the latter tomb, with its ghostly herd, their bodies aligned within long trenches divided into pits and sealed simultaneously. Osteological remains from other funerary complexes of the First Dynasty confirm the practice of human sacrifice on a large scale

Figure 1.3 Painted decoration of Mastaba tomb fac¸ade (after Emery, Great Tombs, III, pl. 7).

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 37 12.1.2007 2:19pm Compositor Name: SJoearun

Enchantment and Sacrifice in Early Egypt 37 in association with prestigious burials.16 A direct relationship between ritual killing and the preparation of the tomb is indicated by the contents of a well-preserved subsidiary grave, in which the individual was buried together with pots containing residues of green, red, black, and yellow paint, traces of which were also found on the mastaba fac¸ade.17 Other sacrificial victims were buried with copper tools, weapons, and various types of storage and serving vessels.

From Oceania to Egypt The path of the historical imagination that leads from latter-day Oceania to early Egypt is no virgin trail.18 As my point of departure, and in considering the agent-like properties of mastaba tombs, I will take the final case study of Art and Agency. There Gell discusses a phenomenon more typically archaeological than ethnographic in scale: the construction of Maori meeting-houses on North Island between 1850 and 1930. During that period, growing investment in the construction and decoration of these buildings was in part a response to European colonial incursions, forming an alternative to traditional modes chiefly of competition through violence. Maori meeting-houses encompassed the living and the ancestral dead as parts of a continuum. Gell describes the gradual absorption of living inhabitants into the material fabric of the house, where they were rendered as carved ancestral images, like so many ‘‘fixtures’’ and ‘‘furnishings.’’19 Nicholas Thomas has further described the cumulative effect of these carvings, and the mesmerizing patterns that enfolded them, as creating a politicized environment for the conduct of ongoing hospitality between rival groups: A focused grasp of the whole carving is perceptually difficult; this optical complexity and restlessness reinforces the awe-inspiring effect that is certainly often intended by expressions of challenge and defiance. These figures cannot be gazed upon with any confidence; they tend rather to crowd out the viewer’s own sense of presence and power.20

He goes on to explain how the acuteness of the viewer’s susceptibility to carved designs increases with social distance from the ancestral collectivity objectified by the house. Outsiders are unable to go beyond a sense of awe and intimidation, which is retained as a memory trace upon leaving the confines of the building. By contrast, the visitor who identifies

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 38 12.1.2007 2:19pm Compositor Name: SJoearun

38 David Wengrow genealogically with the carved figures ‘‘can move beyond this state of uncertainty . . . what may be initially confusing and threatening can be recognized as the ancestral source of self, and can be incorporated into it.’’21 Where the psychological violence of house decoration has replaced the physical violence of warfare, the experience of the ‘‘insider’’ is presumably analogous to that of the male initiate, granted status through possession of decorated weaponry and body tattoos. The discernible similarities between Maori meeting-houses and Egyptian mastaba tombs of the First Dynasty are twofold. Firstly, they relate to their common function as objectifications of, and containers for, ancestral power. Secondly, they relate to the appearance and spatial distribution of surface features, notably the dazzling decoration of walls and the predominance of rectilinear relationships and flat fac¸ades. The foregoing discussion suggests that these ‘‘family resemblances’’ should not be dismissed as superficial, although in a purely technical sense that is exactly what they are. As Vedia Izzet has pointed out, with reference to ancient Etrurian architecture, flat-fac¸aded buildings limit the hypotheses (or ‘‘abductions’’) formed in the mind of an approaching visitor as regards what lies behind the surface. While curvilinear buildings betray something of their volume, flat fac¸ades afford no such clues, rendering the distinction between inside and outside ‘‘more immediate’’ and ‘‘more impenetrable.’’22 From this perspective, the surface decoration of mastaba tombs constitutes a double paradox. Boldly painted imitations of vividly decorated reed matting and endless rows of sunken recesses advertised the permeability and animacy of a structure that was in reality solid, static, and protected. Equally suggestive are the differences between these two types of ancestral building, which – as this last observation suggests – are in some senses the reversal or mirror-image of one another.23 Maori meeting-houses brought the dead into the domestic world of the living, where they shared a common dwelling-space. Their visual effects were woven into the rhythms and movements of everyday life (entrances, exits, meals, exchanges, and other group activities), and took form gradually in relation to individual processes of aging, dying, and achieving ancestral status. By contrast, mastaba tombs provided the focus for a finite ritual performance that reversed the normal flow of life and material goods through orchestrated acts of sacrifice and closure. Incorporation into the ancestral unit objectified by the tomb was achieved as a controlled and collective process, which superseded the ordinary timing of individual death and created a space that was sealed off from further intercourse.

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 39 12.1.2007 2:19pm Compositor Name: SJoearun

Enchantment and Sacrifice in Early Egypt 39 As visible monuments, however, the agency of mastaba tombs extended beyond the ritual process of their creation. Their simultaneous capacity to captivate and repel, to generate life and to kill, would have had ongoing implications for their role as containers, not just for human bodies, but also for objects consigned to them by the living. For the privileged few who participated in such transactions and survived, the world behind the sealed fac¸ade remained accessible as a layered structure, with internal parts as well as external surfaces, receding inwards from the tomb fac¸ade via a chain of replications, toward the intimate confines of the burial chamber and body within. The same ritual artistry that sought to protect the tomb from the harmful intentions of outsiders made possible its penetration, not as a physical movement, but as a movement of mind, anchored in the shared memory of past sacrifice.

The Enchantment of Authority In discussing Gell’s ideas in relation to ancient Egypt, I have avoided some of the more obvious remarks that could be made about the ritual animation of images and the consubstantiality of objects, persons, and gods. Central though these topics are, it seems to me that the more novel and provocative aspect of Gell’s theory, particularly from the standpoint of early Egyptian art, lies elsewhere. Although he made little explicit reference to politics, Gell’s ‘‘art nexus’’ is an inherently political concept, constituted by the tensions and harmonies of social interaction. Insofar as all the ‘‘art’’ in Art and Agency may be said to possess a common kind of value, it is a value which expresses the operation of relationships and regimes of control between social actors, of which the sacrifice of life is an extreme manifestation. By highlighting the non-trivial role of patterns, linking motifs, and the physical layering of images in constituting such relationships, Gell encourages us to look beyond the more explicit visual articulations of political and sacred power, and to consider the full variety of material strategies through which states have pursued the enchantment of authority.

Acknowledgments I am very grateful to the editors of this volume and to the other contributors for a stimulating day’s discussion in Cambridge. My thanks also to

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 40 12.1.2007 2:19pm Compositor Name: SJoearun

40 David Wengrow Stephanie Moser for supplying a valuable quotation, and to John Baines for comments on a much earlier attempt at grappling with Gell and Egyptian art, which helped to bring this later attempt into clearer focus.

Notes 1 2

3 4 5

6 7

8 9

10

11

C. Eire, Waiting for Snow in Havana: Confessions of a Cuban Boyhood (London, 2003), p. 97. E. Durkheim, Emile Durkheim: Selected Writings, ed. A. Giddens (Cambridge, 1972), p. 64 (trans. Giddens, from Les re`gles de la me´thode sociologique [1895]). Ibid., 59. A. Gell, Art and Agency: an Anthropological Theory (Oxford, 1998), p. 253. A. Gell, ‘‘The technology of enchantment and the enchantment of technology,’’ in J. Coote and A. Shelton, eds., Anthropology, Art and Aesthetics (Oxford, 1992), p. 46. Gell, Art and Agency, p. 69. S. Ku¨chler, ‘‘Why knot? Towards a theory of art and mathematics,’’ in C. Pinney and N. Thomas, eds., Beyond Aesthetics: Art and the Technologies of Enchantment (Oxford, 2001), p. 64. Chambers Edinburgh Journal 8 (1839), p. 91. H. Scha¨fer, Principles of Egyptian Art [1919], English edition, trans. J. Baines, foreword by E. H. Gombrich (Oxford, 1974); J. Baines, ‘‘Theories and universals of representation: Heinrich Scha¨fer and Egyptian art,’’ Art History 8 (1985), pp. 1–25. The primary publications upon which I draw are: W. B. Emery, Excavations at Saqqara: The Tomb of Hemaka (Cairo, 1938); Excavations at Saqqara 1937–1938: Hor-Aha (Cairo, 1939); Great Tombs of the First Dynasty, vol. 1 (Cairo, 1949); Great Tombs of the First Dynasty, vol. 2 (London, 1954); Great Tombs of the First Dynasty, vol. 3 (London, 1958). For accounts of the architectural evolution of mastaba tombs, heavily oriented toward the development of later pyramid complexes, see also E. Lehner, Wege der architektonischen Evolution: die Polygenese von Pyramiden und Stufenbauten: Aspekte zu einer vergleichenden Architekturgeschichte (Vienna, 1998); R. Stadelmann, Die a¨gyptischen Pyramiden: vom Ziegelbau zum Weltwunder (Mainz am Rhein, 1997). For the much-debated issue of tomb-ownership, and associated problems, see B. J. Kemp, ‘‘The Egyptian 1st dynasty royal cemetery,’’ Antiquity 41 (1967), pp. 22–32, and, with further references, W. Stevenson Smith, The Art and Architecture of Ancient Egypt, 3rd ed., revised with additions by

Osborne / Art’s Agency and Art History 1405135379_4_001 Final Proof

page 41 12.1.2007 2:19pm Compositor Name: SJoearun

Enchantment and Sacrifice in Early Egypt 41

12

13 14

15 16

17 18

19 20 21 22

23

W. K. Simpson (New Haven, 1998), pp. 17–18, 255 n. 14; D. Wengrow, The Archaeology of Early Egypt: Social Transformations in North-East Africa, 10,000–2,650 BC (Cambridge, 2006). In connection with the present article, it is highly recommended that the reader consults the three-dimensional color reconstructions of First Dynasty mastabas at www.digitalegypt.ucl.ac.uk/tarkhan/tarkhangreattombs/index.html. See, e.g., A. J. Spencer, Death in Ancient Egypt (Harmondsworth, 1982), p. 166. W. M. F. Petrie, The Royal Tombs of the Earliest Dynasties, Part II, Egypt Exploration Fund Memoir 21 (London, 1901), pl. 1; idem, Petrie, Gizeh and Rifeh (London, 1907), pl. 3. Emery, Great Tombs, vol. 3, pls. 6–8, 16. And see www.digitalegypt.ucl.ac.uk/ tarkhan/tarkhangreattombs/colour.html. G. Dreyer, ‘‘Umm el-Qaab. Nachuntersuchungen im fru¨hzeitlichen Ko¨nigsfriedhof. 3/4 Vorbericht,’’ in Mitteilungen des deutschen archa¨ologischen Instituts, Abteilung Kairo, 46 (1990), pp. 67, 81–6; W. Kaiser, ‘‘Ein Kultbezirk des Ko¨nigs Den in Sakkara,’’ Mitteilungen des deutschen archa¨ologischen Instituts, Abteilung Kairo 39 (1985), pp. 61–87. See also J. Baines, ‘‘Origins of Egyptian kingship,’’ in D. O’Connor and D. Silverman, eds., Ancient Egyptian Kingship (Leiden: Brill, 1995), pp. 136–7. Emery, Great Tombs, vol. 2, 146–7. J. de Morgan, Ethnographie pre´historique et Tombeau royal de Ne´gadah. Recherches sur les origines de l’Egypte, 2 (Paris, 1897); J. Capart, Primitive Art in Egypt (London, 1905). Gell, Art and Agency, pp. 251–8. N. Thomas, Oceanic Art (London, 1995), pp. 63–4. Ibid., p. 64. V. E. Izzet, ‘‘Putting the house in order: the development of Etruscan domestic architecture,’’ in J. Rasmus Brandt and L. Karlsson, eds., From Huts to Houses: Transformations of Ancient Societies (Stockholm, 2001), pp. 41–9. It is worth noting, in this context, that the design and decoration of early mastaba tombs has long been thought by some scholars to derive from Mesopotamian architectural prototypes, also constructed in mud-brick, which were temples (‘‘Houses of the Gods’’) rather than tombs, and therefore accessible to the living. The best exposition of this view remains H. Frankfort, ‘‘The origin of monumental architecture in Egypt,’’ American Journal of Semitic Languages and Literatures 58 (1941), pp. 329–58; and see, more recently, P. R. S. Moorey, ‘‘On tracking cultural transfers in prehistory: the case of Egypt and Lower Mesopotamia in the fourth millennium B.C.,’’ in M. Rowlands, M. Larsen, and K. Kristiansen, eds., Centre and Periphery in the Ancient World (Cambridge, 1987), pp. 36–46.

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 42 12.1.2007 6:04pm Compositor Name: SJoearun

2

Agency Marked, Agency Ascribed: The Affective Object in Ancient Mesopotamia Irene J. Winter

Ancient Mesopotamia offers a useful test case for Alfred Gell’s ascription of agency to works we call ‘‘art’’: that is, that artworks may be considered the equivalent of persons, capable of acting on and for their social universes, both psychologically and physically.1 At the most literal level, the textual and artifactual record permits identification of several classes of material objects explicitly intended to exert force on the world (amulets, apotropaic gateway figures). At a deeper level, additional classes of material objects were conceived as animate, hence as having the same agency as living entities. This was achieved largely through ritual consecration, in a system of belief that allowed for the transfer of personhood and/or divinity into physical matter. Such ritual enlivenment is analogous to that found in the practices of Buddhist and Hindu Asia, for example, where matter, once consecrated, becomes the locus of divine manifestation, as distinct from the merely representational that is theologically rationalized (with considerable anxiety) in Christianity and explicitly denied in Judaism and Islam. It will be my argument throughout the present paper that one must distinguish between agency ascribed by the analyst of a given work from agency marked by cultural practice, and even grammar, within the

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 43 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

43

originating culture, if we are to fully understand the historical role(s) accorded the artwork. I stress this point in order to avoid conflating indigenous and analytical perspectives until and unless they may be shown to be congruent. I do so also to reduce ambiguity in the semantic range of the very term ‘‘agency.’’ The agentive role is grammatically marked in early Sumerian, particularly of the third millennium BC, and can be implied in later Sumerian along with the Semitic dialects of Akkadian, for which relevant texts are preserved from the third millennium to the end of the first millennium BC. Ascription of agency in these two linguistic and cultural traditions of ancient Mesopotamia extends beyond the animate person to anthropomorphic images, buildings and artifacts; indeed, in Sumerian, the agentive marker can shift in the course of narrative – as, for example, when Gudea, ruler of the early city-state of Lagash (c.2110 BC), recounts his (re)building of the Eninnu, a temple for the god Ningirsu. In the first part of the narrative, preserved on the clay artifact known as ‘‘Cylinder A,’’ Gudea himself is indicated as the agent, instrumental in carrying out the construction and outfitting of the temple. However, in the succeeding ‘‘Cylinder B,’’ once the building is complete and has been ritually consecrated, with the deity installed in residence, the temple itself receives the agentive marker, a morphological sign that the temple has taken on an active role in emitting its affective properties to the community. In other instances, images are expected to take on the animate properties of their referent – attested for statues of this same Gudea, for example – following specific rituals that effectively open the sensorium of the image. These rituals result in the image’s putative ability to respond actively to stimuli such as smell, taste, hearing, and sight, along with speech and thereby the ability to engage in direct discourse with the gods. The Mesopotamian case therefore supports Gell’s construct, by offering concrete examples of the agency of the material object or work of art. At the same time, it also gives rise to a critique of Gell: namely, whether he distinguished sufficiently between agency literally ascribed within a given folk tradition or grammatically marked as a coded property of the work, and agency figuratively inferred or metaphorically projected for certain artworks under certain circumstances or in general, largely by analysts. The importance of this distinction is pursued in the present chapter, in terms of both the locus of agency and the complementary concept of the ‘‘affective’’ as an alternative for the second, figurative condition.

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 44 12.1.2007 6:04pm Compositor Name: SJoearun

44 Irene J. Winter

The Locus of Agency ‘‘Agency’’ (i.e., the affective or instrumental force exerted by a source of energy or action upon a recipient) can be exercised not only by individuals, but also by social institutions and material objects. An obvious category of objects considered ‘‘affective’’ would include talismans, amulets, and apotropaic figures or emblems, all of which are thought to exert power and may be attested in any number of traditions.2 Such works are not only accorded the power to affect their respective universes, they also participate in an underlying belief system permitting such objects to exercise force. However, these clearly designated ‘‘affective objects’’ are only a minor subset of works exercising agency in Alfred Gell’s construct of the role of art in society. In the preface to Gell’s Art and Agency, Nicholas Thomas makes clear that, for Gell, any form of ‘‘doing’’ is ‘‘theorized as agency.’’3 It is the next step, however, that has significant consequences for the analysis of the artwork. Thomas glosses Gell’s view that ‘‘the anthropology of art is constructed as a theory of agency or of the mediation of agency by indexes,’’ with these indexes then ‘‘understood . . . as material entities which motivate inferences, responses or interpretations.’’4 Unfortunately, the ‘‘or’’ in the previous sentence tends to get lost, particularly by those who would apply Gell’s study to specific cases, with the artwork itself accorded autonomous agency, even on those occasions when in effect it serves largely as the index for an agent lying behind it. In my opinion, the concept of agency is most useful when a distinction is made between animate or human agency exerted as physical action and the impact derived from the affective properties of a work, which must then be inferred (Gell’s abduction of agency) by the entity upon whom or which the affect is imprinted. With such a distinction, in which ‘‘affect’’ is a necessary if not sufficient consideration within an anthropological theory of art (see below), the artwork under discussion most often requires the adjectival formulation ‘‘distributed agency,’’ even when, as Gell notes, the work has been subject to ritual transformation according it an agency of its own.5 It is this ritual transformation of the material that empowers the image ‘‘to speak, or to see, or to act, through various culturally-subscribed channels,’’6 after which it may be said to exercise its own agency. In Mesopotamia, the principal sensory organ addressed in the enlivening

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 45 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

45

process is related to speech, while in Hindu and Buddhist practice it is related to vision. In all cases, lengthy and complex ritual processes are engaged to ‘‘open the mouth’’ or ‘‘open the eyes,’’ so that the image then, within the cultural/belief system, can act upon or interact with its audience.7 Objects as well can be ritually charged with such power to channel divine forces, hence be endowed with agency.8 These are special cases of agency, however, dependent upon ritual transformation of substance. When one shifts to more normative, languagebased indicators of agency, purposeful action and causal consequence must be able to be indicated for individuals, supra-human forces, spirits, ancestors, or the divine in virtually all languages and cultures. In most IndoEuropean and Semitic languages, agency can be indicated directly in transitive sentences such as ‘‘the king [active subject] built [transitive verb] the temple [passive object],’’ or ‘‘the king killed the enemy.’’ An intransitive sentence permits one to shift subject–object relations, as in the statements ‘‘the temple [now the subject] was built by the king’’ and ‘‘the enemy was killed by the king,’’ while retaining understanding that it was the king who accomplished the act.9 In the early Sumerian texts we will look at, by contrast, there is no intransitive construction. In the language of linguistics, subject::object construal gives way to that of agent::patient – the very vocabulary adopted by Gell. The agent (i.e., the king) remains constant, as the originator of the action, and so does the recipient or product of his action, the patient (i.e., the temple, the enemy). The building paradigm is the more interesting for our purposes, because the patient in this case is not another person, but a product that can be construed as an artwork: the temple. The role of the ruler as builder has a long history in Mesopotamian texts,10 and the visual trope of an active ruler engaged in the building process has an equally long history, from the Sumerian Early Dynastic period of the mid-third millennium BC to the neo-Assyrian period ending in the mid-first millennium BC. On the wellknown plaque of Ur-Nansˇe, founder of the Early Dynastic dynasty at the city-state of Lagash, for example, the ruler is depicted with a brick- or mud-bearing basket on his head – that is, as if having actively participated in construction – while in the accompanying text, beginning just to the right of the ruler’s head (figure 2.1), he is equally clearly identified as agent, the builder of the temple: ˇIR . BUR . LA. . . . e2-dnin.gir.su mu-du3 ˇe.ur lugal S nans Ur-Nansˇe, king of Lagash . . . built the temple of (the god) Ningirsu11

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 46 12.1.2007 6:04pm Compositor Name: SJoearun

46 Irene J. Winter

Figure 2.1 Plaque of Ur-Nansˇe of Lagash, found at Tello, c.2600 BC; limestone, ht. 40 cm. Courtesy of the De´partement des antiquite´s orientales, Muse´e du Louvre.

The tense/aspect of the Sumerian verb tells us that this is a completed, not an ongoing, act. Depicting the ruler with a basket on his head is in effect a trope for the class of act commemorated on the plaque. Yet, to match the verbal epithet, Ur-Nansˇe is not ‘‘carrying the basket’’ in a perpetual imperfective, iconic of a ‘‘type.’’ Rather, it is to be inferred that he ‘‘carried’’ the basket (perfective), a way to visually declare his having built the temple.12 A later ruler, Ur-Namma of Ur (c.2110 BC), on the obverse of his stele found within the temple precinct of the moon god Nanna/Su’en, participates in the same class of activity. The king is here depicted walking behind a deity, with that same basket and building-tools slung over his back, to indicate his direct participation in the construction of the ziggurat complex of the moon god, initiated during his reign (figure 2.2) The culmination of the process is then depicted in the registers above, through the ritual performance being enacted in the completed sanctuary.

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 47 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

47

Figure 2.2 Detail, registers 2–3 of the Stele of Ur-Namma of Ur, found at Ur, c.2110 BC; limestone, ht. as restored c.2.85m. As originally restored (disassembled 1988). Courtesy of the University of Pennsylvania Museum.

Ur-Namma’s Neo-Sumerian contemporary, Gudea of Lagash, continued the building practices of his Early Dynastic predecessor, Ur-Nansˇe, by further (re)building of a temple to Ningirsu in the satellite city of Girsu. Images and texts of Gudea commemorating his undertaking have been well studied.13 In particular, the seated statue known as Gudea B preserves for us the large tablet on the ruler’s lap, on which the plan of the temple and precinct has been engraved, with a stylus aligned along the left-hand edge (figure 2.3). An account of the king, the statue, and the building project is also inscribed around the physical mass of the statue, a shortened version of what appears in Gudea’s lengthy Cylinders A and B. At the obvious level of action, Gudea is clearly the agent through whom the project is realized, just as Ur-Nansˇe and Ur-Namma had been. However, three additional aspects of this account are useful in the present context. First, in the building account, we are told that in anticipation of the project Gudea spends a night in a (prior) temple of Ningirsu, and receives a dream or omen directly from the deity authorizing him to undertake the new building project. One could argue, therefore, that there is an agency behind that of Gudea: the god without whose blessing, and indeed vision,

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 48 12.1.2007 6:04pm Compositor Name: SJoearun

Figure 2.3 Statue ‘‘B’’ of Gudea of Lagash, found at Tello, c. 2115 BC; diorite, ht. 93 cm. Courtesy of the De´partement des antiquite´s orientales, Muse´e du Louvre.

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 49 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

49

the enterprise cannot be carried forward. The temple is then physically executed by the god’s earthly agent, the ruler. Thus Gudea, given grammatical agency in the narrative, must be construed as an example of ‘‘distributed agency’’ in Gell’s terms.14 Second, the role of the image of Gudea is elaborately described in the text written on the statue, which through ritual consecration would have been empowered to have its own agency when installed in the temple – yet another sort of distributed agency: that of the referent through his empowered image.15 And third, the temple itself, once completed, takes on an agency of its own. This agency is explicitly marked or claimed in text, and projected as if able to be inferred or perceived (be ‘‘abducted’’) by its public. The first aspect of distributed agency, from the god to Gudea, is paralleled in a number of cases, even depicted in Sumerian art. One such instance is seen on official seals and sealings of the Ur III period (c.2110–2000 BC), which contain lengthy formulaic inscriptions naming the seal-holder as the servant of the king, along with imagery of that individual standing before or being presented to the ruler (figure 2.4). The king, like the god for Gudea, is primary agent to the extent that, both

Figure 2.4 Clay Tablet with seal of Ur-Lisi, ensi of Umma, citing ruler Amar-Su’en of Ur, Ur III period, c.2045 BC. Courtesy of the Pierpont Morgan Library.

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 50 12.1.2007 6:04pm Compositor Name: SJoearun

50 Irene J. Winter visually and verbally, the legitimacy and power of the official flows from his empowerment by the state or the ruler. In the imagery, the ruler occupies the dominant, seated position, while in the text, the king and his titles are mentioned before the official is named.16 However, through the transfer of agency (and sometimes even through the explicit statement that the seal has been gifted by the ruler), the official functions as the distributed person of the ruler in the exercise of his office. The seal imagery, therefore, records this doubling of agency, the visual juxtaposition of the figures paralleled by the verbal juxtaposition of king and official in the inscription. At the same time, the imagery allows the seal to signal its mediation of that agency, through cascading indexical relationships: from ruler to official, from official to his seal, from seal to its impression.17 Yet another instance may be seen on the Stele of Eannatum of Lagash of the Early Dynastic Period, often referred to as the ‘‘Stele of the Vultures.’’ On the obverse of the stele (figure 2.5), the god Ningirsu smites the head of an enemy of Lagash with his mace, at the same time as he has gathered the enemy in his battle net. Both in the lengthy text engraved on the stele and visually by virtue of placement on the obverse, the god is accorded prime agency in the achievement of victory. At the same time, on the reverse (figure 2.6), the ruler of Lagash, Eannatum, is seen leading his troops into battle and ultimately victorious. It is his physical agency that has effected the outcome of the battle, even as he declares himself the recipient of the distributed agency of the deity through selection and empowerment.18 The artwork provides the instantiation of this flow of agency through its visual message. Indirectly, one can also argue that the stele itself performs politically, exerting its agency upon the viewing public which responds to and is shaped by the rhetoric of the effective ruler as the god’s agent. The stele would then represent Gell’s ‘‘index’’ of royal agency in its role as affective monument making visual the verbal account of the ruler’s ability to act.19 These examples help elucidate the hierarchical chain from principal agent to distributed agent/index, using illustrations from the Sumerian corpus. The flow of agency from deity to ruler, or from ruler to official, can be demonstrated, beginning with Gudea’s account of his receipt of empowerment from the deity and the actions that follow, as recounted above. It is the second and third aspects of Gudea’s accounts, however – those concerning his seated statue and the temple he claims to have constructed – which fall more directly into Gell’s literal domain of the

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 51 12.1.2007 6:04pm Compositor Name: SJoearun

Figure 2.5 Detail, obverse of the Stele of Eannatum of Lagash, found at Tello, c.2560 BC; limestone, full ht. as restored 1.88 m. Courtesy of the De´partement des antiquite´s orientales, Muse´e du Louvre.

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 52 12.1.2007 6:04pm Compositor Name: SJoearun

Figure 2.6 Reverse of the Stele of Eannatum of Lagash. Courtesy of the De´partement des antiquite´s orientales, Muse´e du Louvre.

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 53 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

53

agency of the artwork exerting force on its social environment. An examination of each is useful in the present context.

The Statue The text inscribed on the seated Gudea ‘‘B’’ (figure 2.3) tells the reader explicitly that the image (Sum. alam) was intended for the ki-a-nag˘, a commemorative, funerary chapel (literally a ‘‘place for the pouring of water,’’ or libation), dedicated to a ruler and, in this case, located within the Eninnu temple. Food offerings of grain, beer, and bread to be regularly presented to the consecrated, hence animate, image, are enumerated across the upper back of the image.20 Other known images of Gudea were intended for other shrines, according to the inscriptions indicating the various deities to whom each was dedicated.21 But what is also learned from the inscription on the seated Gudea is that once the image is brought into the temple, the living Gudea, ruler of Lagash, commands the now consecrated, installed, and presumably activated statue: ‘‘Image, to my lord [the god Ningirsu], speak!’’22 By the end of the section it is clear that at least one of the reasons the statue has been installed is precisely in order to recite Gudea’s own pious deeds and words to the deity. I have argued elsewhere that one should distinguish seated statues, which presumably occupied shrines of their own, from standing statues placed before divine images in the gods’ own shrines.23 As the dyadic partner in this latter interaction, the cult image of the deity would have been similarly enlivened, able to both receive and send communications or sensory stimuli. At least, such agency is attested in text, although we do not have preserved the cult images themselves, inscribed with explicit attestations to such empowerment. Nevertheless, a recent study of omen texts from Nineveh by Walther Sallaberger makes clear that the tradition of according agency to the image continued in southern Mesoptamia well into the first millennium BC. Sallaberger demonstrates how at the New Year’s festival, the cult image of the Babylonian deity Marduk is expected to play an active role, mediating through communication acts between heavenly intention and the earthly community.24 That statues are accorded agency specifically through their ability to speak is not limited to polytheistic traditions. A primary aspect of the Christian cult of images, particularly within Roman Catholicism, is the miraculous intervention of images, especially of the Virgin Mary. Even in iconoclastic phases, words can be put into the mouth of the cult image, if

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 54 12.1.2007 6:04pm Compositor Name: SJoearun

54 Irene J. Winter only to disavow devotion. Such a case has been studied by Shirin Fozi, who has discovered a sixteenth-century treatise against devotion directed to the image of the Virgin of Le Puy in the Auvergne, in which the image, deemed an idol (idole) in the text, is said to address the idolaters (idolatres) in direct speech, counseling them not to adore her through her image: . . . n’est pas moi que tu dois reverer Je ne suis rien . . . sinon un . . . malhereux idole . . . Dieu seul . . . te peut secourir au besoin . . . it is not I whom you should revere I am nothing . . . but a . . . miserable idol God alone . . . can take care of your needs.25

It is ironic that the agency accorded the image in the text should thus be put to disclaiming agency for the image! But since the text would only have been written if the populace did accord some degree of agency to the image, it is a useful addition to Gell’s discussion of images, where he articulates the line between the non-idolatrous use of images as aids to piety and the idolatrous empowerment of them as ‘‘physical channels to access divinity.’’26 The Le Puy text provides a hint of the degree to which cults of miraculous images can expose the tensions within Christianity between the power accorded such images and the theological anxieties about practice deriving from the Old Testament commandment to make no graven images and prophetic pronouncements on the subject. Indeed, a number of pioneering studies in this vein have been undertaken over the past 10 to 15 years, largely by art historians.27 At the same time, the negative case from the Auvergne, as with the ‘‘Lamentation over Sumer and Ur’’ discussed in the following section, provides us with an additional perspective on those instances when an image is directly accorded power. For, with the true Virgin behind her cult image, as with the living Gudea instructing his consecrated image, what we are seeing is indeed a case of the ‘‘distributed person’’;28 the agency of the image is only able to be autonomous once the agency of the referent behind has been (ritually or by belief) transferred into the image in a chain linking the originating source or person to the extended or distributed material person. What this suggests, therefore, is that one should not lose sight of the referent, all too easy when the same word, ‘‘agency,’’ is applied to both

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 55 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

55

Gudea and his statues, or to the Mesopotamian deity and his cult image. It argues for avoiding the shorthand of referring simply to the agency of the image in such cases, and for maintaining the modifier ‘‘distributed,’’ or at least a clear articulation of the indexical relationship graphed by Gell.29 Only with such distinctions can the cline from empowering source to empowered manifestation be made transparent. This hierarchy from the inherently agentive to the delegated agent is important when theorizing the agency of ‘‘art’’ – whether established through performative sacralization or merely by social accord. If left undistinguished, Gell’s expressed goal of determining the nexus of social relations occasioned through the performance of the work in a specific socio-cultural, historical context is itself undermined.

The Temple At the end of the text known as Cylinder A, in which Gudea recounts the planning and construction of the Eninnu for Ningirsu, the completed temple is described with all of its affective properties. We are told that the great awe-inspiring radiance (ni2-gal) emanating from the temple was cast over the land ‘‘like the sun(god)’’ (dutu-gim), that the temple itself was exuding sumptuousness or allure (hi-li gur3-a) and ‘‘stood to be marveled at’’ (u6-di-de3 gub-ba).30 That one should wish to translate the final statement in such a way as to portray the temple as marvelous or wondrous is clear from the context. The actual phrase, however, includes the Sumerian ideographic sign u6, which, while it can be rendered as marvelous, actually is comprised of component signs for both eye, igi, and house/temple, e2, in such a way as to suggest the intensified or augmented visual experience that has induced the sense of the marvelous. A more literal translation, then, would preserve the underlying visual experience, as in the original derivation of ‘‘admire’’ from the Latin augmentative prefix ad- joined to the verb mirari, ‘‘to see, look at.’’ Hence, the temple literally ‘‘stood to be adþmired.’’ This semantic loading is clearly implied in another, negative, case: the text of the ‘‘Lamentation over Sumer and Ur.’’ In that account, the temple of the moon god Nanna at Ur, described as once fragrant with cedar and decorated with gold, silver, and lapis lazuli, has been destroyed. In the closing summary statement, the author laments: . . . u6-di-bi ba-gul (The temple) . . . its admirableness was extinguished.31

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 56 12.1.2007 6:04pm Compositor Name: SJoearun

56 Irene J. Winter We see here the same literary pattern as in the Gudea cylinder’s summary of the Ningirsu temple: after the attributes contributing to the building’s original affect have been described, the culminating statement relates to the visual response the temple (once) evoked. Previously, the overall affect had been strong and positive; its loss is to be lamented. Through the negative case, then, the opposite is articulated as well. The earlier temple had been charged with positive affect, which then elicited adþmiration; that is, through the act of looking, a heightened visual response had been induced. In later Akkadian texts of the neo-Assyrian period in the early first millennium BC, there is an exact lexical equivalent, sometimes even written with Sumerian logograms, to be read: ana tabra¯t (kisˇˇsat) nisˇeˆ. Here again, an intensified form of the now Akkadian verb ‘‘to see’’ (baruˆ) is used, with the whole phrase to be translated: ‘‘for the intensified viewing experience [hence adþmiration] of [all] the people.’’32 In the rhetorical structure of official texts, these phrases, in both Sumerian and Akkadian, are generally used to conclude descriptions of major undertakings, calling attention to the work or project in question by claiming for it a powerful response. What is interesting for Gell’s formulation of agency exerted by the work is the active role the enterprise or product itself can be shown to take in disseminating its properties. In a Sumerian temple hymn of the Ur III/Old Babylonian Period (late third to early second millennium BC), we are told that a temple’s interior is ‘‘a beaming light issuing forth.’’33 Use of the verb ‘‘to go out, issue forth’’ suggests the outward movement (read extromission) of the properties of the sanctuary from their source into the perceiving world. It lays the foundations for our understanding of the active way in which the finished work, once fully endowed and ritually consecrated, was thought to emit its qualities affectively. This brings us to a discussion of early Sumerian as a language demonstrating ergativity: that is, one that distinguishes agent and patient, rather than subject and object.34 It will be recalled that Gudea’s Cylinder A describes the antecedents to and construction of the temple, by the end of which the building has been physically completed. In the continuing account preserved in Cylinder B, the temple has been fully appointed, the deity formally (and ritually) installed. Indeed, once this is accomplished, it is no longer Gudea as builder but the temple that is described in an active mode:

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 57 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

57

ˇu mi-ni[e2]-e me-gal-la [s]ag mi-ni-ib2-il2 ni2-me-lim5-ma s ib2-du7

The temple [þagentive marker -e] . . . lifted its head, unparalleled in radiant splendour35

Significant is the post-position -e attached to the noun e2, ‘‘temple/ house,’’ which marks the fact that the temple itself takes on an agency directly comparable to that indicated in the Sumerian temple hymn cited above, in which the temple actively emits its light-bearing properties. The agency performed by the Ningirsu temple fits well into Gell’s construct of an active artwork distinguished from an essentially passive spectator, and in fact helps us to inflect the Sumerian viewer who, as a result of his or her intensified viewing experience, is ultimately the recipient of the temple’s affective power. Gell states that the individual who permits him- or herself to be ‘‘attracted to an index, and submits to its power . . . is a patient, responding to the agency inherent in the index.’’36 But from where does he derive his analytical terms? Ironically, he rejects a semiotic approach to ‘‘decoding’’ the artwork according to linguistic models,37 while simultaneously appropriating the vocabulary, along with the ‘‘polarity of agent/ patient relations,’’ directly from linguistics.38 Agent::patient relations (ergative-absolutive constructions) are seen as alternatives to subject::object relations (nominative-accusative). The literature is extensive, reflecting work done in linguistics especially from the late 1970s and 1980s, but continuing to the present. Analyses range from expositions of the nature of ergative systems39 to case-studies of particular languages,40 including a substantial body of work on early Sumerian.41 Ergative-absolutive and nominative-accusative constructions can occur within the same language, with a variety of strategies determining which construct will be deployed in a given usage. Interesting for our purposes, as discussed by F. R. Palmer, is when to separate agent::patient roles, which can be communicated in most grammatical structures independent of language group – often by word order and/or verbal construction, from specifically ergative systems in which agency is marked, most often in the nominal case.42 Clearly, Gell uses the terms agent and patient to indicate roles, independent of any specific language group and communicable no matter what the linguistic structure. Nevertheless, it seems important to uncover the linguistic model, precisely because it does make a difference whether the role is to be

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 58 12.1.2007 6:04pm Compositor Name: SJoearun

58 Irene J. Winter inferred or is actually marked. In the former case, as employed by Gell, the agency is abducted43 through a series of signs; in the latter, it is marked grammatically, and so directly signaled to the literate, rather than inferred. In the specific case of Sumerian, ergativity ‘‘manifests itself not only in nominal case marking, but also in pronominal systems and verbal agreement, as well as in syntax.’’44 Early Sumerian, corresponding chronologically to the third millennium BC works of Eannatum and Gudea, actually manifests a ‘‘split ergative’’ system: ergative in the perfective or completed aspect of the verb, nominative-accusative in the imperfective or ongoing.45 Agentivity and perfectivity are thus conjoined, and argued to be related to ‘‘the notion of pragmatic foregrounding in the narrative.’’46 This fits well with the observation of sumerologist Thorkild Jacobsen, who noted that the ergative is used especially when the agent is to be stressed semantically.47 The point is that one sees the use of the ergative marker for the third person nominal plus a verbal form indicating the perfective or accomplished act precisely in rhetorical language such as is used on public monuments and in official accounts of the third millennium BC, and I would argue that this is so because it is the entity manifesting agency in such cases (the king, the work) that is meant to be foregrounded, whether in text or image.48 One could object, of course, that an association of ergativity in the grammatical structure with agency in narrative confuses morphology with semantics, especially as every language is likely to have ways of identifying agency or causation (as in who built the temple). However, recent studies have demonstrated that the early acquisition of the ergative-absolutive distinction is ‘‘learned through the semantic feature agentivity’’49 – that is, that the two are cognitively intertwined. For our purposes, in the Sumerian case, I would argue that morphology and semantics are not discrete categories precisely in those instances where grammatical marking provides stress that is congruent with narrative intent. At such moments, grammar can signal meaning (e.g., stress) as well as structure. This claim is in fact supported by work in linguistics that discusses the semantics of case markings. In particular, the work of Suzanne Kemmerer and Arie Verhagen insists that ‘‘the knowledge underlying grammar is not qualitatively different from other aspects of human understanding’’; they suggest that viewing causative constructions as cognitive schemata helps to explain why languages with case markings ‘‘to express the causee are rare, if not non-existent, as distinct from those marking the causer’’ (i.e., the agent).50

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 59 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

59

In Sumerian, too, I suggest it is not insignificant that the agent carries the post-position marker -e, while the patient is unmarked. The agentive -e serves not just as a morphological, but also as a semiotic, marker of stress. The strategy of choosing ergativity to mark the ruler as agent in the performance of certain acts, accompanied by the perfective aspect of the verb to indicate that the act is completed, directly complements the imagery that also foregrounds the acts of the ruler and means that the viewing and reading audiences are brought to an identical conclusion, one consistent with the goals of the social constructor of the work. In short, especially in a split-ergative system such as early Sumerian, ergativity becomes part of the semantic signaling system of the language/text/culture/society. Recognition of this distinction is again important for Gell’s own goal of probing social relationships through the agency of the artwork, for what I am arguing is that social relations are no less encoded than the cultural, linguistic, iconographic ‘‘meaning’’ Gell claims to eschew. That a particular grammatical voice can be chosen – in Samoan, for example, or in Sacapultec Maya, as in Sumerian – emphasizes the fact that ergativity is discourse-based, carrying semantic and/or pragmatic import and marked agency therefore reflects a socially determined as well as socially deployed empowerment.51 Furthermore, to the extent that ‘‘agentive indicates agency and patientive [indicates] affectedness,’’52 the properties or attributes causing affectedness, which Gell dismisses as mere aesthetics,53 in fact cannot be so dismissed if one is to be successful in establishing those social relationships operative within his ‘‘art nexus.’’ Rather, I would say that only if the values attached to affective properties of works or objects are understood can one successfully reconstruct the social system in which the works are actively engaged.54

The Affective Object That a given artwork exists within a social ‘‘art nexus’’ needs little argument. By Gell’s definition, ‘‘the theoretical study of ‘social relations in the vicinity of objects mediating social agency’’’ constitutes the mission of the anthropologist of art.55 One is forced to ask, however, whether that is the only job of the anthropologist of art. Whose job then is to show what the affective properties of the work are, how those properties may operate and change in relation to society, or how those properties may themselves act to effect change in society, hence in social relations?

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 60 12.1.2007 6:04pm Compositor Name: SJoearun

60 Irene J. Winter Gell dismisses attempts by anthropologists and others to pursue indigenous aesthetics as misguided efforts that merely ‘‘equate the reactions of the ethnographic Other . . . to our own.’’56 Robert Layton, in his review article unpacking Art and Agency,57 suggests that Gell has flattened the discourse on art within anthropology to date (see also Tanner and Osborne, ‘‘Introduction,’’ this volume). This is no less the case for art history, and even philosophy and psychology, where both agency and the affective properties of work have been under discussion for well over a decade.58 Within anthropology, Howard Morphy, among others, has made significant contributions to the discussion of affective properties such as light in an indigenous context; and he has directly engaged the aesthetic as part of anthropology’s mission, precisely by seeking to uncover socio-cultural conventions and values within a tradition, rather than by imposing our values upon it.59 That the art historian’s role should, conversely, be limited to aesthetics and formal analysis is equally reductive. The social history of art has been well theorized at least since the 1950s.60 Underlying Gell’s reduction of the art historian’s enterprise is an old-fashioned definition of aesthetics that demands the work be alienated from its originating context, and an equally outdated notion of formal analysis tied to attribution studies. ‘‘Style,’’ for example, that important concept and tool for formal analysis to which Gell devotes an important chapter, can be shown to have been consciously deployed as an affective property in its own right, intertwined with iconography toward meaning.61 In addition, art historians have themselves been grappling with the issue of agency for some time: from patronage, to the possibilities for subjective expression on the part of producing artists, to the power of the work as socially affective, not merely reflective.62 For anthropology no less than art history, a larger set of dimensions within which agency is imputed to artworks seems called for: dimensions that consider not only ‘‘social relations in the vicinity of objects,’’ but also underlying authority structures, systems of belief, notions of history, and systems of value – symbolic, material, and aesthetic. If, with Robert Wicks, part of the response to the artwork, hence its potency, is ‘‘how . . . intention has been realized,’’63 then the work cannot be fully understood in its social dimension unless one can decode what is affective, what is causing the social force,64 along with the hierarchy of agentive power discussed above. This criticism notwithstanding, Gell’s volume has opened the debate to a more public forum on the nature of the artwork and its role in society, as

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 61 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

61

understood by anthropologists and art historians alike. Conceivably, had Gell lived to reflect further on the importance of his insights, he might have resisted the temptation to represent such reductive and essentializing views of either anthropology or art history in order to make his points – particularly as so many of the issues he raises are challenging and had already begun to generate vigorous inquiry in their respective disciplines. The Sumerian case has been useful as an illustration of the purposeful marking of agency, both in text and in image. This has implications for how one may understand the power of the artwork within that society, the work itself serving as a vehicle with agentive force. Other languages and other visual repertoires will have devised equally effective signaling systems permitting their speakers or social members and subsequent analysts to infer agency, causation, and transformative power. What I am suggesting is that how a tradition and its language(s) mark or ascribe agency is not a trivial component in any analysis of the weight to be given to the social power of the agent, including the agency accorded the artwork. The Sumerian case also asks us to distinguish text from image in the articulation of agency, so that represented power or authority may be interrogated as either consistent or contested before the work, the period, or the nature of agency itself is assessed. Particularly in the case of inscribed artworks, things may be said to happen because of the semiotic relationship between word and image. Furthermore, if the potentially ambiguous image (as the basket-bearing ruler on the ‘‘Ur-Nansˇe Plaque’’) is in fact anchored by the accompanying text, then it suggests that one has to have language to conceive or perceive the image: that, in short, the image is understood by its originating society in language, so that when the art historian is talking about a social history of art or attempting to reconstruct the affective power of the work, correspondences in language, grammatical structure, and idiom need to be part of the analysis. In the end, it is this art historian’s view that what one cannot do is simply take the unmodified ‘‘agency’’ of the artwork as given. However, once modified – as with ‘‘distributive’’ or ‘‘abducted’’ agency (which demands awareness and analysis of the properties of the work that generate the affect and so permit appropriate inferences) – then Gell’s concepts as articulated in Art and Agency have much to offer. But I would also insist that the art historian and the anthropologist working on aesthetics – indigenous, historically specific systems of value – have much to contribute to the discussion as well. Once the relational nexus surrounding and generating the artwork is seen to function as part of a holistic system,

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 62 12.1.2007 6:04pm Compositor Name: SJoearun

62 Irene J. Winter affective properties and meaning intersect with social agency. That is, instead of having to choose ‘‘art as a system of action intended to change the world rather than encode symbolic propositions about it,’’65 one is enjoined to see art both as a system of meaning encoding propositions about the world and as a system of action intended to change the world, precisely because the excitation generated by the work lies in the interaction between the two.

Acknowledgments I am particularly grateful to all of the members of a graduate seminar on ‘‘The Cult of Images’’ in Christianity, taught by my colleagues Thomas Cummins, Ioli Kalavrezou, and Hugo van der Velden at Harvard University in the Fall of 2004, which I had the privilege of auditing, for stimulating discussions on a number of the issues raised here. I am also grateful to Miguel Civil, Mark Geller, Nicholas Postgate, and the late Jeremy Black, for cautions and perspective regarding the reading of ergative construction, which have helped to shape my argument; to Richard Davis, Judith Irvine, Piotr Michalowski, Erika Naginski, Michael Silverstein, Christopher Woods, and Kapila Vatsyayan for conversations on related topics; to the editors, Jeremy Tanner and Robin Osborne, for pertinent comments; and to Robert Hunt, as always, for his acute reading of an early draft.

Notes 1 A. Gell, Art and Agency: An Anthropological Theory (Oxford, 1998), pp. 9, 66, 96. 2 See, for example, C. Faraone, Talismans and Trojan Horses: Guardian Statues in Ancient Greek Myth and Ritual (Oxford, 1992). For the ancient Near East, see E. Reiner, ‘‘Plague amulets and house blessings,’’ Journal of Near Eastern Studies 19 (1960), pp. 148–55. 3 Gell, Art and Agency, p. ix. 4 Ibid., emphasis mine. 5 The notion of distributed agency Gell seems to have derived from the work of Marilyn Strathern on the extension of the person through the social processes of material exchange (see The Gender of the Gift (Berkeley, 1988), esp. p. 273 where she defines personal agency as ‘‘mindful action’’). Gell only got to this part of the discussion in ch. 7, ‘‘The Distributed Person,’’ well after his chs. 2 and 3 on the ‘‘Art Nexus.’’ Yet his argument for the extension from animate

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 63 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

6 7

8

9

10 11 12

13

63

to distributed agency (e.g., with respect to consecrated works, Art and Agency, pp. 133–53) needs the former prior to articulation of the latter. The reader must therefore evaluate the robustness of each of Gell’s terms or concepts as his argument progresses. I. J. Winter, ‘‘Idols of the king: royal images as recipients of ritual action in ancient Mesopotamia,’’ Journal of Ritual Studies 6 (1992), 13. J. P. Waghorne and N. Cutler, eds., Gods of Flesh, Gods of Stone: The Embodiment of Divinity in India (Chambersburg, PA, 1985); C. Walker and M. B. Dick, ‘‘The induction of the cult image in ancient Mesopotamia: The Mı¯s Pıˆ Ritual,’’ in M. B. Dick, ed., Born in Heaven, Made on Earth: The Making of the Cult Image in the Ancient Near East (Winona Lake, IN, 1998); I. J. Winter, ‘‘Opening the eyes and opening the mouth: the utility of comparing images in worship in India and the ancient Near East,’’ in M. W. Meister, ed., Ethnography and Personhood: Notes from the Field (Jaipur and New Delhi, 2000), pp. 129–62. For Mesopotamia, see G. Selz, ‘‘The holy drum, the spear, and the harp: towards an understanding of the problem of deification in the third millennium Mesopotamia,’’ in M. Geller and I. Finkel, eds., Sumerian Gods and their Representations (Groningen, 1997), pp. 149–94; in general, see R. R. Dipert, Artifacts, Art Works, and Agency (Philadelphia, 1993). For the relationship between active voice and transitive verb, as distinguished from passive voice and intransitive verb, see The Chicago Manual of Style, 15th ed. (Chicago, 2003), pp. 172–3, 176–7. S. Lachenbacker, Le roi baˆtisseur: les re´cits de construction assyriens des origines a` Teglat-phalasar III, Etudes assyriologiques, Cahier 11 (Paris, 1982). H. Steible, Die altsumerischen Bau- und Weihinschriften, Freiburger altorientalische Studien, Bd. 5 (Wiesbaden, 1982), pp. 82–4. The implications of this correlation for art-historical studies of text::image relations are not trivial, but cannot be pursued fully here. Alone, the action could be ambiguous, understood either as ongoing, indicating a royal trope of ruler as perpetual builder, or as tied to a specific historical act. With the use of the past tense or perfective aspect of the verb in the inscription, however, the reader is cued to understand that the image, too, has a temporal aspect. Such instances, in which an associated text exerts authority over an ambiguous image by limiting its semantic range, merit further consideration. For sculpture, see A. Parrot, Tello. Vingt campagnes de fouilles (1877–1933) (Paris, 1948), and F. Johansen, Statues of Gudea – Ancient and Modern (Copenhagen, 1978). For statuary texts, see H. Steible, Die neusumerischen Bau- und Weihinschriften, Freiburger altorientalische Studien, Bd. 9 (Stuttgart, 1991), esp. pp. 157–79. For Cylinders A and B, see D. O. Edzard, Gudea and his Dynasty, Royal Inscriptions of Mesopotamia, Early Periods, vol. 3/1 (Toronto, 1997), esp. pp. 69–101. For a recent synthesis of text and

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 64 12.1.2007 6:04pm Compositor Name: SJoearun

64 Irene J. Winter

14

15

16

17

18

19

20 21 22

23 24

image, see C. E. Suter, Gudea’s Temple Building: The Representation of an Early Mesopotamian Ruler in Text and Image (Groningen, 2000). Cf. G. Zo´lyomi, ‘‘Directive infix and oblique object in Sumerian: an account of the history of their relationship,’’ Orientalia 68 (1999), p. 242, citing Cylinder A, I:20–1: ‘‘In it [the dream], he [Ningirsu] made him [Gudea] see Eninnu [the temple].’’ See in general L. Marin, Portrait of the King, trans. M. H. Houle (Dexter, MN, 1988); and for the ancient Near East, W. Sallaberger, ‘‘Das Erscheinen Marduks als Vorzeichen: Kultstatue und Neujahrsfest in der Omenserie Sˇumma a¯lu,’’ Zeitschrift fu¨r Assyriologie 90 (2000), pp. 227–62, esp. p. 256. I. J. Winter, ‘‘The king and the cup: iconography of the royal presentation scene on Ur III seals,’’ in M. Kelly-Bucccellati with P. Matthiae and M. Van Loon, eds., From Image to Insight: Studies in Honor of Edith Porada (Malibu, 1986), pp. 253–68. See J. Rappaport and T. Cummins, ‘‘Between images and writing: the ritual of the king’s Quilca,’’ Colonial Latin American Review 6 (1998), pp. 7–32, and I. J. Winter, ‘‘Introduction: glyptic, history and historiography,’’ in W. W. Hallo and I. J. Winter, eds., Seals and Seal Impressions: Proceedings of the XLVe Rencontre Assyriologique Internationale (Bethesda, MD, 2001), p. 3. The full text is published in Steible, Die altsumerischen Bau- und Weihinschriften, pp. 120–45, especially 128–9. Visual deployment of the deity on the obverse and the ruler on the reverse has been glossed in terms of a distinction between icon and narrative (I. J. Winter, ‘‘After the battle is over: the ‘Stele of the Vultures’ and the beginning of historical narrative in the art of the ancient Near East,’’ in H. L. Kessler and M. S. Simpson, eds., Pictorial Narrative in Antiquity and the Middle Ages, Studies in the History of Art 16 (Washington, DC, 1985), pp. 11–32). There is redundancy in text and image regarding the agency attributed to the deity on the obverse; however, the way the ruler pulls away to exercise an agency of his own on the reverse of the stele suggests that text and image must be analyzed separately before social agency is inferred. See further I. J. Winter, ‘‘Sex, rhetoric and the public monument: the alluring body of Naram-Sıˆn of Agade,’’ in N. Kampen, ed., Sexuality in Ancient Art (Cambridge, 1996), pp. 11–26. Steible, Die neusumerischen Bau- und Weihinschriften, SB.1, lines 1–20; discussed in Winter, ‘‘Idols of the King,’’ pp. 21, 29–30. Winter, ‘‘Idols of the King,’’ pp. 24–6. Steible, Die neusumerischen Bau- und Weihinschriften, SB.7, lines 21–5. These statues were named and there is likely to have been a correlation between naming and works intended to exercise agency. Winter, ‘‘Idols of the King,’’ pp. 30, 34. Sallaberger, ‘‘Das Erscheinen Marduks,’’ pp. 255–6, 258.

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 65 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

65

25 Shirin Fozi, ‘‘A miracle of darkness and the Madonna of Le Puy’’ (unpublished). A number of instances are attested in the New World, where books on cult are written in the first person as if the narrative were told by the cult image itself (Tom Cummins, personal communication). 26 Gell, Art and Agency, pp. 133ff, esp. p. 135. 27 See, for example, K. M. Ashley and P. Sheingorn, ‘‘An unsentimental view of ritual in the Middle Ages or, Sainte Foy was no Snow White,’’ Journal of Ritual Studies 6(1) (1992), pp. 63–85, and other papers in that issue, devoted to art in ritual context; also M. Caviness, ‘‘Reception of images by medieval viewers,’’ in C. Rudolph, ed., A Companion to Medieval Art: Romanesque and Gothic in Northern Europe (London, 2005). 28 Gell, Art and Agency, pp. 96ff. 29 I persist in use of the term ‘‘referent,’’ signaling the associative presence behind the work (as discussed in Winter, ‘‘Le palais imaginaire : scale and meaning in the iconography of Neo-Assyrian cylinder seals,’’ in C. Uehlinger, ed., Images as Media: Sources for the Cultural History of the Near East and the Eastern Mediterranean, Orbis Biblicus et Orientalis 175 (Fribourg, 2000), esp. pp. 75–8), rather than Gell’s term ‘‘prototype,’’ which has its own ambiguities. For Gell, see Art and Agency, p. 54. 30 Edzard, Gudea and his Dynasty, p. 87: Cyl. A, cols. xxix, 14–xxx, 12. 31 P. Michalowski, The Lamentation over the Destruction of Sumer and Ur, Mesopotamian Civilizations 1 (Winona Lake, IL, 1989), pp. 62–3. 32 See on this P. Garelli, ‘‘La conception de la beaute´ en Assyrie,’’ in T. Abusch, J. Huehnergard, and P. Steinkeller, eds., Lingering over Words: Studies in Ancient Near Eastern Literature in Honor of William L. Moran (Atlanta, 1990), pp. 173–7, and I. J. Winter, ‘‘The eyes have it: votive statuary, Gilgamesh’s axe, and cathected viewing in the ancient Near East,’’ in R. S. Nelson, ed., Visuality Before and Beyond the Renaissance: Seeing as Others Saw (Cambridge, 2000), pp. 22–43. 33 A˚. Sjo¨berg and E. Bergmann, The Collections of the Sumerian Temple Hymns, Texts from Cuneiform Sources 3 (Locust Valley, NY, 1969), pp. 24, 77. 34 Akkadian has related grammatical features, probably due to its proximity to Sumerian (Nicholas Postgate, personal communication; H. P. Muller, ‘‘Ergative construction in early Semitic languages,’’ Journal of Near Eastern Studies 54 (1996), pp. 261ff). 35 Edzard, Gudea and his Dynasty, p. 97: Cyl. B, col. xvi, 3–4. 36 Gell, Art and Agency, pp. 31 (emphasis mine), 66. 37 Ibid., pp. 6, 25, 66. 38 Equally masked is the important work of Anthony Giddens: for example, Central Problems in Social Theory: Action, Structure and Contradiction in Social Analysis (Berkeley, CA, 1979), which offers an agency-oriented framework for social analysis; and The Constitution of Society: Outline of the Theory

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 66 12.1.2007 6:04pm Compositor Name: SJoearun

66 Irene J. Winter of Structuration (Berkeley, CA, 1984), in which the agent, agency and power are delineated in terms of the flow of social action. See also I. Karp, ‘‘Agency and social theory: a review of Anthony Giddens,’’ American Ethnologist 13 (1986), pp. 131–7, esp. p. 134. 39 E.g., B. Comrie, ‘‘Ergativity,’’ in W. P. Lehmann, ed., Syntactic Typology: Studies in the Phenomenology of Language (Austin, TX, 1978), pp. 329–94; R. M. W. Dixon, ‘‘Ergativity,’’ Language 55 (1979), pp. 59–138; F. Plank, ed., Ergativity: Toward a Theory of Grammatical Relations (London, 1980); and A. Cooreman, B. Fox, and T. Giv’on, ‘‘The discourse definition of ergativity,’’ Studies in Language 8(1) (1984), pp. 1–34. See also, more recently, R. D. Van Valin, Jr., ed., Advances in Role and Reference Grammar (Amsterdam/Philadelphia, 1993); R. M. W. Dixon, Ergativity (Cambridge, 1994); C. Manning, Ergativity: Argument, Structure, and Grammatical Relations (Stanford, CA, 1996); R. D. Van Valin, Jr. & R. J. LaPolla, Syntax: Structure, meaning and function (Cambridge, 1997); Alana Johns, D. Massam & J. Ndayiragije, eds., Ergativity: Emerging Issues (Dordrecht, 2006). 40 E.g., W. H. Eilfort, P. D. Kroeber, K. L. Peterson, eds., Papers from the Parasession on Causatives and Agentivity at the twenty-first Regional Meeting, Chicago Linguistic Society, 1985 (Chicago, IL, 1985). Also, M. Silverstein, ‘‘Hierarchy of features and ergativity: grammatical categories in Australian languages,’’ in R. M W. Dixon, ed., Grammatical Categories in Australian Languages, AIAS Linguistic Series 22 (Canberra, 1976), pp. 112–71 and ‘‘Cognitive Implications of a referential hierarchy,’’ in M. Hickman, ed., Social and Functional Approaches to Language and Thought (Orlando, FL, 1987), pp. 125–64. S. Chung, Case Marking and Grammatical Relations in Polynesian (Austin, TX, 1978); J. W. Du Bois, ‘‘The discourse basis of ergativity,’’ Language 63 (1987), 805–55, for ancient Maya. More recently, M. J. Ezeizaburena and M. P. Carran˜aga, ‘‘Ergativity in Basque: a problem for language acquisition?’’ Linguistics 34 (1996), pp. 955–89, note that ‘‘the acquisition of the ergative/absolutive distinction can be learned through the semantic feature agentivity’’ (p. 965). 41 E.g., D. Foxvog, ‘‘The Sumerian ergative construct,’’ Orientalia 44 (1975), pp. 395–425, P. Michalowski, ‘‘Sumerian as an ergative language,’’ Journal of Cuneiform Studies32 (1980), pp. 86–103; M. Yoshikawa, ‘‘Ergativity and temporal indication in Sumerian,’’ in M. Mori, H. Ogawa, M. Yoshikawa, eds., Near Eastern Studies Dedicated to HIH Prince Takahito Mikasa on the Occasion of his 75th Birthday (Wiesbaden, 1991), pp. 491–507; G. Zo´lyomi, ‘‘Voice and topicalization in Sumerian,’’ Ph.D. thesis, Eo¨tvo¨s Lorand University, Budapest, 1993; C. Woods, ‘‘Deixis, person, and case in Sumerian,’’ Acta Sumerologica 22 (2000), pp. 303–34; also J. Braun, Sumerian and Tibeto-Burman (Warsaw, 2001).

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 67 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

67

42 F. R. Palmer, Grammatical Roles and Relations (Cambridge, 1994), esp. pp. 22–87. See also B. Byrne and E. Davidson, ‘‘On putting the horse before the cart: exploring conceptual bases of word order via acquisition of a miniature artificial language,’’ Journal of Memory and Language 24 (1985), pp. 377–89, which argues that it is ‘‘natural’’ to refer first to agents, then to patients. 43 Art and Agency, p. 14, re abduction as inference and induction in the service of explanation. 44 Michalowski, ‘‘Sumerian as an ergative language,’’ pp. 88–9. Michalowski’s point is that ‘‘it is not enough to state that Sumerian is an ergative language because it marks nouns in a particular manner; ergativity must be investigated throughout the grammar,’’ especially as, in some cases, ergative markings can be correlated with specific ‘‘semantic classes of verbs or on the basis of tense or aspect of the verb.’’ 45 Ibid., pp. 93, 101; also Woods, ‘‘Deixis,’’ esp. pp. 305, 327–8. 46 Ibid., pp. 101–2. Note that the reader’s cue for agency is signaled from word order and other (semantic) signals in nominative-accusative languages such as English or Latin, no less than by the ergative marker in early Sumerian. However, I stress the agentive marker as a particularly powerful and unambiguous sign. In addition, in a split ergative system such as Sumerian, agency can be indicated more richly precisely because it has been selected to be marked in the perfective. 47 T. Jacobsen, ‘‘The Sumerian verbal core,’’ Zeitschrift fu¨r Assyriologie 78 (1988), pp. 161–220, esp. pp. 204–5 on the ‘‘Intrusion of Ergative.’’ 48 This also fits well with Karp’s discussion of Giddens’s social definition of agency, in which the agent is one ‘‘engaged in the exercise of power in its primary sense of the ‘bringing about of effects’ ’’: ‘‘Agency and social theory,’’ pp. 136, 137 n. 1. 49 E.g., Ezeizaburena and Carran˜aga, ‘‘Ergativity in Basque,’’ p. 965. 50 S. Kemmer and A. Verhagen, ‘‘The grammar of causatives and the conceptual structure of events,’’ Cognitive Linguistics 5 (1994), pp. 115–56, esp. p. 147. See also S. Wallace, ‘‘Figure and ground: the interrelationships of linguistic categories,’’ in P. J. Hopper, ed., Tense and Aspect: Between Semantics and Pragmatics (Amsterdam, 1982) and A. Alexiadou, Functional Structure in Nominals: Nominalization and ergativity (Amsterdam/Philadelphia, 2001). 51 Clark, Language 57(1), p. 201; Du Bois, ‘‘The discourse basis of ergativity,’’ p. 806; Chung, Case marking . . . in Polynesian, ch. 3; Michalowski, ‘‘Sumerian as an ergative language,’’ p. 89. It is not insignificant (as noted by Judith Irvine, personal communication) that ergativity is sometimes related to social distance, hence reinforcing the public as distinct from personal context of usage. See also F. C. Southworth, ‘‘Linguistic marks for power: some relationships

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 68 12.1.2007 6:04pm Compositor Name: SJoearun

68 Irene J. Winter

52 53 54

55 56 57 58

59

60

61

62

between semantic and social change,’’ Anthropological Linguistics 16(5) (1974), pp. 177–91. What I would emphasize, then, is the diacritical nature of ergative marking, of particular interest for those who would study (inscribed) public monuments. Palmer, Grammatical Roles, p. 85. Gell, Art and Agency, p. 3 and passim. As a model for such an integrative approach, see J. Tanner, ‘‘Social structure, cultural rationalisation and aesthetic judgement in Classical Greece,’’ in N. K. Rutter and B. A. Sparkes, eds., Word and Image in Ancient Greece (Edinburgh, 2000), pp. 183–205. Gell, Art and Agency, p. 7. Ibid., p. 6. Robert Layton, ‘‘Art and Agency: a reassessment,’’ Journal of the Royal Anthropological Institute 9 (2003), pp. 447–64. See, for example, Dipert, Artifacts, Art Works and Agency. Certainly, some recent work, published after the death of Alfred Gell, has risen beyond his thumbnail sketches of the disciplines: e.g., W. van Damme, ‘‘Universality and cultural particularity in visual aesthetics,’’ in N. Roughley, ed., Being Human: Anthropological Universality and Particularity in Transdisciplinary Perspectives (Berlin/New York, 2000), pp. 258–83; N. Carroll, ‘‘Art and the domain of the aesthetic,’’ British Journal of Aesthetics 40 (2000), pp. 191–208. H. Morphy, ‘‘From dull to brilliant: the aesthetics of spiritual power among the Yolngu,’’ Man 24 (1989), pp. 21–40. See also Morphy’s comments in T. Ingold, ed., ‘‘1993 debate: aesthetics is a cross-cultural category,’’ in Key Debates in Anthropology (London, 1996), p. 249–94. E.g., A. Hauser, The Social History of Art (New York, 1957). More recent studies concerned with such issues are M. Baxandall, The Limewood Sculptors of Renaissance Germany (New Haven, CT, 1980); M. Camille, The Gothic Idol: Ideology and Image-Making in Medieval Art (Cambridge, 1989). R. Klein, Form and Meaning: Essays on the Renaissance and Modern Art, trans. J. Jay and L. Wieseltier (New York, 1979), esp. p. 149; also I. J. Winter, ‘‘The affective properties of styles: an inquiry into analytical process and the inscription of meaning in art history,’’ in C. A. Jones and P. Galison, eds., Picturing Science/Producing Art (New York/London, 1998), pp. 55–77; and M. B. Garrison, ‘‘Seals and the elite at Persepolis,’’ Ars Orientalis 21 (1992), p. 17. Norman Bryson, ‘‘The gaze in the expanded field,’’ in H. Foster, ed., Vision and Visuality (Seattle, WA, 1988), p. 91, discusses how the work actively erupts into the visual field of the viewer. A panel on agency and ‘‘The affective object in art history’’ was mounted at the annual meetings of the College Art Association in 1995, chaired by medievalist Madeline Caviness, which included papers that dealt with artworks as active and interactive components

Osborne / Art’s Agency and Art History 1405135379_4_002 Final Proof

page 69 12.1.2007 6:04pm Compositor Name: SJoearun

The Affective Object in Ancient Mesopotamia

69

in their social environments. More recently, Brian Just has taken on the issue of agency regarding pre-Columbian sculpture, suggesting that Gell ‘‘downplays the complex multiplicity of social roles art objects can play’’ (‘‘Ars longa, vita brevis: beyond agency in ancient Maya sculpture,’’ paper given at 2003 CAA session, ‘‘Between creation and destruction: the aesthetics of iconoclasm’’). Nor is the issue of the agency of the work particularly new in discourses on art. See, for example, discussion of an eighteenth-century case in E. Naginski, ‘‘The object of contempt,’’ Yale French Studies 101 (2002), pp. 32–53, esp. pp. 34–5: ‘‘The High Enlightenment recognized the potential agency of things . . . ,’’ moving beyond ‘‘Hobbesian mechanistic explanations of agency’’ to the notion that ‘‘matter was radically active,’’ and things could be ‘‘endowed with an intrinsic . . . vitality.’’ 63 R. Wicks, ‘‘Dependent beauty as the appreciation of teleological style,’’ Journal of Aesthetics and Art Criticism 55 (1997), pp. 387–400, esp. p. 395, emphasis mine. 64 Cf. M Tamen, Friends of Interpretable Objects (Cambridge, MA, 2001), pp. 2–4, on works becoming interpretable only in the context of ‘‘a community of agreement,’’ within which certain objects may be said to speak. 65 Gell, Art and Agency, p. 6.

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 70 12.1.2007 7:33pm Compositor Name: SJoearun

3

Portraits and Agency: A Comparative View Jeremy Tanner

This paper explores the uses of Alfred Gell’s art nexus for comparative art history. It focuses on three images: a statue of an East Asian Buddhist patriarch, Eizon, dedicated in 1280 AD, from the Saidaiji in Japan (figure 3.1), containing relics, from the ashes of his parents to sutras and lists of the disciples of Eizon who were responsible for the creation of the portrait (figure 3.2); a portrait of Epicurus – a Roman copy of a statue of the Greek philosopher originally created c.270 BC (figures 3.3a, b); a sixth century AD Christian icon from Sinai, portraying Christ, the Virgin Mary, and two saints (figure 3.4). Separated by three continents and some 1500 years, these three images can nevertheless be described in terms of Gell’s formula for portraits – the physical facts of the prototype’s appearance causally constrain the artist in forming an index by which the viewer is affected in virtue of inferring the prototype’s appearance and character from the index: [[[Prototype-A] ! Artist-A] ! Index-A] ! Recipient-P

Put like this, Gell’s art nexus seems like a rather reductive, flattening analytical device: it elides the manifest and considerable formal differences between the images, and their specific social contexts and cultural backgrounds. But the basic formula provided by the art nexus is intended by Gell only as the beginning of the analysis, not the end. Following a deductive strategy, one could manipulate the set of basic terms to see how this redefines the ways one might consider a portrait to function, and

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 71 12.1.2007 7:33pm Compositor Name: SJoearun

Portraits and Agency: A Comparative View

71

which of the range of variants or extensions of the basic portrait formula is actually instantiated in practice. Thus, by placing the artist before the prototype in the above formula, one would generate an expression which more or less corresponds to the role of the portrait in modernism’s artistcentered art-world: the artist’s agency is prioritized over the role of the sitter, as in Picasso’s Kahnweiler. Alternatively, adopting an inductive approach, we can explore the character of the ‘‘causal’’ relations between each element of the basic formula in a number of different cases of portraiture: what Anne d’Alleva has referred to as ‘‘the mechanics of representation.’’1 What specifically is the character of the prototype’s agency in portrait-like situations? How exactly does the prototype causally determine the character of the index, and with what consequence for the portrait’s effect on the recipient/patient? By what material technology, enchanted or otherwise, does the artist fix, bind or in some other way imprison the prototype within or on the surface of the index? To what extent is the recipient able to act as agent in relation to both artist, index, and prototype (thus rendering Gell’s formula in certain respects recursive), and how is such agency realized? A model for such a comparison is offered by Gell’s earlier work on tattooing, Wrapping in Images. Gell argues that tattooing can be meaningfully studied comparatively and cross-culturally insofar as it involves a universal technical schema (puncturing the skin, blood, scars) to realize certain modes of social and political being. Of course this schema is not realized everywhere, but it has a patterned epidemiology, most characteristically being found in ‘‘preliterate tribal societies’’ and ‘‘repressed or marginalized minorities within more complex state systems.’’ Further, the kinds of political identities and modes of being to which tattooing gives rise bear a family resemblance to each other, as instantiations of a single basic schema, but vary insofar as certain potentialities inherent in (or permitted by) the schema are ‘‘foregrounded’’ or ‘‘repressed’’ according to the demands of the environing social and cultural systems.2 Portraiture, as formulated in Gell’s art nexus, can be understood as a similar kind of basic representational schema, permitting the search for intelligible covariation between particular practices and institutions of portraiture and contextual (social, political, and cultural) factors.3 This is illustrated by the shared cognitive issue faced by the producers and consumers of the portraits with which I am concerned, namely to what extent one can really say the index is like the prototype, and the differences in the ways in which they culturally articulate that issue. All are concerned with

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 72 12.1.2007 7:33pm Compositor Name: SJoearun

72 Jeremy Tanner the fact that a portrait is both like and unlike its prototype. For the Greek philosophers this is primarily an epistemological issue, the soundness or unsoundness of images as a basis of knowledge.4 For early Christians the relationship between image and prototype was a theological issue, and incorrect understanding of the nature of that relationship was both itself heresy and the basis of heretical (whether iconophile or iconoclastic) practices in relation to images – a matter of life and death in the here and now, salvation or damnation in the hereafter.5 For Buddhist patriarchs it was the occasion for playful paradoxes: inscribing and thus authenticating painted portraits which they gave to their disciples as certificates of dharma-transmission, they teasingly call into question the very possibility of a portrait authentically representing, let alone transmitting, the essence of its subject: ‘‘This image is alas not true (or: not a portrait – zhen); truth (zhen) alas is not an image (or likeness xiang).’’6 The three cases chosen lend themselves to comparison. They are closely parallel in terms of certain fundamentals of socio-cultural setting and function. They are all commemorative portraits of pre-eminent religious figures, often religious leaders of some kind, set up by religious communities in part in order to transmit the charisma or ‘‘agentive capacity’’ of those figures beyond the limits of their own lifetime.7 The religious communities – the Epicurean philosophical school in the ancient Greek world, Byzantine Christians, Chinese and Japanese Buddhists – are specifically religious, structurally differentiated and relatively autonomous within their larger host societies. Each community was the bearer of one of the cultural orientations of the axial age, characterized by a relatively strong sense of tension between the mundane world and some transcendental vision which the leaders are concerned to transmit to their followers and, in varying degrees, implement in the world.8 Furthermore, these three traditions have been cross-referenced in the kind of casual, and generally somewhat misleading, comparisons which Gell’s framework permits us to progress beyond. Some such comparisons have been highly particularistic. Bernard Faure´ has suggested the Ch’an Buddhist patriarch portraits are like the Greek kolossoi analyzed by JeanPierre Vernant, insofar as each provides a kind of substitute body for the soul of the deceased, thus fixing a point where this world and the beyond may meet.9 Beyond the parallelism of the concept of the double, however, the kolossos and the Ch’an portrait have little in common: the former need have little if any iconic elaboration and is not necessarily the recipient of ongoing cultic attention – on the contrary kolossoi may be designed to be

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 73 12.1.2007 7:33pm Compositor Name: SJoearun

Portraits and Agency: A Comparative View

73

ritually destroyed or buried in a deserted location, precisely in order to keep the psyche or soul of the deceased at a safe distance from the living.10 Once one moves beyond the point of contact – the concept of the double – the differences in social setting, ritual practice, and material representation are so great that comparison ceases to be fruitful, and in fact calls into question whether the initial comparison is anything more than a weak, decontextualized analogy. Overly global comparisons also have relatively little explanatory value. Helmut Brinker, in a study of Ch’an reliquary portraits, compares them in passing with late-medieval reliquary statues in north-western Europe and theurgy (the magical animation of statues) in late-antique Greco-Roman cult-images.11 Hans Belting makes the same comparison, starting from a focus on the Christian reliquary statues, and infers from the parallelism that the practice is ‘‘so general in religious history’’ as to constitute a ‘‘world-wide phenomenon,’’ and thus something natural, not requiring any special explanation.12 As Gell, drawing on Foder and Boyer, has argued, the widespread occurrence of such practices does indeed depend on an innate mental module whereby we conceptualize the character of human agency in terms of a homunculus, a mind inside a body.13 Statues with relics activate that module and ‘‘animate’’ the statue, but not all representational regimes exploit that particular potential of the mind in relation to images, or, even if they do seek to activate that module, necessarily use the kernel-within-the-core strategy to do so. The Greeks of Epicurus’s era seem to have been familiar with the practice of placing ‘‘magical’’ substances like magnetite, or papyrus texts, inside little clay or wax images of gods such as Hermes, which would then be hidden in their producer’s shop as a phylactery, intended to improve business.14 However, the scope of the practice seems to have been very limited, probably not extending to the monumental statues of civic cult, let alone portrait statues, like those of Epicurus. Only systematic comparative analysis of the social and cultural environments of portraits can help us to explain the similarities and differences in the technologies of enchantment deployed to mediate the prototype’s agency to the recipient in different cases of the same basic type, the portrait formula, identified by Gell.

Portraits and Agency in East Asian Buddhism Buddhist holy-men and abbots were charismatic leaders, ‘‘reservoirs of spiritual power,’’ ‘‘awakened Buddhas’’ who, like their Christian counterparts,

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 74 12.1.2007 7:33pm Compositor Name: SJoearun

74 Jeremy Tanner took on the role of the Buddha in ritual contexts. Their death rent the fabric and threatened the continuity of their monastic communities. Portraits, whether manufactured before their death or after, served to substitute for the abbot until his successor was installed, living in the abbot’s quarters, and receiving offerings of food and incense. After the installation of the new abbot, the effigy would be removed to a portrait hall. Here, continuing to receive ritual attention and to transmit its prototype’s charisma as one amongst a set of portraits representing the genealogical lineage of the monastery, the portrait loaned authority to the doctrinal tradition adhered to by the monastery, and served to attract continued support from both laity and government officials.15 As a document describing the commissioning of a portrait of one eighth-century monk asserts: ‘‘Although his wisdom dies with him, his merits will continue to exist along with his portrait.’’16 (In what follows I treat the Chinese and the Japanese Ch’an/Zen Buddhist portraits as a unity, as also the sets of practices associated with them, which are in varying degrees characteristic of East Asian Buddhism more generally, and seem to have been imported into Japan with Ch’an Buddhism, brought by Chinese monks.17) None of these features fundamentally differentiates the Buddhist institution of portraiture from its Epicurean and early Christian counterparts, in which portraits were also the objects of ritual attentions, if somewhat less elaborate ones. But the Chinese Buddhist concept of a portrait implies a very particular understanding of exactly what might be involved in ‘‘preserving the merits’’ of a subject alongside his portrait, and thus the ways in which such images were understood to mediate their prototype’s agency. Greek concepts of likeness and mimesis define images primarily in negative terms, the gap between the image or likeness (eikon) and its prototype, the true reality. The Chinese concept of portrait or likeness, xiang, has a rather different set of connotations. Influential strands of Chinese thought, from the Book of Changes onwards, conceptualized the relationship between different levels of reality in terms of ‘‘correspondences.’’ Thus the auspicious images on an emperor’s robes – dragons, the sun and the moon, the constellations – were the iconic ‘‘doubles’’ of their heavenly counterparts. As such they effected a harmonious correlation between the ritual movements of the emperor on earth and the celestial movements of the planetary bodies, which were both indexes and determinants of the good governance of the Middle Kingdom. Whereas in Plato the analogy of the mirror can be used to devalue painting as a childish plaything, merely replicating

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 75 12.1.2007 7:33pm Compositor Name: SJoearun

Portraits and Agency: A Comparative View

75

Figure 3.1 Portrait of Eizon, 1280 AD. Saidaiji, head temple of the Shingon Risshu, Nara Prefecture. Photograph: Nara National Museum.

superficial appearances, in Chinese thought the iconic reduplication effected through mirrors permitted its users to ‘‘discern and manipulate the underlying structures of reality.’’18 Thus, alongside the ambivalence concerning the authenticity of portraits as a medium (the basis of patriarchs’ playful teasing of disciples desperate for signed portraits), there was also a belief that a portrait could in some sense embody and be capable of

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 76 12.1.2007 7:33pm Compositor Name: SJoearun

76 Jeremy Tanner transmitting the truth, the ‘‘living spirit’’ (shen) of its prototype. This was recognized in the use of the term zhen (‘‘truth’’) as an alternative for xiang (‘‘likeness’’) in describing what we would call a portrait.19 The ability of portraits to share in the nature of their prototypes, and thus effectively to mediate their agency, was realized through specific technologies of portrait production, which sought to elide the difference between portrait and prototype. Apart from high levels of mimetic verisimilitude realized by such details as eyes of colored glass and rock-crystal fingernails, the prototype could be quite literally incorporated in the image in a number of different ways. The subject’s own hair might be used for beard, eyebrows, and head hair. Sometimes the ashes of the deceased, and exuviae such as hair and nail-clippings, were placed in the interior of the statue. One text describes how the ashes of the interestingly named Wuxiang (‘‘Without distinguishing features’’) ‘‘were mixed into clay [to become] a true form of the original appearance (zhen xing).’’20 Identity between image and prototype was realized most fully in the socalled rou-shen (flesh-body icon) or zhen-shen xiang (true body portrait), self-made indexes in which the prototype/agent formed the index through acting on his own body as patient. Attested as early as the fifth century AD,

Figure 3.2 Objects discovered within the portrait of Eizon. Saidaiji, head temple of the Shingon Risshu, Nara Prefecture. Photograph: Nara National Museum.

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 77 12.1.2007 7:33pm Compositor Name: SJoearun

Portraits and Agency: A Comparative View

77

and as recently as the 1950s in Taiwan, flesh-body portraits were autoicons produced by a sustained process of self-starvation, while seated in a meditating position, relying on a diet including pine kernels and cinnabar to dehydrate the body and prevent flesh and inner organs from rotting. When the subject died he would normally be stored for a period of years, in a stone chamber or a sealed jar. If the body was still intact on retrieval, the process of preservation would be completed by lacquering, either directly applied or by draping the body with lacquer-soaked cloth, a technique in certain respects continuous with that used in the production of Buddhist statuary from the Tang dynasty onwards. Once equipped with appropriate regalia, and sometimes gilded, such images were placed in portrait-halls or zhen-shen yuan, ‘‘abodes of the true body,’’ as privileged objects of cult and pilgrimage.21 The terminal incorruptibility of the patriarch’s corpse was in itself evidence of spiritual enlightenment, by means of which the duality of soul and body is overcome, and the flesh body transformed into a true-body (zhen-shen), identical with the indestructible dharma-body (fa-shen).22 The recipients of portraits played an active role in animating them as living presences. Both true-body portraits and their more conventional counterparts could be the objects of external animation, being clothed, fed, or ritually processed to communities facing natural calamities such as floods or drought, which the portraits were held to be able to alleviate.23 Animating congruence between the external relational context, within which the portraits were set, and the internal ‘‘mind–body’’ structure of the portrait was also effected by relics placed within portraits by their dedicators and worshippers.24 Some such relics established correspondences between the portrait as microcosm and the wider universe as macrocosm: cloth models of the five viscera (heart, liver, lung, kidney, spleen), not only enlivened the image but pointed toward correspondences between the viscera, the five cosmic elements (fire, metal, wood, water, earth), and the ‘‘five transcendental wisdoms,’’ which were the ultimate object of meditative practice, a set of relationships also invoked by the Five-Ring Pagoda, or gorinto, another common relic placed inside such statues, including Eizon’s.25 Other relics mediated the patriarch-portrait’s capacity to establish a harmonious equilibrium between the portrait, its immediate social environment, and the larger cosmos. This was accomplished by the mutual encompassment of patriarch and disciple within the portrait. The contents of the statue of Shinchi Kakushin (c.1276) include a gilt-bronze cylinder, to contain the ashes of the master, inscribed with the

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 78 12.1.2007 7:33pm Compositor Name: SJoearun

78 Jeremy Tanner names of 68 of the monk’s closest disciples and followers, who were thus able to ‘‘surround the holy substance of his body . . . beyond the boundaries of life and death ‘in harmony with the Dharma and the totality of all Holy Scriptures’,’’ as related by the inscription.26 Eizon’s statue (figure 3.1) contained, in addition to lists of names of his disciples, who gained merit by participating in the creation of the statue, the ashes of his parents (figure 3.2). Like Gell’s fractal gods, such representational strategies blurred the boundaries between the individuals who formed the links in the genealogical chains that constituted Buddhist traditions, thus enhancing the sense of intimacy between image and worshipper.27 In the statue of Eizon, external and internal strategies of animation fuse: behind the crystal eyes is installed a small copper coil, representing the urna, invisible as an indicator of Eizon’s enlightenment, and a scroll inscribed with dharani, Sanskrit incantations esteemed as the quintessence of the Buddha’s wisdom. These devices focus the extromissive power of the statue, as transmitter of the charismatic agency of Eizon, at the point where it will meet the receptive gaze of the viewer.28

Portraits’ Agency and Greek Philosophers Philosophical life in ancient Greece was a religious life, oriented toward the purification of the soul, and organized on a religious basis, through thiasoi, religious troupes or bands. The souls of the charismatic leaders of philosophical movements were held to have a special affinity with the divine, and, after their death, some were honored in memorial rites in the sanctuaries where their schools were located, and where portrait statues of them were erected.29 A focus on images as an element in the cultic memorialization of their leaders was particularly pronounced in the case of the Epicurean school. Pliny tells us that the Epicureans ‘‘display images of Epicurus in their bedrooms and carry them about with them; they offer sacrifices on his birthday.’’ A character in one of Cicero’s philosophical works, reminiscing about walking in the Gardens of Epicurus with a friend, exclaims ‘‘I could not forget Epicurus even if I wanted: our associates have his likeness not only on panels but even on their drinking cups and rings.’’30 The original portrait of Epicurus was in all probability set up in the Gardens or kepos, where the Epicurean school was institutionalized as a cultic community on its founder’s death.31 Alongside ancestral cults of his parents and brother, Epicurus made provision not

Figure 3.3 Portrait of Epicurus. Cast of Roman copy after Greek original of c.270 Institute of Goettingen University, photograph Stephan Eckardt.

BC.

Reconstruction by Klaus Fittschen. Archaeological

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof page 79 12.1.2007 7:33pm Compositor Name: SJoearun

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 80 12.1.2007 7:33pm Compositor Name: SJoearun

80 Jeremy Tanner only for celebrations on the date of his birthday, according to the convention for a hero-founder, but also on the twentieth of each month, the eikades, as was customary for a god, and particularly Apollo. These festivals involved sacrifices, shared eating and drinking (probably using the cups decorated with Epicurus’s visage, a practice attested in other hero-cults), and the reading of memorial literature, celebrating the lives of the foremost disciples of Epicurus, and of course the master himself.32 Something of the terms in which Epicurus was praised can be gathered from the eulogies in Lucretius’s Epicurean poem On the Nature of Things (V.7–12): ‘‘If I am to suit my language to the majesty of his revelations, he was a god, a god indeed . . . , who first discovered that rule of life which now is called philosophy, who by his art rescued life from such a stormy sea, so black a night, and steered it into such a calm and sunlit haven.’’33 Functionally, then, Epicurus’s portraiture closely parallels that of Buddhist patriarchs like Eizon. But the Greek ontology of the image differs significantly from its Chinese counterpart. This has entailments for exactly how the prototype causally determined the index, for the nature of the artistic technologies by which the artist bound the prototype to the index in such a way that it might transmit Epicurus’s agency to recipients, and for the character of the agency transmitted. Both the concept of eikon and the institution of portraiture developed out of a process of fundamental change in the status of the image from the archaic to the classical periods. This change has been characterized as a shift from ‘‘the presentification of the invisible to the imitation of appearance.’’ The archaic image, like the kolossos, functioned as a double, the place of irruption into this world of an otherwise invisible world beyond, a real making present of an absent prototype.34 This concept of the image as a real presentification of its prototype, and the intrinsic efficacy of the image which it entailed, was challenged in the late sixth century BC. It was suggested that conventional apotropaic images, Gorgons figured on shields for example, should hold no real fear for their viewers, since they were merely the human product of ordinary craftsmen, and thus not mediators of their prototype’s agency.35 Plato further emphasized the gap between image and reality. The world of images was a world of pale reflections of reality, like the shadows cast on the wall of a cave by objects caught in the flickering light of a fire, a world of dreaming and delusion. Far from sharing in the character of their prototype, artistic representations are the antithesis of truth, ‘‘really unreal nonbeing’’ accurately reflecting not even the objects of the natural world which they

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 81 12.1.2007 7:33pm Compositor Name: SJoearun

Portraits and Agency: A Comparative View

81

imitate, let alone the world of the Forms, the ultimate true reality.36 Plato can conceptualize the possibility of a portrait, of Cratylus, in which the artist is able to represent not only his exterior features, but also the ‘‘elements within him’’ and the corresponding ‘‘warmth and softness’’ of the body, analogous to the flesh-images of Ch’an patriarchs or their portraits enlivened with textile viscera; but he does so only to raise the idea as a logical adunaton, an impossibility: for then there would not be Cratylus and an eikon of Cratylus, but two Cratyluses.37 In comparison with the Chinese concept of xiang (and archaic Greek concepts of the image), Epicurus’s concept of the image is in the same horizon as Plato’s. The image no longer functions as a double, making present a transcendent invisible. But it differs from Plato’s insofar as it conceptualizes the image positively, as a source of knowledge about the world, if rightly understood. According to Epicurus, all knowledge is ultimately derived from sense impressions, in the form of visual, acoustic, or tactile images.38 Objects emit immensely fine films of atoms, simulacra or eidola, which flow through the air and impress the sensory organs, for example the eyes, of recipients. Such images have an inherent material power or dynamis and engender a kind of atomic ripple effect, through which they reach into and ultimately affect the mind by rearranging its atomic configuration. The diathesis or mental constitution of the recipient, and the material operations which he performs on these images by virtue of this diathesis, determine the ultimate effects of such images at a psychological or intellectual level. The mentally disturbed Orestes, for example, mistakes eidola of women and snakes which have accidentally fused for real Furies, which scare him out of his wits. Conversely, the philosopher operates on impressions incoming to and stored within the mind ‘‘by means of confrontation, analogy, similarity and combination, with some contribution from reasoning also,’’ to determine the status of impressions and to use them as the basis for elaborating abstract knowledge: thus, the idea of the center of the earth (not observable) can be constructed on the basis of analogy from the experience of a small ball.39 Not surprisingly, since the Epicurean poet Lucretius is one of the sources Gell uses in constructing his own theory, this conceptualization of the nature of the image lends itself very nicely to a purely ‘‘causal’’ account of the formation and agency of Epicurus’s portrait. Bernard Frischer has shown that Epicurus’s portrait (figure 3.3) incorporates a series of iconographic references. In addition to a standard philosopher

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 82 12.1.2007 7:33pm Compositor Name: SJoearun

82 Jeremy Tanner type (indicated by Epicurus’s clothing, coiffure, and bookroll), Herakles is referenced by the narrow, elongated shape of the head, with sunken cheeks, Asklepios, the god of healing, by the lightly furrowed forehead, arched eyebrows, and intent gaze, the father-figure, according to some copies and reconstructions, by the right arm stretched out toward the viewer, in a gesture of teaching, reminiscent of the father-figure on Attic stelai stretching his arm forward to greet his relatives in the gesture of dexiosis.40 These references served to suggest that, through his philosophical teaching, Epicurus was a savior of mankind, like Herakles, bringing healing of the mind, as a counterpart to Asklepios’s healing of the body, and offering fatherly reassurance in the face of death. How did Epicurus come to be visualized in this way? According to Epicurean theory, we should think of the artist as synthesizing the image of Epicurus, which he subsequently objectifies in a portrait, on the basis of visual impressions received from Epicurus, as well as from Herakles and Asklepios (in the form of their statues), and from the archetypal father-figure represented in many fourth-century funerary reliefs. Translated into Gell’s algebra this might look something like: [Prototype1 A ! [[Prototypes2–4 A ! Artist A] ! [Prototype5 A ! Artist A]] ! Index A] ! Recipient P

Epicurus thus appears twice as prototype: first, in person impressing the sculptor with his special powers, whether by virtue of his savior-like qualities or by specifying in the commission the image types he wishes to be deployed as models for his own portrait; second, in the sculptor’s imagination as the image of Epicurus synthesized with the features of three secondary prototypes, the ‘‘father-figure’’ Asklepios and Herakles (prototypes 2–4 in the formula). This final prototype (no. 5) determines the executive activity of the sculptor who objectifies it as a portrait of Epicurus. The formula differs from Gell’s Mona Lisa formula – emphasizing the mysterious creativity of the artist in relation to the prototype – by virtue of the fact that it is only by twice being passively acted on as patient, first to the agency of Epicurus himself and second to the agency of the images of Herakles and Asklepios, that the artist is able actively to form the prototype which shapes the character of the index. Thus the artist’s agency is very much secondary to that of Epicurus himself. In terms of its technical character, the artistic fixing of the prototype to the index is understood as a purely mechanical process, having little in common

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 83 12.1.2007 7:33pm Compositor Name: SJoearun

Portraits and Agency: A Comparative View

83

with the enchanting magical technologies which Gell invokes, exemplified in the Buddhist portraits of the first case-study. What of the recipient? It has been argued that Epicurus’s portrait functioned primarily as an advertisement for his philosophy, standing in public space and emitting psychologically reassuring eidola.41 In this case the relatively passive viewer-recipient simply abducts the agency of Epicurus as a heroic savior-healer by virtue of the philosophical activity which he is represented as performing, transmitting knowledge by teaching from the bookroll open on his lap. The pleasure associated with these reassuring images will have induced the recipient to seek out their source, in order to augment and safeguard that tranquillity of mind which the images engender. More probably the primary contexts for the display of images of Epicurus were private: in the first place in the Garden, his philosophical school; secondly in the villas of educated members of the HellenisticRoman elite, where libraries and peristyle gardens might be equipped with sculptural decorations, including portraits, designed to evoke the settings of the schools of the original Greek philosophers.42 Here, on the occasion of the festivals discussed at the beginning of this section, the recipient may have played a more active role in ‘‘animating’’ the image, garlanded with wreaths, and ‘‘worshipped with a divine ritual,’’ as Seneca describes his birthday celebrations before the images of his cultural heroes – Socrates and Plato, Zeno and Kleanthes – preserved and reverenced as stimuli for his mind (incitamenta animi).43 Epicurus himself asserted that such ‘‘worship of the philosopher is a great good to his votary.’’44 And Seneca explains how the benefit is supposed to accrue, on the basis of Epicurus’s own instructions: ‘‘Cherish some man of high character, and keep him ever before your eyes, living as if he were watching, and ordering all your actions as if he beheld them.’’45 Seneca reinforces the contrast with traditional religion by deliberate parallelism, referring to the innermost recesses of the mind or soul in which the image of the revered philosopher is held as the secretum, or adyton, of a temple, where the cult-statue might be housed. The agency exercised by the portrait image of Epicurus is thus logically parallel with that of the Buddhist patriarch portraits and Christian icons, all also instances of Gell’s ‘‘idol formula,’’ insofar as through the image the prototype as agent acts on the recipient as patient.46 But notwithstanding the additional parallels of substance in the ritual framing of the portrait, the locus, scope, and nature of the image’s agency – rectifying

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 84 12.1.2007 7:33pm Compositor Name: SJoearun

84 Jeremy Tanner the character and calming the mind of the viewer – are significantly different from that of Buddhist patriarch portraits or icons of Christian saints, as a result of the different ontologies of the image characteristic of each case. There is no expectation that the image may be empowered to intervene in the natural processes of the world, like the flesh-portrait of the Buddhist patriarch Huineng, or that the absent prototype may step forth from the image as a real presence, to engage in interaction with the viewer, as Christian saints sometimes do from their icons (or indeed the gods and their statues of traditional Greek civic religion so much criticized by Greek philosophers like Epicurus). The rhetorical and agentic force of the Epicurean portrait lies in Epicurus being able to exercise agency over and on behalf of his adherents by purely natural material processes, even in his definitive absence, annihilated by death. ‘‘Happy is the man,’’ as Seneca writes, ‘‘who not only in his actual presence [non praesens tantum], but even when in their thoughts can make others better.’’47

Early Christian Portraits: The Agency of Icons Like portraits of Buddhist patriarchs and Epicurean philosophers, icons of Christian saints were the objects of ritual attention in the context of the festival cycles which defined the religious year of their community of worshippers. Like portraits of Buddhist patriarchs, but unlike those of Greek philosophers, images of saints often had an association with their relics, though, as we shall see, the exact ways in which image and relic were related to each other and mediated their prototype’s agency were significantly different. In the early Christian period, the ritual uses of portraits of Christ and the saints closely paralleled, because they were partly modeled on, the practices characteristic of pagan and philosophical sects in honoring and venerating holy-men like Epicurus. In his sermon on the martyr Meletius of Antioch, John Chrysostom recalls how ‘‘everyone rejoices to put his image everywhere, on rings, goblets, dishes, and on bedroom walls, so that they can not only hear his holy discourses, but also gaze everywhere on his bodily image, thus gaining a double consolation to make up for his departure from us.’’48 From an early date, the boundaries of licit practice were patrolled by the church authorities, anxious on the one hand to monopolize legitimate access to the sacred, on the other to maintain distinction from pagan and secular practice.49 The final settlement of the

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 85 12.1.2007 7:33pm Compositor Name: SJoearun

Portraits and Agency: A Comparative View

85

iconoclastic controversies sought to regulate the cult of images more closely within the context of official church liturgy. The scope of the portrait representation of saints – hitherto extending to household objects such as textiles and pottery dishes – thereafter became more restricted, focused primarily on portable and mural icons, in both painted and mosaic media.50 Visitors to churches routinely showed respect to icons, by lighting candles, touching and kissing the icons, and, as need demanded, making specific requests or prayers to the saint.51 In accordance with the festival cycle of the year, wall-icons might be lit up by candles, or portable icons hung from the iconostasis on their saint’s special day, in order to receive the ritual attentions which as a rule were reserved to the titular saint of a church or to Christ and Mary. Like the images of Buddhist patriarchs, the icons of the primary saints of church or monastery, or of Christ and Mary, might perform a role in the political organization of their community, acting as guardians of the keys which symbolized the authority of specific officials, to whom they entrusted the keys of office in return for three genuflections and a kiss.52 The exact nature of the relationship between image, prototype, and recipient entailed by such ritual engagements is a preoccupation both of narrative accounts of the origins and functions of specific icons and of Byzantine theology more generally. In terms of Gell’s art nexus, the issue of the status and character of icons may be said to have hung on whether the supposed prototype, Christ or a saint, could be held to impress their true nature on the index or image, which could thus stand in for them as an agent in the world, and reciprocally whether in venerating the material index the recipient could be truly said to act upon the prototype. According to iconoclastic thought, the only form in which Christ’s true nature as savior could be represented, and mediated to human beings, was in the bread and the wine of the Eucharist. A humanly executed icon represented only the fleshly human aspect of Christ, and claiming through it access to the prototype confused that fleshly aspect with the divine uncircumscribable godhead – both heresies.53 Iconophile stories sought to minimize the gap between image and prototype, eliding the mediating figure of the artist in order to enhance the directness of the reciprocal action of prototype and recipient. A number of images were claimed to be acheiropoieta, made without the intervention of human hands, through the direct impress of the prototype. The Mandylion of Edessa was a sudarium or handkerchief on which Christ had impressed a permanent image when he used it to wipe his face.

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 86 12.1.2007 7:33pm Compositor Name: SJoearun

Figure 3.4 Icon of Virgin and Child with saints. Sixth century AD. After Kurt Weitzmann, The Monastery of St. Catherine at Mount Sinai: the Icons, vol. 1, plate XLIII. ß Princeton University Press, 1976, renewed 2004 PUP. Reprinted by permission of Princeton University Press.

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 87 12.1.2007 7:33pm Compositor Name: SJoearun

Portraits and Agency: A Comparative View

87

The cloth’s efficacy in transmitting the agency of its prototype was demonstrated during a siege of Edessa by the Persians. When the city walls were already aflame, the bishop Eulalios processed the sudarium around the battlements, producing a wind which blew the fire back against the besiegers, and saved the city. A mosaic icon representing the Virgin and the Child manifested its unity of nature with its prototype by bleeding when shot in the knee with an arrow by an Arab. Similar guarantees underwrote the authenticity of portraits of saints, strengthening the link between index and prototype to the point of identity in nature, whereby the image could take on the agency of its prototype. The saints who appeared to believers in healing-dream visions could be identified on the basis of their likeness to their icons, mutually validating both icon and vision. One story tells how a dry well gave water when the icon of a saint was lowered into it; another, how a woman was restored to health and relieved from pain after drinking a potion concocted from pigments scraped from a wall-painting of saints Cosmas and Damian.54 Claims to such a close relationship between image and prototype, and the veneration of images which it entailed, ran the risk of an accusation of idolatry, offering worship to a graven image, and thus contravening the second commandment. Iconophiles replied by claiming that it was not the material, the wood and pigment, that they venerated, but the prototype, and the woman who consumed the icon-pigments was not cured by the powder as such but by the ‘‘intervention’’ (ephoitesis) of the saints.55 Faith on the part of the viewer was required to activate the icon. Conversely, outrages against an image, spoken or performed as acts of polemical skepticism, provoked retribution, not by the icon itself, but by the injured prototype – though the distinction sometimes seems very blurred. When a priest, possessed by a demon, used a knife to scratch out the right eye of an icon of St. Andrew, his own right eye immediately sprang from its socket to make good the loss in the icon!56 But how, if the person of the saint or Christ was not literally embodied in the image, could the icon be the medium of both the viewer’s real access to the saint and the saint’s agency in relation to the viewer? The nearest Byzantine theologians get to a solution to this apparent contradiction is in the writings of St. John of Damascus, which perhaps best represent the intellectual consensus formed after the second iconoclastic controversy, though to what extent these nice theological distinctions were of interest to or maintained by ordinary believers is open to question. While

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 88 12.1.2007 7:33pm Compositor Name: SJoearun

88 Jeremy Tanner accepting that the divine, God the father, is invisible, uncircumscribed and without form, thus unrepresentable, John argues that God enabled ordinary human beings to experience his divinity through the person of Christ, being of one substance with the father. Icons of both Christ and the saints, although not embodiments of the invisible divine, nevertheless share a link of substance with it. Just as shadows are linked to the body which casts them – and reputedly, in the case of shadows of saints, could heal the sick or put demons to flight – so the icon has an internal link to its prototype, only more so since it is a ‘‘more distinct portrayal of its prototype,’’ and thus can take on the agency of the prototype in putting demons to flight. Playing on the synonymy of ‘‘shadow’’ and ‘‘eikon’’ inherited from Plato’s aesthetics, John reverses the value placed on both shadows and likenesses by Plato. John develops an idea later put in even stronger terms by Theodore of Studion – arguing against a fictitious ‘‘heretic’’ interlocutor – namely that far from images being the realm of non-truth, they are the only means through which limited humans can gain a glimpse of the light of the infinite divine. This vision of heavenly light was perceived through the icon as through a ‘‘dark glass’’ or, more literally, ‘‘as through a looking glass in an enigma’’ (di’ esoptrou en ainigmati). Some interpret this phrase as an analogy to observing celestial phenomena, such as eclipses, through the dulled reflection of a metal mirror.57 Others invoke the background of Old Testament thought, where not only is God not to be known directly, but even the reflected brightness of someone like Moses, who was in close proximity to God when receiving the commandments, must be veiled to avoid harm to lesser mortals.58 Either basis for the metaphor implies that sight or knowledge of the object in question is only possible by virtue of mediation, without which the brilliance of the object of knowledge would be harmful to the subject. Thus, St. John’s implication is that the icon allows our sight to reach beyond the natural perception characteristic of our worldly existence to a splendid window onto the unbearable brilliance of the world beyond, the light of God. Gazing through this window, like Alice entering the world of the looking-glass, the recipient as agent could engage the saintly prototype through acts of faithful veneration; conversely the icon could quite literally come alive as the saint stepped forth from it to act in this world on behalf of the viewer or to journey to heaven in search of still more potent aid.59 These ideas concerning the ontology of the image inform the techniques selected from the array of those available in late antiquity to materially represent saints and mediate their agency through icons. We can perhaps

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 89 12.1.2007 7:33pm Compositor Name: SJoearun

Portraits and Agency: A Comparative View

89

best understand their specific character if we start by considering some possibilities that were not pursued. Christian images of saints might have been much more like Ch’an portraits. When Daniel the Stylite died in 493, the local clergy filled the gap left by his absence by standing the corpse upright, attached to a plank, to be displayed to crowds who ‘‘with cries and tears besought him to be an advocate with God on behalf of them all.’’60 That this model of Christian-style flesh-body portrait failed to catch on is not simply a function of a technological deficit: after all the closest models, both visually and geographically, for early Christian icons are the late Roman portraits from the Fayum in Egypt, which originally covered the faces of mummies.61 It was selected against by specific historical factors, and more general cultural ones including the ontology of the icon we have just discussed, which gave the early Christian and Byzantine institution of art its characteristically ethereal features. Christian icons achieve an almost immaterial materialization of their prototypes’ agency. In this respect, they stand in polemical opposition to the pagan tradition of statues of gods, and perhaps also for this reason have a less intimate relationship with relics than was characteristic of Ch’an portraits. Christian icons are associated with relics, but they maintain their mutual autonomy in a different way from the Ch’an images. Icons of saints mounted over their tombs might dribble healing oil from their faces and hands, modeled in metal-relief.62 At Hosios Loukas in Greece, the icon of St. Luke is placed on the west wall of the northern arm of the church. Opposite on the east wall, above the tomb in which the healing relics of Luke were preserved, was an icon of the Virgin and Child. Luke’s icon was thus an autonomous means through which the saint’s intercession was sought, mediated then through the Virgin and finally, in the vault above Mary, Christ in heaven, through whose grace, one supposes, the relics ultimately functioned.63 A statue containing a relic, or a body-portrait, like the Ch’an patriarchs, would have seemed all too close to pagan cult-statues, animated by theurgists’ packets of noxious substances and inhabited by the demons which these substances attracted. Demons were above all material, creatures of darkness, ‘‘mirror images of matter and evil.’’64 Saints and angels, by contrast, were fundamentally immaterial, pure spirit, reflections and emanations of the divine light of God.65 Monastic asceticism was effectively a practice by which would-be saints achieved a state of grace by dematerializing themselves: their fleshless, shadow-like character becomes a topos of descriptions of saints and their icons, like that of St. Mary of Egypt:66

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 90 12.1.2007 7:33pm Compositor Name: SJoearun

90 Jeremy Tanner She had the fleshless quality of angels, and it is a marvel how she turns weight into shadow and how she both is substance and remains shadow, and how the thinness of a shadow is embodied [in her] . . . It seems to me that they paint the Egyptian woman as the nature of angels was often seen. For she is not inferior to the angels, even by a little.

These conceptions of the nature of their prototypes, and the Christian ontology of the image, are reflected in the technologies of enchantment through which the immaterial presence of the saints was realized in icons like our example from Sinai (figure 3.4). The image is quite flat, and its surface transparent, a window, as some Byzantine texts express it, onto another, more ethereal world, set apart from the everyday world by the golden splendor associated with the light of heaven.67 The figures inhabiting this ethereal world vary in their degree of dematerialization. The two saints, standing either side of Mary are rendered in an abstract style, their bodies strangely insubstantial in drapery whose decorative pattern suggests a surface which neither folds nor recedes into depth at the contours of the figures. More mobile, but still more immaterial, like light itself, are the two angels in the background, standing before a relatively substantial architectural niche, but themselves painted white on white with transparent nimbuses, flanking the shaft of light which falls from heaven to form the aura of Mary and the Christ child. The most substantial figure is Mary herself, with her open lap, one knee raised above the other, providing a secure support for the Christ child, and her figure endowed with bodily substance by the careful modeling of her drapery, all contributing to suggest the effectiveness with which Mary mediates the faithful viewer’s petitions by virtue of her shared humanity.68 The faithful viewer, lighting a lamp or candle before the image, activates it; the flickering flame brilliantly reflected in the gold of the haloes and Mary’s throne brings the figures to life, as the icon casts its own light in turn back into the space of the viewer, the real emanation of God’s grace into our sinful world, mediated by Mary, the saints and the angels.

Conclusion I have tried to show how Gell’s art nexus offers a language for comparing artistic institutions as systems of material representation. This has

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 91 12.1.2007 7:33pm Compositor Name: SJoearun

Portraits and Agency: A Comparative View

91

necessitated qualifying Gell’s approach in certain respects. Most notably, I have suggested that the representational schemata that can be described in terms of Gell’s art nexus need to be embedded in the two abstractions of ‘‘culture’’ and ‘‘social structure or institutions,’’ which Gell eschewed in favor of an exclusive focus on social interaction (itself as much an abstraction as culture or social structure, pace Gell). If culture and the language-like properties of certain cultural systems are admitted into the analysis, representational practices and interpretive regimes that focus primarily on the communication of symbolic meaning – as in classic iconographical analysis and, amongst our case-studies, the Epicurean portraits – can be interpreted as technological systems intended to transmit agency and affect viewers in a manner parallel to Gell’s preferred examples, but within the context of a different cultural ontology. Gell’s emphasis on the causal relations between the elements of the art nexus serves two valuable functions. First, it defamiliarizes the conventional semiotic categories of post-structuralist art history, heavily influenced by linguistic theory and with the default assumption that systems of meaning consist in networks of culturally arbitrary symbols. Thinking in terms of causation at least makes us explore in some detail the character of the links between the elements of the art nexus, as conceptualized in particular cultures and as realized through specific technical practices – levels of analysis often short-circuited in traditional iconographic analysis. Second, Gell’s art nexus offers a metalanguage which makes institutions of artistic representation comparable across cultural boundaries, on the basis of a conceptual model which does not privilege Western ontologies of art in the way characteristic of iconography or formalism. In certain respects, the analysis which results has some of the characteristics of what Panofsky referred to as the iconological level of analysis, where he analyzes the ‘‘basic principles,’’ which underlie ‘‘the production and interpretation of images’’ and ‘‘give meaning even to the . . . technical procedures employed.’’69 But in place of the appeal to ‘‘intuition,’’ or the vague and often undertheorized links between the work of art and its larger milieu characteristic of general appeals to ‘‘context,’’70 Gell’s art nexus offers a matrix for elaborating a set of models in terms of which the similarities and differences between artistic systems can be systematically described, and then perhaps explained, in terms of interrelationships between social structure, culture, and the specific eventualities of history.

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 92 12.1.2007 7:33pm Compositor Name: SJoearun

92 Jeremy Tanner Acknowledgments For their helpful comments on an earlier version of this paper, my thanks to Jas´ Elsner, Lukas Nickel, Peter Stewart, and to Oliver Moore and his students and colleagues in Leiden; also to Robin Cormack for assistance in tracking down a photograph of the Icon from Sinai.

Notes 1

2 3 4 5 6

7 8 9 10

11 12 13 14 15 16

A. D’Alleva, ‘‘Captivation, representation and the limits of cognition: interpreting metaphor and metonymy in Tahitian Tamau,’’ in C. Pinney and N. Thomas, eds., Beyond Aesthetics: Art and the Technologies of Enchantment (Oxford, 2001), p. 81. A. Gell, Wrapping in Images: Tattooing in Polynesia (Oxford, 1993), pp. 8–9, 18, 289. Gell, Wrapping, p. 289. M. Barasch, Theories of Art: from Plato to Winckelmann (New York, 1985), pp. 4–14. R. Cormack, Byzantine Art (Oxford, 2000), pp. 86–102, esp. p. 87. Hong-zhi, quoted in T. Griffith Foulk and Robert H.Sharf, ‘‘On the ritual use of Ch’an portraiture in medieval China,’’ Cahiers d’Extreˆme Asie 7 (1993–4), p. 206. A. Gell, Art and Agency: an Anthropological Theory (Oxford, 1998), pp. 21, 212. S. Eisenstadt, ‘‘The axial age: the emergence of transcendental visions and the rise of clerics,’’ European Journal of Sociology 23(2) (1982), pp. 294–314. B. Faure´, The Rhetoric of Immediacy: a Cultural Critique of Zen/Ch’an Buddhism (Princeton, NJ, 1991), pp. 171–3. J.-P. Vernant, ‘‘The representation of the invisible and the psychological category of the double: the colossos,’’ in Myth and Thought in Ancient Greece (London, 1983), pp. 305–20. H. Brinker, ‘‘Facing the unseen: on the interior adornment of Eizon’s iconic body,’’ Archives of Asian Art 50 (1997–8), pp. 42–4. H. Belting, Likeness and Presence: a History of the Image before the Era of Art (Chicago, 1994), p. 302. Gell, Art and Agency, pp. 126–33. K. Preisendanz and A. Henrichs, eds., Papyri Graecae Magicae (Stuttgart, 1973–4), IV. 2359–72, 3125–71. Foulk and Sharf, ‘‘Ritual use,’’ pp. 171–80, 191–2. H. Mori, Japanese Portrait Sculpture (Tokyo, 1977), p. 96.

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 93 12.1.2007 7:33pm Compositor Name: SJoearun

Portraits and Agency: A Comparative View

93

17 J. C. Dobbins, ‘‘Portraits of Shinran in medieval pure land Buddhism,’’ in R. H. Sharf and E. H. Sharf, eds., Living Images: Japanese Buddhist Icons in Context (Stanford, CA, 2001). 18 Foulk and Sharf, ‘‘Ritual use,’’ pp. 158–63, quotation p. 159; Bernard Faure´, Visions of Power: Imagining Medieval Japanese Buddhism (Princeton, NJ, 1996), pp. 237–9. 19 Foulk and Sharf, ‘‘Ritual use,’’ pp. 160–1. 20 H. Brinker and H. Kanazawa, Zen Masters of Meditation in Images and Writings, Artibus Asiae Supplementum 40 (Zurich, 1996), pp. 87–8, 92 quotation from Song gaoseng zhuan, ch. 19; D. Croissant, ‘‘Der unsterbliche Leib: Ahneneffigies und Reliquienportra¨t in der Portra¨tplastik Chinas und Japans,’’ in M. Kraatz, J. Meyer zur Capellan, and D. Seckel, eds., Das Bildnis in der Kunst des Orients (Stuttgart, 1990). 21 Brinker and Kanazawa, ‘‘Zen,’’ pp. 89–91; Croissant, ‘‘Der unsterbliche Leib,’’ p. 237; R. Scharf, ‘‘The idolization of enlightenment: on the mummification of Ch’an masters in medieval China,’’ History of Religions 32(1) (1992), pp. 14–15. 22 Sharf, ‘‘Idolization,’’ pp. 22–3; Brinker-Kanazawa, Zen Masters, pp. 88–90; Faure´, Rhetoric, pp. 133–5, 148–78, esp. p. 170. 23 Scharf, ‘‘Idolization,’’ p. 8; Croissant, ‘‘Der unsterbliche Leib,’’ p. 243. 24 Gell, Art and Agency, p. 137. 25 Croissant, ‘‘Der unsterbliche Leib,’’ p. 253; Brinker, ‘‘Eizon’s iconic body,’’ p. 45. 26 Brinker and Kanazawa, Zen Masters, pp. 93–4. 27 Gell, Art and Agency, pp. 137–43. 28 Groner, ‘‘Eison’s religious activities,’’ pp. 148–9; Brinker, ‘‘Eizon’s iconic body,’’ pp. 53–6. 29 P. Boyance´, Le Culte des Muses chez les philosophes grecs (Paris, 1937), pp. 267–72, 327. 30 Pliny, Natural History 35.5; Cicero, De Finibus 5.1.3. 31 H. Wrede, ‘‘Bildnisse epikureischer Philosophen,’’ Mitteilungen des deutschen archa¨ologischen Instituts 97 (1982), p. 237. 32 D. Clay, ‘‘The cults of Epicurus,’’ Cronache Ercolanesi 16 (1986), pp. 11–28; C. Williams, ‘‘Late Hellenistic portrait bowls from Mytilene,’’ Classical Views17(2) (1998), pp. 321–36. 33 Trans. R. Latham; Boyance´, Culte, pp. 322–3. 34 J.-P.Vernant, ‘‘The birth of images,’’ in Mortals and Immortals: Collected Essays (Princeton, NJ, 1991), pp. 164–85. 35 P. O’Sullivan, ‘‘Image and artifice in early Greek literature’’, Ph.D. dissertation, Cambridge University, 1998, pp. 82, 125–55. 36 Sophist, 240b11; Vernant, ‘‘Birth,’’ pp. 168. 37 Cratylus, 432b–c.

Osborne / Art’s Agency and Art History 1405135379_4_003 Final Proof

page 94 12.1.2007 7:33pm Compositor Name: SJoearun

94 Jeremy Tanner 38 Frischer, Sculpted Word, pp. 80–5, and A. A. Long and D. N. Sedley, The Hellenistic Philosophers, vol. 1 (Cambridge, 1987), pp. 72–90. 39 Diogenes Laertius 10.52. 40 Frischer, Sculpted Word, pp. 199–240. 41 Frischer, Sculpted Word. 42 Wrede, ‘‘Bildnisse,’’ p. 237; R. Neudecker, Die Skulpturenausstattung ro¨mischer Villen in Italien (Mainz, 1988), pp. 67–77. 43 Neudecker, Skulpturenausstattung, p. 72. 44 Epicurus, Frg. 32; Clay, ‘‘Cults,’’ p. 25. 45 Epicurus, Frg. 210; Seneca, Epistle XI. 46 Gell, Art and Agency, p. 40. 47 Seneca, Epistle XI.9. 48 In St. John of Damascus, On the Divine Images, 2nd Apology, Appendix, trans. D. Anderson (New York, 1980). 49 A. Grabar, Christian Iconography (London, 1969), pp. 67–8, 83–5. 50 H. Maguire, The Icons of their Bodies: Saints and Images in Byzantium (Princeton, NJ, 1996), pp. 137–42. 51 R. Cormack, Painting the Soul: Icons, Death Masks and Shrouds (London), pp. 24–7. 52 Belting, Likeness, pp. 172, 518. 53 R. Cormack, Byzantine Art (Oxford, 2000), p. 87; Barasch, Icon, p. 210. 54 R. Cormack, Writing in Gold: Byzantine Society and its Icons (London, 1985), pp. 46–7, 124–30. 55 Maguire, Icons, p. 138, quoting the Miracles of saints Cosmas and Damian, sixth century. 56 Cormack, Writing, p. 129, ninth-century text. 57 Barasch, Icon, p. 218. 58 Norbert Hugede´, La Me´taphore du miroir dans les e´pıˆtres de saint Paul aux Corinthiens (Paris, 1957). 59 Barasch, Icon, pp. 204–20; Cormack, Writing, p. 46 for the Alice analogy. 60 Life of Daniel the Stylite, ch. 99; P. Brown, Society and the Holy in Late Antiquity (London, 1982), p. 268. 61 Cormack, Painting, pp. 69–73. 62 Belting, Likeness, pp. 187–8. 63 Maguire, Icons, pp. 93–7. 64 Barasch, Icon, pp. 100–1: Minucius Felix and Tatian. 65 Barasch, Icon, pp. 238–41: Dionysius Areopagiticus, St. John of Damascus. 66 Maguire, Icons, pp. 66–74: John Apokaukos. 67 Cormack, Byzantine, p. 40. 68 Belting, Likeness, p. 131; Cormack, Byzantine, p. 77; Maguire, Icons, pp. 87–8. 69 E. Panofsky, Studies in Iconology (Oxford, 1939), p. 14. 70 N. Bryson, ‘‘Art in context,’’ in R. Cohen, ed., Studies in Historical Change (Charlottesville, VA, 1992), pp. 18–42.

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 95 12.1.2007 6:10pm Compositor Name: SJoearun

4

The Agency of, and the Agency for, the Wanli Emperor Jessica Rawson

The Ming emperor, Shenzong (1573–1620), known today by the title of his reign, Wanli, is the only Ming emperor whose tomb has been excavated.1 This chapter will take advantage of that fact to consider a ritual portrait of the Emperor with reference to the robes and other artifacts discovered in that tomb. In this way it will be possible to explore some aspects of the concept of agency as developed by Alfred Gell. In this portrait (figure 4.1) the Emperor holds us in his gaze, or does he? The oval face, staring out, even beyond us, to sights we ourselves cannot see, is the only trace of personality. The Emperor is otherwise wrapped, or even encased, in a yellow Imperial Robe, embroidered with dragons in roundels. This robe is known as a gunfu. On the skirt of the robe, below the Emperor’s belt, are two rows of six emblems. These are, reading from top to bottom, the ritual vessels, aquatic grass, fire, millet, ax, and fu characters. To these are normally added another six. These include the sun and moon, here on the proper left and proper right of the shoulders of the garment, the emblem of the constellations, which is likely to be found on the back of the robe, and three other motifs, the mountain, the paired dragons, and the flowery fowl, also likely to be on the back and so out of sight of the viewer. Such figures are mentioned in ancient texts, most notably one of the late chapters of the text known as the Shu Jing, dated to the third century BC.

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 96 12.1.2007 6:10pm Compositor Name: SJoearun

Figure 4.1 Anonymous, ritual portrait of the Wanli Emperor. By permission of the National Palace Museum, Taipei.

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 97 12.1.2007 6:10pm Compositor Name: SJoearun

The Agency of and for the Wanli Emperor 97 [My] ministers constitute my legs and arms, my ears and eyes. I wish to help and support my people; you give effect to my wishes. I wish to spread the influence of [my government] through the four quarters; you are my agents. I wish to see the emblematic figures of the ancients – the sun, the moon, the stars, the mountain, the dragon, and the flowery fowl, which are depicted [on the upper garment]; the temple cup, the aquatic grass, the flames, the grains of rice, the hatchet and the symbols of distinction (fu), which are embroidered [on the lower garment: I wish to see] all these displayed with the five colors, so as to form the [official robes]; it is your duty to adjust them clearly.2

Although the text claims antiquity, it was probably assembled no earlier than the Han period (206 BC–AD 220). Therefore, we do not know whether anything similar was actually employed on the robes of rulers before the Han. Around the throne and the back of the figure is a heavy textile that appears to carry a lotus design on one side and a pattern of clouds on the other. On the carpet, the same clouds alternate with designs known as the precious treasures on the carpet.3 In the Chinese tradition, as will be discussed below, such motifs were expected to ensure auspiciousness. Thus the ornament alone indicates to an informed viewer that some sort of agency is at work, but who is the agent and what is the objective of the agent? I shall look at these motifs later on. This portrait and all like it were objects of ritual. Paintings of this category were deployed at the funeral of the person and above all in the regular, often daily, rites, when food and drink were offered to nourish him or her after death.4 Before the Song dynasty (AD 960–1279), offerings to ancestors, a fundamental Chinese religious practice, were made in front of tablets inscribed with the names of the person. The rituals in which food and drink were formally prepared and offered, accompanied often by music, were deemed to attract the spirit of the dead individual to be present in the tablet. From the Song period, sculptures and paintings were made and, as with the spirit tablet, the spirit was thought to be present in their image at the moment the offerings were made. As Craig Clunas has put it, the ancestors ‘‘were certainly understood as being continuously ‘present’ with their descendents, and never more than when present in pictorial form.’’5 A reciprocal relationship was understood. In return for continued attention to the dead members of the family with offerings and filial respect, it was expected that the ancestors, the dead, would contribute to the benefits that the living hoped would

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 98 12.1.2007 6:10pm Compositor Name: SJoearun

98 Jessica Rawson continue throughout their own lives. Such portraits were also made for a few other occasions.6 A ritual painting had to present a likeness, and the detail of the robes, belt, and hat had to be entirely correct. Otherwise the painting might not achieve the objective of making present the emperor. Indeed, the use of paintings was criticized by the Song neo-Confucian philosophers on the grounds that if incorrect features were represented the offerings would be received by the wrong ancestor!7 Thus, although these ritual portraits closely resemble one another, undoubtedly extreme concern and care was taken with all details to ensure not only the correct resemblance, but also the correct dress, headdress, and other paraphernalia, most especially belts, which were emblems of rank. These are the features that would have preoccupied the viewers, if they had reflected upon their experiences at all. Artists do not figure in this account, as the system of portrait production was institutionalized, undertaken by workshops, both private and official. Many craftsmen would have been involved in any one painting. Indeed, as Jan Stuart has discussed, at many non-court workshops, a generalized outline would be prepared before any portrait had been ordered. When a painting was required by a family, the workshop would have taken instructions as to the age, rank, and suitable dress of the proposed subject.8 Painting of these different details, such as the face, or embroidery on the dress, or headgear, was almost certainly carried out by several specialized artisans. The same working procedure would have been true, likewise, in the court workshops. The paintings are, indeed, skillful pieces of work that, in Gell’s terms, might be expected to ‘‘enchant’’ as a consequence of their technical virtuosity. But agency is highly unlikely to have been imputed to an artist or a group of artists by the viewers. To the officials concerned with commissioning appropriate formal or ritual portraits of the emperors and their relatives, it is almost certain that correct presentation according to set practice would have been of overwhelming importance. For the family members and ritual specialists, who would have seen these paintings on set occasions, the effects of the painting would have been interpreted as making the person pictured, here the Emperor, present. None of the viewers would have given the workmen any thought at all. If the formulae developed by Gell are appropriate in this instance, we can set up several alternative patterns. Were we first to look at the painting in Western terms, as having a subject being painted by an artist or artists, we would arrive at the following: the Emperor, the subject, would be, in

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 99 12.1.2007 6:10pm Compositor Name: SJoearun

The Agency of and for the Wanli Emperor 99 Gell’s terms, a prototype, exerting agency on the artists, who created the painting as an index, which in turn has agency on the viewers. This is the formula: Prototype A ! Artist A ! Index A ! Patient P

and not the alternative Artist A ! Prototype A ! Index A ! Patient P,

where the artist is the primary agent.9 However, a different formula might be more appropriate in the Chinese context, where the painting, as Index A, exerts agency, so that the ancestor, here the Wanli Emperor, as Prototype A exerts agency directly on the descendents as recipients or Patients, to be shown in the sequence Index A ! Prototype A ! Descendent/Patient P

But the agency is reciprocal, as mentioned above. The Index A allows the Descendent P to exert agency on the Prototype by making offerings to him. This would give the formula Descendent P ! Index A ! Prototype A

While the formulae are rather deadening in a discussion like the present one, they do enable us to ‘‘map’’ many different forms of effect and efficacy that a work like a Chinese ritual portrait might achieve. But the different formulae also reveal that the notion of what constitutes agency in Gell’s framework shifts according to the topic in question. In the West, where artists, patrons, and subjects have distinct and known identities, at least from the Renaissance period, it makes some sense to talk about the artist having agency. It is also the case that the viewers might, from that time, have been conscious of the role of the artist. But in the case of a Chinese ritual portrait, we have a very different context. The artists have no role in the minds of the viewer. The subject, the Emperor, on the other hand was thought to have had direct agency, much more effectively than the subjectmatter of many Western paintings of the Renaissance period and later.10 These differences arise as a consequence of cultural assumptions of peoples in the two regions, however reluctant Gell may have been to examine cultural difference in his work on agency.

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 100

12.1.2007 6:10pm Compositor Name: SJoearun

100 Jessica Rawson Gell would have found Chinese ritual portraits very congenial topics in his account. They were certainly never viewed as works of art within the Chinese canon, which was limited to calligraphy and painting categories derived from calligraphy. Westerners have often included such portraits in the Western category of art, presenting them in art exhibitions for example. Such portraits were made for particular functions in specific contexts. They had social roles indeed and would no doubt have been the sort of work to interest Gell greatly. For in effect he replaces discussions of works of art based upon aesthetic qualities with a socially grounded approach. He inserts the work of art in a relation with the viewer, as though it was a member of that viewers’ society, as indeed the Wanli Emperor would have been deemed to be. The social role of the Emperor was mediated by his portrait. For the portrait made possible a form of engagement between the descendents and their ancestor, the Emperor. In addition, as mentioned, the Emperor was deemed to be actually present, especially when the descendents viewed the portrait. While in practice the descendents had given the portrait its power, by expecting it to be efficacious, they had turned the situation on its head, assuming that the efficacy of the Emperor had given the portrait power.11 In other words, the different interpretations of where the agency lies depend intimately on the nature of the viewers’ perception of cause and causality, as the works of Pascal Boyer indicate.12 We shall return to this issue below when discussing the motifs of the carpet in the painting. In foregrounding the social roles of works, Gell was allowing the aesthetic to serve the social. At the same time he dismisses other forms of analysis, including concerns with culture, symbolism, and iconography. Further, Gell discounts analogies with language. In their place Gell elevates the role of technology and its capacities to enchant. This forms a vital component of his discussion. The intricacy of all workmanship, Gell argues, resists interpretation and thus allows viewers to assume that the works were produced by or had magical properties. He regards the so-called magical properties attributed to the canoe-prows of the Trobriand Islanders as a transcription into magical language of the sense that the ways in which their extraordinary forms and colors were produced cannot be readily understood.13 The physical character of the actual painting is indeed important. Gell’s fear of traditional Western art history approaches and terminology would have led him to reject the idea that the painting had any sort of aesthetic effect. Yet how can we see the bright colors of the robes and carpet as

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 101

12.1.2007 6:10pm Compositor Name: SJoearun

The Agency of and for the Wanli Emperor 101 anything other than intended to excite visual perceptions and our emotional responses? These are what we call aesthetic interests. Gell avoids the obvious interest in form, pattern, line, and color by turning to technology. Had he considered the painting under discussion, he might have pointed to the extraordinary qualities of the execution that entrance the viewer and were intended to enthrall the descendents and reinforce their beliefs about the powers of the dead Emperor on their lives. However, it would not have been to the visual sensations but to the qualities of execution that Gell would have pointed. For in the words of Nicholas Thomas, his argument is that ‘‘art objects mediate a technology to achieve certain ends, notably to enmesh patients in relations and intentionalities sought and prescribed by agents.’’14 However, the Chinese elite who viewed such portraits were quite familiar with the art of painting and above all with writing, that is, with calligraphy. As all emperors were calligraphers and some could also paint, it seems unlikely that the ways in which a painting was made were unknown to them. But they simply had no direct involvement with the offices that did such work. The workshops that produced ritual portraits for the Chinese court were part of the complex bureaucracy of that court and must have been known, at least to some parts of it. Indeed in all societies, craftsmen and artisans, even those engaged on secret religious activities, were unlikely to have been completely unknown to their peers, except in some exceptional cases. Furthermore, in most societies, the arts and crafts most widely employed developed often fortuitously, but then survived for long periods, their features and qualities becoming well known at least to the major figures in that society.15 Seen in this light, Gell’s emphasis on the technology may need some adjustment. It seems that Gell’s preoccupation is simply his way around the difficult question, Why do we find engaging skilled work in almost any medium that excites our visual perceptions? Many would indeed say that fine works in almost any material baffle those who are not members of the group who practice such arts. But only in some societies do people explicitly consider the production of works. And we might turn to another form of explanation. Perceptual psychologists would probably respond by suggesting that we are constantly seeking information from the world around us, as Semir Zeki has argued.16 The richer the information and the more readily interpreted, the more we pay attention to it. Complexity of material, color, line, and form simply offers us a richer information source than less complex variants. Gell’s emphasis on technology seems

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 102

12.1.2007 6:10pm Compositor Name: SJoearun

102 Jessica Rawson slightly arbitrary here. He could just as well have approached this issue through the perceptual and cognitive sciences. But we might see this interest in technology as a product of the time in which Gell was writing and of the areas which he studied. In the 1970s to the 1990s, the technology of the production of artifacts was a focus for scholars and students of material culture. Ceramic glazes, casting-molds for bronzes, and sculpture techniques seemed firmer bases for discussion than the subjective assessment of authenticity and beauty about which disagreement was inevitable.17 China is in fact a long way removed from the societies, such as the Maori, that Gell studied. In China, the workshops producing ceramics or paintings were very large with much subdivision of labor. They were highly controlled and in many cases commoditized. Whereas in the Pacific the products of very skilled workman were likely to have been interpreted as having magical powers, the products of Chinese enterprises probably were not. Therefore, an emphasis on technology in a Chinese context is in effect an emphasis on the skills available to a workshop in an environment in which objects and paintings were produced to a very high technical level on a large scale. Whether it is the enchantment of the technology or the dazzling qualities of the colors, the visual aspects of the work engage the viewers and influence their reactions. What these viewers infer or abduct from these engagements is a complex issue, as I suggest below. To get away from the visual qualities, Gell indeed focuses on abduction. In his discourse, abduction refers to inferences about the causes that bring about the visual engagement. A concern with abduction is a concern with causality. Here we approach Semir Zeki’s account of the search for information and explanation. If people explain works in terms of supernatural causes, which is what Gell argues, then they also use the same artifacts and representations to supply information about the supernatural. This is another way to express the preoccupation that lies at the heart of Gell’s account, the human tendency, as we search for information, to impute cause to what we see, to believe that what we see must be directed at us for some purpose. In the case of the Wanli portrait, it would be more appropriate to say that the viewers, rather than searching for causes, had certain expectations about the portrait, that offerings to the Emperor present in his portrait would provide benefits. What Gell does not do in Art and Agency is explore, in any theoretical way, the contexts in which such expectations or abductions develop. Indeed, because he wishes to leave aside cultural conditions as part of

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 103

12.1.2007 6:10pm Compositor Name: SJoearun

The Agency of and for the Wanli Emperor 103 his equations, Gell is not concerned with what seems to me the nub of the matter. For in the case of the Wanli portrait, the expectations of the viewers depended on a set of beliefs about the family and the role of the ancestors within the family, a set of beliefs about nourishing the dead, and a set of beliefs about the Emperor’s position in the universe. Insofar as these sets of ideas embraced the social and institutional structures within which the portrait functioned rather than just a literal account of symbolism and iconography, Gell could not have been dissatisfied with them. His primary concern was, indeed, to remove the consideration of artifacts from discussions dependent on particular cultures and to seek an overarching theory in social interactions that were more fundamental than these individual manifestations. Had Gell known about Chinese ancestor portraits he would probably have emphasized the social relations between the Emperor and his descendents, or his subjects, rather than the notions about life after death and the transmission of food to the dead that I shall argue provided the cement in this set of ideas. He might have suggested that the latter two ideas were independent of the social relations. Yet the production and reception of the painting were completely dependent, not only on the social institutions, but on these closely interlinked beliefs. And a further set of beliefs was in play, that concerning the decorative motifs on the robes and the carpet. The dragons, the cosmic emblems, and the patterns of clouds and precious treasures were deemed to have auspicious effects. This was and is a distinctly Chinese aesthetic dimension of agency. For the Chinese have sustained to this day the notion that decorative motifs, most especially of creatures and flowers, but also of many other items, ensure good fortune for those who wear clothes so decorated and live in buildings in which these designs are reiterated on architecture, furniture, and utensils. From at least the Han period (206 BC–AD 220), motifs such as dragons and written characters, embellishing dress, artifacts, and buildings were expected to have agency and to ensure for the inhabitants a good outcome. This notion illustrates Gell’s efficacy very directly. We can explore the origins of these ideas in an account by the skeptic thinker, Wang Chong (second century AD), of the practice of setting up dragons to bring rain. At all times in China, dragons were linked with clouds and rain, a benign fertilizing force. Dragons could be expected to appear and to be seen when the rains came. Conversely if rain was desired, clay dragons would bring the rain, for the two necessarily appeared together. Now Wang Chong was, indeed, dismissive of many religious beliefs and practices, but he did

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 104

12.1.2007 6:10pm Compositor Name: SJoearun

104 Jessica Rawson consider that images of dragons, strategically placed, were likely to be efficacious and cause rain to fall. 18 An inversion had occurred; a consequence has been turned into a cause. The rain brings the dragons, so that dragons and pictures of dragons have agency and cause the rain. In the same way, when the emperor enjoyed good fortune, strange colored birds, or tall ripe grains, would appear spontaneously. Like the clay dragons, propitious motifs and emblems on the robes of the emperor had agency to ensure that he too would enjoy the benefits of good fortune in the future. Thus an alternative formula can be set out as: the Index A, the robe, has agency on the Prototype A, the emperor, who in consequence becomes a Prototype to exert agency on those who commission portraits and those who exploit them in offerings to the ancestors, that is, the descendents again. We can consider a further dimension by looking at the contents of the Wanli Emperor’s tomb. The Emperor was buried in the Imperial tomb complex north of the capital in 1621.19 With him were buried his Empress, Xiaoduan, and the mother of the Wanli Emperor’s first son, known after her death as the Empress Xiaojing, previously entitled the Honorable Imperial Concubine. The Emperor and the Empress Xiaoduan died within a short time of each other in 1620. The Honorable Imperial Concubine, mother of Zhu Changluo, the Emperor Guangzong, whose reign title was Taichang, had died in 1611. It was the Taichang Emperor who ensured that his mother was recognized as an empress and that she too was buried in the tomb with the Wanli Emperor.20 This grouping in the tomb gave formal recognition to the precedence accorded by tradition to an empress and to the mother of the heir. Not recognized and not buried with the Emperor were his other favorite concubines, including Lady Zheng, with whom he had established a long and close relationship.21 The tomb thus did not reflect in any way the personal inclinations of the Emperor. It was intended to present in a formal manner the institutional position of the Emperor and those of the two most important women at court in light of the succession to the throne.22 In keeping with the full institutional arrangement of these burials were the clothes interred with the Emperor. The tomb contained a remarkable set of robes for formal court rites and for general court occasions. These were matched by belts for the Emperor and by the headdresses for the Emperor and the two women and a large range of jade and jewel-encrusted hairpins. It might, perhaps, be suggested that these were simply the finery in which the Emperor and the women were buried. But the funeral alone could not have required 61 dragon robes!

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 105

12.1.2007 6:10pm Compositor Name: SJoearun

The Agency of and for the Wanli Emperor 105

Figure 4.2 Yellow robe embroidered with dragons from the tomb of the Wanli Emperor (1621).

The tomb contents were much more interesting than seemed at first sight. The vast underground chambers contained large chests in which were not just multiple sets of robes for the occupants, but 177 bolts of cloth, most of extremely fine weaves. These bolts were in some cases woven with the exact designs required for formal robes, but had not yet been cut up and sewn into their final forms. It would appear that, in the afterlife, the Emperor was expected to wear a whole set of different robes appropriate for the court occasions and for the sacrificial rituals that he might attend. In addition, to provide for the longer term, fresh robes could be made from the bolts thus provided.23 A yellow robe embroidered with dragons and ancient emblems similar to those seen in the portrait stands out (figure 4.2). It is one of five in the tomb of the gunfu type seen in the painting. These were intended for major ritual occasions such as the offerings to Heaven and to the Earth. Like the robe in the painting, the gunfu in the tomb carried dragons in roundels and the twelve ancient emblems. In both cases, the robes effected a transformation, fitting the Emperor not just for court life but for specific rituals. Transformation is, indeed, a major contribution of all dress, artifacts, and building structures. Here, they turned the naked, dead man into an Emperor, an empty hall into a court reception room,

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 106

12.1.2007 6:10pm Compositor Name: SJoearun

106 Jessica Rawson and an underground shaft and chambers in the earth into a palace for the afterlife. In the case of the robe, the colors, the dragons, and the emblems were essential to effect this transformation and were deemed necessary to the efficacy of the Emperor in both life and death. On some occasions in life, the Emperor wore more informal clothes, and these too were included in the tomb, as were the robes and other clothes for the women. If we summarize the discussion of the forms of agency at work in Gell’s terms, we might reach the following description. Here, the robe is the Index A, and the Emperor wearing the robe the Agent, with agency on a viewer or Patient P. In the case of the offerings to Heaven and to the Earth, among the patients would have been deities and spirits. In this instance, the strong aesthetic characteristics, terms with which I would replace Gell’s interest in technologies of production, mediated not between an artisan and a viewer, but between the Emperor and his court and the Emperor and the spirits. And these visual characteristics made possible the Emperor’s agency on those who participated in these events. For this reason, there are in effect two agents, the Emperor as empowered by his robe and the robe as an agent working on the Emperor as patient. The tomb is a very special environment and raises further interesting issues about the effectiveness of artifacts of all kinds in human society. The tomb is in modern terms simply a large vault with several rooms in which three seventeenth-century human beings were laid to rest. They were interred with this wonderful wardrobe and a whole series of artifacts. These included fine jades and porcelains and also small replicas in tin. An entourage of several hundred wooden tomb figures and mounted horsemen supported and provided for the three occupants of the tomb. Chinese Imperial and elite burials have, from the time of the First Emperor (221–210 BC) and earlier, comprised a similar mixture of items used in the daily lives of the tomb occupants and of models. The models have sometimes led commentators to suggest that the Chinese did not really believe that the tomb provided a full afterlife for the tomb occupant. On the other hand, the clothes in the Wanli Emperor’s tomb suggest quite the reverse. The 61 dragon robes would appear to provide for a full afterlife, while the bolts of cloth ready woven were seemingly to assure the Emperor of a future suitable wardrobe. The robes were specifically decorated and cut for designated official roles and occasions. To ensure proper ceremonies in the afterlife, it may have seemed essential to put

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 107

12.1.2007 6:10pm Compositor Name: SJoearun

The Agency of and for the Wanli Emperor 107 them in the tomb. On the other hand, the servants were simply generic and would, if properly numbered, apparently have been available to serve the Emperor after death. And the same contrast is common in many of the most elite tombs of all periods. In such other tombs, it is the specific that is provided in fine materials and a high level of craftsmanship, while the everyday comes in the form of models. Some famous Classical authors, most especially the Warring States period philosopher, Xunzi, and the Han-dynasty skeptic Wang Chong, mentioned above, described at some length inconsistencies, as they saw them, in burial practice.24 These led them and their followers to dispute the efficacy of fine burials. And some later writers have followed them. But contrary to such views, it would appear from the abundant material evidence that most Chinese of all periods took the view that life after death was a continuation of this life. To ensure that the afterlife was properly provided for, tombs were equipped with the needs of daily life, either as models and as pictures or in the form of actual utensils. Clothes too have survived from quite a number of major tombs excavated in the last few decades.25 What is more, the use of portraits described above as the focus for rituals of offering food for the dead suggests, indeed, a continuity between life and the afterlife that allowed for the continuing presence of the dead through the proper provision of tombs and the observation of offering rituals with ritual portraits. A rather special form of agency could be deemed to be at work here. Tombs, whether equipped with ‘‘real’’ objects or models were in themselves models of life that were deemed to make possible an afterlife. They had in effect the power, or, in Gell’s terms, the agency, to provide the tomb owner with a further and different, yet similar, life to the one enjoyed in what we call life. The agency of the tomb is a consequence of the engagement neither with technology nor with the primarily aesthetic characteristics of artifacts or works of art. Rather, this is an example of the power of artifacts themselves. This power seems to exist in several spheres. First of all the artifacts provide a way of living. The robes are those that enable the Emperor both to be clothed and keep warm in everyday life and, most importantly, to carry out certain ceremonies. The vessels enable certain types of meal to be prepared. They allow for some activities and by extension deny the possibility of others. In other words, they determine actions and processes. In this respect the artifacts directly affect, indeed determine, some human action.26 Secondly, these artifacts ensure a good outcome. That is, the

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 108

12.1.2007 6:10pm Compositor Name: SJoearun

108 Jessica Rawson motifs on the robes and the decoration of the porcelains were intended to be auspicious. For instance, one of the women’s jackets carries a design of children, a decorative type intended to ensure many descendents. Presumably the efficacy of such decoration was to be present both in life and in the afterlife. In this arena, the artifacts affect the workings of the universe as a whole. Thirdly, the artifacts allow a difficult and inaccessible area, the life after death, to be thought about and provided for. They are in effect a tool for thinking about and affecting the unknown. But the artifacts in such a tomb cannot alone and without a complex intellectual framework make possible, be agents for, such a complex range of outcomes. For their efficacy depends critically, as we have already observed, on the expectations of the viewers. If we are to make progress in understanding the ritual portrait and the tomb with its abundance of fine robes, it would be helpful to consider a more complex model of the world than the unitary one provided in Gell’s work. Simply to insert works of art, as Gell seems to do, into a discourse that encompasses not only everyday society, but also its complex religious and political institutions is, in my view, to risk confusing categories of agency. Instead, I wish to propose three levels of the world, or three worlds, each with the same three components: a physical presence, a social structure, and a set of beliefs. It is within this complex structure, I suggest, that it is possible to elucidate the sources of belief that gave the portrait agency. Level one is what we might call the everyday world. The everyday life of any individual in any society has a physical side, in which the person lives in a structure and consumes certain foods and wears certain clothing. That person inevitably also has a social context, a group of people with whom he spends his time and a set of beliefs current in an everyday context. The Wanli Emperor could, perhaps, be seen as having a daily life in his palace, which we now call the Forbidden City, and wearing the sorts of informal robe included in the tomb. As an individual, rather than an Emperor, his social relations included in particular Lady Zheng. In addition, his beliefs included both simple ones about day-to-day existence and notions of auspiciousness and, on the periphery from the point of view of the everyday, more complex ones about his role in the state, his ancestors, life after death, and the universe. While people tend to regard their everyday lives as ‘‘natural’’ they are every bit as artificial and constructed as the institutions that govern this everyday.27 In any complex society, and possibly in all societies, there is an institutional world, that is, level two. This level, likewise, can be seen as having

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 109

12.1.2007 6:10pm Compositor Name: SJoearun

The Agency of and for the Wanli Emperor 109 three broad components. First of all the institution has a physical presence with its organizational roles and people filling those roles – emperors, ministers, magistrates. It also has a social structure by which the roles are related to one another and to the population of the society. In addition, the institution is underpinned by a set of beliefs. In this instance, we have the Wanli Emperor filling a major institutional role, and his Empress and the mother of the heir to the throne likewise filling set roles. In this institutional setting, the Emperor had another set of social relations with ministers for example, rather different from those of an individual living a daily life. He also had a set of beliefs, shared by all in the court and many in Ming society, about the central relationship of the Emperor to the universe as a whole. When in life the Emperor was seen in court audiences wearing the ceremonial robes described above, he embodied these beliefs for the courtiers.28 The auspicious motifs on the robes were a physical expression of these beliefs also. The third level is the one that has concerned us most here. I would term this level the conceptual world. To this one belong the rituals and the tombs of the Chinese, ancient and more recent. The ritual portrait and the Wanli Emperor’s tomb are of this level also. This conceptual world is the one central to what we have come, with hindsight, to call art. For to it belong many physical manifestations of very complex concepts in religious buildings, in paintings, and in performances. For many centuries, and in most corners of the globe, people made artifacts with a high degree of skill primarily for religious or political functions, many of them with reference to the supernatural. Within this level too are social relationships of those who participate and those who do not. And above all, the paintings and the tombs we have discussed are underpinned by beliefs. These beliefs include the notions of the efficacy of offerings to the ancestors and of the possibility that a tomb will provide an afterlife for an emperor. I would, therefore, go as far as to say that the third world, like the previous two, has a physical manifestation, a set of social relations, and a set of beliefs. It is of course artificial to separate the discussion into three levels. All exist simultaneously and all interact in the minds and perceptions of people, such as the members of the Ming court, who would have been concerned with the production of the ritual portrait and the construction of the tomb. All the worlds, and all the sets of beliefs, would have influenced the expectations of members of the Ming court and determined the ways in which they responded to the portrait of the Wanli Emperor and to his tomb.

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 110

12.1.2007 6:10pm Compositor Name: SJoearun

110 Jessica Rawson However, only if we can momentarily see the three worlds as separate levels of activity within a society, in which the same participating people may have rather different concerns, can we identify the different types of agency at work. In suggesting that works of art have agency, Gell collapsed these levels into one. Technology and manufacture, which Gell argues are important instruments with which to enmesh the viewer, or patient, belong to the everyday world, level one. In some societies, particularly those studied by Gell, the ideological beliefs permeated such activity and linked them inextricably to level three. Canoes embellished by fine carving were given status as war canoes and indeed even magical powers in level two, in the institution of warfare. The makers would have receded in importance. But in China the makers of ceramics or of ritual portraits would have been less readily seen as having magical powers. There was nothing mysterious about the commissioning or painting of a ritual portrait of the Wanli Emperor. However, when a portrait was viewed at a ceremony of offerings to the ancestors, the viewer was no longer merely in the institutional world, but had shifted into the conceptual world, where the relationship between the viewer and the ancestor, the belief in the value of filial piety shown in offerings, and a belief in the presence of the ancestor were all-important. At that moment, the ways in which the portrait was produced had likewise receded. Thus the agency of the painting was sustained as much by a set of beliefs as by correct application of paint in a skilled and accurate manner. Belief systems were also behind the use of decorative motifs on the carpet, on the actual robes in the tomb, and on the painted robe. The beliefs that concerned the use of auspicious decoration, such as those on the carpet, to ensure a good outcome could be seen as everyday beliefs, though ones that had, perhaps, migrated over time from the institutional sphere into the everyday, even the popular, world. However, the particular use of dragons and antique motifs on the actual robes belonged, of course, to the institutional world in which the emperor exercised his state powers. But when these robes were painted in a ritual portrait and deployed in the tomb, quite different beliefs were at work, in these cases beliefs about the life after death. The agency of the carpet, of the robes used in court life, and of the robes in the tomb depended, therefore, on somewhat different beliefs. Indeed, it would be possible to argue that in the third conceptual world, all three sets of beliefs were involved. Of course the degree to which these beliefs were significant to the viewers varied from person to person and from rank to rank. They could be expanded or curtailed, blended or separated, as the experiences of the persons dictated.

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 111

12.1.2007 6:10pm Compositor Name: SJoearun

The Agency of and for the Wanli Emperor 111 Gell did not take this sort of approach, in part because he was, as he claimed, uncomfortable with iconography and symbolism. He would in all probability, if asked, have played down the role of the dragon motifs on the robes. Yet these are not simple isolated examples of symbols. They belong to a large set of decorative motifs that formed a system held together by the belief that if certain creatures, such as dragons, occur at auspicious moments, these auspicious circumstances can be achieved by designing robes with dragons on them. This system of motifs was not a dictionary of iconography but the material presence of a set of beliefs fundamental to Chinese peoples over many centuries. What is missing in Gell’s approach is an understanding that systems of symbols or of iconography, and even traditional methods of painting and carving, are not isolated systems. They are integrated with, and are maintained in use by, complex and, usually unquestioned, quite other systems of both practice and belief. While Gell’s emphasis on efficacy must imply some recognition of intellectual backgrounds, he gives very little space to their consideration. However, if we are to understand the powers of artifacts and paintings to engage us and our predecessors in different societies, we must recognize that it was the belief systems that enthralled viewers and provided them with the expectations that enabled them to see an emperor in a painted surface and a palace in a tomb. The skills of the painters and carvers were tools to bring those beliefs to the forefront of the viewers’ minds. Gell was correct in his use of the term ‘‘enchant,’’ but he left out some essential elements that contributed to that enchantment. The skill he so much admired and documented could only work its magic if sustained by a system of beliefs as well as by a society that held to those concepts intuitively and without question. Notes 1

2 3

For the excavation report see ‘‘Zhongguo shehui kexueyuan kaogu yanjiusuo, Ding Ling bowuguan, Beijing shi wenwu gongzuodui,’’ Ding Ling, 2 vols. (Beijing, 1990). J. Volmer, Ruling from the Dragon Throne: Costume of the Qing Dynasty (1644–1911) (Berkeley, CA, and Toronto, 2002), p. 112. A variety of motifs, such as ling zhi fungus, ingots and chime stones were regarded as auspicious emblems. They could be combined either in groups of eight or in different combinations; see Wang Qingzheng, A Dictionary of Chinese Ceramics (Singapore, 2002), p. 254.

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 112

12.1.2007 6:10pm Compositor Name: SJoearun

112 Jessica Rawson 4

5

6

7

8 9 10

11

12

13 14 15

16 17

For a range of different aspects of such ritual portraits, see J. Stuart and E. S. Rawski, Worshipping the Ancestors: Chinese Commemorative Portraits (Washington, DC, 2001). C. Clunas, ‘‘ ‘Not one hair different . . .’: Wen Zhengming on imaging the dead in Ming funerary portraiture,’’ in R. Shepherd and R. Maniura, eds., Presence: The Inherence of the Prototype within Images and Other Objects (Aldershot, 2006), pp. 31–46. For an account of the use of portraits in a variety of religious uses during the Song (960–1279) see P. Ebrey, ‘‘Portrait sculpture in Imperial ancestral rites in Song China,’’ T’oung Pao 83 (1997), pp. 42–92. M. Siggstedt, ‘‘Forms of fate: an investigation of the relationship between formal portraiture, especially ancestral portraits, and physiognomy (xingshu) in China,’’ in International Colloquium on Chinese Art History (1991), 2 vols. (Taipei, 1991), vol. 2, pp. 717–48; Clunas, ‘‘ ‘Not one hair different . . .’.’’ Stuart and Rawski, Worshipping the Ancestors, pp. 93–115. A. Gell, Art and Agency: An Anthropological Theory (Oxford, 1998), ch. 4. The principal discussion of the agency imputed to or abducted from Western paintings and sculptures is found in D. Freedberg, The Power of Images: Studies in the History and Theory of Response (Chicago, IL, 1989). However, this way of using the word agency is particular to Gell and differs somewhat from other uses in which the notion of intentionality plays a larger role; see discussions in A. Gardner, ed., Agency Uncovered: Archaeological Perspectives on Social Agency, Power and Being Human (London, 2004), pp. 1–15. P. Boyer, The Naturalness of Religious Ideas: a Cognitive Theory of Religion (Berkeley, CA, 1994); P. Boyer, Religion Explained: the Human Instincts that Fashion Gods, Spirits and Ancestors (London, 2001). Gell, Art and Agency, p. 17. C. Pinney and N. Thomas, eds., Beyond Aesthetics: Art and the Technologies of Enchantment (Oxford, 2001), p. 5. See for example the paintings and carvings of the indigenous peoples in the north-west of the North American continent; B. McLennan and K. Duffek, The Transforming Image: Painted Arts of Northwest Coast First Nations (Vancouver, Toronto, and Seattle, 2000). Semir Zeki, ‘‘Art and the brain,’’ Journal of Consciousness Studies, Controversies in Sciences and the Humanities 6, June/July (1999), pp. 76–96. Research in Chinese clays and glazes first came to prominence in the First International Conference on Ancient Pottery and Porcelain (ICAPP) held at Shanghai in November 1982. For bronze casting-molds the extraordinary production of the foundry at Houma is an exemplar and was discussed by B. Keyser, ‘‘Decor replication in two late Chou chien,’’ Ars Orientalis 11 (1979), pp. 127–62, and R. Bagley and Institute of Archaeology of Shanxi

Osborne / Art’s Agency and Art History 1405135379_4_004 Final Proof

page 113

12.1.2007 6:10pm Compositor Name: SJoearun

The Agency of and for the Wanli Emperor 113

18

19 20 21 22 23 24

25 26

27

28

Province, eds., The Art of the Houma Foundry (Princeton, NJ, 1996). For sculptural techniques in the West see J. Montague, Roman Baroque Sculpture: the Industry of Art (New Haven, CT, 1989). The discussion about the efficacy of clay models of dragons in producing the benefit correlated with dragons, namely rain, goes back to the Han dynasty (206 BC–AD 220), see M. Loewe, Divination, Mythology and Monarchy in Han China (Cambridge, 1994), ch. 7, ‘‘The cult of the dragon and the invocation for rain.’’ Qianshen Bai has argued that the use of motifs on roof tiles to ensure auspicious outcomes goes back to the late Warring States period, 4th–3rd centuries BC; see Qianshen Bai, ‘‘Image as word: a study of rebus play in Song painting (960–1279),’’ Metropolitan Museum Journal 34 (1999), pp. 57–72. See Ding Ling. Ding Ling, pp. 6–8. Ray Huang, 1587, A Year of No Significance: The Ming Dynasty in Decline (New Haven, CT, 1981), pp. 30–3. Huang, 1587, A Year of No Significance, pp. 125–9. Ding Ling, pp. 43–150; Wang Yan, Wanli dihoude yichu (Taipei, 1996). J. Knoblock (trans.), Xunzi: a Translation and Study of the Complete Works, 3 vols. (Stanford, CA, 1988); A. Forke (trans. and annotated), Lun-heng Philosophical Essays of Wang Ch’ung, 2nd ed. (New York, 1962). See especially Jiangxi sheng wenwu kaogu yanjiusuo, ‘‘Nanchang Mingdai Ningjing Wang furen Wushimu fajue jianbao,’’ Wenwu 2 (2003), pp. 19–34. For a discussion of the ways in which the ancient Chinese ritual bronzes could be thought to control or determine action, see J. Rawson, ‘‘Late Shang bronze design: meaning and purpose,’’ in Roderick Whitfield, ed., The Problem of Meaning in Early Chinese Ritual Bronzes, Colloquies on Art and Archaeology in Asia, 15 (London, 1992), pp. 67–95. The same argument is illustrated more extensively with reference to a longer time span in J. Rawson, ‘‘Chinese burial patterns: sources of information on thought and belief,’’ in C. Renfrew and C. Scarre, eds., Cognition and Material Culture: the Archaeology of Symbolic Culture (Cambridge, 1998), pp. 107–33. For the constructure of culture see Philip Chase, ‘‘Symbolism as reference and symbolism as culture,’’ in R. Dunbar, C. Knight, and C. Power, eds., The Evolution of Culture: an Interdisciplinary View (Edinburgh, 1999), pp. 34–49. The complex and almost contradictory effects of an ordinary person wearing robes and sometimes masks that transform them into a deity or ruler has been described as an embodied paradox: M. Blackmun Visona, R. Poynor, H. M. Cole, and M. D. Harris, A History of Art in Africa (London, 2000), pp. 336–7.

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 114

12.1.2007 6:12pm Compositor Name: SJoearun

5

The Material Efficacy of the Elizabethan Jeweled Miniature: a Gellian Experiment Jessen Kelly

In the early seventeenth century, Nicholas Hilliard, goldsmith and formerly painter of portrait miniatures of Queen Elizabeth I, composed a treatise on the art of limning. Never published during his lifetime, this document comprised a kind of book of secrets, undoubtedly intended for a select circle of gentlemen who would protect knowledge that, as the author stipulated, was ‘‘not for comon mans vsse.’’1 For Hilliard, the miniature’s singular, elite status lay in its apparently mysterious origins. Limning ‘‘excelleth all other Painting what souer in sundry points, in giuing the true lustur to pearl and precious stone, and worketh the metals Gold and Siluer with themeselfes, which so enricheth and innobleth the worke that it seemth to be the thinge it selfe, euen the work of god and not of man.’’2 The miniature, according to this assessment, poses questions regarding the agency responsible for its making.3 Although limning was a figurative art devoted to portraiture, Hilliard here makes no mention of content. Instead, the miniature’s rather cryptic roots result from its physical composition, its ability to incorporate, shape, and suggest precious materials and their properties – the ‘‘lustur’’ of pearls and the actual application of valuable substances. The treatise maintains this emphasis. To be sure, Hilliard provides theoretical ruminations on portraiture, but

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 115

12.1.2007 6:12pm Compositor Name: SJoearun

The Elizabethan Jeweled Miniature 115 he is primarily concerned with the stuff of his craft and the technical knowledge it entails: the mixing of pigments, the perfection of certain hues, and the relation of colors to precious gems.4 Seen through Hilliard’s remarks, the miniature’s ambiguous origins constitute an instance of what Alfred Gell defines as captivation. For Gell, captivation is a mode of response that stems from an artwork’s capacity to mask its inception through spectacular technical virtuosity. Viewers cannot envisage the process of the object’s manufacture from its immediate appearance, and thus they are unable to imagine creating such an object themselves. The marvel induced by this illegibility creates a crucial inequity between spectator and object; the work ‘‘traps’’ its viewer in a manner that ultimately contributes to the reproduction of social hierarchies.5 Captivation is, in this respect, an instrument for the exercise of power. As Robert Layton has noted, however, the notion of captivation calls attention to the significance of power relations in the Gellian anthropology of art more broadly.6 This anthropology stresses the agentive dimensions of art objects as mediators of social relations rather than their communicative or symbolic import. Art objects are indexes that enable the abduction of prior social agency while initiating still further activity on the part of users (recipients) and/or makers (artists). The abduction of cause or intentionality from the index continually reconfigures social relationships in the Gellian ‘‘art nexus,’’ where artists and recipients variously occupy the position of agent or patient.7 Captivation, in halting the activity of abduction, places the viewer in the patient position. The power disparities at stake in captivation indicate the extent to which this position may be identified with subordination, and the agency of the work (and its maker) with authority. In this paper, I use Gell’s framework to examine the jeweled miniature’s mediation of power relations between Queen Elizabeth and her privileged subjects at court. If Hilliard’s work could indeed captivate viewers, then it is hardly surprising that portrait miniatures constituted an important medium for royal representation.8 Indeed, Queen Elizabeth seems to have surpassed Hilliard’s other patrons both in the number of miniatures commissioned and in their degree of embellishment: though the luster of the miniature was sought by other customers, the queen consistently extended its scope by having her image set in jeweled cases that opened to reveal the royal likeness. These objects were typically presented as gifts from the queen to her most esteemed courtiers and counselors, a practice

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 116

12.1.2007 6:12pm Compositor Name: SJoearun

116 Jessen Kelly that integrated the jeweled miniature into the social relations of a gift economy in which the queen was afforded a unique position as sovereign. While Marcel Mauss’s model of generosity and obligation, when mapped onto the Gellian art nexus, suggests a changeable hierarchy of agent-givers and patient-receivers, in theory the queen’s power was such that no gift could raise its giver to a status of advantage over or even equity with her.9 The presentation of Elizabeth as a ‘‘prime mover’’ and ‘‘fountain of grace’’ rendered her a source of largesse to which her subjects were ceaselessly obligated as patients.10 The aim of the jeweled miniature as captivating gift, then, would seem to be to replicate the eternal indebtedness of subjects and, with it, their subsidiary status within the social order. Given the threats faced by the Elizabethan government, these gifts could serve a crucial unifying and disciplining role. Roy Strong has suggested that the jeweled miniature became fashionable in the mid-1580s, precisely when alleged international and domestic plots against Elizabeth appeared to multiply.11 In this interpretation, Hilliard’s royal portraits are affirmations of loyalty that efface the queen’s political and physical imperilment by evidencing her ability to command the alliance of her subjects. Strong’s appraisal of the jeweled miniature, which is continuous with his notion of a propagandistic ‘‘cult of Elizabeth’’ constructed through cultural representation, demonstrates the affinities between some of the previous scholarship on these objects and the Gellian idea of captivation.12 In employing certain terms and concepts of Gell’s anthropology of art, however, I want to suggest a different picture of the jeweled miniature’s social agency. Taking Hilliard’s lead, I privilege the material composition and effects of such items as their most vital characteristics.13 I argue that the technical elements of Hilliard’s practice can be seen in relation to early modern proto-scientific discourses which highlighted the agency of matter and its potential applications. If Hilliard was able to extol the inscrutable origins of the miniature in terms of precious metals and stones, this was due at least in part to the problems posed by such entities in alchemy, medicine, and natural philosophy – fields of knowledge often associated with European court culture and whose close connections to early modern artistic practice have recently been examined by Pamela H. Smith.14 Viewed from this perspective, the properties of gems, metals, and pigments submits the social relations of the royal gift to instability in ways that do not guarantee an unproblematic reproduction of power within this uneasy political environment. Thus, while I value the ways in which Gellian captivation affords unique insight into the technical and social

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 117

12.1.2007 6:12pm Compositor Name: SJoearun

The Elizabethan Jeweled Miniature 117 efficacy of the miniature (particularly in a context in which art was not so strongly distinguished from other technical practices), I also question the extent of captivation’s relevance as a critical category. I focus my discussion on the most elaborate extant example from Hilliard’s oeuvre: the Heneage Jewel (sometimes called the Armada Jewel), traditionally held to have been given by Queen Elizabeth to her favored Privy Councilor, Sir Thomas Heneage, in the 1590s.15 As not much is known regarding the specific history of this work, I rely principally on the object itself, tracing its unfolding and multiple dimensions within Gell’s framework. Measuring 5  7 cm, the Heneage Jewel exemplifies the Elizabethan miniature’s tendency toward marvelously circumscribed richness in materials and artistry. The complexity of the work derives not only from the intricacy of its physical construction, but, as with other jewel-enclosed miniatures, the multiple facets on which its varied stuff and motifs have been inlaid. No surface is left vacant, and both interior and exterior, obverse and reverse, make references to the giver, Queen Elizabeth. The oval jewel case presents a profile portrait of the monarch embossed in gold, set against a background of brilliant blue enamel (figure 5.1). A small golden inscription in Roman capitals encircles this representation along the upper edge of the blue backdrop, declaring the scope of the queen’s power: ELIZABETHA D.G. ANG.FR.ET.HIB.REGINA (Elizabeth, by the grace of God, Queen of England, France, and Ireland). The image is in turn enclosed under a cover of transparent rock crystal, framed by a shaped band of gold that is enameled in white and light blue and encrusted with alternating diamonds and rubies. On the reverse, gold is the ground for enameled strata of sea and clouds in which a depiction of Noah’s Ark is set. The reverse, too, is framed by a Roman inscription, one that designates the scene as emblematic of Elizabeth’s guarding of the English Church (SAEVAS TRANQVILLA.PER.VNDAS – peaceful through the storm waves). In the jewel’s open state, its lid displays a painted image of a red rose contained within a pattern of intertwining vines and leaves (figure 5.2). The lid echoes the two exterior fac¸ades in that its visual elements are similarly encompassed by inscription, though here the phrase is rendered in calligraphic script whose lines reverberate in the curved vegetal motifs: Hei mihi quod tanto virtus perfusa decore non habet eternos inviolate dies (Alas that virtue suffused with so much beauty is not eternally inviolate). Yet the lid’s concentric configuration appears as but a caption to Hilliard’s lush limned portrait of the queen. Uncontained by the inscriptions that embellish the other facets of the jewel (the date of 1580 at left is spurious),

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 118

12.1.2007 6:12pm Compositor Name: SJoearun

118 Jessen Kelly

Figure 5.1 Nicholas Hilliard, the Heneage Jewel (exterior). Gold with white, black, opaque pale blue, green, red, blue, grey, orange, and yellow enamel, set with table-cut diamonds and rubies; c.1595. Courtesy of V&A Images, Victoria and Albert Museum, London.

the miniature gives free rein to the visual splendor of its construction. Elizabeth is portrayed against an indigo background that is virtually obscured by the profusion of ornamentation on her collar, itself laden with flowers and gems. In its delicate construction, this miniature seems to confirm Hilliard’s statement on limning’s miraculous qualities, harnessing these potentially captivating attributes in the service of the queen’s image. This image, however, was not encountered in the manner of a largescale portrait. Not destined for display on a wall, the jeweled miniature was an object of adornment to be worn on one’s person. Accordingly, the apex of Sir Thomas’s royal gift terminates in an ornamental loop suitable for a ribbon or pin for fixing the object at the belt, around the neck, or on clothing, a manner in which some of the queen’s favored courtiers were portrayed during the period.16 The incorporation of miniatures into more

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 119

12.1.2007 6:12pm Compositor Name: SJoearun

The Elizabethan Jeweled Miniature 119

Figure 5.2 Nicholas Hilliard, the Heneage Jewel (interior). Courtesy of V&A Images, Victoria and Albert Museum, London.

traditional forms of portraiture is indicative of their role in courtly selfpresentation: despite their scale, these items were intended to be seen. But portraits containing jeweled miniatures invariably show them in a closed state, suggesting that the conventions for such self-presentation involved accenting the actual stones and metals of the exterior rather than the painted luster of the miniature. In the case of the Heneage Jewel, the three-dimensionality of the queen’s embossed portrait implies that this surface was to be oriented outward, toward Sir Thomas’s interlocutors. The jeweled visage is thus the initial means by which the object mediated Heneage’s social relations, transposing the gentleman’s interactions into ‘‘the vicinity of art.’’ Visually, the queen’s presentation on the exterior interjects an image of imperial power between Heneage and those he encounters, her strict

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 120

12.1.2007 6:12pm Compositor Name: SJoearun

120 Jessen Kelly profile evoking official forms of state sanction. Indeed, the profile portrait is strongly reminiscent of the imprimatur of medals and seals. More specifically, the portrait nearly replicates that on a medal of the Order of the Garter of the early 1580s, given to knights – among whom Sir Thomas may be counted – upon investiture.17 The queen’s status as living prototype is thus mediated by a second, representational prototype whose modes of agency the jewel assumes through verisimilitude, as the object legitimately ‘‘seals’’ its wearer as knight and honored subject. This index of sovereignty thereby manifests a transformative capacity as an agent that, through its sealing action, converts the wearer-recipient into an index. The loyal courtier becomes a corporeal frame and surface for the queen’s image, a vehicle for the distribution of her agency and inscribed omniscience as ruler. Elizabeth’s textured profile is foregrounded against the wearer, who would seem to recede into subordination in the face of the monarch’s physical precedence. But the exterior fac¸ade is not a direct replica of an Elizabethan medal or seal. Whatever resemblance the object may bear to these political exemplars is somewhat diluted through the mixed materials of its construction. The jewel is no monochromatic medal; it articulates the queen’s ascendancy through the assorted properties of gold, multicolored enamel, diamonds, and rubies. The use of such substances suggests an additional prototype for the work, one that both extends its agentive role and introduces more subtle relations than those created by its legitimizing, sealing activity. Adorned with gems and worn on the body, the Heneage jewel also invokes the protective amulet of precious stones.18 The belief that gems could be worn as healing and apotropaic devices has its roots in the classical lapidarian tradition. Refracted through the adaptations of medieval theologians such as Albertus Magnus and Isidore of Seville, this tradition persisted into seventeenth-century Europe as the province of alchemists, physicians, and natural philosophers.19 Like their ancient predecessors, Renaissance lapidarian texts mingled the visual, physical, and agentive properties (or ‘‘virtues’’) of individual gems within a single discursive frame. How a material looked, for instance, was often considered an indication of its distinctive ability to act. However, the level of scrutiny evident in these texts did not necessarily resolve crucial ambiguities regarding the precise origins of the gems’ effects. While God was almost invariably posited as the ultimate source for such powers, his designs were also somehow mediated by Nature (his handmaid) and, in particular, the heavens.20 According to some writers, stones received their

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 121

12.1.2007 6:12pm Compositor Name: SJoearun

The Elizabethan Jeweled Miniature 121 virtues from the celestial spheres in an intricate pattern of correspondences between gems and planets that allowed the latter’s potency to ‘‘emanate’’ from jewel to wearer.21 Yet there remained something powerfully inaccessible about the chain and type of causality affecting the potentiality of these things. As one author conceded, ‘‘mans coniecture is farre from the true operation of hidden properties.’’22 This inability to abduct causality echoes Hilliard’s remarks regarding limning’s true luster and seems to set the stage for the Heneage Jewel’s captivating effects both within and without, in appearance and operation. In its amulet-like display of precious stones, the Heneage Jewel might induce reflection on the curative ‘‘hidden properties’’ that could confound lapidarian authors. Yet the object’s exterior exploits this ambiguity of causality to accord the queen an agentive role in the efficacy of precious materials. Elizabeth’s portrayal in gold, for instance, identifies the monarch with alchemy’s most perfect and ennobled metal, which, participating in the cosmic fluid and heat of the sun, embodied the solar sphere’s generative faculties.23 These celestial affiliations are upheld not simply by the gold’s luminosity, but also by the structure of the fac¸ade itself. The enameled border of the medallion portrait is not continuous with the surface of the royal depiction, and instead forms a separate ornamental perimeter. Tiny bars of gold encircle the central image, connecting it to the enameled frame and coinciding with the diamonds and rubies along the major axes of the ovoid form. The linear activity of these bars is accentuated by their projection slightly beyond the frame’s inlaid nodes, evoking the conventional imagery of heavenly rays that here seem to be cast from the royal representation. In Gellian vocabulary, the exterior facet of the jewel implements formal relationships ‘‘internal to the index’’ in a way that suggests a generative process.24 Through the pattern of celestial rays, the Heneage Jewel installs the queen as a decisive origin of the gems surrounding her, an intercessor who channels the powers of the heavens. Although the inscription D[EI] G[RATIA] duly attributes the source of Elizabeth’s agency to God, the relatively small, peripheral text is no match for the visual prominence of the queen’s embossed portrait. Indeed, the precious stones – and, by extension, their marvelous properties – seem to emanate from the sovereign. To her we may ascribe the ability of the rubies on the diagonal axes to, for example, comfort the wearer and heal his sight.25 The diamonds, whose ornamental frames give them special emphasis, are dispersed by the queen to ward off a variety of conditions (including lunacy and fear) and

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 122

12.1.2007 6:12pm Compositor Name: SJoearun

122 Jessen Kelly as a defense against enemies (especially when worn on the left shoulder), poison, and venom.26 Though the compositional structure enables the abduction of the English queen as a cause of mysterious, hidden properties, the imperial connotations of Elizabeth’s portrait define her abilities as attributes of sovereignty and thus as privileges that courtiers cannot commandeer. By means of the gems, the queen’s seal positions her as a kind of physicianprotectress of the jewel’s wearer, who bears a tactile connection to the indexes of her potency. While, for Sir Thomas, this connection presents a means for gaining privileged access to Elizabeth’s agency, it also reiterates the conditions of reciprocity that the jewel has set in motion. The protective emanations of the jeweled exterior can be seen as part of the object’s status as a royal gift and therefore as an extension of Elizabeth’s power.27 Her protection, insofar as it is continuously exerted on the wearer, amplifies the queen’s original gesture of generosity in conferring the object as a whole. Heneage is in this sense constantly positioned as a recipient-patient to the monarch and her singular dominion over precious materials. This obligation might be seen to delineate the wearer’s subordination to the monarch via captivation. As the corporeal frame and surface for the jewel, Heneage himself is located within the path of linear emanations spanning from the imperial portrait: like the gems, he is an extension and index of the ambit of the queen’s authority. Yet, we must ask how this jeweled miniature, as a mediator of social agency, structures and defines the forms of obligation that the exterior appears to prescribe. Even as the frontal fac¸ade of the object promulgates Queen Elizabeth as the prime origin and channel of power over precious materials, the jewel in its entirety suggests (quite literally) that this is not the whole picture. The existence of the jewel’s interior – and, with it, intentions and indexes beyond what is immediately shown – unsettles the declaration of causality on the exterior and repositions agents and patients within the art nexus. Whatever subordination is exerted upon the wearer by the exterior, the interior dislocates him from the uncertainty that the notion of captivation implies. In its hiddenness, the interior renders the wearer a proprietor of knowledge, a privileged keeper of secrets.28 To Sir Thomas’s interlocutors, the jewel appears simply as an admixture of talisman and medal; there is no visible indication of its underlying construction on the frontal fac¸ade. The jewel hides its mode of entry since it can only be opened by turning the object over to access the lid. The viewer-interlocutor is subject to the

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 123

12.1.2007 6:12pm Compositor Name: SJoearun

The Elizabethan Jeweled Miniature 123 indexes of royal authority conveyed through the queen’s dominion over gems and courtiers alike. By withholding the interior of the miniature from the viewer, the wearer becomes an agent in this process. To the extent that the courtier’s withholding constitutes an enactment of obligation to the queen, he assumes a role akin to that of Elizabeth herself, exercising a special dominion over precious jewels and protecting her interior representation from view. The queen’s gift thus returns to her in equivalent form, through a correspondent mode of agency.29 But the wearer’s power of withholding is also that of revealing, creating a dynamic that resonates with the procedures and premises of early modern natural inquiry.30 Historically, the response of awe at the inexplicable that Gell identifies with captivation might be considered akin to wonder, or the astonishment produced by the mysterious or perplexing. For specialists in proto-scientific fields, such responses were not a barrier to agency or abduction but rather served to compel investigation and use. As Lorraine Daston and Katherine Park have argued, later sixteenthcentury proto-scientific practitioners distinguished themselves from their medieval precursors by their insistence that the causes of marvelous and resistant natural materials could indeed be ascertained through human intellect.31 Published investigations of natural materials typically imputed a structure to the object of scrutiny, one that apprehended these materials as surfaces beneath which knowledge was obscured.32 The aim of the inquirer was to pass beyond these surfaces and bring hidden properties and virtues to light through acts of physical transformation. The famed alchemist-physician Paracelsus, for example, understood the attainment of knowledge as an ‘‘unveiling of nature,’’ a movement from what is corporeal and perceptible to a concealed spiritual ‘‘essence.’’ Through chemical processes of extraction and distillation, the alchemist converted this essence into concrete forms.33 In its material composition and clandestine structure, the jeweled miniature invokes this context of investigation. Echoing the epistemological process of investigating natural matter, opening the Heneage Jewel enables the attainment of concealed knowledge through the puncturing of an external countenance. Attaining this knowledge entails a dramatic physical transformation, as the single face of metal, gems, and enamel gives way to a double-faceted interior dominated by pigments. Yet this transmutation is perhaps most strikingly manifested in the contrast between the stately exterior profile and the softened frontal visage of the multicolored limned portrait that meets the viewer’s gaze.34 Devoid of

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 124

12.1.2007 6:12pm Compositor Name: SJoearun

124 Jessen Kelly creases or shading, the queen’s portrait departs not only from the external depiction on the jewel but also from her living prototype, as she was likely over sixty years old at the time of the image’s making. The reduction of Elizabeth’s age in the realm of pictorial representation – the so-called mask of youth – has been characterized as a mystification designed to insist on the continuing vitality of an elderly, heirless monarch.35 Yet iconicity need not constitute the mode of resemblance by which the limning is measured. The gap between image and prototype accentuates the work of a transforming agent in the portrait’s appearance, one that extracts an ‘‘essence’’ beyond external likeness. Elizabeth’s unblemished, youthful countenance, in conjunction with the sumptuous floral imagery that surrounds her, indexes her luxuriant but chaste virtues as the Virgin Queen. Similar transformations are evident with regard to precious gems: the diamonds and rubies that encompass the exterior image are absent from the internal enameled frame, yet they remain operative as motif and effect within the miniature itself. Hilliard has characteristically built up the queen’s ornate accouterments through a thick, complex layering of paint virtually encrusted on the surface of the vellum. The use of diverse, burnished pigments, which undoubtedly shimmered in their original condition, conveys the glistening, faceted properties of precious materials – the luster of which Hilliard wrote. Not confined to the jewelry Elizabeth wears, the effulgent qualities of gems have been distilled and isolated from both actual and depicted jewels and are distributed to the queen herself as a characteristic of her essence.36 Materially, then, the miniature transfigures the precious metals and stones on the exterior into the chemical constituents of varied pigments, revealed through the unfolding of the object. This unfolding is enacted by the jewel’s wearer, who mediates access to the technical splendor of the limned portrait in opening the queen’s gift. Hilliard’s miniature may bear an ambiguous causality through its captivating properties; yet, through possession of and proximity to the index, the wearer emerges as the visible and physical origin of the limned portrait’s appearance. The owner’s agency in the advent of the miniature parallels both Hilliard’s calculated manipulation of materials in creating the object and the alchemist or natural philosopher who transforms natural substances in unveiling their secrets. Within the field of protoscientific inquiry, this type of agency is of significant consequence. For natural philosophers, the ability to transfigure matter was an act of special virtuosity – one that was sometimes specifically linked to the craftsman’s agency over natural matter – and the knowledge thereby revealed could

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 125

12.1.2007 6:12pm Compositor Name: SJoearun

The Elizabethan Jeweled Miniature 125 comprise a means to attain considerable power over the natural world.37 The proposed equivalence between queen and wearer as protective agents is perhaps not so equivalent after all. Initiating the appearance of what is concealed beneath the gold surface of the jewel, the wearer potentially trumps the queen’s claims to the position of origin and dispenser of such wondrous material properties, an artist-agent who manipulates matter to induce her hidden essence into form and knowledge. But the jeweled miniature ordains divergent modes of viewing and access, and thus the queen’s image is not the only patient to the wearer’s agency. There is a marked tendency toward display entailed in the wearer’s manipulations. The virtuosity of unveiling is a performance of knowledge and power that requires a viewer for its greatest impact.38 This accretive net of social relations is evident in an account by the English ambassador to France, Sir Henry Unton, regarding a viewing of Queen Elizabeth’s limned image in 1595. Though the narration does not specify whether Sir Henry’s miniature was contained in a precious setting, the incident is nevertheless bound up with the rhetoric and process of uncovering inherent in encased examples such as the Heneage Jewel. Recorded in a letter to the queen, Unton’s story appeals to certain social conventions of miniature viewing and thus provides a basis for broader analysis regarding the potential inherent in the jeweled miniature and the Heneage Jewel in particular.39 Unton reports a conversation with the king regarding the caliber of the French monarch’s mistress. In response, Sir Henry discusses the splendor of his own so-called mistress, Queen Elizabeth herself, asserting, if, without Offence I might speake it, that I had the Picture of a farr more excellent Mistress, and yet did her Picture come farr short of her Perfection of Beauty. As you love me (sayd he) shew it to me, if you have it about you. I made some Difficulties; yet, uppon his Importunity, offred it unto his Viewe verie secretely, houlding it still in my Hande.40

Despite the diplomat’s eventual acquiescence to the king’s appeals, his obstruction of royal desire through ‘‘Difficulties’’ underscores his withholding agency. Unton’s continuing authority in compliance is conveyed by his refusal to relinquish the miniature from his grasp, and he makes special note of the fact that he held the object ‘‘still in my Hande.’’ The persistent, tangible link between image and ambassador accentuates his status as the only one who may orchestrate the appearance of the material

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 126

12.1.2007 6:12pm Compositor Name: SJoearun

126 Jessen Kelly form. Unton’s artistic agency over materials, however, also extends to the social relationships that define the scene. Although the ambassador stresses the inadequacy of the limned index when compared with its regal prototype, his responses to the king’s entreaties function to build an anticipation for and valuation of the object, casting the eventual display of the portrait as a revelation that holds special knowledge. This revelation is offered to the royal gaze as an act of visual largesse that expands the dynamics of generosity and obligation that constitute the miniature’s social origin. By presenting the king with the gift of valuable access to the portrait, Unton construes him as a recipient who must reciprocate. But the king himself is also the object of special scrutiny for Unton’s artistry. The unveiling of Elizabeth’s visage ostensibly proves the superiority of the Englishman’s nominal mistress to that of the French sovereign. Unton’s divulgement of the portrait as proof of such an assertion presumes a similar disclosure on the part of his regal interlocutor. Unton can persuade the king to confirm or deny the ambassador’s contention regarding the queen’s virtues and thereby gain access to the French monarch’s own affinities. The ‘‘Passion and Admiration’’ that King Henry reportedly expressed were such that Unton was able to declare that ‘‘the dombe Picture did drawe on more Speache and Affection’’ than an afternoon’s worth of ‘‘Arguments.’’41 Just as the alchemist transfigures materials to distill knowledge of a hidden essence, Unton the artist-agent deploys the limning to extract a confession of the king’s alliances, thus locating him as patient within a Gellian schema. Indeed, according to Unton, the French king ‘‘kissed [the miniature] twice or thrice, I detayning it still in my Hand.’’42 The ambassador’s persistent physical control over the object of the king’s professed longing evokes an especially striking image of Henry’s subordination to both miniature and diplomat. By means of the miniature, Unton transforms (at least temporarily and in certain respects) the social hierarchy; the skillful revealing of limned secrets enables a kind of power, even over kings. If King Henry is situated as a patient, however, then what of Queen Elizabeth? Unton’s gift is a means of further distributing the queen’s agency to elicit confessions of ‘‘Admiration.’’ Yet the king’s professed affinity for Elizabeth is not the only relationship at stake in this interaction. As a gift, presenting the miniature to ‘‘Viewe’’ also aids in delineating the relationship between king and diplomat as a deliberate affirmation of the ‘‘love’’ that the monarch perceives on the part of Sir Henry. If the luster of the miniature constitutes a source of wonder for its

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 127

12.1.2007 6:12pm Compositor Name: SJoearun

The Elizabethan Jeweled Miniature 127 viewers here, this sentiment is not exclusively in the service of the queen’s power. Though Unton is an ambassador (and thus an index of his ruler), Elizabeth herself is in this case transformed into a rather passive intermediary through her own image, as an object that is held, handled, and exchanged to sustain other social – and in this case male – relationships.43 As such, however, her image becomes the potential vehicle for and victim of varied intentions. Though Elizabeth is clearly not imperiled in this (rather flattering) account, the social relations that coagulate around her portrayal enable a projection of other possible actions. For the ability of the Heneage Jewel to render its wearer an index sealed by the queen is only provisional. If her seal may be used to signify loyal alliance, opening and displaying the jewel requires its separation from the surface of the wearer’s body since, as Hilliard specifies, the small size of the miniature renders it seen ‘‘of nesesity close in hande neare vnto the eye.’’44 In this viewing, wearer and subordinating seal are no longer located on the same plane. The transformation of the wearer’s agentive capacities in opening the jewel makes it an object through which he might wield his own judgments and actions, constructing his own coalitions outside the queen’s medallic sanction. For the French ambassador Jean Hotman, these coalitions were precisely the danger of gifts in diplomatic contexts, as those who accept presents ‘‘become slaues to those that giue them,’’ thus forcing political figures into alliances outside of or even against their own state’s interests.45 In this sense, Unton’s gift to the king is potentially laden with treachery. By licensing viewers and owners to reflect on and manipulate material properties, the jeweled miniature renders the queen-as-index subject to instability. This instability is not simply a consequence of the viewer or possessor’s agency but is additionally dramatized in the limned miniature of the Heneage Jewel through the stuff of its making. From one perspective, Queen Elizabeth’s portrait affiliates her with the valuable virtues and secrets of precious materials, depicted in luminous, varied pigments. But the miniature’s physical condition is not exclusively salutary or steady; what is revealed under scrutiny may prove problematic. The same chemical substances shaped in the image of a lustrous royal image are themselves mutable agents that add a different cast to the gift. For example, silver pigments, used by Hilliard in so many of his miniatures, have inevitably oxidized and darkened over time. The indigo used in the backdrop of the Heneage limning is notoriously impermanent and must be kept from strong sunlight to avoid fading.46 Similarly, the white lead

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 128

12.1.2007 6:12pm Compositor Name: SJoearun

128 Jessen Kelly that makes up the queen’s collar was closely managed in the production process, finely ground and decanted in an effort to prevent it from turning black in use.47 But as the application of white lead intimates, the dangers of the limner’s pigments exceed those of chemical mutability. The miniature in the Heneage Jewel contains perfidious substances known in the sixteenth century to have potentially toxic effects: in addition to white lead, red lead, vermilion, and orpiment (a form of arsenic) have been identified as components of this work.48 Though sometimes taken in small doses as physic or worn around the neck in amulets to ward off the plague, the therapeutic value of such substances remained highly contested.49 Medical manuals warned of the dangers of orpiment, for instance, in creating welts and ‘‘burn[ing] the Skynne’’ on contact.50 Neither white lead nor orpiment, it was advised, should be ‘‘taken inwardes’’ due to their poisonous and corrosive properties.51 The use of such materials here snares the miniature in conceivably apostate intentions. The dangers emanated by these substances are not purely physical. Hilliard’s condemnation of orpiment as among the ‘‘ill smelling’’ and ‘‘ill tasting’’ pigments that are ‘‘naught for Limning’’ was based not only upon its poisonous capacity but also on its cheap cost and wide availability, which, to his mind, compromised the patrician ‘‘sweetness’’ of his gentlemanly art.52 In the Heneage Jewel, Hilliard used orpiment in portraying the reflection on the queen’s pearls – an effect that he specified in his treatise should be limned with fine, burnished silver.53 The appearance of this material can be seen as a corruption of the queen’s own sweetness, an indication of the ways in which the artist-agent’s transformations of matter might ennoble or debase. And, in the gift’s status as an extension of Elizabeth herself, these effects might pass to her living prototype in a kind of ‘‘volt sorcery’’: potentially poisoned and poisoning, the queen is subject to the virulent qualities of the agents from which she has been wrought. To be sure, the internal relations of the index within the miniature, in which the queen’s patterned collar and jewels seem to issue from her visage, could imply the emanations of these chemical entities to the wearer. Yet there are elements of the Heneage Jewel that uphold the queen as object and not agent of these relations. In its open state, the jewel’s painted lid and limned portrait form a diptych in which the queen’s image is echoed in the large rose on the cover’s interior. The rose, as a floral specimen, is bound to the natural world from which the pigments that represent it also derive. In full bloom, the rose’s state of

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 129

12.1.2007 6:12pm Compositor Name: SJoearun

The Elizabethan Jeweled Miniature 129 verdant vitality reiterates Elizabeth’s own virtues and expands beyond this immediately visible analogy, via its traditional symbolic signification, to the Tudor dynasty as a whole. As such, the rose connotes longstanding political power. But the rose is simultaneously a potent vanitas symbol: its properties encompass growth as well as the threat of inevitable deterioration and demise. The flower’s connotations are reinforced by the lid’s inscription: ‘‘that virtue suffused with so much beauty is not eternally inviolate.’’ References to fugitive inviolability, though multifarious in their possible interpretation, evoke the vulnerability of the sovereign to the corrosion of physical and temporal processes. The royal virtues revealed in the miniature’s hidden interior are not, as the inscription makes clear, interminable, as even the most ennobled forms are apt to decay. The delicate and changeable nature of the miniature’s luster therefore finds additional representation in the textual and symbolic facets of the Jewel interior. As they pertain to material properties and processes, these facets are not distinct from the issues of agency that have been under consideration here. Whereas Gell would debar such communicative forms of signification from the anthropology of art as a ‘‘system of action,’’ in this case symbolic decoding is not anathema to social agency but rather participates in its generation by affirming the precariousness of the queen’s material position. Not even the queen is fully immune to the activity of time, and both owners and viewers are asked to assume a relation to an English monarch whose parlousness renders her in need of protection. Through the vulnerabilities conveyed in the open jewel, the viewer is afforded a choice of loyalty to an imperiled sovereign or treacherous exploitation of her fragility – that is, to accept or refuse the obligations of her gift. This act of choosing highlights the agency of the beholder and renders his relation to the monarch a subject of conscious selection and exchange.54 Thus, in establishing the possessor of the object as an artistagent by virtue of his transformation of and dominion over its physical composition, the jeweled miniature proposes that choosing loyalty to the sovereign is not a matter of passive subordination or immobilizing astonishment. Instead of suggesting the reproduction of Queen Elizabeth’s power through captivation, the Heneage Jewel potentially engenders commitment by granting the courtier an experience of authority that, as I have argued, in some sense exceeds that of the queen herself in the transfiguration of and control over material forms. The Elizabethan jeweled miniature is a royal gift framed and overlaid by declarations of prodigious power, from the queen’s imperial profile on the

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 130

12.1.2007 6:12pm Compositor Name: SJoearun

130 Jessen Kelly Heneage Jewel, to Unton’s exaltations of the queen’s beauty, to Hilliard’s assertions regarding limning’s seemingly mysterious origins. Though such pronouncements constitute hallmarks of Gellian captivation, the jeweled miniature ultimately dramatizes Queen Elizabeth’s political instability rather than imposing an unproblematic image of authority or an exclusively propagandistic cult.55 The physical components of the Heneage Jewel both exert and provoke agency as elements of proto-scientific and medical, as well as artistic, knowledge and manipulation. Indeed, it might be said that the apparently ambiguous causality of the miniature and its properties leave its origins open to appropriation by artists, monarchs, owners, and viewers. By tracing the multifarious activities of one jeweled miniature and its materials, enmeshed in technical-epistemological discourses and a royal gift economy, I have sought to show that the media of the queen’s authority are also the means by which this authority may be undone. But the material undoing of the monarch’s authority is also its latent sustenance. If Elizabeth’s power is reproduced, it is not through captivated subjection but through the assumption of agency, the agency of poisoning and protecting, of revealing and withholding, that mediates the relationship between sovereign and courtly subject. Nevertheless, some of the social implications of these instabilities – namely treason and subterfuge – remain hypothetical rather than verifiable. The queen’s physical vulnerability in the jeweled miniature is not without correspondent threats to her person and political authority, but neither Sir Thomas Heneage nor Sir Henry Unton (to our knowledge) engaged in treasonous or questionable alliances. Whereas Gell’s own diagrams often depict relationships already known or transpired, in this case the Gellian framework has provided a means for intimating the possibilities and paths for social action that, even when not performed, are part of the miniature’s distinctive mediations and help to delineate the stakes of deeds in fact actuated.56 But if the notion of captivation cannot account for the jeweled miniature’s agency, the inconstant elements of the Heneage Jewel are nevertheless bound up with its identity as an object of interpretive ‘‘difficulty.’’ At the conclusion of this study, I admit that I am unable to say to what extent the miniature’s complex social relations should be abducted as part of the deliberate intentionality motivating its design. Thus in my own critical unfolding of the Heneage Jewel, I have not necessarily escaped some of the entrapments of captivation with which I have taken issue. In this sense, the jewel and its luster still generate and encompass causal

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 131

12.1.2007 6:12pm Compositor Name: SJoearun

The Elizabethan Jeweled Miniature 131 ambiguities that may lend credence to Hilliard’s praise for the miniature’s equivocation. Acknowledgments I am grateful to Whitney Davis, Elizabeth Honig, Matthew Hunter, and the editors of this volume for their comments and suggestions. Notes 1 2 3

4 5

6 7 8 9

10 11 12 13

Nicholas Hilliard, A Treatise on the Art of Limning, ed. R. K. R. Thornton and T. G. S. Cain (Ashington, 1981), p. 62. Hilliard, Treatise, p. 62. For another account of the miniature’s problematic origins, see J. Yates, Error, Misuse, Failure: Object Lessons from the English Renaissance (Minneapolis, 2003), pp. 36–60. A. R. Jones and P. Stallybrass, Renaissance Clothing and the Materials of Memory (Cambridge, 2000), pp. 34–5. Gell, ‘‘The technology of enchantment and the enchantment of technology,’’ The Art of Anthropology: Essays and Diagrams (London, 1999), pp. 159–87; Gell, Art and Agency (Oxford, 1998), pp. 68–72. Layton, ‘‘Art and Agency: a reassessment,’’ Journal of the Royal Anthropological Institute 9 (2003), p. 451. Gell, Art and Agency, chs. 1–5. For limning and royal portraiture, see R. Strong, The English Renaissance Miniature (London, 1983), p. 8. M. Mauss, The Gift: the Form and Reason for Exchange in Archaic Societies, trans. W. D. Halls (New York, 1990). P. Burke also applies Maussian theory to jewelry in ‘‘Renaissance jewels and their social setting,’’ in Princely Magnificence: Court Jewels of the Renaissance, 1500–1630 (London, 1980), pp. 8–11. L. Klein,‘‘ ‘Your humble handmaid’ ’’: Elizabethan gifts of needlework,’’ Renaissance Quarterly 50(2) (1997), pp. 461–8. R. Strong, Gloriana: the Portraits of Queen Elizabeth I (London, 2003), pp. 121–2. Strong, The Cult of Elizabeth: Elizabethan Portraiture and Pageantry (London, 1977). See also J. Murrell, The Way Howe to Lymn: Tudor Miniatures Observed (London, 1983); P. Fumerton, ‘‘ ‘Secret arts’: Elizabethan miniatures and sonnets,’’ Representations 15 (1986), pp. 57–97; and Jones and Stallybrass, Renaissance Clothing, pp. 34–5, 40–3.

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 132

12.1.2007 6:12pm Compositor Name: SJoearun

132 Jessen Kelly 14 P. H. Smith, The Body of the Artisan: Art and Experience in the Scientific Revolution (Chicago, 2004). For an interpretation of Hilliard’s Drake Jewel, given by Queen Elizabeth to Sir Francis Drake, as symbolic of ideas promoted by the queen’s alchemist-philosopher, John Dee, see K. C. C. Dalton, ‘‘Art for the sake of dynasty: the Black Emperor in the Drake Jewel and Elizabethan imperial imagery,’’ in P. Erikson and C. Hulse, eds., Early Modern Visual Culture: Representation, Race, and Empire in Renaissance England (Philadelphia, 2000), pp. 178–214. 15 Burke, Princely Magnificence, p. 60. 16 See, for example, the National Maritime Museum’s portrait of Sir Francis Drake shown wearing the Drake Jewel at his belt. 17 The connection to the Garter medal is observed in Burke, Princely Magnificence, p. 60. 18 Strong notes the talismanic appearance of the miniature in Gloriana, p. 121. 19 See J. Evans, Magical Jewels of the Middle Ages and Renaissance, Particularly in England (Oxford, 1922). 20 On the celestial origins of virtues, see Evans, Magical Jewels, pp. 143–4. The notion that Nature serves as God’s handmaid is presented by, for instance, Joseph du Chesne, The Practise of Chymicall, and Hermeticall Physicke, trans. Thomas Tymme (London, 1605). 21 Evans, Magical Jewels, p. 172. 22 Anglicus Bartholomeus, Batman vppon Bartholome his booke De Proprietatibus Rerum, trans. J. Trevisa, ed. and rev. S. Batman (London, 1582), pp. 258. 23 See J. Proust, ‘‘L’aurum potabile au temps de la Renaissance, en France et en Angleterre,’’ in L’Or au temps de la Renaissance: du mythe a` l’e´conomie (Paris, 1978), pp. 15–17. 24 Gell, Art and Agency, ch. 6. 25 On the virtues of the ruby, see Anon., Here begynneth a lytell boke of the xxiiii. stones pryncipalles that profyteth most to mans body (London, 1528). 26 Here begynneth, pp. 18–20; Batman vppon Bartholeme, p. 256; Albertus Magnus, The boke of secretes of Albertus Magnus, p. C.iii. 27 For gifts as containing the essence of the giver, see Mauss, The Gift, p. 12. 28 On the miniature and the ‘‘secret self,’’ see Fumerton, ‘‘Secret arts.’’ 29 For Mauss, the gift ‘‘seeks to return to . . . its ‘place of origin’ or to produce, on behalf of the clan and the native soil from which it sprang, an equivalent to replace it’’ (The Gift, p. 13). 30 Fumerton (‘‘Secret arts’’) analyzes this dynamic in relation to the ‘‘public’’ and ‘‘private’’ self. 31 Daston and Park, Wonders and the Order of Nature, 1150–1750 (New York, 1998), p. 160. 32 This practice is evident in the writings of the widely read natural ‘‘magician,’’ Giambattista della Porta, discussed by W. Eamon, Science and the Secrets of

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 133

12.1.2007 6:12pm Compositor Name: SJoearun

The Elizabethan Jeweled Miniature 133

33

34 35 36

37

38 39 40

41 42 43

44 45 46 47 48

49

50 51 52

Nature: Books of Secrets in Medieval and Early Modern Culture (Princeton, NJ, 1994), pp. 194–233. See Mario Luigi Bianchi, ‘‘The visible and the invisible: from alchemy to Paracelsus,’’ in A. Clericuzio and P. Rattansi, eds., Alchemy and Chemistry in the 16th and 17th Centuries (Boston, 1994), pp. 17–50. On this contrast, see Fumerton, ‘‘Secret arts,’’ pp. 62–3. Strong, Cult, pp. 47–54. On the jewel-like appearance of the miniature, see Murrell, Way Howe, pp. 13–14; Fumerton, ‘‘Secret arts,’’ pp. 66–7; Jones and Stallybrass, Renaissance Clothing, p. 41. For the connection between artisans and natural philosophers, see Smith, Artisan, and Eamon, Science, pp. 217–21. For the power assumed through their activity, see Daston and Park, Wonders, p. 171, and Eamon, Science, p. 212. This performative element is also suggested in Fumerton’s notion of the miniature as a ‘‘public’’ staging of the ‘‘private’’ (‘‘Secret arts,’’ p. 56). Fumerton, ‘‘Secret arts,’’ p. 61. ‘‘Sir H. Unton to her Majesty, from Coucy, Feb. 3, 1595–6,’’ in William Cecil Burghley, transcribed W. Murdin, Collection of State Papers Relating to Affairs in the Reign of Queen Elizabeth, vol. 2 (London, 1759), p. 718; also cited in Fumerton, ‘‘Secret arts,’’ 61. ‘‘Unton,’’ p. 719. ‘‘Unton,’’ p. 718. C. Le´vi-Strauss analyzes the exchange of women between men in The Elementary Structures of Kinship (trans. J. Harle Bell and J. R. von Strumer, and ed. R. Needham (Boston, 1969)), to which Gell acknowledges his debt in Art and Agency, p. 9. Hilliard, Treatise, p. 86. Jean Hotman, The Ambassador, trans. J. Shaw (London, 1603), p. D7v. R. D. Harley, Artist’s Pigments, c. 1600–1835: A Study in English Documentary Sources (London, 1982), p. 70. Ibid., pp. 166–7. A. Derbyshire and R. Withnall, ‘‘Pigment analysis of portrait miniatures using Raman microscopy,’’ Journal of Raman Spectroscopy 30 (1999), pp. 183–9. Peter Turner, The opinion of Peter Turner Doct: in physicke, concerning amulets or plague cakes (London, 1603); and Francis Herring, A modest defence of the caueat giuen to wearers of impoisoned amulets (London, 1604). Thomas Newton, Approoved medicines and cordiall receipts (London, 1580), p. 76. Ibid., pp. 75–6. Hilliard, Treatise, p. 88.

Osborne / Art’s Agency and Art History 1405135379_4_005 Final Proof

page 134

12.1.2007 6:12pm Compositor Name: SJoearun

134 Jessen Kelly 53 Ibid., p. 88. 54 For an analysis of Elizabeth’s relationship to her subjects that also addresses courtier agency and an unsteady royal regime, see L. A. Montrose, ‘‘Idols of the Queen: policy, gender, and the picturing of Elizabeth I,’’ Representations 68 (1999), pp. 108–61. 55 Montrose critiques the idea of a ‘‘cult of Elizabeth’’ in ‘‘Idols,’’ esp. pp. 132–48. 56 For the already known aspects of Gellian diagrams, see W. Davis, ‘‘Gellograms, or five questions for Alfred Gell,’’ unpublished MS, 2003.

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 135

12.1.2007 2:25pm Compositor Name: SJoearun

6

Representational Art in Ancient Peru and the Work of Alfred Gell Jeffrey Quilter

Alfred Gell’s call for a distinctly anthropological approach to the study of art is audacious in the same way that anthropology boldly presumes to claim all of human behavior and its products as its subject of study. Almost every discipline within the social sciences makes or has made the same pronouncement as anthropology, however. Or, at least, individuals who call themselves practitioners of those disciplines may be found who, at one time or another, have claimed that the theories and methods of their fields can serve to elucidate what might, at first glance, appear to be subjects far removed from their everyday practice. As Gell notes, ‘‘art’’ would seem to be easy prey because it manifests itself as so completely an artifice of human thought and actions. Contrast anthropology’s claim for universal sovereignty and art history’s narrow focus on ‘‘art’’ and ‘‘history.’’ In this field, as in many of the humanities, an object is focused upon through which the larger human condition can be viewed and appreciated. Art history’s gaze has tended to move from the object outwards toward society while anthropology’s perspective wraps the artifact within the embrace of culture. In the everyday practice of these disciplines, though, we may find anthropologists acting like art historians and vice versa, but these greater tendencies are fairly widespread and lie embedded in the rites of passage of graduate school training.

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 136

12.1.2007 2:25pm Compositor Name: SJoearun

136 Jeffrey Quilter Gell makes it quite clear that his approach is distinctly anthropological in the Social Anthropology mode. But with its reliance on social agents and their interactions, how can his proposed analytical techniques be employed for past societies that had no written records? The Moche of ancient Peru is an interesting test case for Gell’s theories because an important issue in his writing is that of representation: it forms one of the longer entries in the index of Art and Agency. ‘‘Moche’’ (or ‘‘Mochica’’) refers to an archaeological culture that existed in the river valleys of the north coast of Peru from about AD 1 to 800. The Moche inherited three thousand years of cultural developments which resulted in a society with extensive irrigation agriculture and sophisticated craft traditions. Moche society was hierarchical, with elite warrior and priest sectors. The Moche used representation to a degree exceeded by no other South American art style. At the same time, however, they engaged in permutations of geometric designs similar to the elaborations of the Marquesan tattoos which also captured Gell’s imagination and attention. Although temporal and regional variations are in evidence in Moche art, there are distinctive techniques and features that exist throughout time and space. First, there is a preference for bichrome ceramics, usually employing red-brown and cream-white colors. Second, there is a penchant for veristic or representational art in which figures representing gods, mythic heroes, humans, animals, plants, and objects were molded or painted on vessels. Third, in contrast to ceramics, the walls of temples were decorated in bright, polychrome murals and friezes, often in highly geometric and, to our eyes, ‘‘abstract’’ styles. The notable exception to this, however, is the last phase of Moche art in which representational styles also appear on temple walls.1

Applying Gell’s Theories to Moche Art To attempt to evaluate how Gell’s ideas might be applied to the study of Moche art as a whole would require many more pages than are available for this chapter. I will take a different tack, therefore, and discuss two forms of Moche art media within the larger corpus: portrait head ceramics and wall art in the form of painted murals and friezes. Consideration of each of these will necessarily require reference to other examples of similar and dissimilar artworks, enlarging the overall discussion. In attempting to evaluate Gell’s theories, I will review his three major concepts for each type

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 137

12.1.2007 2:25pm Compositor Name: SJoearun

Representational Art in Ancient Peru 137 of object: the art nexus, the enchantment of technology/technology of enchantment, and the distributed person. The order in which I will review these ideas will vary, because different considerations obtain for each example. I conclude with a general consideration of Gell and Moche art.

Portrait Vessels Moche portrait head vessels are among the most ‘‘representative’’ of all Andean art (figure 6.1). At least one illustration of a ‘‘typical’’ portrait head vessel almost always finds its way into introductory textbooks or chapters on the art of pre-Columbian America. Presumably, the popularity of portrait heads today is due to their appeal to Western tastes for

Figure 6.1 A Moche portrait head vessel. Peabody Museum, Harvard University, Photo 46-77-30/5050 N30119.2.

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 138

12.1.2007 2:25pm Compositor Name: SJoearun

138 Jeffrey Quilter portraiture: we can gaze upon one of these vessels and imagine that we are seeing how an ancient Moche person actually appeared in the flesh. It seems safe to say that the enchantment of technology/technology of enchantment likely were significant factors in the appreciation of these vessels in antiquity. Many of the pottery visages are so lifelike that they are startling even to modern eyes jaded by cheap mirrors and a multitude of devices for the reproduction of images. For people who had no easy means to see reflections of their own faces, these representations must have been very powerful indeed. Although lifelike, none of the portrait heads is fullsized, even taking into account the small stature of the ancient Moche. The head sections of vessels usually are between 10 and 20 cm in height, less than full-scale. The most veristic of portrait vessels serve as technologies of enchantment by momentarily convincing viewers that they are looking at a living person. Skin hues likely were matched in the colors of paint applied to the faces while the cream-colored slip of the headdress resembled naturally hued cotton textiles. Shadows of nose and lips and in eye orbits produced by the three dimensionality of the sculpture would have added an additional aura of realism in viewing the face of a portrait head vessel. At the same time that the vessels were technologies of enchantment, the enchantment of technology was at play. It has long been known that the majority of Moche vessels were made through the use of molds and it was generally assumed that molds were employed because the technology was more efficient than other methods and allows the production of multiples of the image in question. A recent study of refuse at a Moche ceramic workshop suggests that the issue is more complicated than previously thought.2 The sum of the evidence and interpretations from it suggest that the use of molds was more difficult and complicated than if vessels had been made more straightforwardly, by hand building in the case of portrait vessels, for example. Many separate molds were necessary to make a vessel. Furthermore, it seems that, for a complicated vessel, the chances of destroying mold sections might have meant that producing more than one vessel was not guaranteed. How might Gell’s concepts of art nexus and its component parts be applied to Moche portrait vessels? If we are to consider these pots as social agents we must know something about the social contexts in which they acted. We do not know much about Moche society beyond some rather general concepts and suppositions. If we knew about the ‘‘life cycle’’ of

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 139

12.1.2007 2:25pm Compositor Name: SJoearun

Representational Art in Ancient Peru 139 these pots we might be able to infer some aspects of their agency, since we would be able to identify the social environments through which they traveled. In other words, if we knew when and where these pots appeared in Moche social contexts we might be able to infer how they interacted with other social agents. We can state that Moche pots were produced in specialist ceramic workshops and, very likely, these were recognizable to others. We also know that the pots were not simply grave-goods. At least some of them had use-lives above ground, judging by wear patterns on their bases. But how they were used is uncertain. Self-reflexively, pots are sometimes depicted on pots. They are shown accompanying curers engaged in their skills; they are shown next to dignitaries engaged in administrative or ritual activities; they are placed near weavers in a workshop. They never are shown being used in any clear way, however. Rather they appear, emblematically, to announce their presence and apparently nothing more. Again, in their insistence on making themselves known as present, and in their low utility value, these objects register as works of art. In and of themselves they moved as social actors. Since many contemporary depictions of these pots lack details of the designs on them, especially when they are painted at small scales, we can infer that in some realms of social interaction the symbolic messages of their decorations were less important than their status as objects that held distinct and special values. In other words, their roles as social agents – affecting other agents by their presence in particular social contexts in which agency is expressed differently than in others – may have been more important than the symbolism they carried. If we are unclear as to how ceramic vessels interacted in social settings in ancient Moche culture, we are equally unclear as to the prototypes to which portrait head vessels refer. I began this discussion by noting the lifelike nature of these pots. But to state that the ceramics appear to accurately portray a human visage does not explain much. The most common, traditional explanation of these vessels is that they are portraits, hence the name of the type. One or even a few examples of portrait heads selected and presented for their apparent portrayals of specific individuals in a text book masks the problem of the range of variation within the type. There are many pots depicting what appear to be humans. Many seem to be quite ‘‘generic’’ without distinct personalities while others may depict specific individuals but are not finely made.3 We can look at paintings in a modern art museum and distinguish between poorer and better examples of

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 140

12.1.2007 2:25pm Compositor Name: SJoearun

140 Jeffrey Quilter portraiture or different schools of painting. But we can’t make such distinctions very easily within the corpus of Moche art. In some cultures, minimally carved wooden logs or lumps of clay may be cited as the ‘‘portraits.’’ In such cases there is no requirement for veristic representation of the details of the physiognomy of the prototype; often, apparently, all that is required is that there is general agreement or someone in authority states that an object is a person or god and it is taken as so. In other times and places, highly individualized portrayals may represent generic types. Through the ages many different individuals posed for portraits of Hercules and other demigods, gods, and heroes, but they all referred to an idealized type whose original visage was unknown or irrelevant. Christopher Donnan4 has conducted a detailed study of portrait head vessels and noted several interesting facts about them. First, there are known examples of more than one rendition of the same portrait. Apparently, the same mold was used to produce the face but different molds were relied upon to produce varying elements such as stirrup spouts or jar necks. Potters also varied the face paint or tattoos, headdress, ear spools, and other ornaments surrounding the face itself. Another observation is that individuality was marked not only by the unique physiognomy of the face but by distinctive features such as scars. Remarkably, individuals were portrayed at various stages of life. One example is of a male with a scar near his lip, shown as a boy of around ten, a young man of about 20, and an adult probably 30 years of age. The most intriguing of Donnan’s observations is the depiction of distinctive characters as captives. He notes two examples, one of a man with a distinctive bushy mustache and beard (facial hair is infrequently shown in Moche art) and the other of a person with a narrow face, heavylidded elliptical eyes, and a distinctly large, pointed nose. In both cases, these individuals are known as portrait heads, fulfilling social roles, and as bound, naked prisoners. The mustachioed man is shown as a warrior with club and shield while the other figure holds a bowl, in some cases with a spatula in the other hand. Donnan interprets the depiction of these distinctive individuals as representing events that actually took place – that specific men were captured and then sacrificed and that portraits were made of them as captives. He raises the issue of who made these portraits and offers two possibilities. The first is that it was the captors who made portraits to commemorate their victories. He rejects this, however, because he assumes

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 141

12.1.2007 2:25pm Compositor Name: SJoearun

Representational Art in Ancient Peru 141 that the three valleys in which portraits were made were unified during the period in which portraiture was practiced. He proposes, instead, that combat was ceremonial and that capture was planned. Therefore, the portrait vessels could have been made ahead of time because who was to be a sacrificial prisoner was preordained. Some kind of close tie between the Chicama, Moche, and Viru´ valleys appears to have been in effect late in Moche times, given their virtually identical decorative program of the major ceremonial center in two valleys – Chicama and Moche – and the sharing of many traits between the three. But we do not know the reasons for these similarities, whether they were through confederation, alliance, conquest, or some other factor. Different possibilities might lead to different conclusions about the bellicose nature of relations between the people of the different valleys and, therefore, the likelihood that warfare was in earnest or staged as ritual. The core of Donnan’s argument for the portraiture of elite prisoners is that who was to be prisoner was prearranged. But many other possible explanations for prisoner ‘‘portraits’’ may be proposed. The problem we face is that there is no clear way to test one proposal against another. One aspect of the distinctive prisoner portraits is that there are other examples of them depicting the individuals at various life stages. The incompleteness of the archaeological record makes it hard to assess whether this was done for all the subjects of portrait vessels or only a few, although the latter seems more likely. If young boys were selected to one day play the roles of prisoners and sacrificial victims then portraits of them could have been made throughout their life. But there are simpler explanations. The two simplest explanations eliminate the necessity of having to account for the making of portraits in synchrony with human life stages. One possibility is that the portraits indeed are based on real people who were captured in battle, whether ritual or agonistic, but that all of the series was made after the warrior had been caught and was destined to be sacrificed. Physical anthropologist John Verano has suggested that two weeks or more may have separated the days of capture and sacrifice, based on the healing of wounds likely sustained in battle, enough time for potters to make representations of the captives in various life stages.5 The other possibility is that the images are not portraits of humans but of mytho-historical heroes or demigods. The distinctive facial features may be marking the equivalents of Hercules, Ajax, or Achilles. Perhaps the victims took on the roles of ancient combatants to be sacrificed in ways

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 142

12.1.2007 2:25pm Compositor Name: SJoearun

142 Jeffrey Quilter similar to that in which Aztec sacrificial victims were forced to impersonate gods. This discussion of portrait heads and sacrificial rituals in light of Gell’s ideas has allowed us to see some of the apparent relationships between prototypes and agents in the Moche art nexus. But as to prototype, we do not have a very clear idea as to who or what these may be for portrait vessels, and the prototype is an essential element for any further analysis. There are many questions of agency in regard to Moche portrait vessels that remain unanswered. Who commissioned the portraits and why? It seems that the likely patrons of the potters were members of the group who did the sacrificing, perhaps members of the warrior class who had captured the victim on the battlefield, but other possibilities certainly could be proposed. What were the social motivations for making pottery likenesses of the victims? Why kill a man twice, once in the flesh and again in clay? If the victims were taking on the roles of others, then the pottery portraits might have been seen to add to the verisimilitude of killing an ancient foe. Given the notion that to portray a likeness of someone is to capture part of that person’s essence, however, we might speculate that the figurine helped to extinguish the personality of the sacrificed in some kind of spiritual dimension in addition to his physical existence. The multiplicity of representations of specific individuals returns us to the notion of the distributed person. For reasons unclear, the Moche desired to portray specific individuals in more than one place. We must suspect that this is so because the many examples of portrait vessels probably were found in graves. We do not have precise information on the origins of so many vessels, but the fact that quite a few are not broken suggests that they never had been placed in plazas to be fractured by thrown rocks. They likely were excavated whole from graves. While many of these issues might be more fully comprehended through future research, these problems are likely to remain current for some time. Gell’s ideas have opened up ways to consider Moche portraiture and have given us food for thought along several dimensions. The problem remains, however, that with a corpus of art that only represents a small portion of the totality, with poor chronological and temporal controls over stylistic variations, with no written records made by or about the people who made the art, and no way to securely know how the Moche valued one pot in relation to another, we are limited in our abilities to apply Gell’s theories. Looking at a different medium of expression employed by the Moche may offer us some different possibilities, however.

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 143

12.1.2007 2:25pm Compositor Name: SJoearun

Representational Art in Ancient Peru 143

Immobile Agents: Tattoos and Wall Art The word ‘‘agency’’ carries with it the concept of movement. In his definition of the Social Agent, Gell uses words such as ‘‘outcome,’’ ‘‘sequences,’’ ‘‘events,’’ and ‘‘actions,’’ and he writes of the index ‘‘behaving.’’ Being ‘‘social’’ is about behaving, and behaving largely concerns physicality. The more ‘‘social’’ one is, whether in a colloquial or in an academic sense, the more one interacts with others, and such interactions occur through language which requires little physical movement but also through moving about in the physical world. The roles of agent-patient and the diagrams presented by Gell of their relationships express social dynamics. As social agents, paintings and other artworks may hang on the same museum walls for decades or centuries and social agency can involve more transmission of influences than their physical movement. Until recent times, however, with the advent of relatively easily reproducible art (printing, film, and computers), people – the ‘‘gallery goers’’ frequently cited by Gell – had to physically move to see art. Even today the true colors (variable in changing light conditions) and brush strokes of a Van Gogh or a Leonardo can only be appreciated by a viewer going to see the painting. So too, and especially in societies without writing, the physical movement of people is essential to social agency because it is primarily through movement through space that agency is enacted. In the case of archaeological remains uncovered through excavation, the history of movement of objects, especially of portable objects, is of primary concern. At best, archaeological context documents the last place an object was left by its prior owners. Its prior movements may be traced, at best, to a limited degree through various forms of analysis such as the study of ceramic pastes, but such studies usually do not provide fine-grained details of social agency. Movement of art-as-social-agent may be of less importance, depending on the level of analysis or the nature of the art itself. The design motifs found within the physical boundaries of the frame of a painting or the body proportions used to carve a tiki can be examined without much concern about how the larger object moved in the larger social world. Thus, etua motifs may be analyzed within and among themselves. But how a tattooed body expresses its agency as it physically moves through space, engaging in different social situations, is an important consideration to

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 144

12.1.2007 2:25pm Compositor Name: SJoearun

144 Jeffrey Quilter understanding that agency. But without knowing how objects such as portrait heads moved through the physical and social worlds in which they existed, we are limited in understanding their agency. Compared to studying portable objects, the advantage of investigating wall art is that we know where it belongs. Unlike ceramics made in one place, used in others, and deposited elsewhere, wall art remains exactly where its makers intended it to be. If enough of the art and its architectural contexts survives, we can observe which elements of art were visible from close or far away, which were in direct lines of sight, and which could only be seen after others had been viewed. Moche murals and friezes, painted in brilliant colors on adobe surfaces in temple precincts, thus offer opportunities to understand ancient art in their original contexts. They are immobile social agents and some issues of movement and agency are reduced in examining them.

Enchantment and Moche Wall Art For purposes of economy, I will concentrate my discussion on the Maritime Frieze, found at the Huaca Cao Viejo in the Chicama Valley, a masterpiece of geometric art. The Maritime Frieze dates to late in the use life of the huaca, the penultimate Phase D. About 30 meters of the frieze has been exposed (figure 6.2). The Peruvian archaeologists in charge of research at the site believe that the room was about 3 to 4 m high and 30 m square. Preliminary studies suggest that all walls in this space were decorated with the maritime frieze in its last remodeling, before it was filled with adobes in a very late building stage. In the southeastern corner of the patio, a small room is decorated with two large front-facing deities holding trophy heads in one hand and a spatula-like object in the other. The Maritime Frieze received its name because of the dominant decorative motif of what has been interpreted as aquatic life. It is hard to know exactly what the prototypes for the fish-forms were, however. A large form, with a deltoid shape, resembles a ray, while a smaller form has a split tail and lacks fins. Perhaps a careful study of the ichthyology of the eastern Pacific might reveal the source of inspiration for these designs, or they may simply be generalized aquatic creatures. Recent investigations by Peruvian archaeologists working at the huaca have suggested that a similar though notably different aquatic creature represented in some murals

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 145

12.1.2007 2:25pm Compositor Name: SJoearun

Representational Art in Ancient Peru 145

Figure 6.2 The Maritime Frieze on the interior wall of a patio on the summit of Huaca Cao Viejo.

actually is a freshwater fish found in irrigation canals. Thus, the reference in that case is not to the ocean but to agricultural fields. The creatures in the Maritime Frieze are different enough, however, to leave the question of their prototypes in doubt. For the sake of economy and ease of discussion, I will refer to examples of the larger form as ‘‘rays’’ and the smaller ones as ‘‘fishes.’’ The concepts of enchantment and distributed person can be of great use in considering this frieze. Rather than being two concepts, employed separately, they are intertwined in this case. The frieze appears to employ a permutation of geometric motifs in a manner similar to the elaborations of etua motifs so thoroughly elucidated by Gell. On close inspection, however, we will find that the principles of repetition and invention employed by the Moche are quite different than those used in the Marquesan corpus. I suggest that the Maritime Frieze enchants because, looking at it, the viewer is struck with a powerful sense of color and movement. This was so even though the Moche who looked at this art are now very far removed from us in time and culture. Despite this gulf, I believe that the employment of color and form in this frieze is such that we and the Moche have

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 146

12.1.2007 2:25pm Compositor Name: SJoearun

146 Jeffrey Quilter generally similar reactions, due to common, basic, and shared physiology and psychology in the reception of images so skillfully rendered as in the Maritime Frieze. To press my case further, we must first examine the elements and their composition in the frieze. The basic motif in the frieze (figure 6.3a) consists of two rays facing in opposite directions joined at their tails by a serrated volute flanked by two fish, also opposed. A section of the serrated volute near the final curve before it joins the tail of the ray is marked as a fish head by eyes of the same color as those of the flanking fish. It could be argued that there is a more basic motif than the one just described. It would consist of one ray, a volute segment, and a fish (figure 6.3b). The completed figure as seen in the frieze would then consist of a motif (figure 6.3b) joined to a mirror image, an image rotated 180 degrees and joined to the original (to appear as in figure 6.3a). I think that this is not the case, however, and that we must take the complete symmetrically arranged composition of opposing rays with connecting volute and opposed fishes as the elemental motif. The commonly cited Andean concept of ‘‘two inseparably joined as one’’ may be at play in this motif. The design also plays with notions of pairs of pairs which, again, is employed in Andean thought today and was in operation at the time of the arrival of the Spanish in the sixteenth century. Geometrical games are rampant in the frieze. The two elementary forms that comprise the motifs are the triangle and curve. The motifs are arranged in panels separated by diagonal frames. The volutes, of course, curl and spin.

Figure 6.3 (a) Possible ‘‘basic’’ unit of the motif of the Maritime Frieze (duplication and rotation of ‘‘B’’). (b) Possibly, a more elemental ‘‘basic’’ unit of the Maritime Frieze.

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 147

12.1.2007 2:25pm Compositor Name: SJoearun

Representational Art in Ancient Peru 147 The panels are formed by the boundaries of the upper and lower horizontal borders and the diagonal bands that define each panel at angles of 35 degrees. The basic maritime motif is arranged in a repeating series within the diagonals, appearing to pass beyond the horizontal boundaries of the frame producing a sense of an infinite design field (figure 6.4). Segments of the frieze are on the edges of those panels in which the complete motif of rays, volute, and fishes appear, while in other panels the motif is only partially present and these pieces of the larger motif create the illusion that the image repeats itself infinitely, suggesting movement. That the Moche wanted to achieve the effect of movement is underscored by the fact that they violated strict geometry when they needed to emphasize the continuity of the design beyond the limits of the frieze. The expanded motifs in figure 6.4 were made by tracing the motif as it appears on a photograph, copying it, and then pasting it repeatedly to expand the design. The darkened fish in the figure marks the actual placement of the image in the frieze but the fish above it marks where the fish should appear if strict geometrical replication were followed. It thus appears that the fish was squeezed into the frame in order to enrich the visual experience for the viewer and to register more strongly the continuation of the motif above the upper border of the frieze.6

Figure 6.4 The patterns of the Maritime Frieze drawn beyond its borders.

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 148

12.1.2007 2:25pm Compositor Name: SJoearun

148 Jeffrey Quilter The longest single straight line in the motif is the outer edge of each ray. This line helps to define the limits of the motif, relative to those above and below it. The proximity of the fish in the next motif draws the eye in a line perpendicular to the diagonal frame. As the eye crosses the border, the next motif encountered is shifted one fourth its length relative to the motif with the nearest ray in the frame of first reference. The patterning follows through the frieze even for fragments of motifs on the edges of the borders of panels, creating a vivid sense of movement. The motif sequence also plays tricks with the eye because of the way in which the two fish heads contained in the volute are placed mouth to mouth. Facing each other, they create their own design and draw attention to themselves because of the eyes that designate them as fish heads. Thus the viewer’s attention shifts from one thing to another: the motif as a whole, its rotated parts, the ‘‘new motif ’’ created by the volute fish heads, and the appearance of motifs extending above and below the frieze. While several geometrical tricks are at work in the frieze, color patterns add another level of enchantment for the viewer. The primary colors dominate the frieze although, in addition to red, yellow, and blue, pink and black are employed. Black mostly appears as a frame element, while white is used in the eyes of some rays and fishes as well as in the borders. The top border and the diagonal frames are in black and white while the basal border is a simple white band. The use of black and white as frames highlights the color of the patterns that appear to be behind them. The colors of the frieze have not been uniformly preserved while some colors may have degraded. In particular, it is hard to determine if the blacks and blues we see today were the same hues 800 or more years ago. Starting with panel 1, on the left (not quite complete due to missing its lower left-hand section), there are eight panels available for study and a ninth, fragmentary one that still reveals its color scheme (figures 6.5, 6.6). In every panel all rays were painted in the same color while all fishes were in a different color. The volute was colored differently as was the background, for a total of four different colors for each panel. In panel 1 the background is yellow, the volute is red, the rays appear to be blue, and the fish appear to be painted black. In panel 2, red, which had colored the volute in panel 1, becomes the background color while the volute itself is colored blue. Panel 3 then repeats the color sequence of panel 1. At this point, it might be expected that a pattern of alternating sequences would follow through the rest of the panels, but this is not the case. Instead, panel

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 149

12.1.2007 2:25pm Compositor Name: SJoearun

Representational Art in Ancient Peru 149

Figure 6.5 The Maritime Frieze. Panels are numbered from left to right. The incomplete panel at the far upper left designated ‘‘1A’’ serves the purposes of illusion of extension of the design field.

3 serves as the beginning of a series. If the color arrangement in panel 1 is labeled ‘‘A’’ and that of panel 2 ‘‘B,’’ then a third combination, ‘‘C,’’ begins with panel 4. Panels 1, 2, and 3 follow a sequence of A, B, A. But panels 3–9 follow a repeating sequence of A, C, B two times, ending with A, as if to begin again, in the ninth, fragmentary panel while a small section of a tenth panel has the blue background of C, continuing the sequence.

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 150

12.1.2007 2:25pm Compositor Name: SJoearun

150 Jeffrey Quilter Panel 1A 1 2 3 4 5 6 7 8 9 10

Ground B Y R Y B R Y B R Y B

Volute Y R B R Y B R Y B R --

Rays -B Y B R Y B R Y B --

Fish R BL B BL P B BL P B BL P?

Pattern C? A B A C B A C B A C?

Key: B = Blue, Y = Yellow, R = Red, BL = Black, P = Pink

Figure 6.6 Color patterns in the Maritime Frieze at Huaca Cao Viejo.

For the A, C, B pattern, the colors of the three largest elements – background, volute, and ray – appear to cycle in a distinct pattern: the color of the background (yellow in panel 3) is applied to the volute in the next panel (4, in this example) and the ray in the second panel to its right (here, panel 5). The color of the ray (blue, in panel 3) becomes the background in the next panel and then the volute in the second panel to the right. While the three main elements are cycling through the three colors, the fish appear to be following a different cycle. Starting in panel 3, the fish appear to repeat a sequence of black–pink–blue, although the designation of blue fish is somewhat uncertain. Because of the use of pink, the fish pattern is not integrated into the sequence of the three major elements. But the fish pattern is linked to the larger sequence in that pink fish are on blue backgrounds, black fish are on yellow backgrounds, and blue fish are always on red backgrounds. If the above analysis is correct, it reveals a sophisticated appreciation of and use of permutations hitherto unknown for the Moche. The length of wall now visible is 30 m. If the room was square, then about 90 m more of frieze was decorated, and, as it is likely that the room was rectangular, even more wall-space was decorated with the Maritime Frieze as these patterns are commonly repeated on all of their walls. Because we do not have the rest of the frieze available for study, we also do not know if the A, C, B pattern is maintained throughout the courtyard. Creating short

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 151

12.1.2007 2:25pm Compositor Name: SJoearun

Representational Art in Ancient Peru 151 patterns that are replaced by new, different ones would have made for a dazzling, confusing environment, but perhaps the employment of the A, C, B pattern produced enough of an effect for the purposes of those who supervised its making.7 The combined use of color and geometry created a dazzling, moving scene to the viewer confronting the frieze who likely did not have the time or the inclination to stop and analyze the formal organization of what was being witnessed. Indeed, the experience of being in a multicolored room must by itself have been astounding to people whose lives mostly were lived in the dun colored landscape of the Peruvian desert. The effect of shadow playing through the frieze likely intensified the experience of fascination, as did any psychotropic drugs that may have been consumed prior to entering the precinct and which we know were part of at least some rituals. The formal analysis just presented suggests how a technology of enchantment operated. The etua motifs discussed by Gell ‘‘evolve’’ from a few elemental forms into complex ones. In the case of the Moche Maritime Frieze the starting motif (whether the entire motif or ‘‘half’’ of it) is quite complex to begin with. Further complexity is not produced by a transformation of the motif itself. To the contrary, the basic motif is a constant. The technical means to produce complexity and its ability to enchant was to astound with color, pattern, and movement. A principle of what might be called ‘‘elusive symmetry’’ was employed in which symmetry in different components of the composition is present in different ways, often ones that are at odds with one another. But while the walls ran riot with color and motion, the underlying pattern was the same, though complex. The Marquesan tattoos wrap bodies, covering a biological being to make it social and projecting itself outwards to other bodies with other tattoos. Carried on the flesh, the designs are mobile, moving through social and physical space. The Moche frieze encloses social units (the group participating in rituals) providing an intense experience, circumscribed by limited space and time, to a limited audience. Despite their very different social roles, both Marquesan tattoos and the Maritime Frieze are works of art that are technologies of enchantment and enchant us through the skillful use of technique. In terms of distributed personhood, the frieze shares some of the features of an ‘‘interlocking’’ style that was widespread throughout Peru during Moche times. The distributive aspect of the frieze lies mostly in the seemingly endless chains of motifs that expand beyond the frames that

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 152

12.1.2007 2:25pm Compositor Name: SJoearun

152 Jeffrey Quilter contain them in the room. Indeed, when the enchantment of the frieze is working, the frame serves merely as a barrier separating the viewer from a larger world beyond, much like a long, narrow window at the bottom of a huge aquarium tank.

Conclusion The enchantment/technology and the distributed person perspectives suggested by Gell have offered ways in which to think about Moche art, especially regarding how Moche art may have worked – how it enchanted and how social agents interacted. The idea of the art nexus was less satisfactorily applied in the examples discussed above. Gell’s ideas of art as social agents are best applied, I believe, in cases in which we know how other social agents were constructed and how they acted in their social worlds. We do not have much knowledge of these things when it comes to the Moche. An anthropological approach to art is only as good as the anthropology employed. We can see large-scale agency at work. In the final building stages of Huaca Cao Viejo and at Huaca de La Luna, high abstraction in wall art was abandoned in favor of representational art. On the plaza-level wall of each huaca, friezes were made of bound, naked prisoners marching under escort of military guards while on the level above, officials were depicted, holding hands as if cordoning off the parade (figures 6.7 and 6.8). The prototype of the prisoner frieze was an actual parade of prisoners through the plaza. We have plenty of archaeological evidence that prisoners were taken and sacrificed and their depiction as naked and suffering maltreatment (broken noses, cut fingers and genitals) conforms to other archaeological information. Such scenes are also portrayed on ceramic vessels, so we cannot be certain if depictions on pots or the wall art serve as intervening prototypes: that the specific ways in which the scenes are rendered are derived from another art form rather than from the real-life parade. This subject is too complex to discuss at length here, but the implementation of representational styles in public art late in the Moche era, when the society was facing severe economic, environmental, and social problems, is of interest and is an expression of art as a social agent. There is a tyranny of the word in our understandings and our academics. All of our conclusions and musings are ultimately written out in books

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 153

12.1.2007 2:25pm Compositor Name: SJoearun

Representational Art in Ancient Peru 153

Figure 6.7 The Prisoner Frieze at Huaca Cao Viejo. At the time it was made, the base of the frieze was at ground level. Excavations have uncovered early phases of architecture at lower levels.

or in articles such as this one. Art history, itself, by its very title is historical and therefore intimate with the written word. But what about art prehistory? Gell’s discussions of Marquesan tattoos, Trobriand canoe prows, and Duchamp’s paintings ultimately need texts. This is not to sell him short, because his analyses extend well beyond the text to look at designs and objects themselves and how they transform or act as agents. But texts nevertheless enter into analyses to greater or lesser degrees in Gell’s various case-studies. So too, it is ironic that while Gell criticizes formal analysis in the early chapters of Art and Agency he employs the method at length in the later chapters of the book. Perhaps the grave state of his health

Figure 6.8 Roll-out photograph of the Prisoner Frieze.

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 154

12.1.2007 2:25pm Compositor Name: SJoearun

154 Jeffrey Quilter prohibited him from explaining this contradiction as he rushed to complete the work. To use the theory of the art nexus, or any theory that attempts to look at art as a social phenomenon, we need to know as much as possible about indexes, artists, recipients, and prototypes. Texts provide this information, generally, in greater abundance than does archaeology. In the Moche case, we have the indexes in the form of portrait head vessels or representational friezes. We know that there were artists who were responsible for creating the indexes but we don’t know much about their motivations for making them or why they decided to imbue them with some characteristics as opposed to other possible choices. We know even less about the recipients. We know that there were recipients, of course, and wall art is at least immobile, allowing us to consider how the recipients physically (mostly through vision) received the agency of the index, but anything else we may say is rather tentative. Although wall art doesn’t move, the people who view it do. The question then shifts from how the art moved through space to how people moved through space to view it. In terms of social agency, the question becomes who was permitted to view the art. We can say with some confidence that the inner sanctums of temples were probably viewed only by a few high-ranking members of society, while the art in the plazas was probably seen by a wider range of the members of society. That the representational style of the Prisoner Frieze was carried out in the most public place in which temple art was shown is therefore indicative that its agency was different – likely, more propagandistic – than the art in the remote precincts on the temple summit, such as the Maritime Frieze. Representational art makes the claim that it captures what the portrayed object is in a way that is different than other art styles that may make the same claim but achieve it differently. It claims that the portrayal is the same as the thing portrayed and that the sameness lies in the appearance of the thing more than (as much as?) in unseen qualities. It is the insistence on the visual aspects and semblances of the portrayed and the portrait that can justify the techniques employed and allow outsiders, perhaps, to judge whether it is successful or not. It is, perhaps, for this reason that representational prehistoric art has a special appeal for contemporary viewers. Gell opposes approaching art as symbolic communication. His protest carries the force of argument in the Andean case. Most studies of Andean art, today and in recent times, follow one of two lines of investigation. Some scholars drift steadily toward late prehistory where they can employ

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 155

12.1.2007 2:25pm Compositor Name: SJoearun

Representational Art in Ancient Peru 155 the accounts given by the Inca to their Spanish conquerors regarding their society, customs, and art. This, then, is art ethnohistory, relying upon historical records to project back in time, beyond the reach of documents but within sight of them. The other school relies upon the interpretation of symbolism, drawing mostly upon well-worn Andean concepts of dualism and complementary opposition, among others, to attempt to ascribe how images communicated messages or expressed Andean ideas. Again, the scholar invariably reaches toward the written records of the early Colonial Period or even recent ethnographies to find the codes that will provide the reading of the symbols of art created centuries before such records were made. We may not be able to be very successful in applying Gell’s theories to truly prehistoric art which has no or only very tenuous connections to written sources. In the absence of texts, formal analyses do offer us a way to see how art was put together, how it was displayed, how some choices were made and not others. While Gell’s approach dismisses formal analyses, I believe that they do provide an important baseline of understanding that can then be placed within the matrices of his approach, and, if we are to attempt forge an art prehistory we must use as many tools as we can find to do so.

Acknowledgments I could not have done this study without the very kind support of my Peruvian colleagues who have given me access to their sites and their data. In particular, Re´gulo Franco Jordan, Ce´sar Galvez Mora, and Segundo Vasquez, co-directors of the Huaca Cao Viejo project, were most generous in inviting me to participate in research at the site. Marco Aveggio of the Fundacio´n Wiese (which has funded most of the research at the site) and his wife Marina also offered me many kindnesses while I was in Peru. The National Geographic Society funded the research that allowed me to take many of the photographs that appear in this article. There, John Echave, illustrations editor for the magazine, was very kind in many ways in supporting my work. The magnificent ‘‘roll-out’’ photographs of the friezes (not fully done justice here, because of scale) were made by project archaeologist and photographer, Hal Starratt, and I thank him very much for his fine work. William Doonan also provided essential service, in the field, by attending to matters that allowed me to study the friezes.

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 156

12.1.2007 2:25pm Compositor Name: SJoearun

156 Jeffrey Quilter Thanks to Nicholas Saunders for first introducing me to the work of Alfred Gell. Joanne Pillsbury greatly helped in focusing my attention on Gell’s work. Thanks to Irene Winter for encouraging me to explore Gell’s theories in relation to Moche art, and to the editors of this volume for inviting me to write an article for it. Robin Osborne was particularly helpful in staying in touch and offering useful comments that strengthened the revised version of this manuscript. Michel Conan, Director of Garden and Landscape Studies, has helped me to clarify my ideas and offered many stimulating ones of his own. Tom Cummins provided extensive and useful comments on the first draft of this chapter. Stephen Zwirn, Assistant Curator for the Byzantine Collection at Dumbarton Oaks, was very kind in advising me on Roman examples of head-flasks, as was Ralf Behrwald, Fellow in Byzantine Studies for 2003–04. Bryan Just, Junior Fellow in Pre-Columbian Studies, 2003–4, is greatly thanked for his ideas and his enthusiasm in discussing Gell’s theories. As the only person in the hallowed halls of Dumbarton Oaks to possess a protractor, Mike Steen, Project Manager for New Library Construction, was most generous in lending me his protractor to measure the angles of the Maritime Frieze.

Notes 1

2

3

4 5 6

For details on the fine points of Moche artistic conventions see C. B. Donnan, Moche Art of Peru: Pre-Columbian Symbolic Communication (Los Angeles, 1978). G. S. Russell and M. A. Jackson, ‘‘Political economy and patronage at Cerro Mayal, Peru,’’ in J. Pillsbury, ed., Moche Art (Washington, DC, New Haven, and London, 2005), pp. 159–76. Christopher Donnan claims that the most lifelike portrait vessels were made only in phases III and IV. See Donnan, ‘‘Moche ceramic portraits,’’ in J. Pillsbury, ed., Moche Art and Archaeology in Ancient Peru, Studies in the History of Art 61 (Washington, DC, and New Haven, 2001), pp. 127–39, and C. Donnan, Moche Portraits from Ancient Peru (Austin, TX, 2004). Donnan, ‘‘Moche ceramic portraits.’’ J. Verano in Pillsbury, Moche Art, pp. 111–26. The small unnumbered triangle to the left of panel 1 appears to represent a section of a panel behind the small room in the corner of the patio. The motifs in this triangle occupy a small space yet hint at the larger pattern of rays and volutes.

Osborne / Art’s Agency and Art History 1405135379_4_006 Final Proof

page 157

12.1.2007 2:25pm Compositor Name: SJoearun

Representational Art in Ancient Peru 157 7 I suspect that the A, C, B pattern is maintained, however. I also suspect that the frieze was created moving from left to right, starting with panel 1 or the short sections of panels on the adjacent small room to the left of panel 1. I have no reason to believe this other than a hunch. The hunch is partly based on the sense of the diagonals moving upwards, from left to right. If correct, this would mean that the Moche worked from left to right, at least in this case.

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 158

12.1.2007 2:25pm Compositor Name: SJoearun

7

Gell’s Idols and Roman Cult1 Peter Stewart

The relationship between classical studies and anthropology has been close for a century or more.2 In particular anthropology has long had an enormous influence on the study of Greek religion, mythology, and society. This no doubt largely explains why it has also affected classical art historians interested especially in the rich and enigmatic repertoire of Athenian vase-painting, with its countless representations of myth and cult, and in its heyday structuralism played a prominent role.3 Yet despite these contributions, the impact of anthropological approaches on other areas of classical art, even religious art, is not pronounced.4 This paper examines a few of the ways in which Alfred Gell’s ideas might be used to view Greco-Roman cult images – a somewhat neglected arthistorical subject – from his particular, sometimes idiosyncratic, anthropological perspective. I am concerned primarily with their application to iconic (in this context that means anthropomorphic) and aniconic cult statues in the Roman Empire (though generally speaking the relevance is the same for earlier Greek religious art). Here we are dealing with objects of great ritual significance, which have usually been interpreted very selectively within an essentially art-historical or archaeological framework, for example with attention to date, iconography, typology, and artistic genealogy (this applies almost as much to the aniconic images as to the iconic, for the antiquity and origins of such forms are of primary interest).5 In that sense their interpretation has been highly rationalistic, occasionally acknowledging but showing little interest in antiquity’s treatment of such images.6

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 159

12.1.2007 2:25pm Compositor Name: SJoearun

Gell’s Idols and Roman Cult 159 For in the ancient world these objects were ‘‘idols’’ (to use Gell’s terms) – the original ‘‘idols’’ (eidola) in fact – and although Gell himself does not deal much with classical material,7 cult images of the Roman world have a great deal in common with, for example, the Hindu images that he does discuss as illustrations of his theories. Greco-Roman cult images were revered as divine; in various cases they were regularly clothed, fed, given offerings and sacrifices, decked with garlands, kissed, consulted for opinions, and asked for favors; they were painted, tended, and anointed; occasionally they were considered to move, speak, sweat, and bleed; some were bound or chained up as if to keep them in their place or under control.8 However, while I wish to offer some examples of the assistance Art and Agency can provide in the understanding of Roman cult images, it is not exactly my intention to criticize the aims and interests of conventional studies. In fact, the confrontation of anthropology – in this case an anthropology which relies heavily on traditional anthropological, ‘‘ethnological’’ material but has more global ambitions – and the conventional territory of the history of classical art raises important questions about the general applicability of some of Gell’s ideas. The confrontation is certainly salutary for the classicist or classical art historian, but ultimately a key question must be: to what extent is Art and Agency adapted for the analysis of cultures which have their own sophisticated awareness of the sorts of issues the anthropologist himself is pursuing? Are there historical and cultural limits to some of the ideas Gell expounds?

Anthropology of Images and History of Art Gell’s stated aim is to produce an anthropological theory of art which looks like an anthropological theory.9 What this amounts to, in his words, is a theoretical study of ‘‘social relations in the vicinity of objects mediating social agency,’’ and this is carried out with a ‘‘biographical depth of focus.’’ Yet in other, less clearly articulated ways Gell’s book looks anthropological. It relies most heavily on ‘‘non-Western’’ material, especially on ‘‘art production in the colonial and post-colonial societies anthropologists typically study.’’10 To a greater extent than his previous work, Art and Agency does attempt to assume a global relevance. Yet it remains very obviously oriented toward the ‘‘ethnological’’ material which was Gell’s specialism, notably Polynesian imagery.11

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 160

12.1.2007 2:25pm Compositor Name: SJoearun

160 Peter Stewart Like other anthropologists, Gell employs ‘‘non-ethnological’’ examples from complex societies in an interesting way. They are naturally used to test (or rather demonstrate) the broad applicability of the theories concerned across cultures (Western culture, complex Asian cultures). At the same time, a number of the examples, especially those that concern the modern Western experience of life and social interaction, assume a contrast between the ‘‘primitive’’ cultures under examination (Gell uses that term) and ‘‘our’’ experience. The point is partly to blur the contrast: to show in unexpected ways that the concepts required to explain Polynesian art (etc.) are entirely relevant to our own lives: that we are not as different as we might assume. This seems to suggest an assumption that, in fact, modern Western art and society generally resist interpretation in the terms Gell proposes. They occupy their own, historically defined domain. They offer illuminating analogies but there is often an implication that these remain merely illustrative. In fact there is an inherent tension in Art and Agency between its generalizing ambitions and its frequent, explicit, and somewhat provocative acknowledgment of the differences between academic disciplines and the methods they generate for themselves. The message is often that suchand-such an approach may be fine for art historians, but not for the anthropologist.12 So we are surely invited to ask how easily Gell’s own framework is transferred to the classical world. This question relates to longstanding debates about historicity in anthropological studies.13 But the issue in the case of Art and Agency, with its shorter ‘‘biographical’’ depth of focus and its distrust of grand schemes, is rather simpler. It is partly a matter of rhetoric perhaps: a sometimes tacit construction of boundaries which serves to organize and simplify the material under consideration, but which deserves to be questioned. Moreover, even while Gell resists the idea of culturally specific theories of art and readily employs work from other disciplines, including the history of art, one of his main objectives is to challenge or avoid those theoretical models devised with reference to Western civilization and inappropriately transferred to the cultures that concern him. In particular, he rejects the use of Saussurean semiotics, linguistic analogies, and, implicitly, iconography.14 Other distinctions are maintained between Gell’s ethnological cultures and Western cultures on less theoretically developed grounds and at the risk of impoverishing the non-Western cultures. It is perhaps unfair to pick on a parenthetical comment in the book, but one of the most

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 161

12.1.2007 2:25pm Compositor Name: SJoearun

Gell’s Idols and Roman Cult 161 revealing comes in reference to the beliefs of the reformist Hindu Arya Samaj sect. The founder, Dayanand Saraswati, is said as a child to have been struck by Shiva’s lack of response when a mouse ran over the Shiva linga and ate the offerings – an experience that informed his opposition to images.15 Gell suggests that the early use of this story in arguments against idolatry ‘‘could have been convincing only to one who had already (no doubt as a result of Christian-Protestant ascendancy in British India) decided that idol-worship was backward and futile.’’16 It is true that Arya Samaj originated and flourished in the political and cultural context of the Raj, and in reaction against it, yet it is misleading to imply that opposition to idol-worship was in some sense an import. In doing so Gell manages to rescue mainstream, authentic Indian responses to cult images for use in the anthropological theory that looks like an anthropological theory. But, as Richard Davis describes, intellectual opposition to imageworship within Hinduism (opposition often reminiscent of some GrecoRoman writers) has an ancient history.17 Paradoxically, a certain kind of historicism seems to emerge from such underlying assumptions. There is an understanding that the conventional territory of art history, with its rational attitudes to images, is itself historically bounded. Gell ventures into this territory, enthusiastically (and successfully) seeking stimulating illustrations of his theoretical principles, but it remains fundamentally foreign. In the end, his own ethnographic material provides a more appropriate ‘‘innocent’’ subject for analysis in anthropological terms. In fact, to an extent it is analogous to medieval art as viewed by Hans Belting in his influential work, Likeness and Presence, although that is a consciously untheoretical book.18 For Belting art in the modern sense is also a historically circumscribed construct. Before the Renaissance we find the ‘‘era before art’’ – an era of images (such as Byzantine icons) which were powerful or charismatic and which, like Gell’s indexes, can be seen as active participants in medieval society. Belting shares some of Gell’s concerns and both display an interest in the related work of David Freedberg on that ‘‘power of images’’ which is belied by conventional art history.19 Belting pays little attention to art and attitudes to art in classical antiquity, so that his historical framework is in fact partially limited. It would be somewhat harder to construct a modern ‘‘era of art’’ if his attention to the Roman world extended back beyond late antiquity. For while Roman responses to images were pre-modern – they may have been ‘‘irrational’’ in many ways and they certainly do not correspond precisely

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 162

12.1.2007 2:25pm Compositor Name: SJoearun

162 Peter Stewart to the modern Western concept of art – they nevertheless have too much in common with that perspective to be confined in a kind of anthropological crucible. The same problem applies in part to Gell’s approach, so that an examination of Roman cult images will help to show how far Art and Agency can contribute and where some of its own limitations lie.

The Agency of Images in Roman Cult In spite of my preliminary comments, it is clear that the Roman Empire does indeed offer cases that invite consideration in Gell’s terms. Roman religion was polytheistic, and central to it were images of gods which were effectively treated as if alive and were certainly positioned as partners in social relationships. (Greek and) Roman cult has much in common with Hinduism and we can probably derive a rather better impression of what Roman cult images were actually like from the modern Hindu treatment of divine images than from the bleached, classical, marble sculptures elevated as works of art in museum collections. As I have already indicated, these images were treated ‘‘as if ’’ they were in some sense alive – as if they embodied the deities represented and made them present. It is possible to explain away such irrational responses to images as ‘‘play-acting’’ or ‘‘make-believe.’’20 At the opposite extreme one can imagine the ancient users of images really believing that the images are animate. There are certainly some sources that attest to both attitudes, though the reality of Romans’ beliefs was undoubtedly more complex and marked by apparent inconsistencies and contradictions.21 Gell’s approach is rather different. By abolishing the practical distinction between works of art and other, living participants in the nexus of social interaction, he helps us to see how the question of the ontological beliefs concerned may simply hold little relevance to the way images are regarded in the context of social life.22 This means that the authenticity of beliefs about the images’ animacy is not an issue. At the same time, however, Gell does not ask us to assume that idolaters are unaware of the ‘‘genuine’’ nature of the images venerated. This too is an important point, but one easily overlooked.23 The similar position of an artifact and a person (or a god) within social relationships does not make them the same thing. Those who worship a cult image may readily accept that it is not a living creature. But in certain exchanges it is considered to act as one. So it would be misleading to present the attitudes of idolaters as the projection

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 163

12.1.2007 2:25pm Compositor Name: SJoearun

Gell’s Idols and Roman Cult 163 of animacy onto inanimate objects. Nor is ‘‘make-believe’’ the right term for this behavior. We need to consider the extent to which all interaction, even between humans, is based on the abduction of agency. Humans are usually more convincing as agents because they do a wider range of more complicated things and respond to each other with spontaneity. But generally speaking objects like cult images are not expected to be either complicated or spontaneous. Anthropomorphism was usual in Roman cult images, and perhaps it made them more convincing as social partners, partly because it made them resemble living human beings who are accepted to perform as agents. Yet Gell usefully disengages the pursuit of artistic naturalism or illusionism from the basic functioning of cult images:24 When particular statues bleed, or perspire, or move about, these are ‘‘miracles.’’ But such happenings would not be miracles if the expectation was that all idols should behave in this way; in fact, they are generally expected not to. Idols may be animate without, in other words, being endowed with animal life or activity. In addition, there are automata, real or imaginary ones, which do really move, speak, and perform various human-like actions, but these are remarkable, not because they are alive, but because they are not alive, while maintaining the appearance of being so. It follows that ‘‘ritual’’ animacy and the possession of ‘‘life’’ in a biological sense are far from being the same thing.

Virtuoso artistic illusionism was often admired in Greco-Roman antiquity, but the numerous ancient expressions of wonder at the lifelikeness of works of art always start from the premise that artworks are not lifelike.25 A useful illustration of this paradox is provided by the various cases of ‘‘agalmatophilia’’ (love of [cult]-statues) described in classical literature. Most famous is the story of Pygmalion himself, who (in Ovid’s account) fell in love with an ivory female statue of his own making.26 She was subsequently brought to life by the miraculous intervention of Aphrodite. The circularity of agency in the story surely lends itself to one of Gell’s formulae (the only one in this chapter):

[[Artist-A] ! Index-A] ! Artist-P The story of the Aphrodite of Cnidus is somewhat similar. This renowned cult statue was the work of the famous Praxiteles in the fourth

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 164

12.1.2007 2:25pm Compositor Name: SJoearun

Figure 7.1 Roman imperial, marble copy of the Aphrodite of Cnidus by Praxiteles, original c.340s BC; ht. 2.04 m (Vatican Museums). Courtesy of Deutsches Archa¨ologisches Institut, Rome, Inst. Neg. 66.2278 (Koppermann).

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 165

12.1.2007 2:25pm Compositor Name: SJoearun

Gell’s Idols and Roman Cult 165 century BC (figure 7.1). It was notorious for its sensuous representation of the goddess’s nudity, and ancient sources dwell on its sexual power, which was a departure in the history of classical art, though entirely appropriate for the embodiment of the goddess’s own affective influence. Two of the sources reproduce a story that illustrates the image’s power. Pliny tells us of a lover who hid himself in the temple at night and attempted to make love to the statue (leaving a stain on it as evidence of his lust). PseudoLucian’s Amores adds some detail, including the tradition that the youth finished by throwing himself into the sea.27 Such tales are typically used as mythical examples of the power of mimetic art, and in the case of the Aphrodite of Cnidus at least they make an important point about the potency of classical developments in naturalism.28 And yet, in fact, the stories simultaneously insist on the limits of mimesis. For instance, Pygmalion’s statue is animated not by art but by divine intervention; its illusionistic allure was sufficient merely to captivate the artist himself, as illustrated by the diagram.29 The Cnidian Aphrodite’s thwarted lover is ruined by his frustration. At the same time one can also find other, more surprising evidence of lovers supposedly besotted with aniconic or semi-iconic objects. For example the first-century Roman former-consul Passienus was said to have adored a beech tree sacred to Diana, near Tusculum.30 Such examples illustrate Gell’s contention that iconic and aniconic objects can occupy the same space in social relationships and that iconic form is not essential for this perception of art. There is, in fact, a little evidence for the further artificial ‘‘animation’’ of Roman cult-images through mechanical devices or by magic. Such statues as were deemed capable of speech or movement were not necessarily taken more seriously as cult images: quite the contrary. Theurgy, the magical animation of statues, occupies a special space in Roman religious thought. It most certainly does not belong to the mainstream activities of cult.31 Meanwhile, mechanical manipulation of cult images, which might be perceived, and perhaps was perceived by some, as extending the mimetic possibilities of art beyond the merely visual and tactile, appears in the literary sources as an illustration of fraud and deceit. Admittedly most of the comments come from the Christian writers, but not all of them. Lucian, for example, is similarly dismissive of the tricks used by the fraudulent holy-man, Alexander of Abonoteichus, in the mid-second century AD. These included the animation of a human-headed divine snake called Glycon by means of a hidden tube within it (a crane’s

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 166

12.1.2007 2:25pm Compositor Name: SJoearun

166 Peter Stewart windpipe), through which an assistant could pronounce the god’s oracles.32 Confirmation of the technique appears to come from an extraordinary marble bust of the venerated philosopher Epicurus, found in Rome and now in Copenhagen.33 The inside of the head has been hollowed out from the rear, as if to permit a similar kind of ventriloquism. In light of such cases Gell’s shift away from iconic representation offers a helpful corrective to conventional assumptions about the ‘‘power of images’’ and the purposes of mimesis. They are perhaps rather exceptional examples, but it is also important to note that in general Roman responses to cult images were not connected in any regular or predictable way to the particular form of the images. It has been demonstrated that aniconic and iconic cult objects were venerated in similar ways in the Roman world.34 Sacred stones (including meteorites) were treated in fundamentally the same manner as highly elaborated iconic cult statues by noted artists. Both occupy the same social position in Roman imperial cult, though they were evidently recognized as very different kinds of representation, mediating differing forms of agency, and their local circumstances and historical origins were naturally distinct. Neither kind of representation had historical priority – indeed some aniconic cults seem to have flourished or even originated in the period of the Roman Empire. So whatever anthropomorphism offers as means of giving substance to absent deities, it does not at first sight seem to be fundamental to central rituals of Roman cult – to the offering of sacrifices or the veneration and cultivation of the god through his or her physical extensions.35

The Problem of Iconography Despite the evident value of Gell’s anthropological models for casting the complex and extensive religion of the Roman Empire in a new light and making it understandable on a human level, they also display weaknesses. For example, Gell’s resistance to the idea of iconography impedes the attempt to explain anthropomorphic cult images. He suggests that any form of cult object might ‘‘represent’’ the god without needing to depict him or her, and this assumption is indeed substantiated by the coexistence of iconic and aniconic cult images in the Roman world. To explain this concept of representation Gell uses the analogy of the ambassador, who represents a country without actually resembling it.36 However, if we use the national flag as the example of the same relationship, we can see that

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 167

12.1.2007 2:25pm Compositor Name: SJoearun

Gell’s Idols and Roman Cult 167 there is an important difference, for the ambassador is to be located only in one place and time, to be approached (rather like a cult image) through rituals and conventions like embassy receptions. The same applies to any one particular flag, but not to ‘‘the flag’’ which can be instantiated in any number of pieces of cloth (not to mention other media). Essentially the flag introduces into the terms of Gell’s analogy a simple concept of iconography. The aniconic cult image, like the ambassador, is frequently unique (figure 7.2). Some such images have features in common, like conical shape, and it is true that aniconic or semi-iconic images can be reproduced – they can have an iconography (the Hindu Shiva linga is a good example). But generally speaking these are local cults and the literary sources consistently emphasize their local peculiarity. In contrast, iconic, anthropomorphic statues of gods, despite the association of particular works with individual places or artists, are infinitely reproducible and distributable, as long as there is adherence to iconographical principles. As Richard Gordon puts it, ‘‘where the aniconic image is utterly particular, non-generalizable and non-reproducible, the iconic image is ‘universal’: it

Figure 7.2 Reverse of bronze coin from Roman Cyprus (reign of Septimius Severus, c.AD 193–211): the conical cult image of Aphrodite in its sanctuary at Paphos. ß The British Museum.

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 168

12.1.2007 2:25pm Compositor Name: SJoearun

168 Peter Stewart can generate as many forms as need to be conceived.’’37 In most of Art and Agency Gell is primarily concerned with localized relationships between people and works of art.38 He dismisses systems of meaning in art as inaccurate abstractions.39 Yet we can see that iconography is not merely a traditional art-historical abstraction, divorced from the social realities of the ‘‘art nexus.’’ In the case of Roman cult images, which in other ways share so much with the ‘‘idols’’ that occupy Gell’s attention, their adherence to an iconographical system has a clear effect on the ways in which they can operate. Iconography has a social reality. The iconography of gods makes them easy to represent and recognize in different places and circumstances: it allows them to be conjured into existence even without the sorts of rituals of animation discussed by Gell.40 But it also makes for significant relationships between different anthropomorphic cult images and, notably, it allows very specific images – such as particular statues – to multiply. This can be seen on a small scale from the story of the Aphrodite of Cnidus itself, as told from Pliny’s Roman perspective.41 For even as his account emphasizes the unique importance and individual aura of the famous cult statue (‘‘superior to everything . . . in the whole world’’), it records Praxiteles’ simultaneous manufacture and sale of ‘‘two [Venuses]’’ – the famous Aphrodite, which was rejected by the Coans who had first refusal, and a presumably similar but clothed version that satisfied their scruples. From its very creation therefore, this special cult image, the work of a master sculptor, with an apparently unprecedented sensual appeal, was a kind of duplicate or alternative. The famous and distinctive, archaizing cult image of Artemis at Ephesus offers a further, larger-scale illustration of how recognizable iconography permitted, as it were, the ‘‘ramification’’ of a divine image. Like many cult statues (including the Aphrodite of Cnidus), the Ephesian Artemis was reproduced in many different formats and media in the Roman period. But already centuries before (we are told), the goddess’s image at Ephesus was consciously replicated, partially at her own bidding, in the Phocaean colony of Massalia (Marseille), and thereafter in its own colonies.42 This sort of (politically significant) replication was possible despite the statue’s schematic archaism, which was no doubt intended to enhance its aura of unique sanctity. Ephesian Artemis was very much a local divinity, but rendered distributable by her striking iconography. So can iconography be reconciled with Gell’s theories? One can of course resort to a cumbersome and uncomfortable elaboration of the ‘‘art nexus,’’ whereby iconography is rendered not as an abstract system

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 169

12.1.2007 2:25pm Compositor Name: SJoearun

Gell’s Idols and Roman Cult 169 but as an infinitely complex network of prototypes as agents, and so on. It is hard, however, to see how this would be an improvement on the existing abstractions of art history or semiotics. But some of the ideas Gell propounds have a closer relevance to the workings of iconography. Indeed the principles of iconography lie just under the surface of his discussions toward the end of the book. Gell’s notion of the distributed person is perhaps of particular help in framing Roman cult images as a collectivity.43 He sees in the idol (for example) not merely a visual representation of the deity, but a bodily extension of its personhood – a sort of prosthesis. In one sense it owes its existence as an idol to the agency of the god, or at least we abduct the agency of the god from such physical forms. Gell offers numerous examples across cultures for this perspective on representations: from the ‘‘simulacra’’ of Epicurean philosophy which are deemed literally to carry the appearance of an object to the eyes of the viewer44 to the ‘‘exuviae’’ of sorcery – the cast-offs that allow the sorcerer to exert control over the (distributed) person from whom they emanate. If all of this sounds like intellectual sleight of hand, then consider a more down-to-earth example. I have suggested elsewhere that Gell’s idea of distributed personhood applies neatly to Roman portraiture, and specifically to the portraits of the emperor that were indeed distributed all over the empire.45 To the inhabitants of the Roman Empire, most of whom would never have seen the ruler, these portraits embodied his presence in their lives. For most Romans the emperor himself, insofar as one can speak of a ‘‘real’’ emperor distinct from these manifestations of his authority, operated ‘‘behind the scenes.’’ His existence was to be discerned from its physical symptoms: letters, laws, the images themselves. It is hardly surprising that the ruler portrait was treated as a proxy, presented with petitions, appealed to for asylum, venerated, and perhaps feared.46 The emperor was analogous to the absent god represented in cult images. Indeed, the living emperor was worshipped in just the same way in many parts of the empire, his portraits serving as the focus of cult.47 Imperial portraits were generally not commissioned or actively distributed by the emperor himself, but the images evidently conformed to official, centrally produced types that were perhaps distributed in the form of models for sculptors. Thus classical art historians have found it relatively easy to indulge their penchant for typology and classification by categorizing imperial portraits within a sequence of types. The typology

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 170

12.1.2007 2:25pm Compositor Name: SJoearun

170 Peter Stewart of Roman imperial portraits is not beyond criticism as a methodology, but it represents a scholarly abstraction that conforms to the social reality of the portraits’ operation in the imperial provinces.48 Adherence to types made imperial images identifiable. In more subtle ways it also implied connections between images. Thus the complicated family relationships of dynasties like the Julio-Claudians of the first century AD were homogenized beneath, in this case, a classicizing veneer that implied their connection with the emperor Augustus and the values associated with him. Conversely, changes in portrait styles and the formulation of new portrait types could naturalize ideological shifts, even within an emperor’s reign (as, arguably, with the successive portrait types of Augustus himself 49). What is important is that all such ramifications were rooted in the putative existence of a real prototype implied by the proliferation of congruent images. Even without typology – even without any adherence to established and identifiable types – imperial portraits by their very nature would have announced themselves as extensions of the emperor’s personhood, simply because it is in the nature of portraits to refer back to a prototype. One might say that portraiture as a genre notionally exists to extend personhood beyond the natural constraints of time and space.50 But the typological, iconographical character of imperial portraits as a collectivity greatly extended their potential for distributing the revered ruler and his family. The same principles apply to more conventional, iconic cult images. The gods have no reproducible facial likeness in the manner of imperial portraits, but since such portrait likenesses were fictionalized and might as well have been fictional, there is little fundamental difference. In summary, therefore, we can extend Gell’s concept of the distributed person, conceiving of gods embodied not merely by individual idols but by whole collectivities of iconographically identifiable cult images which facilitate their distribution and partake in their very identities as omnipresent, multifaceted deities. (Gell himself extends the concept in the last two chapters of his book to help explain styles and corpora as unities which are embodied in each one of their constituent elements.) In connection with all these issues it is worth briefly noting two, quite self-conscious, Roman philosophical comments on (divine) iconography, both of which have a bearing on how we perceive iconography as a system. Firstly, Cicero, in his mid-first-century BC dialogue, De Natura Deorum (1.81), offers an extended acknowledgement of the reality of

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 171

12.1.2007 2:25pm Compositor Name: SJoearun

Gell’s Idols and Roman Cult 171 iconography in people’s lives: ‘‘from childhood . . . we know Jupiter, Juno, Minerva, Neptune, Vulcan, Apollo, and the other gods, by that appearance which painters and artists wanted them to have; and not only their physical appearance, but also by their adornment, age and dress.’’ Cicero, it should be stressed, is here presenting a skeptical, anti-Epicurean argument, which emphasizes the arbitrary nature of such conventions, likening them to language, and divorcing them from the nature of gods themselves. A century or so later, and in a different philosophical context, Seneca uses what we should call statuary iconography or typology as an illustration of classical theories of causes.51 A series of causes can be regarded as leading to the existence of, say, a statue, in the particular form that it takes. For Aristotle and Plato the necessary causes include the material, the maker, the form of the work, and its purpose. To these Plato adds the idea or ideal pattern instantiated by the individual object. According to Seneca’s down-to-earth, artistic analogy, the ‘‘form’’ of the work is presented as its content or iconographical type (specifically, the form without which the statue could not be called ‘‘doryphoros’’ or ‘‘diadumenos’’ – an allusion not only to general subject-matter, but also to two famous and frequently replicated sculptures by Polykleitos). The ideal pattern is presented either as an internal, mental schema or as an external model that the artist can imitate. None of this relates explicitly to the representation or veneration of deities, but it is theoretically significant and useful, for it demonstrates a reconciliation between what we might consider iconography or typology and something like the processes of abduction explored by Art and Agency. Gell resorts little to ancient philosophy in his discussions. But the classical theory of causes, especially as applied by Seneca to Roman art, is so reminiscent of his concept of causation and the ‘‘art nexus’’ that it cannot be left unmentioned.

The Problem of Roman Beliefs The consideration of Roman cult raises a second problem with Gell’s attempt to demonstrate the reasonableness of idol-worship. His theoretical model can help to explain Roman cult-participants’ immediate responses to the objects with which they engage, but it is less obviously suitable – and is not employed – to explain the great variety of often contradictory responses that coexist in many cultures, and which are

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 172

12.1.2007 2:25pm Compositor Name: SJoearun

172 Peter Stewart manifest in the literary record of the Roman world. Gell acknowledges contradictory responses in our own society – children’s animation of toys, drivers’ fury with refractory motor-cars, the fetishism that surrounds the reception of Western High Art – all alongside more rational attitudes. But these examples are intended to help us understand our ‘‘irrational’’ subject-matter. The idolatrous cultures which the book seeks to elucidate are perhaps simplified to exclude inconsistency, skepticism or rationalization, though these are no less important a part of the reception of art in context. As I have suggested above, this problem is betrayed by Gell’s reference to the Arya Samaj sect. In the Roman Empire these contradictions of idol-worship came to be emphasized with derision by early Christian authors. Their rationalizing, iconoclastic logic did not so much move the goal-posts of Roman cult as impose goal-posts on a previously ill-defined field. But Christians were not alone. Greco-Roman philosophical skepticism about idolatry goes back to the sixth century BC, while even sympathetic pagan authors like Dio Chrysostom and Maximus of Tyre in the second century AD find it necessary to excuse the use of images in the worship of deities which were agreed to be heavenly and supernatural.52 They, like modern scholars, found the practices in their midst peculiar and hard to explain. Gell does, in fact, accept this complication implicitly, stating that ‘‘wherever religion exists, it is probable that skepticism also exists, whether or not this is expressed in public.’’53 The question is, how should such responses be dealt with within the anthropology of art when, inconveniently, they are expressed in public? It may be that such expressions are inherently unanthropological and should be left to historians. Gell steers away from them, expressing a certain amount of sympathy with the notion of a ‘‘ ‘great divide’ between the essentially non-sensuous mode of literate thinking and the sensuous, participatory mode of pre-literate thinking.’’54 But this seems an unsatisfactory escape route, especially since there is evidence to suggest that self-conscious questioning of the status of cult images fed back into religious rituals. For instance, in a fragment preserved in the work of St. Augustine, Seneca derides the ritual cultivation of the cult images of the Capitoline Triad in Rome (in every sense the central cult statues of the empire).55 The lifeless statues of Jupiter, Juno, and Minerva were waited upon and assisted in their toilette. Obviously this is an intellectual, skeptical critique, preserved in a Christian attack on idolatry. But Denis Feeney points out a more profound interest in Seneca’s account of the ritual. It is, in fact, conducted as a sort of mime,

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 173

12.1.2007 2:25pm Compositor Name: SJoearun

Gell’s Idols and Roman Cult 173 with the attendants acting out their activities while they deliberately stand away from the images. In this respect it differs from other examples of ritual care for images. Feeney concludes that ‘‘The behaviour of these devotees is a dramatic example of how lucidly the Romans could reflect upon the limits of the mimetic forms with which they honoured their gods.’’56 Similarly, ancient commentators themselves found the more prominent aniconic cult images to be disconcerting and paradoxical elements in a highly figurative religious system. Roman authors frequently comment on these odd objects as exceptions even though they accept that they are perfectly efficacious as cult images. For instance, Tacitus is intrigued by the famous conical cult image of Aphrodite at Paphos on Cyprus (figure 7.2), when he recounts the future emperor Titus’s visit to the island in AD 69: it will not take long to discuss briefly the origins of the cult, the temple rituals and the form of the goddess (for this is not used elsewhere) . . . The statue [simulacrum – a regular word for a cult ‘‘statue’’] of the goddess is not in human form, but it is circular, and it is broader at the base, rising to a small circumference at the top, like a stadium turning-post [meta]. The reason, however, is obscure.57

Lest it appear that we are only dealing with highly self-conscious and intellectual, elite literary texts, it should be stressed that even art-objects themselves sometimes draw attention to the contradictions and paradoxes posed by the coexistence of aniconic and iconic art.58 Arguably the herm is an example of this. The herm was a widespread sculptural form that combined figured elements of deities and individuals (head, genitalia) with an aniconic pillar. The viewer cannot but be aware of which parts of these ‘‘statues’’ are missing.59 More particular ‘‘compromise’’ forms can be found in specific regions of the classical world. Among the most striking are the faceless funerary statues produced in the North African Greek city of Cyrene between about the fifth and first centuries BC (figure 7.3).60 These are, in some respects, rather conventional seated statues of modestly draped women, which probably represent the goddess Demeter or another funerary deity. But their facial features are not represented sculpturally; they are carved smooth, as if a post has been dressed up in a wig and clothes. The symbolism of these bizarre images remains unclear, but they undoubtedly result from a conscious decision to modify

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 174

12.1.2007 2:25pm Compositor Name: SJoearun

Figure 7.3 Marble funerary statue of a faceless goddess from Cyrene, fifth century BC (Cyrene, Archaeological Museum). Courtesy of Susan Kane.

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 175

12.1.2007 2:25pm Compositor Name: SJoearun

Gell’s Idols and Roman Cult 175 anthropomorphic conventions. The makers of such objects were not innocent anthropological subjects, unaware of the problems raised by different modes of divine representation. To an extent, all such criticisms and qualifications are misplaced. One might argue that the purpose of anthropological theory is not to offer universally applicable principles for the explanation of behavior in different cultures, but more precisely to elucidate those kinds of behavior that defy explanation in other terms. Yet this explanation does not do justice to the classical material as a fruitful field for anthropological inquiry and a testing ground for anthropological theory. The challenge is that in Greco-Roman art, embedded though it is in traditional art history, we are confronted simultaneously by many of the ideas and practices which might elsewhere be taken to categorize separate ethnological and ‘‘Western’’ cultures.61 Sometimes these are focused on or embodied in one and the same object. The problem is that this challenge is posed by our anthropological subjects rather than by the academic discourse of the modern historian of Western art.

Notes 1

2

3

4

5

Some of the ideas below appear in a different context in P. Stewart, ‘‘The image of the Roman emperor,’’ in R. Shepherd and R. Maniura, eds., Presence: The Inherence of the Prototype within Images and Other Objects (Aldershot, 2006); cf. also idem, Statues in Roman Society: Representation and Response (Oxford, 2003), esp. pp. 44–5 on agalmatophilia. See e.g. S. C. Humphreys, Anthropology and the Greeks (London, 1978), esp. pp. 17–30; P. Cartledge, ‘‘The Greeks and anthropology,’’ Classics Ireland 2 (1995), pp. 17–28. See e.g. H. Hoffmann, ‘‘Why did the Greeks need imagery? An anthropological approach to the study of Greek vase-painting,’’ Hephaistos 9 (1988), pp. 143–62. Exceptions include R. L. Gordon, ‘‘The real and the imaginary: production and religion in the Graeco-Roman world,’’ Art History 2 (1979), pp. 5–34, reprinted in Gordon, Image and Value in the Graeco-Roman World: Studies in Mithraism and Religious Art (Aldershot, 1996), pp. 5–34; J.-P. Vernant, ‘‘From the ‘presentification’ of the invisible to the imitation of appearance,’’ in idem, Mortals and Immortals: Collected Essays (Princeton, NJ, 1991), pp. 151–63. Gordon’s complaint in 1979 (‘‘The real and the imaginary,’’ esp. pp. 5–7) remains relevant.

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 176

12.1.2007 2:25pm Compositor Name: SJoearun

176 Peter Stewart 6

7 8

9 10 11

12 13 14 15 16 17 18 19 20 21 22

Contrast D. Freedberg, The Power of Images: Studies in the History and Theory of Response (Chicago, 1989). Cf. A. P. Gregory, ‘‘ ‘Powerful images’: responses to portraits and the political uses of images in Rome,’’ Journal of Roman Archaeology 7 (1994), pp. 80–99. On Roman cult images see e.g. H. G. Martin, Ro¨mische Tempelkultbilder: eine archa¨ologische Untersuchung zur spa¨ten Republik (Rome, 1992); Stewart, Statues, esp. pp. 184–222. Note critique in A. A. Donohue, Xoana and the Origins of Greek Sculpture (Atlanta, GA, 1988); and ‘‘The Greek images of the gods,’’ Hephaistos 15 (1997), pp. 31–45. But see Gell, Art and Agency, pp. 96–8, 114, 105–6, 123. See e.g. C. Clerc, Les the´ories relatives au culte des images chez les auteurs grecs du IIme sie`cle apre`s J.-C. (Paris, 1915); H. S. Versnel Faith, Hope and Worship: Aspects of Religious Mentality in the Ancient World (Leiden, 1981), pp. 1–64; Stewart, Statues, p. 263 n. 11 (cf. p. 192 n. 31). Note D. T. Steiner, Images in Mind: Statues in Archaic and Classical Greek Literature and Thought (Princeton, NJ, 2001), esp. pp. 105–20. On talking, weeping, etc. see E. Bevan, Holy Images: An Inquiry into Idolatry and Image-Worship in Ancient Paganism and in Christianity (London, 1940), pp. 25–6; F. Poulsen, ‘‘Talking, weeping and bleeding sculptures: a chapter of the history of religious fraud,’’ Acta Archaeologica 16 (1945), pp. 178–95. On binding: Bevan, Holy Images, pp. 28–9; C. A. Faraone, Talismans and Trojan Horses: Guardian Statues in Ancient Greek Myth and Ritual (Oxford, 1992), pp. 74–93, 136–140; Freedberg, Power of Images, pp. 74–5; Steiner, Images, pp. 160–8. Gell, Art and Agency, p. 10. Gell, Art and Agency, p. 1. Cf. A. Gell, ‘‘The technology of enchantment and the enchantment of technology,’’ in J. Coote and A. Shelton, eds., Anthropology, Art and Aesthetics (Oxford, 1992). Note e.g. Gell, Art and Agency, pp. 1–5; cf. pp. 157–9. Humphreys, Anthropology, pp. 1–13. Gell, Art and Agency, p. 159; idem, ‘‘Technology of enchantment,’’ p. 43. Gell, Art and Agency, pp. 124–5, quoting Bevan, Holy Images, 34. Cf. Lucian, Gallus, p. 24. Gell, Art and Agency, p. 124. R. H. Davis, Lives of Indian Images (Princeton, NJ, 1997), pp. 44–9. H. Belting, Likeness and Presence: A History of the Image before the Era of Art (Chicago, 1994). See e.g. Belting, Likeness, pp. xxi–xxiii; Gell, Art and Agency, pp. 62–4, 97, 105, 113n, 150, 153. Cf. Gordon, ‘‘The real and the imaginary,’’ p. 17. See e.g. D. Feeney, Literature and Religion at Rome: Cultures, Contexts, and Beliefs (Cambridge, 1998); Stewart, Statues, esp. pp. 184–95. Gell, Art and Agency, pp. 122–33.

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 177

12.1.2007 2:25pm Compositor Name: SJoearun

Gell’s Idols and Roman Cult 177 23 24 25 26

27 28

29 30 31

32 33

34 35 36 37 38 39 40 41 42

43 44

Gell, Art and Agency, p. 123. Gell, Art and Agency, pp. 96–154, esp. pp. 96–9. Cf. Gordon, ‘‘The real and the imaginary.’’ Ovid, Metamorphoses 10.243–97. Cf. Clement of Alexandria, Protrepticus 4.50f; Arnobius, Adversus Gentes 6.22. J. Elsner, ‘‘Visual mimesis and the myth of the real: Ovid’s Pygmalion as viewer,’’ Ramus 20 (1991), pp. 154–68; including parallels, p. 167 n. 26. Also M. Bettini, Portrait of the Lover (Berkeley, CA, 1999); Freedberg, Power of Images, pp. 340–44. Pliny the Elder, Natural History 36.21; Ps-Lucian, Amores 15–16; Elsner, ‘‘Visual mimesis,’’ pp. 156–7. On ‘‘Pygmalion’s power’’ see E. H. Gombrich, Art and Illusion: A Study in the Psychology of Pictorial Representation, 5th ed. (London, 1977), pp. 80–98. On Aphrodite cf. A. Stewart, Art, Desire, and the Body in Ancient Greece (Cambridge, 1997), pp. 97–106. See Elsner, ‘‘Visual mimesis,’’ p. 159. Pliny, Natural History 16.242. Stewart, Statues, p. 44. On theurgy see e.g. Steiner, Images, pp. 118–20. Cf. S. Kane, ‘‘Two limestone goddesses from Libya: sculpture from the sanctuary of Demeter and Kore/ Persephone at Cyrene,’’ Apollo 144(413), July (1996), pp. 23–7. Lucian, Alexander 26. Christians: Theodoret, Historia Ecclesiastica 5.22; Hippolytus, Refutatio Omnium Haeresium 4.41. Poulsen, ‘‘Talking, weeping and bleeding’’; F. Johansen, Greek Portraits: Catalogue, Ny Carlsberg Glyptotek (Copenhagen, 1992), pp. 96–7, no. 38 (inv. no. 607). Veneration of Epicurus: Pliny, Natural History 35.5. Cf. Lucian, Alexander 25. Gordon, ‘‘The real and the imaginary,’’ pp. 12–13; cf. Donohue, Xoana, esp. pp. 219–31. On historical priority see Donohue, Xoana, 226; cf. Gordon, ‘‘The real and the imaginary,’’ pp. 12–13. Gell, Art and Agency, p. 98 Gordon, ‘‘The real and the imaginary,’’ p. 13. But note e.g. Gell, Art and Agency, p. 153. Ibid., p. 6. Ibid. pp. 143–53. Classical animation rituals: Steiner, Images, esp. pp. 114–20. Pliny the Elder, Natural History 36.21. Strabo 4.179.4. See R. Fleischer, Artemis von Ephesos und verwandte Kultstatuen aus Anatolien und Syrien (Leiden, 1973), pp. 137–9; also Lexicon Iconographicum Mythologiae Classicae, vol. 2.1, pp. 755–63 on the Artemis in general and esp. pp. 756–7 on its distribution. Gell, Art and Agency, pp. 96–154; cf. pp. 221–58. Cf. use of the words simulacra and eidola for ‘‘idols’’: Stewart, Statues, esp. pp. 19–35.

Osborne / Art’s Agency and Art History 1405135379_4_007 Final Proof

page 178

12.1.2007 2:25pm Compositor Name: SJoearun

178 Peter Stewart 45 See Stewart, ‘‘Roman emperor.’’ Distribution: M. Stuart, ‘‘How were imperial portraits distributed throughout the Roman Empire?’’ American Journal of ¨ ber das Herstellen von Archaeology 43 (1939), pp. 601–15; M. Pfanner, ‘‘U Portra¨ts: ein Beitrag zu Rationalisierungsmassnahmen und Produktionsmechanismen von Massenware im spa¨ten Hellenismus und in der ro¨mischen Kaiserzeit,’’ Jahrbuch des deutschen archa¨ologischen Instituts 104 (1989), pp. 157–257. 46 Stewart, ‘‘Roman emperor.’’ 47 See esp. S. R. F. Price, Rituals and Power: The Roman Imperial Cult in Asia Minor (Cambridge, 1984). 48 Cf. R. R. R. Smith, ‘‘Typology and diversity in the portraits of Augustus,’’ Journal of Roman Archaeology 9 (1996), pp. 30–47. 49 See P. Zanker, The Power of Images in the Age of Augustus (Ann Arbor: University of Michigan Press), esp. pp. 42–3, 98–100. 50 Cf. R. Brilliant, Portraiture (London, 1991), esp. pp. 7–8. 51 Seneca, Epistles 65. 52 N. H. Baynes, ‘‘Idolatry and the early Church,’’ in idem, Byzantine Studies and Other Essays (London, 1955), pp. 116–43; Feeney, Literature and Religion, esp. pp. 76–114; Donohue, Xoana, pp. 122–6 (with extracts from the texts at pp. 281, 350–5). 53 Gell, Art and Agency, p. 115. 54 Ibid. 55 Augustine, De Civitate Dei 6.10. 56 Feeney, Literature and Religion, p. 95; cf. further examples pp. 95–7, and for differing perspective, Gell, Art and Agency, p. 123. 57 Stewart, Statues, pp. 64–72. Tacitus, Histories 2.2–3 (cf. Servius, Ad Aeneidem 1.274). 58 Stewart, Statues, pp. 68–9. 59 Ibid. pp. 62–4; Donohue, Xoana, pp. 228–9. 60 See L. Beschi, ‘‘Divinita` funerarie cirenaiche: storia delle ricerche,’’ Annuario della Scuola Archeologica di Atene 47–8, new ser. 31–2 (1969–70), pp. 133–341; pp. 326–36 on aprosopia. Also Freedberg, Power of Images, p. 72 with fig. 28; Steiner, Images, pp. 180–1. 61 Cf. Gell, Art and Agency, p. 7 for his use of ‘‘prototypical’’ art objects from both.

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 179

12.1.2007 2:27pm Compositor Name: SJoearun

8

Sex, Agency, and History: The Case of Athenian Painted Pottery Robin Osborne

This paper looks at two issues concerning the agency of art, one of which Gell’s analysis offers much help in clarifying, but the other of which Gell’s analysis is unable to address. I argue, first, that Gell’s work offers scope, beyond that which he explored himself, for understanding the distinction between the agency which lies in the ‘‘art’’ of a work of art, and the agency which a work of art may acquire through features not directly connected to its ‘‘art.’’ Second, I turn to the question of how and why works of art change over time, and argue that Gell’s analysis is unable to address the issues of agency involved in such change. I begin with Gell’s own analysis, and in particular its schematization, drawing attention to the logic of his ‘‘tree-structures’’ and to the steps in the argument which Gell elides. I emphasize the value of the treestructures for understanding causal relationships, and raise the question of how the verbs supplied with the arrows that link ‘‘agents’’ and ‘‘patients’’ are to be determined. I then turn away from Gell’s own examples to explore the ways in which Gell’s analysis can help to understand the force and function of images on Greek painted pottery. Greek pots offer a particularly good place to explore Gell’s ideas of art and agency because we know so much about the conditions and context in which the pottery was viewed and used. I apply Gell’s analytical methods to the representation of scenes of sexual

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 180

12.1.2007 2:27pm Compositor Name: SJoearun

180 Robin Osborne activity, whose agency is undoubted but where the degree to which the agency involved is comparable to the agency involved in other works of art is at question. The issue of what might be bundled up in claims about art’s agency is here highlighted, and I argue that Gell’s schematic arrangement helpfully highlights the crucial issues that divide art from pornography. In the final part of the paper I use the distinctive history of sexually explicit images in the history of Athenian painted pottery to raise the question of whether Gell’s analysis can help us to understand change over time. Explicit scenes of sexual activity occur relatively frequently on Athenian pots painted in the last part of the sixth century BC and the first part of the fifth century BC, but then become extremely rare. An exploration of the manner in which such change might be explained raises issues of whether notions of art’s agency take sufficient account of feedback mechanisms to explain adequately the history of images.

Gell’s Analysis Central to Gell’s enterprise of accounting for the production and circulation of art objects as a function of their context in social relations is his analysis of the ‘‘art nexus.’’1 The art nexus is the relationship not only between the artist and the art object (Gell’s ‘‘index’’) but also between that index and its ‘‘prototypes’’ (‘‘entities held, by abduction, to be represented in the index’’) on the one hand, and its ‘‘recipients’’ (‘‘those in relation to whom, by abduction, indexes are considered to exert agency, or who exert agency via the index’’) on the other.2 Artist, index, prototype, and recipient can all be, in Gell’s view, either agents or patients; he illustrates the full range of possible active and passive roles of the four parties in a table, and explains the ways in which they can be active or passive with regard to one another in chapter 3 of Art and Agency, ‘‘The Art Nexus and the Index.’’3 From chapter 4 onwards Gell carries through his analyses by representing the relationships of his four terms graphically, indicating whether each term is ‘‘agent’’ or ‘‘patient’’ by adding the suffix ‘‘–A’’ or ‘‘–P,’’ and indicating relationships between two terms by an arrow. This arrow is long to indicate the relationship between the index-agent and the primary patient (Gell will sometimes call this the ‘‘focal agency arrow’’), and short to indicate subordinate agent–patient relations.4

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 181

12.1.2007 2:27pm Compositor Name: SJoearun

Sex, Agency, History: Athenian Painted Pottery 181 The example which Gell uses to introduce his graphic conventions is Sir Joshua Reynolds’s portrait of Samuel Johnson, where Johnson is seen as the primary agent:

[[[Prototype-A] ! Artist-A] ! Index-A]
!Recipient-P5 This formula is contrasted with the formula accounting for Leonardo’s Mona Lisa, where it is the artist who is seen as the primary agent:

[[[Artist-A] ! Prototype-A] ! Index-A]
!Recipient-P Having introduced his graphic representation of the art nexus in this linear form, Gell introduces ‘‘tree-structures,’’ which are a way of using a second dimension to unpack the square brackets of the linear model and to show that there may be subordinated ‘‘agent’’ relationships on the ‘‘patient’’ side, and subordinated ‘‘patient’’ relationships on the ‘‘agent’’ side. In presenting as tree-structures the agents involved in the Congo nail fetishes and Mary Richardson’s Rokeby Venus (see ‘‘Introduction,’’ figures 0.5–7), Gell spells out the nature of the agency implied by the arrows, by writing (normally) verbs of agency beside those arrows. In the case of the nail fetish, the authority which it exercises over the victim stems from the fact that it derives from a tree, the cutting down of which is understood to have caused the death of a hunter, who has been picked out precisely for his own success in bringing about the death of animals. It is crucial to our understanding of the fetish that the various verbs involved at different stages are in some sense cognate. That is, cutting down the tree has to be parallel to the cutting down of life, both the life of animals cut down by the hunter and the life of the hunter, itself cut down. The nail fetish then acts to cut down, in some sense, the victim. The tree-structure traces back over time the history of the fetish, and that history is, importantly, itself continuously part of the fetish. In the case of Mary Richardson’s Rokeby Venus, that is, the painting by Vela´zquez as slashed by the suffragette Mary Richardson, the situation as represented by Gell is somewhat more complicated. The slashed painting that provokes public outrage is the product of violence done by Mary Richardson to Vela´zquez’s canvas. Mary Richardson’s violence reflects the violence done by prison warders to Mrs. Pankhurst. She chose, on her own admission, the Rokeby Venus as the object of her violence because Venus

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 182

12.1.2007 2:27pm Compositor Name: SJoearun

182 Robin Osborne was the ‘‘most beautiful woman in mythological history’’ and Mrs. Pankhurst was ‘‘the most beautiful character in modern history’’ (Gell, Art and Agency, p. 64). The mythological character, Venus, had inspired Vela´zquez to paint her, and the real-life character, Mrs. Pankhurst, inspires Mary Richardson to slash the painting. Gell chooses to represent the relationship of the prison warders to Mrs. Pankhurst as one of violence, the relationship of Mrs. Pankhurst to Mary Richardson as one of inspiration, the relationship of Mary Richardson to the Rokeby Venus as one of violence, and the relationship of the slashed Rokeby Venus to the public as one of ‘‘sensation, outrage.’’ By doing so, Gell sets up a parallel between the relationship of Venus to Vela´zquez, on the one hand, and the relationship between Mrs. Pankhurst and Mary Richardson, on the other. On both sides of this diagram, however, there are problems with Gell’s representation. The important point about the relationship between Mrs. Pankhurst and Mary Richardson is that the violence to Mrs. Pankhurst causes outrage to Mary Richardson. It is Mary Richardson’s outrage at the treatment of Mrs. Pankhurst that causes her to outrage the Rokeby Venus and in turn to produce outrage in the public. Mary Richardson’s more general inspiration by Mrs. Pankhurst, her coming to regard her as ‘‘the most beautiful character in modern history,’’ is prior to, and independent of, the action of the prison warders. The ‘‘implicit relationship between Recipient–A (the prison warders, agents of the repressive government) and Recipient–P (the outraged public),’’ to which Gell draws attention (p. 65), would be more easily perceived if the arrows of agency consistently indicated the relationship of outrage. But if the left-hand side of Gell’s tree-structure is over-compressed, the right-hand side is over-extended. If we go back to the terms in which Mary Richardson explained her own action, it is important that Mary Richardson herself declares both Venus and Mrs. Pankhurst to be ‘‘most beautiful.’’ That is, if we are to draw out of her declaration that Mrs. Pankhurst is the ‘‘most beautiful character in modern history’’ the statement that she was inspired by Mrs. Pankhurst, so we must draw out of her statement that Venus is ‘‘the most beautiful woman in mythological history’’ a parallel inspiration. What, presumably, makes the Rokeby Venus an appropriate object for Mary Richardson to outrage is her belief that the Rokeby Venus instantiates the beauty of the mythological woman. The history of how Vela´zquez came to depict Venus is irrelevant to the explanation of the production of the slashed Rokeby Venus: what matters is Mary Richardson’s relationship to Venus, not Vela´zquez’s relationship. No abduction of

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 183

12.1.2007 2:27pm Compositor Name: SJoearun

Sex, Agency, History: Athenian Painted Pottery 183 an artist’s agency plays any part in Mary Richardson’s relationship to the Rokeby Venus. And it is the Rokeby Venus as focalized by Mary Richardson that the right-hand side of the tree-structure needs to reveal. I submit that Gell’s tree-structure would more properly take the form of figure 8.1. Why does any of this matter? It matters because dividing the world between agents and patients tells us very little in its own right, although it may be helpful in instances where we are inclined to overlook or disguise agency. Where Gell’s schematic representations of the relationship between agent and patient help are in revealing the nature of the art nexus, that is, revealing precisely how agency works in the case of a work of art. But the explanatory force of the schematic representations depends upon the coherence of the picture of agency which they give. We must be able to understand how the exercise of agency by, for example, a prototype upon a recipient at one level in the tree-structure (or at the left-hand end of a linear diagram) can lead to the exercise of agency by the artist upon the index at the next level up (or in the next places to the right). The coherence to the picture of agency which understanding requires does not necessarily depend upon it being possible to describe the nature of the agency in similar terms at all levels of the tree, but if the terms are not identical then at least their relationship needs to be comprehensible. In the case of the nail fetish, Gell’s tree-structure is helpful because the various verbs which he uses to describe the relationship of agent to patient at the various levels are cognate. In the case of the slashed Rokeby Venus the tree-structure would be more successful in bringing out Mary Richardson’s own explanation of her behavior if it were redrawn to produce a more

Figure 8.1 Reworking of Gell’s fig. 4.4/4.

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 184

12.1.2007 2:27pm Compositor Name: SJoearun

184 Robin Osborne consistent set of agent–patient relationships. But in both cases one might reasonably ask in what sense the tree-structure helps us to understand the agency of art. Gell’s diagrams well represent a world of ‘‘give and take,’’ but while it may be very helpful to our understanding of how agency works to lay out these relationships like this, there is nothing distinct to the art world in a world structured in that way: the whole of human history has sometimes been seen like that.6 In the case of neither the nail fetish nor the (slashed) Rokeby Venus is there anything in the tree-structures that is expressive of the aesthetic content of the work of art involved. Yet the pictures which Gell himself prints suggest that the appearance of the fetishes and the appearance of the Rokeby Venus did matter: the nails are fixed into the body of the fetish in great density, but not into its face; the slashes through the canvas of the painting are across the back of the reclining figure of Venus. It is not simply its derivation from a particular tree that is important for the fetish, but that it has been carved into the likeness of a human figure. It is central to Mary Richardson’s account of her actions that Venus, like Mrs. Pankhurst, is beautiful. When we ask how the aesthetic might be introduced into the treestructures, we can see that in the case of the nail fetish a whole layer has been elided. Gell passes directly from the Muamba tree causing the death of the Brave Hunter, to the Supplicant driving a nail into the fetish. But between the two is the creation of the fetish from the Muamba tree, a process in which, in some sense, the dead brave hunter enters the wood cut from the Muamba tree and the wood of the Muamba tree becomes the index of the hunter, in part at least by resembling him. In the case of the slashed Rokeby Venus, what is elided is the process whereby the idea of beauty, embedded in the name ‘‘Venus,’’ becomes embodied. Vela´zquez, in Gell’s tree-structure, is inspired by Venus and depicts Venus: the idea of beauty is translated into the Rokeby Venus by the artist without there being any intervening ‘‘third term.’’ But it is vital that from the Rokeby Venus Mary Richardson, as viewer, abducts, and assumes that other viewers also abduct, not an artist inspired by an idea, but a beautiful woman who can stand, in some sense, for that other beautiful woman, Mrs. Pankhurst. In Gell’s terms, it is important to Mary Richardson that we can and do abduct another prototype who plays an active role in the creation of the Rokeby Venus alongside Venus herself: a beautiful woman. But if we need to introduce into Gell’s tree-structures explicit acknowledgment of the appearance, and of the agency of the appearance, of the

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 185

12.1.2007 2:27pm Compositor Name: SJoearun

Sex, Agency, History: Athenian Painted Pottery 185 prototypes and indexes involved, those tree-structures valuably alert us to a general truth. This is that the relationship of resemblance that an index has to a prototype can and does have a role to play in determining its agency. Insofar as what the tree-structure does is to bring out the ways in which to account for actions is to understand how those actions parallel the actions from which in some sense they stem, the tree-structure emphasizes that the relationship of prototype to index is central to the agency of the art object. To suggest, however, that that the index exercises its agency by resembling a prototype must remain an insufficient explanation of the agency of a work of art. Insufficient because it does not allow that two indexes, both of which resemble the same prototype, might nevertheless exercise different agency. It might not be by chance that it was the Rokeby Venus, rather than any other image of Venus, that provoked Mary Richardson’s action. It is here that we must return to the question of the verbs which appear above the arrows in the tree-structures. Arguably we need not only to insert into the right-hand side of the tree-structure a beautiful woman to lend her body to the idea of beauty ascribed to Venus, but to acknowledge in the arrows that this depiction of the beautiful woman is also sexually attractive. Mary Richardson’s choice of this Venus depends upon this Venus attracting the attentions of men, just as Mrs. Pankhurst attracted the attentions of men.

Gell and Greek Pottery, Art, and Pornography I want to investigate further the question of the nature of the relationships between index and recipient (and between prototype and artist) by looking at the agency of classical Greek pottery. Pots make a particularly valuable case study because the function of the pot is a product of its shape, and so for any pot of known shape we can, with some degree of confidence, posit a context of use. The particular index on which I want to concentrate my analysis is an oil flask (aryballos) made, and signed, by a potter named Nearchos in sixth-century BC Athens.7 This vessel was made for use in the gymnasium. A sixth-century funerary sculpture shows an aryballos carried by a naked young man on a string round his wrist – the fact that is was normally suspended explains why aryballoi do not have to have flat bases. In the case of Nearchos’s aryballos, the curved stripes on the body of the aryballos

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 186

12.1.2007 2:27pm Compositor Name: SJoearun

186 Robin Osborne emphasize its rounded shape and invite the viewer to run a hand over its smooth surface. Figured decoration on this aryballos is limited to the rim (figure 8.2) and the handle plate (figure 8.3). Aryballoi are not large pots; this one is 8 cm high, and the figures are miniature. But the activities shown are clear. On the rim pygmies fight cranes, with nonsense-writing weaving among them; on the sides of the handle plate are Hermes and Perseus, both labeled; on its top we have Tritons; and on the back of the plate three satyrs, all to some degree crouching, one frontal and two seen from the side, masturbate. The satyrs are given names that allude to the penis and to (sexual) enjoyment: Phsolas (cf. psolos, penis with retracted prepuce), Dophios (cf. dephomai, I masturbate) and Terpekelos (cf. terpsis, enjoyment). This flask has a function: it contains the oil with which the user rubs himself down after exercise in the gymnasium. The decoration on the

Figure 8.2 Nearchos’s aryballos (New York Met. 26.49, J. D. Beazley, Attic BlackFigure Vase-Painters, Oxford, 1956, 83.4); side view. Courtesy of Metropolitan Museum, New York.

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 187

12.1.2007 2:27pm Compositor Name: SJoearun

Sex, Agency, History: Athenian Painted Pottery

187

Figure 8.3 Handle plate of Nearchos’s aryballos (New York Met. 26.49, J. D. Beazley, Attic Red-Figure Vase-Painters, 2nd ed., Oxford, 1963, 83.4). Courtesy of Metropolitan Museum, New York.

aryballos has no direct impact on its function. But the decoration does have an impact on how the user thinks about the function. Oiling oneself down can be construed as an act of cleansing, an efficient way of getting rid of the dust and sweat with which the body is stuck about after the rigors of athletic activity. But it can also be thought of as a sensuous experience, and that sensuousness can be more or less sexually charged. If the stripes on the body of the aryballos attract the touch, the imagery on the back of the handle plate brings the viewer face to face with the sexual effects of touching. The user who sees these satyrs at the point at which he upturns the aryballos to pour oil into the palm of his hand prior to rubbing himself down cannot rub innocent of the sexual pleasure that could be had from rubbing – nor unaware that ‘‘public masturbation’’ is something that marks satyrs off from men.8

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 188

12.1.2007 2:27pm Compositor Name: SJoearun

188 Robin Osborne If it is easy to see how the decoration of this aryballos, both nonfigurative and figurative, can exercise agency, it is much more difficult to answer the question of whether it exercises that agency qua art. The impact of the aryballos upon the user is clearly in part aesthetic: the visual appearance and physical form of the aryballos set up the sensuous and sexual agenda. But the impact of the imagery on the back of the handle plate we might maintain to be aesthetic only in the trivial sense that it stems from perception. Images of masturbation, on this line of argument, achieve their effect more or less as symbols. The more or less overtly sexual names of the satyrs arguably encourage this view. It is as if, as the athlete was rubbing down his thighs with oil, someone were to shout out ‘‘masturbate, masturbate’’ (or some more colloquial version). Reaction to that too might be held to be aesthetic, since it depends upon perception, but it is hardly the aesthetic features of the phrase, as they would normally be understood, that trigger reaction. If we plot out the back of the handle plate of Nearchos’s aryballos in Gellian schematic form, we find the artist as agent determining the appearance of the prototype, since satyrs belong to imaginative art. Similarly both artist and prototype actively shape the index: the prototype dictating its form. The index as agent determines the reaction of the recipient: the ‘‘recipient as ‘spectator’ submits to index.’’9 In linear form we have:

[[[artist-A] ! prototype-A] ! index-A]
!recipient-P which is exactly Gell’s description of the Mona Lisa. But this linear story is not the whole story. In what sense does the story begin the artist? Nearchos’s satyrs were not conjured out of thin air, they stem from a graphic tradition. And, to rephrase the question posed in the last paragraph, does the agency of the index stem from the artist, that is, from the particular way in which this artist has rendered the material he has chosen to use? If we try to map out the way in which the aryballos works through tiered pairs of active and passive parties we are forced to take decisions on precisely those difficult questions. The top tier must bring the shape and all the decoration together and deal with the overall effect of this aryballos on the viewer. But having drawn a long arrow from the aryballos to the athlete, what are we going to write above the arrow? Will we adopt a coy ‘‘enables,’’ flirt with the potential ambiguities of ‘‘stimulates,’’ or go for the out-and-out sexual claim contained in ‘‘arouses’’?

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 189

12.1.2007 2:27pm Compositor Name: SJoearun

Sex, Agency, History: Athenian Painted Pottery 189 When we come to separate the shape of the vessel from the iconography of the scene of satyrs (to pursue only part of the decorative input into the effect of the aryballos), we are faced with another difficult decision. How do we describe the mutual interaction between shape and decoration? We might choose functional priority and suggest that the shape is the active partner, giving a context for and therefore a particular meaning to the scene of the satyrs. But we might equally suggest that the scene of the satyrs is the active partner, turning the aryballos from being any old bit of soap on a string to advertising itself as an instrument of sexual pleasure. Gell himself faces the problem of mapping relationships that are mutually constitutive when he explores the ‘‘intersubjectivity’’ between devotee and idol (his figure 7.7/2), but it is not clear that the regression which he settles for in that case is very helpful if we are trying to understand how the relationship comes about. We must allow, as in a rather special case Gell allows for the slashed Rokeby Venus,10 that relationships of agency may go in both directions at once, put arrowheads at each end, and recognize that both Index 1 (the aryballos shape) and Index 2 (the scene of satyrs) are simultaneously agents and patients. The next tier down on the diagram is relatively unproblematic on both sides (Nearchos paints the satyrs/produces the aryballos – for simplicity I assume that he was potter as well as painter). On the side of the aryballos shape, whether we think that the production of the flask was to order or speculative, no major issues arise. But issues lurk in the question of how Nearchos came to paint this imagery. Gell resorted to traditional language of inspiration to describe how Vela´zquez came to depict Venus, but he gets away with that only because Venus is both a pagan deity and the personification of female beauty. We ordinarily talk of people being inspired by the natural world (by a sublime view, for instance) or by beauty (‘‘inspired’’ is not altogether knowing in the phrase ‘‘inspired by his muse’’), but it is far more difficult to apply inspiration to the relationship between a fictional entity and an artist: painters of life on Mars are hardly to be described as inspired by little green men. There are circumstances in which an artist is influenced by another work of art, and in these circumstances we might indeed use the word inspired. But at some stage, and probably here with Nearchos himself, an artist is responding not to another image but to an idea – an idea which might or might not be embodied in a narrative. To make the agency of the idea of the satyr comprehensible we need to paraphrase the idea of the satyr: what affects Nearchos, we might suggest, is the way in which the

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 190

12.1.2007 2:27pm Compositor Name: SJoearun

190 Robin Osborne figure of the satyr patrols the boundaries of acceptable male behavior. ‘‘Men behaving badly’’ give rise to the satyr, and the possibility of unacceptable male behavior embodied in the satyr provides the artist with resources. Satyrs behave badly in many ways on Athenian pottery, though primarily in ways that have to do with alcohol and sex (homo- and hetero-erotic behavior as well as autoeroticism).11 The idea of men behaving badly embodied in the satyr will not itself, therefore, explain Nearchos’s choice of this type of bad behavior. The diagram must split again at this point as we account both for the satyr and for what Nearchos does with the satyr. Whichever route we explore, we come up against issues which are crucial to the question of pornography. If we try to account for why satyrs were invented, for why a half-human creature is shown engaging in behavior not publicly acceptable in human males, the condemnation of certain types of male behavior as bestial is crucial. But it is not a sufficient condition. To conceive that men’s bad behavior, of whatever sort, can outrage a public and have that public condemn it need not lead to creating ways of showing the behavior condemned. Many acts are so widely regarded as outrageous that they are not to be found in art. The creation of the satyr depends not simply on the condemnation of the behavior that the satyr performs, but also upon the desire nevertheless to display that behavior. What we consider produces the desire to display is a crucial matter, for which there is perhaps no general answer. The problem is best explored in the particular case. In taking this decision to display the bestial satyr masturbating, Nearchos engages with male autoeroticism. But it is crucial here whether we describe his reaction to male autoeroticism as (to cite only a range of possibilities) one of amusement (‘‘These creatures are grotesque!’’), disgust (‘‘Don’t even think about it!’’), attraction (‘‘Wish I could do that!’’), or arousal (‘‘Isn’t it exciting!’’). Answering this question demands putting the scene of the satyrs back in context. Statuettes of sexually excited satyrs have become popular picture postcards, but we know little of their original contexts of display. One of the gates of archaic Thasos had upon its doorpost a relief of a sexually excited satyr – a suitably liminal place to patrol the boundaries of the acceptable. But the vast majority of surviving images of satyrs are painted on pottery, and in particular on vessels made for drinking parties. Nearchos’s aryballos does not belong to that context of drunkenness and the exploration of hetero- and homosexual desire (well described, if in a rather refined

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 191

12.1.2007 2:27pm Compositor Name: SJoearun

Sex, Agency, History: Athenian Painted Pottery 191 version, in Plato’s Symposion), but to a more precise context still: the apre`s-gym activities of naked males in the gymnasium. Put into the context of a vessel containing oil used for rubbing down in the gym, and in particular a vessel whose particular rounded shape and non-figurative decoration encourages enjoyment of tactile sensations, Nearchos’s choice of scene on the back of the handle plate comes to seem rather less open to multiple interpretations. ‘‘Innocent’’ amusement at the sheer grotesqueness of the satyrs becomes impossible, and the choice of noms parlants makes it hard to sustain an interpretation which stresses disgust. Whether or not Nearchos was himself sexually aroused by the thought of satyrs masturbating, this scene does its best to challenge the ability of the naked athlete rubbing himself down to prevent arousal. That challenge can only be issued by an artist who himself knows that such scenes can stimulate arousal. For many readers the above analysis may seem an extremely longwinded way of arriving at an obvious conclusion. I see two gains from taking a Gellian route to my conclusion. The first is that it foregrounds the very question of agency, the work that the object and its imagery does. The second is that it makes us analyze what does that work, and in doing so isolate the crucial questions of where the particularity of this particular object and imagery lie. I want to illustrate further the value of the Gellian analysis with reference to two further images. One is from a slightly later Athenian pot, this time a hydria (water jug) painted in the red-figure technique and ascribed to a painter known as the Kleophrades Painter (figure 8.4).12 This image is one of a number painted about the end of the sixth century which show a female devotee of the god Dionysos, the god most associated with wine and revelry, lying asleep and being approached by a satyr. In this case the satyr, himself sexually excited, lifts the dress of the sleeping woman and moves his hand to her crotch as if to arouse her. Behind him is a second satyr, who kneels and looks down at his own erect penis. My second image is a twentieth-century painting by Raymond Coxon, entitled The Intruder (figure 8.5).13 The painting is dominated by the figure of a naked woman asleep on the grassy bank of a stream. Behind her is a wood, between the trees of which the buildings of a village can be seen. Next to one tree is a figure, dressed in grey and wearing a black hat. He looks out – at the naked sleeping woman and at the viewer. In thematic terms these two images have much in common. In both we are faced with a woman lying asleep in a place where one would not expect

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 192

12.1.2007 2:27pm Compositor Name: SJoearun

192 Robin Osborne

Figure 8.4 Hydria ascribed to Kleophrades (Muse´e des Antiquite´s de Rouen inv 538.3, J. D. Beazley, Attic Red-Figure Vase-Painters, 2nd ed., Oxford, 1963, 188.68). Courtesy of Muse´es De´partementaux de la Seine-Maritime. Photograph: Yohann Deslandes.

to find a solitary woman sleeping. Both women are exposed, and both are approached by intruders of whose presence the woman is ignorant. The interest of both images lies, at least in part, in the sexual relationship which the viewer constructs for the figures. When we come to analyze these images, however, it is clear that they work very differently. The satyr’s fictive origin and explicit sexual interest are extremely important here. We must reconstruct for these satyrs the same basis in the unacceptability of male public display of sexual arousal as for Nearchos’s satyrs. And we must reconstruct for the Kleophrades Painter the same interest in making public sexual arousal that is not acceptably displayed by men and women. But Coxon’s Intruder has to be inserted into a quite different scheme of things. The naked woman who dominates his canvas belongs to the tradition of the western nude. His intruder relates to the figures regularly seen in the background of reclining female nudes, most famously in Manet’s Olympia. The setting of the naked woman in the countryside, and the composition in which the dark trees at

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 193

12.1.2007 2:27pm Compositor Name: SJoearun

Sex, Agency, History: Athenian Painted Pottery

193

Figure 8.5 Raymond Coxon, The Intruder. By kind permission of the Provost and Fellows of King’s College, Cambridge.

the sides frame a more distant sunlit view in the centre, evokes Manet’s De´jeuner sur l’herbe. Whereas in the analysis of the Kleophrades Painter’s satyrs we have to invoke the idea of men behaving badly to explain the male figure, here the presence of the male figure requires only that for a man to come across a naked woman in a public place evokes a potential erotic response. The presence of the female nude in Coxon’s painting may require only that the tradition of artistic presentation of female nudity raises issues of context. Coxon’s inspiration is adequately expressed by an interest in how men and women negotiate their relationships.14 What has made these two analyses different is in part the fictional status of the satyr. Satyrs are freighted with a history that is much more limited than the history of men. Men may behave well or badly; satyrs exist to behave badly. To replace Coxon’s intruder by a satyr would immediately require the terms of the analysis to be changed, since the question would then necessarily be of what sort of bad behavior should be expected. But in

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 194

12.1.2007 2:27pm Compositor Name: SJoearun

194 Robin Osborne part the difference in the analysis comes from the public display by the satyr of behavior not acceptable in a man. The painter of the sexually excited satyr has to think that the display of male sexual excitement is in some respect of interest to the viewer. The painter of the fully clothed man presupposes no particular interest in or reaction to male sexual excitement. But fully to understand the difference between these two analyses we require both the satyr and the sexual excitement. Had Coxon showed his intruder naked and sexually excited, the analysis of his painting would still not require that the painter considered that male sexual excitement as such was of interest to the viewer: the interest would remain interest in how men and women negotiate relationships. The fictional satyr’s history determines the satyr’s behavior to such an extent that no negotiation is conceivable: whatever Coxon’s intruder is going to do, or would be going to do if naked and aroused, the satyr is bound to be going to behave badly. Viewers no doubt relate differently to these two images, and may find both, either, or neither sexually provocative. But the image by the Kleophrades Painter exercises agency primarily in the realm of desire. Although this image is more obviously aesthetically pleasing than that on Nearchos’s aryballos, to account for the choice of subject and manner of depiction we must ascribe to the artist, and presume in the viewer, an interest in sexual acts. By contrast, although sexual activity is one scenario which Coxon’s painting will cause the viewer to contemplate, the choice of subject requires an interest in sexual acts as such in neither artist nor viewer. Coxon’s painting is erotic, but only the images on the Greek vases can be properly, and are necessarily, described as pornographic, that is, images which primarily attract and require a response centered on sexual acts.

Gell and the History of Art So far I have argued that Gellian analysis sharpens our perception of what a work of art is doing to us. In this final section I proceed to argue that Gellian analysis, which is so useful for contrasting different types of images, offers no foothold for those seeking to understand how and why images of a similar type change over time. During the first half of the fifth century images of satyrs on Athenian pots become distinctly less boisterous, and their sexual activities less frequently shown. Shortly before the middle of the century painters create for the first time satyr children, and begin to show satyrs in ‘‘domestic’’

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 195

12.1.2007 2:27pm Compositor Name: SJoearun

Sex, Agency, History: Athenian Painted Pottery 195 scenes – playing piggy-back with their children, bowling hoops etc.15 Although satyrs remain regular companions of Dionysos and his female devotees, the maenads, they are much less frequently shown sexually aroused. Art historians have a number of ways of explaining change over time. In the history of Greek art much has been made of increasing naturalism. The appearance of images changes, in this account, because artists more closely approximate to the appearance of nature. That over the history of art the ways in which images relate to the observed world change is undeniable. The question is: why should one set of features be selected at one moment when a different set had been selected previously? Explanation of the selection of features of the observed world made salient at any historical point might emphasize changes in the world represented or changes in the representation of the world. Imagery may change under the influence of other imagery (e.g. the selection of features displayed may become so restricted as itself to provoke a reaction) without anything else happening to change the way the observer relates to the observed world. We see this most clearly when the presence of an academy or a privileged exhibition space leads to the policing of what art gets displayed. Alternatively, events that have nothing to do with representation may make salient features in the observed world not well represented in existing imagery, and prompt a change in painters’ choices – consider the impact of twentieth-century wars. How people live in the observed world affects the way in which they observe it. How people represent the observed world affects the way in which they live in it. Gell’s tree-diagrams map how the observed world acts upon the imagery (‘‘observed’’ treatment of Mrs. Pankhurst makes Mary Richardson see and treat the Rokeby Venus in a particular way) and the imagery acts upon the observed world (it is Vela´zquez’s Rokeby Venus that brings out Mary Richardson’s violence). But can Gell’s tree-diagrams map not simply the interrelationship between the observed world and the represented world at its various points, but the changing nature of that relationship over time? Gell’s tree-diagrams adapt very well to explaining how a work of art can stand in a tradition (as we have seen with Coxon’s Intruder). They enable us to move from the prototype through factors exercising agency upon the prototype to arrive at the index whose agency is the object of enquiry. But Gell treats the prototype as a single thing. His analysis of Reynolds’s portrait of Dr. Johnson admirably captures the way in which the viewer

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 196

12.1.2007 2:27pm Compositor Name: SJoearun

196 Robin Osborne needs to believe of a good portrait that the prototype has impressed himself more powerfully upon the artist than the artist upon the prototype. Any ‘‘good portrait’’ of Dr. Johnson, however, will fit the same analysis. As with the agency of the prototype, so also with the agency of the index. Gell’s diagram cannot explain why Mary Richardson slashed this particular painting of Venus. The diagrams which are so good at distinguishing different types of relationship between prototype, artist, index, and recipient have little to offer when it comes to distinguishing within those types of relationships. They are unhelpful for our understanding of the history of art for precisely the same reason that they are helpful for our discussion of art and pornography. There is an irony to my claim about the singleness of Gell’s prototypes. For Gell insists that an anthropological theory of art has to be a theory embedded in social relationships, and applies to the work of art the idea of fractal personhood. This would appear to commit Gell to a multiplicity for any person, incompatible with the singleness of which I have complained in relation to Gell’s prototypes. But Gell’s multiplicity lies precisely in the relationships that the prototype sustains with indexes and recipients, and depends upon the recipient’s assumption that there is a single prototype. The process of abduction has the great advantage of giving emphasis to how recipients react, but the disadvantage that to privilege the recipient is to privilege a particular way of observing the world. Any given recipient will abduct in a particular way, but understanding (as opposed to simply describing) change over time demands holding open at the same time a range of different abductions, demands seeing how different recipients might engage differently with the same index. When Gell concludes that ‘‘Marquesan art, considered as a whole, can be conceptualized . . . as a ‘distributed object’ ’’ he also notes that the only unity it has is ‘‘from a stylistic point of view.’’16 What makes Marquesan art conceptually a single distributed object, is, I suggest, that it involves the same abduction. The ‘‘isomorphy of structure between the cognitive processes . . . and the spatio-temporal structures of distributed objects,’’ which he goes on to assert, is what I have referred to as the particular selection of features of the observed world treated as salient at any one time (and place).17 But the analysis that is so good at contemplating the multiple forms in which a particular take on the observed world may manifest itself is ill adapted to consideration of the ways in which that take itself may change over time. Coming to see what Gell’s analyses are and are not good for demands that we grapple with some very basic problems in accounting not just for how

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 197

12.1.2007 2:27pm Compositor Name: SJoearun

Sex, Agency, History: Athenian Painted Pottery 197 works of art act upon us as viewers but for how we perceive the world. The way in which Gell himself ranges widely over very different sorts of art object and calls upon various different philosophical approaches contributes significantly to the difficulty of his book. The rewards, however, are themselves substantial in helping us both to see that visual objects act upon us in very distinct ways which may impoverish, as well as enrich, our encounter with the things and objects around us, and to contemplate the unmappable variety of possible worlds waiting to be observed – worlds whose existence is systematically obscured by our own attempt to understand those features of the observed world which we privilege.

Notes 1 2 3 4 5 6 7 8

9 10 11

12

For the definition of the nature of the enterprise see Gell, Art and Agency, p. 11. A. Gell, Art and Agency: an Anthropological Theory (Oxford, 1998), p. 27. The table from Art and Agency, p. 29 is reproduced above as ‘‘Introduction,’’ figure 0.2. Ibid., p. 51. Ibid., p. 52. See J. P. Gould, Myth, Ritual, Memory, and Exchange: Essays in Greek Literature and Culture (Oxford, 2000), ch. 12, ‘‘Give and take in Herodotus.’’ New York Metropolitan Museum 26.49, J. D. Beazley, Attic Black-Figure VasePainters (Oxford, 1956), 83.4; H. Immerwahr, Attic Script (Oxford, 1990), p. 97. I exceed the evidence here in two ways. First, this image stands chronologically at the head of our images of explicit sexual activity on Athenian painted pottery, and so the reactions of the original viewer are strictly hard to guess. Second, Athenian black-figure pottery does include masturbating figures who are not satyrs, but sexual excitement in general, and masturbation in particular, are more frequently shown for satyrs than for ‘‘ordinary’’ human figures. Public masturbation was notoriously something that marked out Diogenes the Cynic as heedless of social convention. Gell, Art and Agency, p. 29, table 1 (reproduced as ‘‘Introduction,’’ figure 0.2). Ibid., p. 64. F. Lissarrague, ‘‘The sexual life of satyrs,’’ in D. Halperin, J. J. Winkler, and F. Zeitlin, eds., Before Sexuality: The Construction of Erotic Experience in the Ancient Greek World (Princeton, NJ, 1990), pp. 53–81; F. Lissarrague ‘‘On the wildness of satyrs,’’ in T. H. Carpenter and C. A. Faraone, Masks of Dionysus (Ithaca, NY, 1993), pp. 207–20. Muse´e des Antiquite´s de Rouen inv 538.3, J. D. Beazley, Attic Red-Figure VasePainters, 2nd ed. (Oxford, 1963), 188.68.

Osborne / Art’s Agency and Art History 1405135379_4_008 Final Proof

page 198

12.1.2007 2:27pm Compositor Name: SJoearun

198 Robin Osborne 13 King’s College, Cambridge. 14 Coxon’s painting, like the paintings by Manet, makes no reference to any known story. I would maintain, however, that the nature of the analysis would not be different even if the discussion were to concern, e.g., a painting of Susanna and the Elders, where a known story, and one involving sexual desire, was involved. 15 Compare particularly London, British Museum GR1856.12–13.1, ascribed to the Niobid Painter (Beazley, Attic Red-Figure Vase-Painters, 601.23). 16 Gell, Art and Agency, p. 221. 17 Ibid., p. 222; see also p. 236.

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 199

13.1.2007 2:10pm Compositor Name: SJoearun

9

Abducting the Agency of Art Whitney Davis

Alfred Gell’s Art and Agency: An Anthropological Theory (hereafter AA), posthumously published in 1998, offers an intriguing proposal for an ‘‘anthropology of art.’’ According to Gell, we should try to identify the beliefs sustained by agents, typically the users or what Gell calls the ‘‘recipients’’ of works of art, when they infer or ‘‘abduct’’ the identity, actions, or motivations and intentions of the agents supposed to have constituted the work of art. In such abductive reception, as we might call it, the recipients of the ‘‘agency of art’’ identify who or what has caused the work of art, as they believe, to have the properties and effects it seems to them to possess or to display or to produce. (I say ‘‘who or what’’ because the inferred generative agents of the work of art need not be human beings; they might, for example, be supernatural entities or natural or inorganic processes.) In this ‘‘art nexus,’’ a work of art can be understood by its makers and other original observers and users as an ‘‘index’’ of the agencies believed to have caused it. Gell’s anthropology addresses the agent-caused properties of artifacts, or, more exactly, what people believe to be the agent-caused properties. Because these properties are believed to be agentively constituted, they can be said to be ‘‘natural signs’’ of their cause in the sense that smoke signals the presence of fire or a bullet-hole in a wall marks the impact of a bullet fired at the wall. Abduction is the mode of knowing – of inferring – causes; abductive knowledge is the understanding of indexes. In Gell’s terms, then, the work of art specifically is an ‘‘art-index,’’ having the power to move its

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 200

13.1.2007 2:10pm Compositor Name: SJoearun

200 Whitney Davis recipients to beliefs about its origins and its efficacy. It is believed to ‘‘do’’ something; whatever its generating agencies, it has, then, its own attributed agency too.

Abducting and Indexing All of these terms – ‘‘agency,’’ ‘‘abduction,’’ ‘‘art nexus,’’ ‘‘index,’’ and ‘‘artindex’’ – have been given special senses in Gell’s anthropology of art; they are conceptually interrelated; and they require careful disentangling. Gell’s emphasis on the agency of art (referring both to its abducted agentive causation and to its power to motivate agents’ actions) is highly instructive. But it is also possible, as I will suggest, that Gell’s model of the abduction of agencies in the art nexus can lead us too far away from the history of the artindex itself. As a consequence, aspects of art’s historicity – its ‘‘art history’’ – could be overlooked. In this sense, Gell’s model of the abduction of the agency of art – as my title has it – risks ‘‘abducting’’ the agency of art. That is to say, it risks neglecting, misrecognizing, or hijacking the historical agency specifically of art – even if and possibly especially because the work of art must be an index of the many historical agents inferred to have caused it. In addressing this matter, at the outset I want to acknowledge that I am sympathetic to Gell’s basic theoretical inclinations. Among these we should count, first, his dogged resistance to any ‘‘textualization’’ of visual culture and of material artifacts; second, his desire to construct a specifically anthropological theory of art (in relation to what he takes to be ‘‘aesthetics’’ and to a lesser extent ‘‘art history’’) that at the same time corrects the unhistorical tendencies of an established ‘‘anthropology of art’’; and third, his continual engagement with art’s intentionality conceived as constituted in people’s abduction of art’s causality. Art and Agency blends the intellectual traditions – and the technical terminology – of communication studies, semiotic and reception theory, intentional analytics, the phenomenology of knowledge, language philosophy, cognitive anthropology, and interactionist pragmatics. A protean writer uniting all these possibilities (at least in nuce) would, of course, be Charles Sanders Peirce – originator of the Index in the sense that Gell deploys the term. Indeed, Gell explicitly cites Peirce’s work as his inaugural authority for the concept of the index and of indexicality. But we should be cautious in supposing that Gell attempted to apply or refine Peirce’s theory of signs. As Peirce noted, any artifact-sign

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 201

13.1.2007 2:10pm Compositor Name: SJoearun

Abducting the Agency of Art 201 triangulates the notional sign-functions of indexicality, iconicity, and symbolicity. These functions must be tracked in turn to further triadic involutions in the continuing recursion of the ‘‘interpretant,’’ the sign as intelligible – resulting in Peirce’s well-known (though still arbitrarily limited) table of 66 logically possible sign-functions. There are no pure indexes.1 Gell himself dismisses the role of symbolic conventions in constituting what he calls ‘‘representation,’’ that is, depiction. But in Peircean terms the iconicity of depiction – in Gell’s terms, its seeming to represent a ‘‘prototype,’’ its denoted and pictured object – rests on a symbolic feature of iconicity: the prototype-resembling feature(s) of the ‘‘index’’ in this case must also (be abducted to) refer to the prototype it putatively resembles.2 And so on: the ‘‘impurity’’ of the index and indexicality – or more generally the mixed (triadic) nature of signs as Peirce classified them in terms of notional indexicality, iconicity, and symbolicity – ramifies in all directions throughout Gell’s model of the agency of art. One way of stating the issue here might be to say that Gell’s anthropology of art investigates the agent-caused identity (as it were the ‘‘metaindexicality’’) of the ‘‘indexicality,’’ ‘‘iconicity,’’ and ‘‘symbolicity’’ of artifact-signs in Peirce’s terms – that is, it examines indexicality, iconicity, and symbolicity as indexes of agency. In this sense, Gell’s model deploys two discriminable definitions of the index – as ‘‘natural sign’’ (Peirce’s ‘‘indexicality’’) and as agent-caused (Peirce’s ‘‘abduction’’). Both concepts of indexing derive from Peirce’s work: ‘‘indexicality’’ refers to one of the notional or logical dimensions of all sign-functions, and ‘‘abduction’’ refers to the cognitive activity of interpreting all signs as having been caused. Gell’s continual passage between the logical and the cognitive levels of analysis of indexing contributes to the richness and power of his work. But it can lead to certain questions and confusions. In particular, Gell’s model continually prompts us to ask about the ‘‘art index’’ in the ‘‘art nexus’’ as he describes it. How do we identify the limits of the intelligibility of the art-index in the indefinite ramification of agentcaused indexicality, iconicity, and symbolicity in the art nexus? Put another way, is it possible to specify the agency specifically ‘‘of art’’ in a nexus that fully includes all the agents inferred to have caused it (i.e., its ‘‘anthropology’’)? As we will see, it will not be possible to answer these questions in a fully satisfactory way on the basis of the arguments provided in Gell’s book. In the end Gell tends to need to say that ‘‘art’’ is abductively opaque or extraordinary – in his memorable phrase, it has ‘‘cognitive stickiness’’ (AA, p. 86; my emphasis) – even though this

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 202

13.1.2007 2:10pm Compositor Name: SJoearun

202 Whitney Davis conclusion would seem to have to undermine his historical ‘‘anthropology’’ of its notional or logical indexicality. It turns out that art is not quite like the smoke telling us that there’s a fire. It’s like a smoke that gets in your eyes.

Gellograms and Art History It will be helpful to understand Art and Agency as incorporating two books. The first book (chapters 1–5, and parts of chapter 6), which I will call AA-1, presents Gell’s substantive theoretical construction – namely, a concept of (and a quasi-formal notation for) the logical relations of the art nexus. This analysis is accompanied by brief ethnographic and art-historical demonstrations and exemplifications. To be sure, for some readers in art history or philosophical aesthetics AA-1 will be strangely distorted by Gell’s failure to engage the fundamental claims of idealist or critical aesthetics. In AA-1, in fact, Gell claims decisively to reject Kantian aesthetics as the presumption of his own project. In the Kantian tradition, however, aesthetic judgment (pace Gell) is conceived as an essentially social process of winning others’ agreements to one’s judgments of taste – that is, of the formation of a sensus communis or what Gell himself might call a ‘‘distributed person’’ or ‘‘extended mind’’ in the second part of his book, AA-2.3 In the end Kantian aesthetics might be thought to require or suggest – certainly it could readily endorse – an anthropology that might look rather like Gell’s own. To some extent this confusing feature of Gell’s presentation can be explained as a historiographic anomaly. In the ‘‘anthropology of art confraternity’’ to whom Gell addresses a good part of his book, especially AA-1, one frequently does find a total reduction of philosophical aesthetics and critical theory, an account of human sensuous knowing in its contexts of bodily finitude and historical social relations, to the question of an individual human being’s ‘‘art appreciation’’ (or, at a slightly higher level, to an ‘‘anthropology of aesthetic judgment’’ in a given historical community).4 To present his analysis in AA-1, Gell developed a suggestive quasiformal notation. It produces what might be called ‘‘Gellograms’’ that permit us to represent the relations of agency and the operations of abduction in the art-index. In turn these representations enable us to make revealing historical and cultural comparisons – comparisons that

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 203

13.1.2007 2:10pm Compositor Name: SJoearun

Abducting the Agency of Art 203 not only suggest particular neglected or new anthropologies or art histories but also, and more deeply, motivate the very project and claims of the ‘‘anthropological theory of art’’ in the first place. To take one of the most prominent examples discussed in Art and Agency, Gell’s notation enables us to identify a logical overlap – specific similarities in the structure and the routes of the abduction of agency – between an ‘‘idol worship’’ (or so-called fetishism) supposedly to be found in many pre-modern or non-Western cultural traditions, on the one hand, and the ‘‘aesthetic appreciation’’ of artworks characteristic of modern Western cultural traditions on the other hand. I will return to the details of this particular example momentarily. But in light of this comparison between idol worship and aesthetic appreciation, Gell’s critique of aesthetics (as he deploys the term) is well taken, though not novel: as an abduction of agency in which works of painting, sculpture, or other media of artifact- and image-making are believed to be essentially self-caused, or created by an artist ‘‘for their own sake,’’ the modern aesthetics specifically of ‘‘art’’ (or Art) reflects early modern and Enlightenment European art theories and idealist philosophies. In particular, it corresponds to Kant’s local doctrine of the ‘‘idealization’’ or ‘‘perfection’’ of disinterested judgments of beauty specifically attained by the subjective universalization of judgment (or aesthetic judgment in the broadest Kantian sense) in the ‘‘fine art’’ of an historical culture. (Needless to say, Kant’s theory of aesthetic judgment was not limited to – indeed, it was not even primarily interested in – these judgments of taste in ‘‘fine art’’; it chiefly addressed itself to judgments of beauty in nature.) As Gell urges, the deployment of this modern aesthetics of Art as a general anthropological theory of human aesthetic judgment outside modern Western cultural tradition has been misguided. By the same token, however, to identify critical aesthetics, or even the Kantian doctrine of the subjective universality of aesthetic judgment, with the ‘‘aesthetic appreciation of (the perfected beauty or the beautiful ideal of) art,’’ whether or not referring to modern Western cultural tradition, is equally misguided. Here, as I have already suggested, Gell indulges a self-serving historiographical amnesia. In a Gellian anthropology and represented or revealed by Gellogrammatic contiguities, ‘‘aesthetic appreciation’’ is nothing more – though also nothing less – than a modern Western cultural instance of the generalized cognitive-social possibility of ‘‘idol worship.’’ If we were to try to explain the efficacies of ‘‘idol worship’’ – or the effects of a particular ‘‘idol’’ – as

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 204

13.1.2007 2:10pm Compositor Name: SJoearun

204 Whitney Davis caused by or generated in ‘‘aesthetic appreciation,’’ as Roger Fry or Clive Bell might have done in the early twentieth century,5 we might entirely invert its social phenomenology and we would probably overlook its cultural-historical genealogies. Perhaps we could superficially describe but we certainly would not explain its agency. Other scholars have noticed generalized similarities between the seemingly disjunct kinds of use and understanding of artifacts and images embodied in ‘‘fetishism’’ and ‘‘aesthetics’’ respectively. Among others, David Freedberg and Hans Belting have observed historical connections between a longstanding devotional iconism in the European Middle Ages (in both Eastern Orthodox and Western Christian traditions) and an emergent aesthetic iconism in the Renaissance and later periods of the European tradition, culminating in peculiarly modern conceptions of aesthetic ‘‘disinterestedness,’’ ‘‘art for art’s sake,’’ and the like.6 But Gell would have us notice a deeper and broader – a structural, logical, and cognitive – affiliation: in Gell’s terms, ‘‘idol worship’’ is cognitively like ‘‘aesthetic appreciation’’ in specific ways made analytically visible to us in the Gellograms. To this extent Gell’s anthropology of art might rewrite the standard or received history of art by discovering hitherto unobserved parallels between different cultures of abducting the agency of art. Whether or not these parallels can be fleshed out in an actual art history – an account of historical interconnections and cultural interactions – was not, perhaps, Gell’s principal interest. Indeed, he drew heavily on existing art-historical portrayals of the transhistorical cultural processes that might have linked a European medieval iconism and a modern (and now global) aestheticism. Still, Gell’s account frequently suggests the possibility of substantive art histories that have yet to be written. The Gellograms ask us to organize and narrate art-historical evidence in new ways and direct us to look for new evidence that might well be mobilized in fulfillment of the Gellogrammatic or notional and theoretical predictions. At the same time, Gell’s account permits a rich anthropology: identifying the historical art nexus of idol worship and/or ‘‘art viewing,’’ as Gell calls it, leads us to the localized socio-cultural determinations of art abducted within one particular tradition. Needless to say, it is not easy to balance the claims of a general anthropology (let alone a cognitive typology) of art and the findings of particular art histories. As Art and Agency proceeds, it becomes increasingly obvious that there are no ‘‘pure cases’’ of agency in the vicinity of the art-index (i.e., a pure case like [Index-A ! Recipient-P] and similar

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 205

13.1.2007 2:10pm Compositor Name: SJoearun

Abducting the Agency of Art 205 logically possible structures in the art nexus [see AA, p. 29, table 1; figure 0.2, this volume]). (Gell identifies 16 logically possible permutations for the notional pure cases, though some of these are said to be cognitively ‘‘empty,’’ ‘‘illegitimate,’’ or non-realized; there will be 36 permutations of the more complex cases he chooses to address.) At best, the pure cases provide the ‘‘general formula’’ – but not a particular art history – of ‘‘artistic genius,’’ for example, or of ‘‘idol worship.’’ The real interest of Gell’s account, then, has to be that the degree and type of affiliation and/or disjunction in logical-cognitive relations (as represented in the Gellograms) might mark a real socio-cultural and historical affiliation and/or disjunction. In this spirit Gell claims that distinctive social practices of viewing art in modern Western nation-states were historically continuous with – if in certain specific ways an inversion of – idol worship; for Gell, idol worship and viewing art are closely related, if not continuous or identical, ‘‘general forms’’ of abductively interpreting agency in relation to artifacts. It is no wonder, then, that late-medieval European history should provide an example of seeming historical development from idol worship to viewing art. In Gell’s view, late-medieval and early-modern European image-makers simply continued to think what they had long ‘‘liked to think’’ about the agency of artifacts – even as these artifacts changed to some degree in their formats and functions as ‘‘icons’’ became ‘‘artworks.’’ Should we be able to find similar histories of development or transfer between idol worship and viewing art – a transfer facilitated cognitively by the putative likeness of the general form of the complex abductions involved – in other cultural traditions, for example, in ancient or nonWestern traditions? And by the same token, should we be able to find an inverse development from viewing art to idol worship within the cultural history of the European tradition itself? The strongly implied answer – virtually the explicit historical claim – of Art and Agency to both of these anthropological and historical questions is ‘‘Yes.’’ The general form of cognition in which idol worship and viewing art are logically affiliated (and can be materially inter-converted) is a kind of anthropological universal: if we set out to look for it in Gellian terms, we should expect to be able to find it in the canonical tradition of ancient Egyptian art in the third millennium BC or in the avant-garde practices of twentieth-century modernist European artists as much as in late-medieval and early-modern Europe. Ancient Egyptian depiction generated both icons and artworks, and its recipients probably moved cognitively between these related

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 206

13.1.2007 2:10pm Compositor Name: SJoearun

206 Whitney Davis possibilities for its abducted agentive constitution. And despite the fact that twentieth-century European artistic culture has supposedly superseded its medieval ‘‘iconism,’’ a ‘‘fetishism’’ or ‘‘idol worship’’ can surely be said to characterize much of the modern artworld. Gell’s anthropology (and the cognitive typology from which it proceeds) would seem to be on firm ground in both of these cases outside the particular historical interaction between devotional iconism and pictorial interpretation in the fourteenth and fifteenth centuries in Europe. Still, the Gellogrammatic approach does risk reducing very different historical types, degrees, or intensities of agency to mere equivalents in a cognitive typology. Consider, for example, Gell’s historical interpretation of the ‘‘Slashed Rokeby Venus’’ (AA, pp. 62–5), Diego Vela´zquez’s Rokeby Venus as slashed by Mary Richardson in 1914 in protest at the incarceration of the suffragette leader Mrs. Pankhurst. The nexus of agencies abducted by the ‘‘outraged’’ viewers of the vandalized canvas in 1914 comprised (1) being caused to be outraged (by Mary); (2) doing violence (twice, or in two different aspects – vandalizing a painting [by Mary Richardson] and assaulting a human being [by the prison guards]); (3) depicting (by Vela´zquez); and (4) being inspired (twice, or in two different aspects – being interested in a mythological figure as a subject for painting [on Vela´zquez’s part] and being interested in a real political actor as a model for society [on Mary’s part]). We might wonder whether all these agencies – for example, the picturing that realized the ‘‘inspiration’’ of Vela´zquez and the politicking that ‘‘inspired’’ Mary Richardson – were essentially equivalent holders of comparable cognitive slots in abductive inference. This is not to dispute the overall plausibility of Gell’s historical conclusion in this case – namely, that the iconoclasm of Mary Richardson seemed so shocking at the time (though as it were perversely appropriate) because of ‘‘the quasi-identity between the ‘mythological’ heroine, Venus, and her ‘modern’ counterpart, Mrs. Pankhurst’’ (AA, p. 65). But we knew this already, without Gell’s analysis of the recipients’ putative abduction of agency in this art nexus and without his demonstration of the way in which Venus and Mrs. Pankhurst could have been inferred to act in parallel ways as agents on Vela´zquez and Mary Richardson. Mary explicitly stated just this much in her own legal testimony in 1914 (‘‘I have tried to destroy the picture of the most beautiful woman in mythological history as a protest against the government for destroying Mrs. Pankhurst, the most beautiful character in modern history’’). At bottom Gell’s formalism in this historical case simply indulges and ratifies Mary’s public claims

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 207

13.1.2007 2:10pm Compositor Name: SJoearun

Abducting the Agency of Art 207 about her motivations. Here the Gellogram might be warranted as a kind of cognitive anthropology – an interpretive reporting of Mary’s own belief, inference, and utterance that offers an analytic schematization of their putative cognitive constitutions. But is it an anthropological art history? To stick with the example broached already, on the face of it Mrs. Pankhurst’s grip on the political imagination of Mary would seem to be quite different from the grip that Venus exerted on the pictorial imagination of Vela´zquez. As historians we could infer this precisely because Venus, so far as we can tell, never caused Vela´zquez to act in the way that Mrs. Pankhurst caused Mary to act – even if Gell’s cognitive typology lumps both of these agencies under one and the same descriptive explication (namely, artistic and political ‘‘inspiration’’). A formal (i.e., cognitive) parallelism in both the type and the degree of agency would seem to be implied in this case (Gell represents the type of agency in his superscript explications of the pure causal arrows of agency in the Gellograms but he tells us nothing about the degree of agency). In particular, the intensity of Vela´zquez’s painting of Venus would have had to be taken to be considerably like (even if it was in some ways also taken to be the inversion of) the intensity of the prison warders’ assault on the person of Mrs. Pankhurst as well as the intensity of Mary’s vandalization of the object of art. On the evidence, historians perhaps cannot say whether these cognitive equivalencings occurred in this case. It might be a virtue of Gell’s model, as I have already noted, that it reveals equivalences or an equivalencing – a history – that we might otherwise overlook. Obviously at the time people had highly to esteem Vela´zquez’s painting to be shocked by Mary’s attack on it. But again, we already knew that art-historical fact.

The Phenomenology of the Art-index and its Agency The second book in Art and Agency, AA-2 (parts of chapter 6, and chapters 7–9), presents Gell’s reflections on the relationships between styles of decoration or of what he calls ‘‘representation’’ (that is, depiction), on the one hand, and, on the other hand, ‘‘persons’’ considered in terms of the social genealogy and dissemination of their agency – terms that cut against certain engrained conceptions of integral personhood. Gell’s views of social personhood and distributed consciousness in AA-2 often repeat notions familiar within anthropology, sociology, and psychology, including but certainly not limited to notions adumbrated in the

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 208

13.1.2007 2:10pm Compositor Name: SJoearun

208 Whitney Davis phenomenology of intersubjectivity and in various schools of psychoanalysis. In particular, Gell’s approach echoes the views expressed (as so-called Melanesian deconstructionism) by the anthropologists Roy Wagner and especially Marilyn Strathern.7 More broadly, as Gell acknowledges (AA, pp. 9–10), his anthropology of art can be seen as a descendent (and as an application) of a conception of the relations between persons and artifacts in a network of exchange(s). (Gell explicitly acknowledges the influence of Marcel Mauss and Claude Le´vi-Strauss on his work – that is, on his formal-theoretical modeling of logico-cognitive possibilities. Thus he courts a parallel between his anthropology of art and structuralist social anthropology: we can have a ‘‘theory of kinship,’’ for example as gendered exchange in alliance and procreation, that is, as a type of ‘‘gift,’’ to account for social formations of intermarriage and lineage in relation to logically available – if abductively excluded or ‘‘tabooed’’ – possibilities.) Compared to the strict and forceful argumentation of AA-1, however, the various sections of AA-2 are not always tightly related analytically; several pieces of AA-2 – such as a long and bravura set-piece on the ‘‘Marquesan corpus’’ (AA, pp. 168–220) – are not entirely congruent with the theoretical or interpretive claims they are supposed to exemplify. In earlier work, Gell had sketched a sophisticated account of the temporality of intentionality based in large measure on Edmund Husserl’s phenomenology of consciousness.8 But its theoretical relation to Gell’s later and quasi-Peircean model of the abduction of agency, though readily apparent in the unfolding of Art and Agency, needs clarification. In my view, Gell’s Husserlian account of the retentive and protentive temporalities of intentionality (developed largely in AA-2 though based on Gell’s earlier writing) must underwrite his account of the abduction of the agency of art (developed largely in AA-1 and representing a newer turn in Gell’s work before his death). Yet the structure and direction of Gell’s argumentation in Art and Agency might lead one to suppose that his analysis of the abduction of agency underwrites his later conclusions about the multiplied intentionalities of art in a ‘‘distributed personhood.’’ Protention and retention, of course, belong as much to an individuated and integral personhood – they establish its coherent, continuous, and unitary ‘‘personal identity’’ over time and in relation to the analogous individuation of objects and other people – as they do to the distributed personhood that seems to stand behind, or to result from, the indefinitely ramified and recursively organized agencies specifically to be identified in the abductions of art’s agency.

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 209

13.1.2007 2:10pm Compositor Name: SJoearun

Abducting the Agency of Art 209 The same point applies to Gell’s pages on the Husserlian doctrine of the ever-moving Now (AA, pp. 238–41) – a crucial element of the traditional analytics of the intentionality of integral personhood or continuous ‘‘personal identity.’’ In his philosophical inclinations Gell clearly remained deeply attracted to it despite his embrace of a ‘‘deconstructionist’’ social anthropology. Of course, certain readers of Husserl – notably Jacques Derrida in his dissertation on Husserl’s history of the intuition of geometry in ancient Greece and in the essays published as Speech and Phenomena – have reached the conclusion that the ‘‘moving Now’’ of temporalized personal self-consciousness inherently produces a ‘‘lack of presence’’ in self-awareness or in ‘‘intentionality.’’ According to them, the ‘‘moving Now’’ constitutes the sign-in-consciousness or meaning, Logos, as essentially indecipherable in terms of its origin – what Derrida called e´criture.9 But if Gellograms are written out, they are not notations for ´ecriture – for at the limit, the ramifications and recursions of e´criture, as Derrida insisted, must defeat the very possibility of causal abduction, of identifying and understanding ‘‘origins.’’ A better substantive – rather than rhetorical – congruence between AA-1 and AA-2 would have obtained if Gell had considered that the Husserlian analytic of the temporality of consciousness analyzes the abductive structure of one index in-the-making (equivalent to one mind-in-modification, whether ‘‘extended’’ and ‘‘distributed’’ or not). In turn, this analytic requires a model of the abductive structures of agency as structures of protention and retention. The latter argument is lacking in AA-1. And oddly, in AA-2 Gell chooses to see Husserlian temporality as a model of singular (seemingly non-distributed) personal identity and awareness distributed in many indexes in turn causally understood to be the product of one agent. This approach (it seems to reflect a bias not required by the underlying philosophy) requires him to discover a factitious set of ‘‘protentions’’ and ‘‘retentions’’ linking up all the artworks in the complete canonical oeuvre of Marcel Duchamp (AA, pp. 242–51). One would have thought, quite conversely, that Gell would consider Duchamp’s oeuvre (one index to the next) to display the multiple, ramified agency of extended mind or distributed personhood. Indeed, art historians have long argued that Duchamp ‘‘intended’’ his work – and was abductively placed as an agent by the recipients of his work – in a critique of existing cultural notions of an individual artist’s integrated and continuous artistic ‘‘originality.’’10 The overall relations between the first book in Art and Agency (AA-1) and the second book (AA-2), then, are difficult to pin down. At points, the

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 210

13.1.2007 2:10pm Compositor Name: SJoearun

210 Whitney Davis analysis in AA-1 seems to warrant or to solicit the reflections offered in AA-2. At other points, the claims advanced in AA-2 (whether or not they are Gell’s original theoretical constructions) are needed to make sense of the analytic decisions in AA-1. And at still other points, the accounts of ramified indexicality in AA-1 and extended mind vested in distributed personhood in AA-2 do not appear to be fully consistent – at least if pushed to their logical conclusions or regarded in historiographical or theoretical terms. Nevertheless, it is clear that Gell himself wanted to see the argument of Art and Agency as continuous and cumulative. In principle, the general argument that should unite AA-1 and AA-2 would see an analogy between the art nexus as the abduction of ramified agency – agency not limited to the individuated personal identities of patrons, artists, or viewers – and the distributed personhoods of social people in culture. Indeed, the one should be the agentive cause (and therefore the index) of the other – or perhaps vice versa. A Gellian model of the causal-temporal imbrication of art’s ramified abducted agency and people’s extended agency in distributed personhood certainly constitutes a significant anthropological contribution to aesthetics and art history. Unfortunately, AA-2 does not directly pursue the analogy as I have just stated it. To be sure, Gell’s project in Art and Agency rests on the proposition, as already noted, that people bring their practical ‘‘inferential schemes,’’ or abductions, to indexical artifact-signs in a way ‘‘very like, if not actually identical to, the ones [they] bring to bear on social others’’ (AA, p. 15). We need not dispute the suggestiveness of Gell’s notion of equating artworks and persons as analogously or equivalently distributed. (Of course, a profound identification of artworks and persons has often been deployed, most notably in the writing of psychoanalytically minded critics of the arts.11 But the artworks and persons in question in these histories – despite psychoanalytic theories of intersubjectivity, transference, and object relations – have often been conceived to be relatively individuated and non-distributed in Gell’s agentive terms.) Still, the equation immediately raises the possibility that artworks, in Gell’s terms, must be just as mysterious, intractable, and resistant to interpretation as other people can be – and for largely the same reasons. Although we can see what they are doing, and we seem to know more or less incorrigibly and irreducibly what they are doing to us, often we do not clearly or fully know what they mean. In AA-2, Gell would have us avoid the ‘‘internalist’’ or mentalist quagmire (once known as the intentional fallacy) of ‘‘meaning.’’ He promotes

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 211

13.1.2007 2:10pm Compositor Name: SJoearun

Abducting the Agency of Art 211 what he calls an alternative ‘‘externalist’’ theory of mind as a ‘‘public’’ or ‘‘extended’’ social fact (AA, pp. 126–33) consistent with his social anthropology of distributed personhood. Nonetheless, Gell’s anthropology of art in AA-1 would track the nexus of agent-derived causality abducted in the vicinity of art-indexes. To this extent, it must be an anthropology of what people believe, and know they believe, and what they know, or believe they know, about the index – an anthropology, in other words, of the most wellestablished mentalist or ‘‘internalist’’ sort. At the same time, a decidedly externalist Gellian approach to the art nexus (if AA-1 were rendered fully consistent with AA-2) might itself be contravened by the internal disjunctiveness, complexity, and self-resistance of the index-as-abducted or in-abduction – what Gell sometimes calls its ‘‘indecipherability’’ (AA, p. 71).

‘‘Indecipherability’’: Plurality, Regress, and Opacity in the Abduction of Agency Gell acknowledges that the abductive structures and histories which he diagrams as ‘‘trees’’ of inference (AA, pp. 53–9) might have to be seen – in the phenomenological (and even quasi-psychiatric) terms to which he reverts – as a ‘‘maze’’ (e.g., AA, p. 86) or even as a ‘‘knot’’ (e.g., AA, p. 62). Indeed, following his earlier influential work on the ‘‘technology of enchantment’’ (see AA, pp. 68–73) Gell would hope to make the supposed ‘‘difficulty’’ of the artwork into the very engine of the abductive work people will expend on it.12 He sees agentive indecipherability chiefly in terms of technical virtuosity and visual-morphological intricacy: the more intricate the work of art seems to be in its formal constitution and effects, and the greater the technical virtuosity we identify and appreciate in it, the higher the degree of its ‘‘cognitive stickiness’’ or ‘‘indecipherability.’’ This is fair enough so far as it goes. At its first exhibitions in 1915–17, however, Kasimir Malevich’s Black Square was ‘‘difficult’’ not in virtue of its painterly virtuosity (which was minimal) or its morphological complexity (it was simple). Malevich’s Black Square was difficult – resistant to interpretation, opaque to a contemporary observer’s abduction of its generating causal agencies – because its recipients did not know what it could possibly mean in terms of any traditions of significance or agent-caused determination traditionally known to them. Nonetheless, by Gell’s terms it must have had some kind of abducted agentive identity – an anthropological intelligibility in and as ‘‘difficult.’’

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 212

13.1.2007 2:10pm Compositor Name: SJoearun

212 Whitney Davis To pursue this question, we must underline the plurality of ‘‘agencies’’ at work in Gell’s model of the agency of art and of art’s agency. Received art-historical models of the causal generation and intentional constitution of a work of art, their ‘‘patterns of intention’’ (to use Michael Baxandall’s phrase), have tended to focus on a small number of interacting agents (often living human social beings) as having been understood (abducted) by the work’s contemporary makers and other original users to have occasioned it – for example, an ‘‘artist,’’ a ‘‘patron,’’ and perhaps an anticipated observer or ‘‘viewer.’’ By contrast, Gell’s model of the agencies of art – and of art’s agency – can specify an indefinitely large number of highly differentiated, ramifying, and nested relations of putatively causal or abducted agency. We might address the agency, for example, of a divinity believed to be acting on a patron, the agency of materials believed to be acting on an artist, and the agency of a depicted object or entity (the prototype) acting on patron, artist, or viewer – or other salient abducted (and as it were primary) agencies of abducted (and as it were secondary) agencies. In particular, in principle Gell’s model elegantly handles recursive or proleptic and retrodictive relations in the resulting nexus. It can, for example, address the anticipated agency of a viewer as it were predictively acting on the agency of a patron or artist. Or to take a different possibility, it can investigate the continuous agency of a divinity or a human ruler differentially acting on the multiple agencies of patrons, artists, and viewers throughout the entire ramified history of the work of art that passes through their different kinds of intentional involvements with it. By the same token, as this mention of the differentiation of agencies will suggest, Gell’s model can readily recognize internally bifurcated, selfdeviating, or seemingly conflictual or contradictory agencies nonetheless believed to constitute art in the entire network or nexus of its causal and intentional relations. For example, the prototype of a depicted ruler (perhaps a mythic, heroic, or ancestral king) acting on the agency of the viewer might be self-evidently disjunct from – even opposite to – the real social personage and identity of the patron (that very same ruler) acting on the agency of the artist and artwork. Individual monarchs in ancient Egypt were typically figured as beings whose divine identity and authority devolved from an ancient past of rulership into each present-day context of rule. Each historical Egyptian king was metaphorically associated with a prehistoric culture of hunting and with immemorial traditions of observing and trying to predict and control the rhythms of the inundation of the

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 213

13.1.2007 2:10pm Compositor Name: SJoearun

Abducting the Agency of Art 213 Nile River and of the sun, moon, and stars. At the same time, monarchs usually asserted themselves as contemporary political actors, sometimes overtly ‘‘usurping’’ or even effacing the monuments of earlier rulers – as it were denying their predecessors’ very historical existence. Nonetheless, the ‘‘usurper’’ might well commission a depiction of himself in terms of the continuous tradition of Egyptian rulership – figuratively linking himself with the metaphors that had been employed by all previous kings as well, including the kings whose monuments were usurped or whose memorials were effaced.13 The perceptual and performative distinction between a continuous divine and a contemporary political kingship might have been readily understood by the working artist-craftsmen commissioned to ‘‘portray’’ the living historical king in terms of his divine identity (they had to balance the ‘‘divine’’ and ‘‘real’’ physiognomy of the king in elaborating an idealized ‘‘portrait’’ of him), and at least certain ordinary viewers as well, even though the prototype (divine king) and the patron (contemporary patron) might be similar or even identical beings. At the extreme, the distinction or disjunction between prototype and patron might lead to the kind of internally contradictory or destabilizing perceptual-cognitive and social-political results ingeniously identified in Louis Marin’s deconstructive analysis of the imagistic or representational – the specifically depictive – quandaries that confronted early-modern image-makers, beholden to emergent concepts of ‘‘portraiture,’’ of the traditional late-medieval European king’s ‘‘two bodies.’’14 Which body does the picture portray? The king’s prototype – in which case the ‘‘real’’ king would seem to be unnecessary? The king’s personage – in which case the prototype would seem to be unnecessary? Both? Neither? Of course, in this kind of complex and bifurcating nexus the prototype could be regarded as an agent acting on the patron even as the patron could be regarded as acting on the prototype in causing it to be depicted by the artist. In the end, viewers might not be able fully to infer and completely to distinguish the agency of prototype and the agency of patron in acting on the work – let alone be able fully to infer and completely to distinguish the agency of the work in acting on the prototype and on the patron, and so on and so forth. It is in these differentiations and recursions – however abductively complex and ambiguous they could be deduced to have been in a particular historical situation – that we might find the social power, the aesthetic ‘‘enchantment’’ and cognitive ‘‘indecipherability,’’ of the work of art. Many of the relations – and interactions – of agency incorporated by Gell’s model have, of course, been identified in established anthropologies

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 214

13.1.2007 2:10pm Compositor Name: SJoearun

214 Whitney Davis and histories of art. But these anthropologies and histories have tended to identify only some of the logically possible – and anthropologically or historically salient – possibilities. And they have privileged those possibilities at the expense of other logical options – options that could be anthropologically or historically salient. To take a simple example, entire literatures of art history have privileged the agency of patrons acting on artists and abducted by viewers as the causation/intention of the work of art. The common (even cliche´d) thesis that works of art serve to legitimate the social authority of the patron – or the patron’s lineage, status group, or class – depends on this limited or selective model. (In turn, of course, these art histories often emerged in order to dispute existing historical interpretations that had tended to privilege the agency of individual master artists.) In some cases, a simplistic model of patronage as the social context of art has led to obvious historical and interpretive distortions – to neglect of the role of the artist’s agency in destabilizing or undermining the legitimist agendas of patrons, of the role of the viewer’s abduction of alternate or proliferated agencies of the work of art in constituting a concept of who or what patrons and artists might be, of the abducted agency of artistic raw materials or of art’s artifactual materiality in constituting viewers’ inferences about the properties that require or solicit such causal understandings – and so on. So far as I know, Gell’s model offers the most comprehensive account of the open range and indefinite recursion of the agencies of art that have been, and equally important should be, considered by anthropologists or historians as the anthropology or the history of art. Gellograms can be indefinitely – perhaps even infinitely – complex. As I have implied already, there must always be a Gellogrammatic ‘‘agent’’ of an ‘‘agent,’’ and an ‘‘agent’’ of an ‘‘agent’’ of an ‘‘agent,’’ and so on – indefinitely and perhaps infinitely. But it is precisely because of this inherent cognitive possibility of regress that we might need to distinguish the logic of abduction and the phenomenology of abduction: what people might know is not necessarily what they do, in fact, actually know. Gell’s ‘‘anthropological theory’’ of the agency of art would have us track abductions of agency, as he put it, ‘‘in the vicinity of art.’’ Gell occasionally put scare quotes around this phrase (e.g., AA, p. 153); clearly it gave him trouble. What, or where, is the ‘‘vicinity’’ of art? At some point, the abduction of agency (as represented in Gellogrammatic terms) must become so complex, attenuated, and ramified – it supposedly identifies agencies of agencies of agencies of agencies for agency . . . – that it is difficult to see how it could have been abducted in the first place by the

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 215

13.1.2007 2:10pm Compositor Name: SJoearun

Abducting the Agency of Art 215 original users of the index (especially without Gellogrammatic assistance!). How, then, do we determine that our tracking of logically possible ramifications – ramifications admitted by the cognitive typology and possible in the cultural tradition in question – matches the purview of the original historical activity of abduction? Simply put, how ‘‘far back’’ or ‘‘deep into’’ the nexus of causal agency are we supposed to go to understand the anthropology and the history of art’s agency? As in the instance of Mary Richardson’s ‘‘Slashed Rokeby Venus’’ of 1914, a number of Gell’s examples rely on predigested histories in which the ramifications of the original abductions have been identified in advance (though not, of course, using Gell’s typology, terms, and notation). But in the end the purview of the ramification admitted for investigation seems to be set by Gell’s theoretical model, despite its elegance, more or less arbitrarily – that is, by his a priori decision that four terms of agency will be treated as nested (i.e., as representing the anthropological and phenomenological purview of actual abductive recursions in the historicity of art) through three removes. This recursion takes the canonical form (e.g., AA, 52):

[[[I-A] ! II-A] ! III-A] ! IV-P Many if not most of the Gellograms conform to this underlying plan. I can find no principled reason, however, why the formula should be definitive in any way. Indeed, it constitutes an ‘‘anthropological’’ vicinity for the artindex – an ‘‘anthropology of art’’ – merely by stipulation. Moreover, and against this very formula, the most productive and intelligible statement of Gell’s anthropology of art requires the possibility of an indefinitely extended and endlessly ramified recursion – a recursion not necessary limited to three removes of four types of agency. There can be no universal answer to the question of – no universal formula for – the actual depth and directions of abduction. In some cases, perhaps it suffices for people to abduct the most proximate causal agency of the artist (i.e., Gell’s ‘‘Artist-A,’’ the ‘‘first’’ agency to the left of final ‘‘patiency’’ or reception) in relation to the index as its final or ultimate agency as well. This, of course, would be the general form of the habitual abduction in the culture of ‘‘artistic genius’’ that Gell describes at one point. In other cases, however, perhaps people abduct causal agents exerting agency through the artist to the Index. Gell provides several cases of this kind (e.g., in his discussions of patronage) and we can readily imagine (and in a cognitive typology we can easily schematize) agent-causes of a

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 216

13.1.2007 2:10pm Compositor Name: SJoearun

216 Whitney Davis patron’s agency and agent-causes of the agent-causes of a patron’s agency, and so on. But do these background agencies in the nested recursions identified in Gell’s notations – Gell calls them ‘‘subordinate’’ in order to designate their relative distance from the proximate agency supposedly abducted for the index – constitute the abducted Prime Mover in the overall abduction? Ordinarily it is a matter of historical or interpretive tact to know where to stop and on what to focus. But Gell wants us to track the actual abductions people really do make in identifying the agency of the art-index. If necessary, then, he would track abduction through indefinitely extended ramifications involving agent-entities quite remote from the ‘‘vicinity’’ of art itself. As he says, he would have no qualms if his anthropology of art converted itself, say, into an anthropology of religion when the abducted causal nexus of inferred generative agency increasingly involves religious agencies as one tracks backward (in his notation, leftward) from the abduction of the proximate artistic agency of the index. This conversion would seem eminently natural in an anthropology of art as a branch of general anthropology and in a phenomenology of art framed in a cognitive typology of all logically possible modes of abducting agency. But it stretches credibility in explicating the particular historicity of the art-index itself. Indeed, in Gell’s own terms the proximate agency (i.e., the rightmost agent specified in the notations) must be the most important causal determination of the index precisely because it mediates the subordinate and increasingly ultimate agents (i.e., the leftmost agencies specified in the notations). As an anthropological proposition it might be perfectly true to say that when people abduct the causal agency of an index they attribute its properties to the unalterable plan of the gods – a plan established for all things, not limited to pictures or artifacts, at the founding of the world. Canonical depiction in ancient Egypt, for example, was authorized by a belief of this kind – a fundamental ideology of the image. But in principle the ideology applied to everything in ancient Egyptian experience. As an art-historical proposition, then, it would be more desirable to say that a canonical pictorial representation in ancient Egypt was manufactured by artists working within the tightly constrained parameters of an academic system of craft-specialized production; despite the ideological mystification exerted by the theory of ultimate agency, the agentive causality of the picture – the icon or artwork – would have been abducted as such by its recipients or viewers. To be more specific, in Gell’s terms canonical depiction in ancient Egypt took the primary ontology (namely, the divine plan) as its prototype. But it remains unclear, I think,

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 217

13.1.2007 2:10pm Compositor Name: SJoearun

Abducting the Agency of Art 217 to what degree the more proximate agencies of patrons, artists, and viewers interfered with – even destabilized – this axis of ultimate agency in the entire nexus.15 In advancing its anthropology, Gellian analysis must track the abduction of agency in the indefinite and indefinitely ramifying regresses of causes in the ‘‘vicinity of art.’’ In advancing an art-historical proposition, analysis tracks the causal nexus within which art – art specifically – was actually made and used. There need be no conflict between the anthropological model and the historical investigations. But because abduction as a cognitive process has no logically necessary endpoint or ‘‘origin,’’ to come to historical terms with the particularity of an index – with the types, degrees, and intensities of agency it is abducted to have as what I earlier called a material meta-index of the ramified indexicality, iconicity, and symbolicity of artifact-signs – we must provide not only an anthropology of art but also an anthropology of art. At the deepest level, Gell’s model does not so much equate ‘‘art’’ – a nexus or theater of potential abductive ‘‘difficulty,’’ ‘‘indecipherability,’’ or ‘‘cognitive stickiness’’ – and ‘‘agency’’ – the habitus of practical abducted instrumentalities – as distinguish them. It is precisely for this reason that we can construct an ‘‘anthropological theory’’ of art, as Gell puts it in the subtitle of his book, that is, an agentive theory of general aesthetic effectivity in art, as well as investigate different and diverse particularized histories of art. Gell himself does not quite make this argument in Art and Agency, in part, I think, because he resists the notion that his project has any kind of ‘‘aesthetic’’ presumption or issues in particular conclusions about ‘‘aesthetics.’’ But it would seem to flow quite naturally from his analytic model. In the sense that Gell fundamentally deploys the term in his book on ‘‘art and agency,’’ ‘‘art’’ always permits and relays recursive, bifurcated, and potentially self-deviating or self-contradictory inferences. In other words, art essentially relays an enigmatic, uncontrollable, or partly unknowable or partly uninferable overall agency – an agency only partly integrated and totalized by its recipients. Presumably such forms of ambiguity have little or no practical cognitive place in non-artistic domains of human socio-cultural life – especially in the instrumental toolmaking that has often been distinguished from aesthetic awareness and activity. To this degree it could be said that the agency of art in Gell’s anthropology – but cutting against the grain of his own statement of his model – lies specifically in the immanent non-abductibilities of its putative agencies.

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 218

13.1.2007 2:10pm Compositor Name: SJoearun

218 Whitney Davis Gell’s anthropology of art, then, issues in an ‘‘aesthetics’’: in Gell’s theory, ‘‘art’’ can be defined as the domain of artifact- and image-making within which the abductibility of agency in artifact-signs tends to reach its limit of cognitive possibility, its phenomenological and historical thresholds of feasibility, transparency, and intelligibility. In dealing with a work of art, as with anything else, it is logically possible to abduct agency to its furthest – its most ultimate – ramifications and regresses of natural, human, and cosmic causation. But as aesthetic objects, works of art do not fully encourage or even permit abducting their agency in this way: the ‘‘agency of art’’ remains partly vested in its own peculiar resistance to its complete abduction. In art, ‘‘agency’’ sticks to the artifact itself.

Notes 1

2

3

4

5 6

7

8

For the mixed nature of signs according to Peirce, see A. W. Burks, ‘‘Icon, index, and symbol,’’ Philosophy and Phenomenological Research 9 (1944), pp. 673–89, and J. Elkins, ‘‘What does Peirce’s sign theory have to say to art history?’’ Culture, Theory and Critique 44 (2003), pp. 5–22. See W. Davis, Replications: Archaeology, Art History, Psychoanalysis (University Park, PA, 1996), pp. 66–94, and ‘‘The presence of pictures,’’ Journal of Aesthetics and Art Criticism 29 (2001), pp. 29–38. Salim Kemal’s expositions of Kant’s aesthetics highlight this point: see Kant and Fine Art: An Essay on Kant and the Philosophy of Fine Art and Culture (Oxford, 1986), and Kant’s Aesthetic Theory: An Introduction (New York, 1992). Of course, some anthropologists have engaged the ‘‘critical theory’’ of Walter Benjamin and Theodor Adorno, and Gell in turn addresses that response (e.g., AA, pp. 99–104). See especially Roger Fry, Last Lectures (Cambridge, 1939), and W. G. Bywater, Jr., Clive Bell’s Eye (Detroit, MI, 1975). See D. Freedberg, The Power of Images (Chicago, 1989), and H. Belting, Likeness and Presence: A History of the Image Before the Era of Art, trans. E. Jephcott (Chicago, 1994), and The Invisible Masterpiece, trans. H. Atkins (London, 2001). Gell’s essay ‘‘Strathernograms, or the semiotics of mixed metaphors,’’ in The Art of Anthropology: Essays and Diagrams by Alfred Gell (London, 1999), pp. 29–75, was clearly a precursor of Art and Agency. See especially The Anthropology of Time (Oxford, 1992), pp. 221–42; cf. AA, pp. 238–41.

Osborne / Art’s Agency and Art History 1405135379_4_009 Final Proof

page 219

13.1.2007 2:10pm Compositor Name: SJoearun

Abducting the Agency of Art 219 9

10

11

12

13

14 15

J. Derrida, Edmund Husserl’s Origin of Geometry: An Introduction, trans. J. P. Leavey, Jr. (Stony Brook, NY, 1978), and Speech and Phenomena, and Other Essays on Husserl’s Theory of Signs, trans. D. B. Allison (Evanston, IL, 1973). See especially R. Krauss, The Originality of the Avant-Garde and Other Modernist Myths (Cambridge, MA, 1985), and T. de Duve, Pictorial Nominalism: On Marcel Duchamp’s Passage from Painting to the Readymade, trans. D. Polan (Minneapolis, 1991), and Kant After Duchamp (Cambridge, MA, 1996). See especially R. Wollheim, The Thread of Life (Cambridge, MA, 1984), Painting as an Art (Princeton, NJ, 1987), and The Mind and Its Depths (Cambridge, MA, 1993). See A. Gell, ‘‘The technology of enchantment and the enchantment of technology,’’ in J. Coote and A. Sheldon, eds., Anthropology, Art and Aesthetics (Oxford, 1992), pp. 40–67. For the ancient – even prehistoric – traditions of representation within which ancient Egyptian kingship was depicted, see W. Davis, Masking the Blow: The Scene of Representation in Late Prehistoric Egyptian Art (Berkeley, CA, 1989). The complex problems of usurpation and effacement have been outlined by G. Bjo¨rkman, Kings at Karnak: A Study of the Treatment of the Monuments of Royal Predecessors in the Early New Kingdom (Stockholm, 1971). For the ‘‘cultural temporality’’ of canonical depiction in Egypt, see W. Davis, ‘‘Archaism and modernism in the reliefs of Hesy-Ra,’’ in John Tait, ed., ‘‘Never the Like had Occurred’’: Egypt’s View of its Past (London, 2003), pp. 31–60. See L. Marin, The Portrait of the King, trans. M. M. Houle (Basingstoke, 1988). See W. Davis, The Canonical Tradition in Ancient Egyptian Art (New York, 1989), pp. 192–224.

Osborne / Art’s Agency and Art History

1405135379_6_index Final Proof

page 220

24.1.2007 12:07pm

Index

abduction, Gell’s notion of 3, 10–12, 14, 16, 19, 44, 49, 58, 102, 115, 163, 171, 180, 196, 199–200, 202–3, 210–11, 214–16, 217–18 Abydos, high-status tombs at 34–5 Achilles 141 aesthetics, agency of, in ancient Athens 188; in China 103, 106–7; Gell on vii, 1, 2, 3, 6–7, 8, 24, 59–60, 99–100, 184, 200, 202, 203, 217, 218; limitations of 16–17; of production 9; in twentieth-century Britain 181–2, 184–5 affective properties, agency of 43–4, 60 agalmatophilia 163 agency, of art object vii, 1, 2, 3, 6, 7, 8, 9, 12, 22–4, 29–30, 39, 42, 44, 49, 50–3, 55, 56–8, 61–2, 78, 85, 99, 110, 115, 125–7, 138, 139, 143–4, 151, 161, 162, 182–3, 184, 188, 194, 196, 199–200, 204–5, 218; and naturalism 163, 165, 166, 184–5, 196, 201; of artist 3, 15–16, 73–8, 83–4, 98–9, 188, 212, 214, 217; of clothes 110;

culturally variously recognized 20, 22, 23, 42–9, 58–9, 61, 91; of god 47–9, 55, 165, 216–17; of humans 2, 11, 47–50; of ideas 184, 189–90; multiplicity of 212; of nature 120–1; operating on three levels 109–10; of patron of art vii, 3, 60, 99, 104, 122, 212, 214, 216, 217; of prototype 14–16, 70, 82, 84, 85–90, 99, 184–5, 189, 212; simultaneous agent and patient 189; of tattoos 8, 143–4, 151; of tombs 31–7; of wearer of portrait miniature 124–6, 129; see also abduction, cult image, culture, enchantment, viewers agency, grammar of, in Sumerian 43–6 Ajax 141 Akkadian, see Sumerian Alexander of Abonoteichus 165 ancestors, power of 38; offerings to 97–8, 110; role of 103; see also death aniconic images 158, 165–6, 173–5 anthropology of art, Gell and vii, 1, 2, 3–5, 24, 31, 44, 59–61, 135–6,

Osborne / Art’s Agency and Art History

1405135379_6_index Final Proof

page 221

24.1.2007 12:07pm

Index 221 159, 199, 200, 202, 203, 211, 214–15 anthropomorphism 163; see also cult images Aphrodite (Venus), of Cnidus 163–4; at Paphos 172–3; see also Vela´zquez apotropaic patterns/objects, Gell on 7; gems as 120–1, 122; in Greece 81; in Mesopotamia 42, 44 archaism, in statues 168 Armada Jewel, see Heneage Jewel art history, comparative 70–94; genealogy of 1, 13, 135, 204, 214; see also Gell ‘‘art nexus’’ Gell’s construction of 1, 10–23, 39, 59, 70–3, 85, 90–1, 115–16, 137, 138–9, 152, 154, 168, 171, 180, 199–201, 205, 210 Artemis (Diana), of Ephesus 168 artist, see agency, of artist Asklepios 82 Asmat shield 3 Athens, drinking parties at 190; painted pottery from 158, 179, 185–95 Augustus 170 Aztecs, sacrificial victims of 142 Bal, M. 5, 9, 10 battle, representation of 50 Baxandall, M. 6, 212 beauty 6, 31, 182, 184–5, 189, 206; see also Kant Bell, C. 204 Belting, H. 76, 161, 204 birthdays, celebration of 80, 83 body-art 5, 143–4, 151; see also tattooing Boyer, P. 11, 73, 100 Brinker, H. 73 Bryson, N. 5, 9, 10

Buddhism (Ch’an/Zen), and portraiture 73, 74, 83–4, 89; and ritual enlivenment 42, 45; see also Eizon calligraphy, in China 100–1; on Heneage Jewel 117 Capitoline Triad 172 China, representation in 78–9; role of ancestors in 97, 103; see also Buddhism, dragons, Wanli emperor Christ, depiction of 17, 29, 70, 72, 84, 85, 86 Christianity, and representation of the divine 42, 53, 72, 73, 83, 84–5, 204; on Roman cult images 172; see also Christ, icon churches, perceptions of 28–9 Cicero, on Epicureans 78; on iconography of gods 170 Coote, J. 6 Coxon, R. The Intruder 191–4, 195 cult image, agency of 53–5, 73, 162; Greco-Roman 158–9, 161–75; and question of belief 162–3; see also theurgy culture, and agency 22–3 Cyrene, faceless funerary statues at 173–4 D’Alleva, A. 71 Daniel the Stylite 89 Daston, L. 123 Davis, R. 161 Davis, W. 14 death, commemoration of, in Buddhist East Asia 73–8; in China 95–7, 103–11; in Egypt 32–8; among Epicureans 78–80, 84

Osborne / Art’s Agency and Art History

1405135379_6_index Final Proof

page 222

24.1.2007 12:07pm

222 Index Dennett, R., on creation of nail-fetish 18 Derrida, J. 209 Dinka 6 Dio Chrysostom, on cult images 172 Dionysos, female devotees of 191, 195; and satyrs 195 distributed agency 30, 44, 49, 50, 55, 61, 62–3 distributed person 19, 21, 23–4, 54, 62, 137, 142, 151–2, 169–70, 196, 202, 208–9 Donnan, C. 140–1 dragons, significance of in China 95, 103, 104, 105, 106, 111 dream, motivation by 47 drunkenness, of satyrs 190 Duchamp, M. 153, 209; Large Glass 1; Network of Stoppages 22 Durkheim, E. 28–9 Edessa, Mandylion of 85–7 Effimova, A. 9 Egypt, art of 31, 39; attribution of divine origin in 216; funerary rites in 31–7; kings of 212–3; sculpture of 31; see also mastaba tombs Eire, C. 28 Eizon, Buddhist patriarch, portrayal of 70, 77, 78, 80 Elizabeth I, Queen of England 114–31 Elsner, J. 14 emperor, Chinese clothes of 95–7, 104–8, 110; images of 74; Roman, images and cult of 169–70 enchantment, by art 3, 9, 10, 12, 24, 29, 36, 37–8, 78–80, 84, 90, 102, 111, 115, 118, 123–7, 130, 137, 138, 145, 151, 152, 161, 211, 213 Epicureans 72, 74, 78–81, 84, 169

Epicurus, Greek philosopher 78, 80; celebration of after death 78–80, 84; garden of 78–83, 84; portrait of 70, 73, 74, 78–84, 91, 166; on representation 82 ergativity, in early Sumerian 56–9, 67 Etruscans, architecture of 38 etua motif 143, 145, 151; Gell on 7–8 Faris, J. 5 Faure´, B. 76 Feeney, D. 172–3 Forge, A., and the meaning of art 2 formal qualities, importance of 6, 7 Fozi, S. 54 Freedberg, D. 8, 161, 204 Freud, S., and limits of conventionalism 9 Fried, M. 9 Frischer, B. 81 Fry, R. 204 Garter, Order of 120 Gell, Alfred Art and Agency applicability of 159–60, 162; on art 2–3, 31, 184, 201–2, 217–18; and art history 2, 3, 6, 30, 60–1, 160, 194–7, 200, 204, 216–17; on culture 2, 21; fractal gods in 80; importance of vii, 10, 60–1, 135; use of Lucretius 81; parallels with work of Bryson 5; on patterns 7, 28, 30, 39; politics of 39; power relations in 115; reactions to 1, 60, 68–9; rejection of aesthetics in vii, 1, 2, 3, 6–7, 8, 24, 59–60, 100–1, 184, 200, 202, 203, 217, 218;

Osborne / Art’s Agency and Art History

1405135379_6_index Final Proof

page 223

24.1.2007 12:07pm

Index 223 rejection of linguistic model of art 2, 3–5, 6, 10, 57, 100, 160, 200; see also abduction, agency, anthropology of art, apotropaic patterns/objects, ‘‘art nexus’’, distributed agency, distributed person, enchantmant, etua, Gellogram, iconography, ‘‘index’’, Maori meeting house, nail fetish, Panofsky, patterns, psychology, semiotics, sociology of art, style; Wrapping in Images 8, 71; ‘‘Technologies of Enchantment’’ 10, 29–30; see also enchantment Gellogram 2, 14–22, 23, 24, 179, 181–5, 188–9, 202–7, 209, 214–15, 216 gift economy, Elizabethan 116–17, 122, 126, 127, 129–30 Gordon, R. 166 Gorgon, on Greek hoplite shield 3, 80 Greeks, ancient, on likeness 74, 75, 78–84 Gudea, ruler of Lagash 43, 47–50, 53, 54, 55, 56, 58 gymnasium, in Athens 185–8; and homosexuality 191 Heneage Jewel 117–31 Heneage, Sir Thomas 117–20, 122, 130 Henri IV, king of France 125–6 Herakles/Hercules 82, 140, 141 Hermes, images of 73 herms 173 Hilliard, N. 114–31 Hindu images 159, 161, 162, 167 Hinduism, and ritual enlivenment 42, 45 homosexuality, among satyrs 190; and Athenian symposion 191 Horus falcon 34

Hosios Loukas, monastery of 89 Hotman, J. 127 Husserl, E. 208–9 icon 70, 72, 84–90; mode of operation of 14, 81; rituals involving 83 iconography, alternatives to 10; and belief 109–10; Gell and 2, 8, 110, 153, 155, 160, 166–7; value of 24, 59, 60, 91, 166–71 icons 161 idolatry 19, 21, 54, 87, 159, 161, 162–3, 168, 170, 172, 189, 203, 204, 205 Inca, accounts of by Spanish conquerors 155 ‘‘index’’, Gell on 2–3, 10–19, 23, 44, 50, 55, 57, 70, 115, 180, 199–202, 210–11, 215–16 Ingres, J., Mme Moitessier 12 inspiration, language of 182, 189, 193, 206 Izzet, V., on Etruscan architecture 38 Jacobsen, T. 58 Johnson, Dr. S., portrayed by Reynolds 15–16, 20, 181, 195–6 Kant, I. 6–7, 202, 203 Kemmerer, S. 58 Kleophrades Painter, hydria by 191–4 kolossoi 72, 81 kula exchange 3 Lagash, city-state of 43, 45, 46, 48, 50, 51, 52, 53 landmines, agency of 2, 12 Layton, R. 60, 115 Le Puy, Virgin of 54 Leonardo da Vinci, Mona Lisa 15–16, 21, 82, 181, 188

Osborne / Art’s Agency and Art History

1405135379_6_index Final Proof

page 224

24.1.2007 12:07pm

224 Index Le´vi-Strauss, C. 208 limning 114, 117–18, 123–8 Lucretius, On the Nature of Things 80, 81 magical powers, and captivating art 3, 9, 12, 30, 73, 100, 165; see also enchantment Malevich, K., Black Square 211 Manet, E., De´jeuner sur l’herbe 193; Olympia 192 Maori meeting house, Gell on 21, 22, 29, 37–8 Marin, L. 213 Maritime frieze, from Huaca Cao Viejo 144–52 Marquesan islands, art of 7–8, 21–2, 145, 151, 153, 196, 208 mastaba tombs 23, 31–7, 38–9 masturbation, on Athenian pottery 186–8, 190, 191, 197 Mauss, M. 116, 208 Maximus of Tyre 172 Mesopotamia, affective objects in 42–69 mind, capacity of for abduction 11; as external disposition 1 miniatures, Elizabethan 24, 114–31; see also portraits Mitchell, W. J. T. 9 Moche, art of 24, 136–56; ceramics of 136, 137–42, 144; murals/friezes of 144–52, 154; portrait head vessels of 137–42; representational art of 136, 152, 154; society of 136; temples of 136, 144, 152, 154; treatment of prisoners by 140–2, 152–4 Morphy, H. 6, 60 nail fetish, Gell on 1, 16–19, 181, 184 Nearchos, aryballos by 185–91 New Grange 7

New Hebrides 7 Ningirsu, Mesopotamian deity 43, 47, 55, 56, 57 Noah’s ark 117 Nuba body art 5 nude, female, tradition of 192–3; see also Praxiteles, Vela´zquez omen texts 53 Ovid, on Pygmalion

163

pairs, of pairs, in Andean thought 146, 155 Palmer, F. R. 57 Pankhurst, E. 20, 181–5, 195, 206–7 Panofsky, E. 91; Gell on 2 Park, K. 123 Passienus 165 patronage, and agency vii, 3, 60, 98 patterns, Gell’s analysis of 7, 28, 30; in Maritime Frieze 144–52; on mastaba tombs 31, 36, 38; in Ming China 103; see also Maori meeting-houses Peirce, C. S. 11, 200–1, 208 Peru, Moche of see Moche photography 14 Picasso, P. 74 pictures, personhood of 9, 207 Plato 171; portrait of 83; on representation 74, 80–2, 88; Symposion 191 Pliny, the Elder 78, 165, 168 Pol Pot 2, 12 Polykleitos 171 pornography 24, 180, 190–1, 194, 196 portraits 13, 15, 19, 23, 70–91; and agency 70–4, 76, 99–100; in Ch’an Buddhism 74–6; in China 95–111;

Osborne / Art’s Agency and Art History

1405135379_6_index Final Proof

page 225

24.1.2007 12:07pm

Index 225 in Garden of Epicurus 83; epistemology of 72, 74, 80–1, 97, 124, 138, 140–1; made from bodily materials 76, 85, 88; miniature, role in courtly selfpresentation 119, 129; role in political history 115–17, 120–3, 125–7, 129–30; technical captivation by 114–18, 121, 123, 125–6; Moche portrait head vessels 137–42; paradox of 72, 113; Roman 89, 169–70; as substitute body 72, 75–6; technologies of 76–7; theology of 72, 87–8; see also Eizon, Elizabeth I, Epicurus, Johnson, Wanli Emperor Praxiteles 163–4, 168 Preziosi, D. 8 primitive art, as classification 1 propaganda, art as 14 psychology, cognitive and art historians 9–10; and Gell’s theory 3, 8, 10, 11–12, 19, 30, 42, 101–2, 146, 201 Pygmalion 163, 165 pygmies, on Athenian pottery 186 realism 14; of Soviet Socialist Realism 9 religion, and abduction 11, 102, 216; and art 8, 17–19, 23–4, 28–9, 30, 74–6, 84–90, 158–74; see also apotropaic patterns/objects, ritual reliquary 73, 77, 89 Reynolds, Sir J., portrait of Dr. Johnson 15–16, 20, 181, 195–6 Richardson, Mary 19–20, 181–5, 195, 196, 206–7, 215 ritual, and art 8, 18–19; in Ch’an/Zen Buddhism 77;

in Egypt 30, 32–3, 37–9; in Greco-Roman world 158, 165–6; in Mesopotamia 42, 44–5, 46, 49; in Ming China 95–8, 104–5, 107, 109; in Moche Peru 141–2; see also enchantment, religion, sacrifice sacrifice, by Aztecs 142; at mastaba tombs 31–2, 36–7, 38–9; in memory of Epicurus 78, 80; of prisoners in Moche Peru 141–2; in Roman cult 166; by Wanli Emperor 105 St. John Chrysostom 84 St. John of Damascus 87–8 Sallaberger, W. 53 Saqqara, mastaba tombs at 32, 33, 36 satyrs, on Athenian pottery 186–95, 197 sculpture, Buddhist 70, 74–6, 82–3; Egyptian 31; Greco-Roman 31, 78–84, 158–9, 161–74; Greek 163–4, 168, 171, 185, 190; Moche 144–54 seals, Mesopotamian, and agency 49–50 semiotics, Gell’s hostility to vii, 2, 3–5, 6, 10, 57, 100, 160; uses of 8, 9, 10, 11, 22–3, 168, 200–1 Seneca 83, 84, 171, 172 serekh motif 34 sexual power, of Aphrodite of Cnidus 165; of Nearchos’ aryballos 186–91; of the Rokeby Venus 185; of scenes on painted pottery 179–80, 186–95 Shelton, A. 6 Shenzong, see Wanli Emperor

Osborne / Art’s Agency and Art History

1405135379_6_index Final Proof

page 226

24.1.2007 12:07pm

226 Index Shinchi Kakushin 77 Shu Jing 95–7 Sinai, icon from 70 social facts, Durkheim on 29 sociology of art, Coote and 6; Gell and 2, 5, 6, 8, 10, 19, 20, 21, 29, 31, 59–61, 91, 100, 102, 108, 115, 143, 159, 180, 196, 214 statue, speaking 53 Strathern, M. 208 Strong, R. 116 Stuart, J. 98 style, feature of society 21; Gell on 21, 60, 196 Sumerian, agentive grammar of 23, 43–6, 55–9, 61 symbolic meaning of art 2, 3, 6–7, 8, 9, 10, 12–13, 23, 30–1, 61, 115, 129, 154–5, 188, 201 Tacitus, on Aphrodite at Paphos 173 Taiwan, portraits from 77 tattooing, significance of 8, 71, 143–4, 151, 153 technical skill, role of realizing agency 3, 12, 71, 98, 114–15 technologies of production, importance of 5, 12, 71, 90, 98, 100–2, 105–6, 110, 115–16, 117–18, 138, 140 temples, agency of 55–8; building of in Mesopotamia 43, 45–9 text, and image 46, 47–9, 50, 53, 54, 61, 63, 153–5, 198 Thasos, gate relief from 190 Theodore of Studion 88 theurgy 165, 167 Thomas, N., on Gell 44, 101; on Maori meeting-houses 37 Trobriand Islanders 3–4, 12, 100, 110, 153

Tusculum, sacred beech tree at Unton, Sir Henry

165

125–7, 130

vagina, representation of 8 Vela´zquez, D., Rokeby Venus 19–20, 181–5, 195, 196, 206–7 Verano, J. 141 Verhagen, A. 58 Vernant, J.-P. 76 viewers, as agents 129, 212, 217; animating portraits 77; historically determined 10; multiplicity of 9, 196; as patients 3, 13, 14–15, 57, 85, 115, 126, 188, 197; physiologically determined 10; producing change in art 14; see also abduction Virgin Mary, images of 53–4, 70, 85, 86, 89–90 visuality, concept of 10 Wagner, R. 208 Wang Chong 103, 107 Wanli Emperor, Shenzong 23–4, 95–111; agency of through portrait 98–9; tomb of 104–11 Western art, as model for looking at other art 2, 6, 17, 60, 98–9, 160–1, 203, 205–6 Wicks, R. 60 workshops, importance of in China 98, 101–2 xiang 72, 74, 81 Xunzi 107 Zeki, S. 101, 102 Zen gardens 14