Studies in Numismatic Method Presented to Philip Grierson

Studies in Numismatic Method

Philip Grierson

Studies in Numismatic Method presented to PHILIP G RIERSON

edited by

C. N. L. BROOKE, B. H. I. H. STEW AR T, J. G. POLLARD andT. R. VOLK

CAMBRIDGE UNIVERSITY PRESS Cambridge London New York New Rochelle Melbourne Sydney

CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo, Delhi Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521225038 © Cambridge University Press 1983

This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 1983 This digitally printed version 2008

A catalogue record for this publication is available from the British Library Library of Congress Catalogue Card Number: 81-15524 ISBN 978-0-521-22503-8 hardback ISBN 978-0-521-09133-6 paperback

Contents

page vii

Acknowledgements

viii ix

Abbreviations

Phi lip Grierson's contribution to numismatics

I

A bibliography of the writings of Phi lip Grierson Introduction Thoughts on the beginnings of coinage

xv xxvi

M. J. PRICE 2

The life of obverse dies in the Hellenistic period

II

OTTO M0RKHOLM

3 Roman coinage of the Cyrenaica, first century

BC

to first century

AD

23

T. V. BUTTREY

4 Roman imperial coin types and the formation of public opinion

47

M. H. CRAWFORD

5 Coin hoards and Roman coinage of the third century

AD

R. A. G. CARSON

6 Belgian finds of late fourth-century Roman bronze

75

J. LALLEMAND

7 The re-use of obsolete coins: the case of Roman imperial bronzes revived in the late fifth century

95

CECILE MORRISSON

8 Interpreting the alloy of the Merovingian silver coinage

II3

D. M.METCALF

9 Carolingian gold coins from the Ilanz hoard

12 7

ERNESTO BERNAREGGI 10

The novi denarii and forgery in the ninth century

137

JEAN LAFAURIE II

On the rejection of good coin in Carolingian Europe

147

ST ANISLA W SUCHODOLSKI

12 JElfred the Great's abandonment of the concept of periodic recoinage

153

MICHAEL DOLLEY

13 King or Queen? An eleventh-century pfennig of Duisburg

161

PETER BERGHAUS

14 Personal names on Norman coins of the eleventh century: an hypothesis F. DUM AS

v

171

Contents 15 The Gornoslav hoard, the Emperor Frederick I, and the Monastery of

Bachkovo

179

M. F. HENDY

16 Coinages of Barcelona (1209 to 1222): the documentary evidence

193

T. N. BISSON

17 Finds of English medieval coins in Schleswig-Holstein

205

G.HATZ

18 Privy-marking and the trial of the pyx

225

C. E. BLUNT

19 Judicial documents relating to coin forgery

231

PIERRE P. COCKSHA W

20 Mint organisation in the Burgundian Netherlands in the fifteenth century

239

PETER SPUFFORD

21 Coinage in Andrew Halyburton's Ledger

263

MARION M. ARCHIBALD

22 Imitation in later medieval coinage: the influence of Scottish types abroad

303

IAN STEW ART

23 Barter in fifteenth-century Genoa

327

CARLO M. CIPOLLA

Index

VI

Acknowledgements

The initiative for a volume to celebrate the scholarship of Phi lip Grierson came from Christopher Brooke and Ian Stewart. Our share of the work has been as follows. Day-to-day editorial responsibility was undertaken at the Department of Coins and Medals, Fitzwilliam Museum, by Graham Pollard and T. R. Yolk; and at every stage the four editors have discussed the development of the book. Individually they have been responsible for the brief appreciation (Brooke and Stewart) and the bibliography (Pollard) of the honorand. The editors are grateful to the contributors for their ready collaboration and patient support; to the University Press for undertaking the publication of a demanding volume; and to the Fitzwilliam Museum not only for permission to illustrate from the University's cabinet, but also for aid and support to the editors. The assistance of students in the Department of Coins and Medals and of others, Countess Antonini, Miss K. M. Brayshaw, Mr T. W. Gallant, Miss E. R. Mullett, Miss S. K. L. Parker, Fr!. R. Sturm, and Mrs B. F. Whiting, principally in the drafting of English translations, is willingly acknowledged, as is the secretarial help of Mrs S. N. L. Lorimer and Mrs H. C. Scotney, Fitzwilliam Museum, and the skilful advice of Mrs E. L. Wetton and Miss A. E. M. Johnston, Cambridge University Press. The editors' greatest debt is, however, to Phi lip Grierson himself, not only as the inspiration for the book, but for much practical help and advice. With the contributors, they hope that this volume will be accepted as a small token of gratitude and affection to a great scholar and dear friend. C. N. L. BROOKE

J. G. POLLARD

B. H. I. H. STEW AR T

YU

T. R. YOLK

Abbreviations

A NS-MN BNJ BCEN BIHR BSFN DAN DOP EHR JHS JMP JRS LMN MA NC NCirc NZ RBen RBNS RBPH RIN RN SEER SSAM THS ZjN

American Numismatic Society, Museum Notes British Numismatic Journal Bulletin du Cercle d' Etudes Numismatiques Bulletin of the Institute of Historical Research Bulletin de la Societe Franraise de Numismatique Dark Age numismatics (reprints of 29 articles by Phi lip Grierson, 1979. See Bibliography no. 15) Dumbarton Oaks Papers English Historical Review Journal of Hellenic Studies Jaarboek voor Munt- en Penningkunde Journal of Roman Studies Late medieval numismatics (reprints of 22 articles by Philip Grierson, 1979. See Bibliography no. 16) Le Moyen Age Numismatic Chronicle Spink's Numismatic Circular Numismatische Zeitschrift Revue Benedictine Revue Beige de Numismatique et de Sigillographie Revue Beige de Philologie et d' His to ire Rivista Italiana di Numismatica Revue Numismatique Slavonic and East European Review Settimane di Studio del Centro Italiano di Studi sull'Alto Medioevo, Spoleto Transactions of the Royal Historical Society Zeitschrift fur Numismatik

Vlll

Philip Grierson's contribution to numismatics

In 1960 there appeared in the Journal of the Economic and Social History of the Orient an article entitled 'The monetary reforms ofl\bd AI-Malik: their metrological basis and their financial repercussions' (Bibliography, no. 116). Many years before, Henri Pirenne had propounded his celebrated hypothesis about the history of early medieval trade and civilisation, one of whose central themes and pillars was the survival OC gold coinage in the West until the early ninth century. For Pirenne its disappearance was the last act in the decline and fall of Rome in the West, and its cause the depredations of Islam. Through all the smoke raised by the ensuing debate 'The monetary reforms of'Abd AI-Malik' shines like a gleam of pure flame. It shows that the caliph's reforms caused a shift in the relative value of silver and gold in Islam at exactly the right moment at the turn of the seventh and eighth centuries to explain the flight of silver to the West and of gold to the East, and so resolved the puzzle of the rise of silver currency in western Europe in the century which followed. 'Obviously this cannot be the whole explanation of so complex and far-reaching a phenomenon as the establishment of the silver monometallism that endured for five centuries in western Christendom, but it must have been a major factor in it' (no. 116, 264). Whatever place the caliph ultimately comes to hold in the economic history of the West, the article on his reforms will remain a fundamental contribution to the problem; and who but Philip Grierson could have commanded the range of learning, the mastery of history and numismatics, the confidence and the daring to see the economy of West and East as a whole, and study their relations, without taking to the wings of fantasy which have seduced some other workers in the field? Yet his range is much wider than this. The main centre of his work lies in the coinage of Byzantium and the West, and its historical context, from the fourth to the fifteenth centuries. He has gone further back, into Roman coinage and counterfeiting; he has travelled further afield, into anthropology and the basic nature of money, into economic and chemical analysis. He can show the historians how little they understood of coins, on any showing a fundamental historical source; and he can show the numismatists how little they know of the world from which their coins come. ix

x

PHILIP GRIERSON

Phi lip Grierson is a native of southern Ireland who came to this country first as a schoolboy and a student; originally destined for a medical career, he arrived in Cambridge with a taste for reading history already formed, and immediately transferred his allegiance. In recent years his adventures in metrology and his wide reading in science fiction are the fruit of his early interest in science; but his interests steadily shifted, down to the early 1940s, in directions very much of his own choosing. He has often expressed his warmth of feeling for those who taught or guided him - Z. N. Brooke, C. W. PreviteOrton, and the eminent Belgian scholar, F.-L. Ganshof; but their influence has never been fundamental. 'To my own surprise and everybody else's,' he modestly claims, 'I won the Lightfoot scholarship in 1931' (no. 206) and this embarked him on his first career, as an ecclesiastical historian. His early research lay in the history of Flanders, first its ecclesiastical history - hence papers on the abbots and relics of Ghent and Bruges, on Grimbald of St Bertin, and on early libraries, and his important edition of the annals of St Peter's Ghent and of Saint-Amand (no. I); then its social and economic history too, as appears in his paper on the relations of England and Flanders (nos. 38, 171). On a side-wind, a visit to Russia in 1932 led to an interest which blossomed in his bibliography of recent books on Soviet Russia in 1943 (no. 2). The most decisive shift in interest came soon after; for he made his debut as a numismatist in 1945, first as a modest collector, then, very soon, as a student of coins: 'It took another piece of happen stance to turn them into my major field of research ' - a lecture in Belgium in 1947 on 'the relevance of numismatic evidence for determining the chronology of the transition from antiquity to the middle ages' led rapidly to his appointment (1948) to the Chair of Numismatics in Brussels, a part-time post he held until 1981 (no. 206, pp. 41-42). It is a sobering thought for those of us who have worked a single furrow for two or three decades that in 1948 the paper on the Caliph 'Abd AI-Malik, with the immense range of learning it reveals, was only a dozen years away. Through all these shifts there remained some strong threads of continuity: in his relation to his parents, to whom he was always closely devoted; in his service to Gonville and Caius College, Cambridge, to which he came as an undergraduate in 1929, of which he has been a Fellow since 1935, and which has been since then, without a break, his home. He has served it as Director of Studies in History, as Librarian, and as President - and over and above all, as a symbol of continuity in the resident fellowship of the College. It is against this background of stabilitas that we view the extension of his research to Byzantium and Islam, of his friendships to Italy, Switzerland, France, Germany, Poland, America and elsewhere; and his travels to many parts of Europe and North America - and in imagination, perhaps, to the moon. His services to history have been large, to numismatics unique. In Cambridge he has lectured in medieval history from 1938 till his retirement in 1978; he was University Assistant Lecturer in History 1938-45, Lecturer, 1945-59, Reader in Medieval Numismatics, 1959-71, Professor of Numismatics, 1971-8; and beyond the normal call of duty in teaching and examining and serving his Faculty, he has been, inter alia, a Syndic of the University Library for many years, latterly Chairman of the Syndicate, in which role he presided over the affairs of one of the world's greatest libraries, and both Honorary

An appreciation

Xl

Keeper of Coins and a Syndic of the Fitzwilliam Museum. In the Museum his own collection resides, and for its comfort the University has provided the Grierson Study Room, which will form in future years one of the world's principal centres of numismatic research. In other ways he has fostered the interests of the Museum and supported the growth of its specialist library. Outside Cambridge he has been Professor at Brussels 1948-81, Advisor in Byzantine Numismatics at Dumbarton Oaks, Washington, D.C., since 1955, Literary Director of the Royal Historical Society, 1945-55, as well as many more temporary or purely honorific appointments: Ford's Lecturer at Oxford, 1956-7, medallist of the Royal Numismatic Society (1958) and the American Numismatic Society (1963), President of the R.N.S., 1961-6, Honorary Vice-President since 1978, member, corresponding or honorary, of the International Numismatic Commission of the Societe Suisse de Numismatique, the Swedish Numismatic Society, the Medieval Academy of America, the American Numismatic Society, the Societe Royale de Numismatique de BeIge, the Istituto Siciliano di Studi Byzantini e Neoellenici; F.S.A. since 1949, Litt.D. since 1971. Finally, he holds honorary doctorates ofGhent (1958) and Leeds (1978); and he is a Corresponding Fellow of the Belgian Academy and a Fellow of the British Academy (1958). Phi lip Grierson is that rare combination, a great collector who is also a great scholar. After he became Advisor to Dumbarton Oaks, he gave up collecting Byzantine coins on his own account, and a large part of his Byzantine collection went there. This apart, he has spent a high proportion of his time and energy over the last thirty-five years in acquiring European and related coins from the fifth to the early sixteenth centuries, with the exception of the British Isles. The collection is probably the most important of its kind in existence today. In range, quality and balance the Grierson Collection is remarkable, and reflects the personal achievement of its collector; these qualities are unthinkable, indeed, save in a scholar's personal collection. One warmly hopes that in due course it will be published; and a catalogue of the Grierson Collection will provide in itself a manual of medieval numismatics, and a contribution to medieval studies of the first rank. It has been said of him, with pardonable exaggeration, that as befits the Life Fellow of a Cambridge College, he rarely allows a year to pass without subscribing his name to thirty-nine articles. In sober truth the number of his articles on numismatic subjects is legion, and in recent years there has been a remarkable swelling in his books. Yet he still contributes articles from time to time of general historical interest, and throughout his teaching career in Cambridge made notable sallies, especially into the theme of medieval Europe and the wider world, looking out far beyond the Caliphate. He is a voracious reader of everything save conventional literature, and something of a polymath, with a good working knowledge of mathematics, statistics, methods of metallurgical analysis, and a range of languages which have enabled him to master the literature of a subject which has generally been pursued on local or national lines. He is uniquely equipped for the task he has undertaken, a thorough-going reappraisal of medieval coinage. As Professor in Brussels, he has been since 1948 concerned not only to provide a

Xll

PHILIP GRIERSON

general introduction to numismatics, but to study and teach the general implications of coinage for the historian, and the methods of the subject. These fields of enquiry have blossomed in a series of lectures, opening with his inaugural at Brussels, in three numismatic bibliographies, and in two general books, both showing a panorama ofiarge areas of the history of coinage from different viewpoints. In his Presidential Addresses to the R.N.S. (nos. 131, 142, 145, 155, 161) and in his Stenton Lecture at Reading (no. 19) the study of weights and measures was set on a new foundation, and many earlier doctrines weighed in his balance and found wanting. The Presidentials also contain some of his contributions to general numismatic theory, on hoards and finds and coin wear; and he has written elsewhere on the manufacture of coins, and entered joyfully into the arguments on the output of mints. His work on metrology and metallurgical analysis is fundamental. It is not only in technical studies of measures that he has made fundamental contributions to economic history: his papers on the nature of commerce in the Dark Ages and the social function of money in early Anglo-Saxon England have set all students of these themes on new paths (nos. 108, 119); the basic structure of Byzantine political history has benefited from his study of the tombs and obits of the emperors (no. 132); a characteristic contribution to a recent Settimana at Spoleto surveys the symbolism of charters as well as coins nell'alto medioevo (no. 199); he has opened a new world in a study of the effects of fresh supplies of bullion on European coinage and economy in the late fifteenth and sixteenth centuries (no. 177). In recent years there has been a major impetus to the study of Byzantine numismatics for which he has been largely responsible. Three of the five volumes of the Dumbarton Oaks Catalogue, published or projected, are his alone, and he has had a major part in the others. He has written on many corners of Byzantine numismatic history, and a substantial handbook on Byzantine coinage has just been published. His main interest in Western coinage has been in the early Middle Ages. Among his most important publications are a series of articles on the coinage of Charlemagne and the gold solidi of Louis the Pious, on the St Martin's Hoard from Canterbury, the Albertini Tablets, and on gold in China, on Anglo-Saxon shillings, on the mancus; he is at present working on a general book on early medieval coins. His contributions to the later medieval period have been more scattered, yet often of great importance. Thus in two papers, on Pegolotti's book of exchange rates (no. 99) and on coins in the Cely Papers (no. 168), he has clarified vital documentary evidence for the international circulation and value of money. Even the Venetian gold ducat has won prestige from his pen. The full sweep of the subject is brilliantly illuminated in his Monnaies du moyen age (no. 13). Equally characteristic of the man are his history of his College in the Victoria History of Cambridgeshire (no. 107), and his memoirs of Z. N. Brooke and H. T. Deas in the Caian (nos. 41, 180), personal testimony to his more parochial friendships and the warmth of his loyalties. For if one had to list briefly his most striking qualities, one might easily say - simplicity and warmth. There is some paradox in this, for great men are never wholly simple. Grierson's friends and colleagues have often delighted in his minor foibles, in his occasional brusqueness or joyous outbursts of candour, in his occasional admission

An appreciation

Xlll

to harmless vanity. Yet at a deeper level the candour is an expression of exceptional integrity - he never resents criticism, and he will listen with equal attention to young and old, research student and professor - and in personal dealings constantly softened by a mellowing sensitivity and warmth; he is a man of many friends. No doubt the scholarly achievement of the last thirty years involved an immense concentration of effort, and economical use of time; all this has meant a restriction on his social life; yet it is hard for his friends to discern. In earlier days his rooms were one of the main social centres of the College, where undergraduates of all disciplines gathered to listen to his records and to read his books; in more recent years they have often met him on the squash court, and even his colleagues incapable of squash can bear witness to the stream of exhausted young Caians who have shown that his vigour in the squash court remained unabated into his seventies. For many of his friends, two settings are especially associated with him: in Hall and Combination Room at Caius, where he was President from 1966 to 1976, giving freely of his time and gifts in looking after his own and his colleagues' guests; and as speaker and participant in numerous meetings oflearned societies and international conferences. He has greatly enjoyed and benefited from his contact with scholars from many lands. He is a generous host, and enjoys entertaining and being entertained; yet this is characteristically united with a strong puritan streak which makes him spartan and austere when on his own, and censorious of other men's extravagance. The financing of his great collection is a mystery beyond the comprehension of economics; suffice it to say that he is very strict in his standards of academic and commercial probity; but equally addicted to the doctrine that money should work as hard as he does - he never likes modern coins to lie idle, or medieval ones uncommented. He is as much at home in an aeroplane as in his modest suite of rooms in College; and this helps to explain how he has come to buy coins in so many foreign sale rooms, how many friends he has among the curators and the conoscenti of the world, how often his friends who cross his path are delighted to find traces in distant parts of his reputation for generosity and learning. His modesty and enthusiasm have greatly helped him to advance his subject. His approach is concrete, and he is sceptical of large generalisations: 'It is the same way in which a scholar who is also a yachtsman may know that particular voyages ascribed to earlier sailors ... could never have been undertaken, either because they were beyond the sailing capacity of such ships as then existed or would have been rendered impossible by prevailing winds or currents whose existence is not apparent from modern maps. Inspired guesses induced by this rather concrete approach have given me as much pleasure as anything else in my studies ... ' - and he went on to cite his observation that the number of small gold pieces in the Sutton Hoo ship treasure, 37 coins and 3 blanks, equalled exactly the number of the forty oarsmen needed to row the boat and so may have been designed to provide a passage for each to the underworld (no. 206, 47; see nos. 176, 187). Thus he sometimes takes more pride in his minor brilliancies than in his major works; and it is this combination of modesty and enthusiasm which helps to explain how deeply conversations with younger numismatists - especially those who have been in Cambridge in his teaching years - have fructified the work of a younger generation, many of whom are contributors to this volume.

XIV

PHILIP GRIERSON

Thus, indirectly, he has made important contributions to British numismatics and to Britain he has 'dedicated a scatter of his articles and studies; and the Syl/oge of early British and Anglo-Saxon coins in the Fitzwilliam Museum (no. 8), which set the pattern for the British Academy's Syl/oge project, is one of his most remarkable achievements. For the rest, he has left Britain to his colleagues; but among the numismatists of Continental Europe, of the east and west, he is a unique phenomenon. Numismatists in other fields have equalled his technical contributions to one series; but no one has shown equal mastery of the monetary, economic, historical and technical aspects of the subject combined in a single view. Nor is it easy to think of anyone who has had such a wide and perceptive grasp of the whole subject outside his specialist period. His small Numismatics (no. 12) has an admirable, concise survey of non-European coinage; he has a good working knowledge of Greek, Roman, Islamic and modern European and world-wide coinage, His comparative knowledge is without rival, and this extra perspective infuses all his work. It is the combination of collector, numismatist and historian which explains the special character of his achievement. This book reflects the width of his interests, and is drawn from a small selection of his many friends; it concentrates on aspects of numismatic method, which has been one of his most characteristic concerns. In it we try to express, in the currency appropriate to friends and disciples, our homage. Note: This brief study is based on personal knowledge and is a small token of a deep admiration and affection. We are also indebted to notes provided by Philip Grierson himself, to the interview with him printed in The Caian, 1978, 33-55 (no. 206: the quotations on xi, xii, xv, above are from The Caian nos. 34,41-42,47), and to the help of several colleagues and friends, especially 1. G. Pollard and T. R. Yolk. Quotations from his own work are cited by the number in the Bibliography in bold type.

Bibliography

xv

A BIBLIOGRAPHY OF THE WRITINGS OF PHILlP GRIERSON

This bibliography ignores reviews and a few ephemeral publications, and has the materials arranged in three groups, of books, pamphlets, and articles. The order is that of the printed date of publication. The place of publication for books is London unless otherwise indicated. The periodicals which are cited on three or more occasions are shown with abbreviated titles.

BOOKS

I937 I Les annales de Saint-Pierre de Gand et de Saint-Amand (Commission royale d'histoire de Belgique), Brussels I943 2

Books on Soviet Russia, 1917-1942. A bibliography and a guide to reading, London (reprinted, Twickenham, I969)

I95 2 3 F. L. Ganshof, Feudalism, translated by Philip Grierson, Foreword by Sir F. M. Stenton, F.B.A., London (2nd edition, New York I96I) 4 C. W. Previte-Orton, The shorter Cambridge medieval history, edited by Philip Grierson, 2 vols., Cambridge I954 5 Coins and medals. A select bibliography (Historical Association, Helps for Students of History 56) 6 Herbert E. Ives, The Venetian gold ducat and its imitations, edited and annotated by Philip Grierson (Numismatic Notes and Monographs I28), New York (published I955) I95 6 7 Studies in Italian medieval history presented to Miss E. M. Jamison, edited by Philip Grierson and John Ward Perkins (Papers of the British School at Rome XXIV) I95 8 8 Sylloge of coins of the British Isles. Fitzwilliam Museum, Cambridge, Part i. Ancient British and Anglo-Saxon coins I9 66 9 Bibliographie numismatique (Cercle d'Etudes Numismatiques, Travaux rr), Brussels (see also no. 17) I9 68 IO

Catalogue of the Byzantine coins in the Dumbarton Oaks and in the Whittemore Collection, vol. rr, Parts i and ii, Phocas to Theodosius Ill, 602-717, Washington, D.C.

xvi

PHILIP GRIERSON

1973 II

Catalogue 0/ the Byzantine coins in the Dumbarton Oaks and in the Whittemore Collection, vol. Ill, Parts i and ii, Leo III to Nicephorus Ill, 717-1081, Washington, D.C.

1975 12

Numismatics

1976 13 Monnaies du moyen age, Fribourg (also German edition, Munzen des Mittelalters) 14 Monnaies et monnayage. Introduction a la numismatique, edition francaise par Cecile Morrisson, Paris (a translation of Numismatics, 1975) 1979 15 Dark Age numismatics. Reprints of 29 articles with 13 pp. of addenda and corrigenda. The reprinted materials are marked in this bibliography by DAN followed by a roman number for the particular section of the book. 16 Late medieval numismatics (I Ith-16th centuries). Reprints of 22 articles The reprinted materials are indicated in this bibliography as LMN followed by a roman number for the particular section of the book. 17 Bibliographie numismatique, 2nd edition (Cercle d'Etudes Numismatiques, Travaux IX), Brussels PAMPHLETS

195 1 18 Numismatics and history (Historical Association, General Series, G.19) 197 2 19 English linear measures, an essay in origins (The University of Reading, The Stenton Lecture, 1971), Reading 1973 20

Byzantine coinage, Exhibition at Dumbarton Oaks (anonymous pamphlet of 16 pp.), Washington, D.C.

1977 21

22

The origins o/money (The University of London, The Creighton Lecture in History, 1970). Translated into Danish by J. Sten Jensen as 'Pengevae senets oprindelse' in separately paginated supplements to Montsamlernyt, 8 iirgang, 7-10, 1977, with illustrations. The lecture has been reprinted, with a few corrections, in Research in economic anthropology I, ed. G. Dalton, Greenwich, Conn. 1978, 1-35 Les monnaies, Typologie des sources du moyen age occidental XXI, Turnhout

xvii

Bibliography ARTICLES

1934 23 Rostagnus of Aries and the pallium, EHR XLIX, 74-83 24 Hugues de Saint-Bertin: etait-il archichapelain de Charles le Chauve?, MA3

XLIV,

241-251

1935 25 Eudes ler eveque de Beauvais, MA3

XLV,

161-198

193 6 26 A visit of Earl Harold to Flanders in 1056, EHR LI, 90-97 27 The early abbots of St Peter's of Ghent, RBen XLVIII, 129-146 1937 28 The early abbots of St Bavo's of Ghent, RBen XLIX, 29-61 29 The translation of the relics of St Donatian to Bruges, RBen

XLIX,

170-190

1938 30 La maison d'Evrard de Frioul et les origines du comte de Flandre, Revue du Nordxxlv, 241-266 1939

3 I The identity of the unnamed fiscs in the' Brevium exempla ad describendas res ecciesiasticas et fiscales', RBPH XVIII, 437-461 32 L'Origine des Comtes d'Amiens, Valois et Vexin, MA3 X-XLIX (continuous), 81-125 33 The translation of the relics of St Amalberga to St Peter's of Ghent, RBen, LI, 292-315 1940 34 35 36 37

Les livres de l'Abbe Seiwold de Bath, RBen LII, 96-rr6 La bibliotheque de St Vaast d'Arras au xn e siecie, RBen LII, 117-140 Abbot Fulco and the date of the 'Gesta abbatum Fontanellensium', EHR Grimbald of St Bertin, EHR LV, 529-561

LV,

275-284

1941 38 The relations between England and Flanders before the Norman conquest, THS4 XXIII, 71 - I 12 (see no. I71) 39 Election and inheritance in early Germanic kingship, The Cambridge Historical Journal VII, 1-22 194 6 40 Books and pamphlets on Russia, 1942-45, SEER XXIV, 133-147 41 Z. N. Brooke: a memoir, The Caian LI, 95-105 42 An abbreviated version of no. 41, printed in The Cambridge Review January 1947), 196- 1 98

XLVIII,

no. 1659 (18

1947 43 Books and pamphlets on Russia, 1946-47 (sic, in error for 1945-46), SEER xxv (1946-:-7), 508-5 1 7

xviii

PHILIP GRIERSON

44 Bibliography of Professor C. W. Previte-Orton, Litt.D., F.B.A., The Cambridge Historical Journal IX (1947---9), 118-119 45 Errata attribuzione alla zecca di Mileto, Numismatica XIII, 119 1948 46 The present position of medieval studies in England, BIHR XXIII (1946-8), 101-106 47 Books and pamphlets on Russia, 1947, SEER XXVI (1947-8), 512-518 48 Earlier medieval history, 500-1200, Annual Bulletin of Historical Literature XXXII (Publications of the year 1946), 12-16 49 Earlier medieval history, 500- I 200, Annual Bulletin ofHistorical Literature XXXIII (Publications of the year 1947), 11-16 50 Ein unediertes 'Kopfchen' von Arnold 11. von Randerath (1290-1331), Hamburger Beitriige zur Numismatik I-ii (1947-8), 68-69 51 La collezione numismatica del Museo Civico di Pavia, Bollettino della Societa Pavese di Storia Patria N.S. 11,111-114 52 Three unpublished coins of Zeno, NC" VIII, 223-226 1949 53 Earlier medieval history, 500- I 200, Annual Bulletin ofHistorical Literature XXXIV (Publications of the year 1948), 13-19 54 Books and pamphlets on Russia, 1948, SEER XXVII (1948-9), 556-562 1950 55 56 57 58 59 60

La numismatique et l'histoire, Revue de rUniversite de Bruxelles N.S. 11, 23 1-248 Books and pamphlets on Russia, 1949, SEER XXVIII (1949-50), 486-492 Dated solidi of Maurice, Phocas and Heraclius, NC" X, 49-70 The consular coinage of 'Heraclius' and the revolt against Phocas of 608-610, NC" A barbarous North African solidus of the late seventh century, NC6 X, 301-305 A follis of Nicephorus Bryennius (?), NC" X, 305-311

X,

71-93

195 1 61 62 63 64 65 66

Oboli di Muse', EHR LXVI, 75-81 (LMN vii) Books and pamphlets on Russia, 1950, SEER XXIX (1950-1), 550-557 Un denier d'Henri 11 d'Allemagne frappe it Dinant, RBNS XCVII, 117-119 The gold solidus of Louis the Pious and its imitations, JMP XXXVIII, 1-41 (DAN xxii) The Isaurian coins of Heraclius, NC" XI, 56-67 (with C. Brooke) Round halfpennies of Henry I, BNJ XXVI, 286-289

195 2 67 The dating of the Sutton Hoo coins, Antiquity XXVI, 83-86 68 The coronation of Charlemagne and the coinage of Pope Leo Ill, RBPH XXX, 825-833 (DAN xx) 69 Pegged Venetian coin dies; their place in the history of die adjustment, NC" XII, 99-105 1953 70 Report on medieval numismatics from 1930 to 1952, Congres International de Numismatique (Paris, 6-11 July 1953), I: Rapports, Paris, 55-101

Bibliography

XiX

71 The Canterbury (St Martin's) hoard of Frankish and Anglo-Saxon coin-ornaments, BNJ XXVII, 39-51 (DAN vi) 72 La trouvaille monetaire d'Ilanz, Gazette Numismatique Suisse IV, 46--48 (DAN xvi) 73 Visigothic metrology, NC6 XIII, 74-87 (DAN xii) 74 A new Anglo-Saxon solidus, NCS XIII, 88-91 (DAN vii) 75 A new Isaurian coin of Heraclius, NCS XIII, 145-146 76 A Byzantine hoard from North Africa, NC6 XIII, 146--148 77 A stray from the Crondall Hoard, NC6 XIII, 148-149 78 Deux fausses monnaies venitiennes du moyen age, Gazette Numismatique Suisse IV, 86-90 1954 79 Cronologia delle riforme monetarie di Carlo Magno, RIN LVI, 65-79 (DAN xvii) 80 Carolingian Europe and the Arabs: the myth of the mancus, RBPH XXXII, 1059-1074 (DAN iii) 81 Nomisma, tetarteron et dinar: un plaidoyer pour Nicephore Phocas, RBNS c, 75-84 82 The debasement of the bezant in the eleventh century, Byzantinische Zeitschrift XLVII, 379-394 83 Zum Ursprung der karolingischen Goldpriigung in Nordwest-Europa, Hamburger Beitriige zur Numismatik 11.8 (1952-4), 199-206 (DAN xxi) 84 A rare Crusader bezant with the Christ us vincit legend, ANS-MN VI, 169-178 (LMN ix) 85 Le sou d'or d'Uzes, MA4 IX-LX (continuous), 293-309 (DAN xxiv) 86 Problemi monetari dell'alto medioevo, Bollettino della Societa Pavese di Storia PatriaLv, 67-82 (DAN i) 1955 87 The thirty pieces of silver and coins of Rhodes, NCirc LXIII, 422 88 Una ceca bizantina en Espafia, Numario Hispanico IV, 305-314 89 The Kyrenia girdle of Byzantine medallions and solidi, NCS XV, 55-70 195 6 90 The Roman law of counterfeiting, in Essays in Roman coinage presented to Harold Mattingly (ed. R. A. G. Carson, C. H. V. Sutherland), Oxford, 240-261 9 I A note on the stamping of coins and other objects, in A history of technology 11 (ed. C. Singer, E. J. Holmyard, A. R. Hall, and T. I. Williams), Oxford, 485-492 92 The Salernitan coinage of Gisulf 11 (1052-77) and Robert Guiscard (1077-85), in Studies in Italian medieval history presented to Miss E. M. Jamison (ed. P. Grierson and J. Ward Perkins), Papers of the British School at Rome XXIV, 37-59 (LMN ii) 93 The silver coinage of the Lombards, Archivio storico lombard0 8 VI, 130-140 (DAN xiv) 94 I grossi 'senatoriali' di Roma, 1253-1363, I. Dal 1253 al 1282, RIN LVIII 36--69 1957 95 On some fol1es of Heraclius and the location of George of Pisidia's ITYAAI, Seaby's Coin and Medal Bulletin, 98 96 Mint output in the tenth century, EHR2 IX (1956--7), 462-466 97 The debasement of the nomisma in the XI century, Congres International de Numismatique (Paris, 6-1 I July 1953),11: Actes, Paris, 297-298 (a summary of no. 82) 98 La moneta veneziana nell'economia mediterranea del Trecento e Quattrocento, in La civilta veneziana del Quattrocento (Fondazione Cini), Florence, 75-97 (LMN xii)

xx

PHI LIP GRIERSON

99 The coin list of Pegolotti, in Studi in onore di Armando Sapori I, Milan, 483-492 (LMN xi) 100 (with A. H. M. lones and l. A. Crook), The authenticity of the Testamentum S. Remigii, RBPH xxxv, 356-373 101 The dates of the' Livre des Mestiers' and its derivatives, RBPH xxxv, 778-783 (LMN xiv) 102 La monetazione salernitana di Gisulfo 11 (1052-1077) e di Roberto il Guiscardo (1077-1085), Bollettino del Circolo Numismatico Napoletano XLII, 9-44 (translation of no. 92) 103 Halfpennies and third-pennies of King Alfred, BNJ XXVIII (1955-7), 477-493 104 The Eagle Crown: a gold coin of the minority of lames V of Scotland, BNJ XXVIII (1955-7), 65 6- 658 195 8 105 The Roman tombs at Vasa. Appendix 11: The coins, in Report of Department of Antiquities, Cyprus, 1940-1948, 61-67 106 Some modern forgeries of Carolingian coins, in Centennial publication of the Amtrican Numismatic Society (ed. H. Ingholt), New York, 303-315 (DAN xxviii) 1959 107 GonviIle and Caius College, in A history of Cambridgeshire and the Isle of Ely III (Victoria History of the Counties of England), 356-362 108 Commerce in the Dark Ages: a critique of the evidence, THS5 IX, 123-140. This work was summarised, without the notes, in The Pirenne thesis: analysis, criticism and revision (ed. A. F. Havighurst), Lexington, Mass. 1969,90-96 and again slightly revised in the 3rd edition, Lexington, Mass. 1975, 146-159. The article was reprinted in full in Studies in economic anthropology (ed. G. Dalton) (Anthropological Studies VII), Washington, D.e. 1971,74-93, and DAN ii 109 Ercole d'Este and Leonardo da Vinci's equestrian statue of Francesco Sforza, Italian Studies XIV, 40/48 (LMN xvii) IIO The Tablettes Albertini and the value of the solidus in the fifth and sixth centuries AD, JRS XLIX, 73-80 (DAN iv) I I I The' Patrimonium Petri in illis partibus' and the pseudo-imperial coinage in Frankish Gaul, RBNS CV, 95-II I II2 Matasuntha or Mastinas: a reattribution, NC6 XIX 119-130 I 13 Solidi of Phocas and Heraclius: the chronological framework, NC6 XIX, 13 1-154 114 Venray 1957, JMP XLVI, 102 19 60 115 Comments on 'Two unpublished Byzantine coins', The Numismatist, LXXIII, 147-149 II6 The monetary reforms of 'Abd al-Malik: their metrological basis and their financial repercussions, Journal of Economic and Social History of the Orient 111,241-264 (DAN xv) I 17 Una moneta d'argento inedita di Teoderico il Grande, Numismatica N .S. I, I 13- I 15 (DAN v) 19 61 118 Monete bizantine in Italia dal VII all'XI secolo, Moneta e scambi nelralto medioevo (Spoleto, 1-27 April 1960). SSAM VIII, 35-55; discussione, 123-163 119 La fonction sociale de la monnaie en Angleterre aux VIIe-VIIle siecles, SSAM VIII, 341-362 (discussione 363-385) (DAN xi)

Bibliography

xxi

120 Coinage and money in the Byzantine Empire, 498-c. 1090, SSAM VIII, 411-453 121 Sterling, in Anglo-Saxon coins. Studies presented to F. M. Stenton on the occasion of his 80th birthday, 17 May 1960 (ed. R. H. M. Dolley), 266-283 (LMN vi) 122 Notes on the fineness of the Byzantine solidus, Byzantinische Zeitschrift UV, 91-97 123 Coins monetaires et officines a l'epoque du Bas-Empire, Gazette Numismatique Suisse XI (1961-2), 1-7 124 The date of the Dumbarton Oaks Epiphany medallion, DOP XV, 221-224 19 62 125 Tetarteron or counterfeit? A note on Mr Uzman's coin, NCirc LXX, 53 126 Die alterations and imperial beards: a note on the early solidi of Constans 11 and Justinian 11, NCirc LXX, 159-160 127 Kiurike I or Kiurike 11 of Lori-Armenia? A note on attributions, ANS-MN X, 107-112 128 A tremissis of the Suevic King Audeca (584-5), Estudos de Castelo Branco, VI, 7-12 (DAN xiii). Translated into Portuguese by L. Pinto Garcia as 'Urn tremissis do rei suevo Audeca (584-5)', Moeda III-ii 19n 27-33 129 An unrecognised florin of Charles the Bad, Count of Evreux and King of Navarre, RN6 IV, 18 7- 192 130 (with R. J. H. Jenkins), The date of Constantine VII's coronation, Byzantion XXXII, 153- I 58 131 Numismatics and the historian (Presidential Address, Royal Numismatic Society), NC7 11, i-xiv (LMN xviii) 132 The tombs and obits of the Byzantine emperors (537-1042), DOP XVI, 1-60, with an additional note by Cyril Mango and Ihor Sevcenko, 61-63 19 6 3 Mint output in the time of Offa, NCirc LXXI, 114-115 (DAN xxv) Carat-grains and grains in 16th century assaying, NCirc LXXI, 139 Some aspects of the coinage of Offa, NCirc LXXI, 223-225 (DAN xxvi) La cronologia della monetazione consolare di Eraclio, Numismatica, N.S. IV, 99-102. The authenticity of the York 'thrymsas', BNJ XXXI, 8-10 (DAN viii) The miliaresion of Leo Ill, NCirc LXXI, 247 A misattributed miliaresion of Basil 11, Recueil des travaux de rlnstitut tf Etudes Byzantines VIII, Melanges G. Ostrogorsky I, Belgrade, I I 1-116 140 La date des monnaies d'or de Louis le Pieux, MAt XVIII (continuous, LXIX) 67-74 (DANxxiii) 141 A new triens of Reccared (586-601) of the mint of Calapa, NC7 Ill, 81-82 142 Coin wear and the frequency table (Presidential Address, Royal Numismatic Society), NC7 Ill, i-xvi (LMN xix) 133 134 135 136 137 138 139

19 64 143 Les foyers de culture en Angleterre au haut moyen age, in Centri e vie di irradiazione della civilta nell'alto medioevo (Spoleto, 18-23 April 1963), SSAM XI 279-285 144 A coin of the Emperor Phocas with the effigy of Maurice, NC7 IV, 247-250 145 Weight and coinage (Presidential Address, Royal Numismatic Society), NC7 IV, iii-xvii (LMN xx) 146 The origins of the English sovereign and the symbolism of the closed crown, BNJ XXXIII, 118-134

xxii

PHILIP GRIERSON

19 65 147 Trace elements in Byzantine copper coins of the 6th and 7th centuries, in Dona Numismatica Waiter Hiivernick zum 23. Januar 1965 dargebracht, Hamburg, 29-35 148 HENRICVS IMP or ALBRICVS PRINCIPS. A note on the supposed denaro of Pope Leo IX (1049-54) and Henry Ill, Numismatiska Meddelanden xxx, 51-56 (DAN xxix) 149 Le gillat ou carlin de Naples-Provence: le rayonnement de son type monetaire, in Centenaire de la Societe Frans:aise de Numismatique, 1865-1965 (catalogue de l'exposition a I'Hotel de la Monnaie, Paris 1965) Paris, 43-56 (LMN xiii) 150 The Great King, Charlemagne and the Carolingian achievement, in The Dark Ages (ed. D. Talbot Rice), 169-298 151 Coinage and currency (The declining Roman Empire 11), The Listener LXXIV, 656--659 152 A supposed Byzantine coin-die, NCirc LXXIII, 232 153 Money and coinage under Charlemagne, in Karl der Grosse: Lebenswerk und Nachleben (ed. W. Braunfels), I. Personlichkeit und Geschichte (ed. H. Beumann), Diisseldorf, 501 -536 (DAN xviii) 154 Two Byzantine coin hoards of the 7th and 8th centuries at Dumbarton Oaks, DOP XIX, 207-228 155 The interpretation of coin finds, I (Presidential Address, Royal Numismatic Society), Ne' v, i-xiii (LMN xxi) 156 The copper coinage of Leo III (717-41) and Constantine V (720--75); Ne' v, 183-196 19 66 157 From solidus to hyperperon: the names of Byzantine gold coins, NCirc LXXIV, 123-124 158 (with M. Thompson), The monogram of Charlemagne in Greek A NS-MN XII, 125-127 (DAN xix) 159 Byzantine gold bullae, with a catalogue of those at Dumbarton Oaks, DOP XX, 239-253 160 Harald Hardrada and Byzantine coin types in Denmark, Byzantinische Forschungen I, 124- I 38 (LMNv) 161 The interpretation of coin finds, 2 (Presidential Address, Royal Numismatic Society), Ne' VI, i-xv (LMN xxii) 162 Entries in Chambers's Encyclopedia I, Amsterdam, History; Antwerp, History; Artevelde, Jacob van; Artevelde, Philip van; Artois, History. 11, Belgium, History, The Middle Ages; Brabant; Bruges, History; Brussels, History; Burgundy. Ill, Charles the Bold. IV, Dauphine. v, Emden, History; Flanders, History; Franche-Comte. VI, Ghent, History; Groningen, History; Hainault, History. VII, Holland, History. VIII, John (Jean sans Peur); Liege, History; Limburg, History; Luxembourg, History (medieval). IX, Namur, History; Netherlands, History (medieval). X, Philip, the Bold; Philip, the Good. XI, Provence, History. XIV, Utrecht, History 19 67 163 The volume of Anglo-Saxon coinage, Economic History Review 2 XX, 153-160 (DAN xxvii) 164 The gold and silver coinage of Basil 11, ANS-MN XIII, 167-187 165 Byzantine numismatics, in A survey of numismatic research, 1960--1965 11: Medieval and oriental numismatics (ed. K. Skaare, G. C. Miles), Copenhagen (International Numismatic Commission), 52-62

Bibliography

xxiii

166 The monetary reforms of Anastasius and their economic consequences, in The patterns of monetary development in Phoenicia and Palestine in antiquity (ed. A. Kindler) (International Numismatic Convention, Jerusalem, 27-31 December 1963), Jerusalem, 183-301 167 Anomalous pentanummia of Justin I, NCirc LXXV, 234 168 Coinage in the Cely Papers, in Miscellanea mediaevalia in memoriam Jan Frederik Niermeyer, Groningen, 379-407 (LMN xv) 169 Byzantine coins as source material (Main Paper x), in Proceedings of the XIlIth International Congress ofByzantine Studies (Oxford, 5- IQ Septembefl966)(ed. J. M. Hussey, D. Obolensky, S. Runciman), Oxford, 317-333 170 Victor Tourneur (obituary notice), Compte-rendu de la Commission Internationale de Numismatique XIV, 27-28 1968 171 The relations between England and Flanders before the Norman conquest, Essays in medieval history selectedfrom the Transactions of the Royal Historical Society (ed. R. W. Southern), 61-92 (a revised and abbreviated version of no. 38) 172 A follis of Leo III from officina r, NCirc LXXVI, 260 173 Variations in die-output, NCirc LXXVI, 298-299 174 Un gros tournois de Borculo, BCEN v, 106-108 1969- 197 0 175 Entries in Encyclopedia Americana. x, Einhard. XII, Fritigern; Gaiseric. XIV, Heruli; Huns. XVII, Lothair; Lothair I (Frankish Emperor); Lotharingia; Louis 11, King of the Western Franks; Louis Ill, King of the Western Franks; Louis IV, King of France; Louis V, King of France; Louis I, Frankish Emperor; Louis 11, Frankish Emperor; Louis Ill, Frankish Emperor; Louis, King of the East Franks; Louis (younger) King of the East Franks; Louis (the child) King of the East Franks 176 The purpose of the Sutton Hoo coins, Antiquity XLIV, 14-18 (DAN ix) 197 1 177 The monetary pattern of sixteenth century coinage (The Prothero Lecture, 1970), THS· 45-60 (LMN xvi) 178 Nummi scyphati. The story of a misunderstanding, NC7 XI, 253-260

XXI,

197 2 179 Numismatic commentary (pp. 235-236) on T. Padfield, Analysis of Byzantine copper coins by X-ray methods, Methods of chemical and metallurgical investigation of ancient coinage (ed. E. T. Hall and D. M. Metcalf) (R.N.S. Special Publication no. VIII) 180 (with G. T. Griffith), H. T. Deas (obituary notice) The Caian, 54-57 181 La cronologia della monetazione salernitana nel secolo XI, RIN LXXIV, 153-165 (LMN iii) 182 The origins of the grosso and of gold coinage in Italy, Numismaticky Sbornik XII (1971-2), 33-44 (discussion 44-48) (LMN x) 183 Notes on early Tudor coinage, BNJ XLI (published 1974), 80--94. I. King Henry VII's dandyprats. 2. Erasmus's lead tokens. 3. The proclamation of 5 July 1504 and its implications. 4. The origin of the portrait groats. 5. The' gold pence' of the proclamation of 1505

xxiv

PHILIP GRIERSON

1973 184 Byzantine numismatics, in A survey of numismatic research 1966-1971, 11. Medieval and oriental numismatics (ed. 1. Yvon, H. W. Mitchell Brown), New York (International Numismatic Commission), 3-21 1974 185 La lettre R au revers de folles de Justinien 11, BCEN XI, 30--32 186 Muslim coins in thirteenth-century England, in Near Eastern numismatics, iconography, epigraphy and history. Studies in honor of George C. Miles (ed. D. K. Kouymjian), Beirut, 387-391 (LMN viii) 187 The Sutton Hoo coins again, Antiquity XLVIII, 139-140 (DAN x) 188 A new early follis type of Leo III (718), Ne' XIV (published 1975), 75-77 189 (with W. A. Oddy), Le titre du tari sicilien du milieu du XIe siecle it 1278, RN'l XVI (published 1975) 12 3- 134 190 A new Audulfus Frisia triens, JMP LX/LXI 1973-74 (published 1977), 153-156 1975 191 The European heritage, in Ancient cosmologies (ed. Carmen Blacker and Michael Loewe), 225-258 192 The monograms on late sixth-century pentanummia of Antioch, NCirc LXXXIII, 5 193 The date and fineness of Byzantine' neatly-clipped' tracha, NCirc LXXXIII 58 194 La date des 'baudekins' de Marguerite de Constantinople, BCEN XII, 7-8 195 Une trouvaille imaginaire: le tresor de Cuenca, BSFN XXX, 810--812 197 6 196 Nicephorus Bryennius or Nicephorus Basilacius? NCirc LXXXIV, 2-3 197 Heraclius's half-follis, Class 4: an anomalous type, NCirc LXXXIV, 51 198 Numismatics, in Medieval studies: an introduction (ed. J. M. Powell), Syracuse, N.Y., 103-150 (LMNi) 199 Symbolism in early medieval charters and coins, in Simboli e simbologia nell'alto medioevo (Spoleto, 3-9 April 1975), SSAM XXIII 2, 601-630 (discussione, 631-640) 200 La signification de De Clementia et les formules semblables sur les monnaies medievales, BSFN XXXI,2-4 1977 201 La monetazione amalfitana nei secoli XI e XII, in Amalfi nel medioevo, Atti del convegno internazionale (Salerno, 14-16 June 1973), 215-243 (LMN iv) 202 A pattern nomisma of Basil 11 (976-1025), NCirc LXXXV, 97 203 Un denier carolingien de Saint-Bavon de Gand, BCEN XIV, 59-61 1978 204 Un denier de l'empereur Arnoul frappe it Milan en mars 896, BSFN XXXIII, 286-289 205 John Caius' Library, in Biographical history ofGonville and Caius College VII (ed. M. 1. Prichard and J. B. Skemp), Cambridge, 509-525 206 Some memories, The Caian, November 1978, 33-55

Bibliography

xxv

1979 207 Notes sobre les primeres amonedacions reials a Barcelona: els termes 'Bruneti', 'Bossonaya' i el Chronicon Barcinonensi, ID Symposium Numismatico de Barcelona (27-28 February) IT, 278- 28 7 208 Letter on 'The Moving Mint', NCirc LXXXVII 248 209 'Coniazione per dispetto' nell'Italia medievale, Quaderni Ticinesi di Numismatica e Antichita Classiche VIII, 345-358 1980 210 Byzantium and the Christian Levant, 717-1453, Coins. An illustrated survey 650 BC to the present day (ed. M. J. Price), 13
Introduction

The range of his learning, his searching curiosity, his flair for setting numismatic problems in their wider context have given Phi lip Grierson's contributions to numismatic method a particular significance. The contributors to this volume were, therefore, invited to submit papers which, whatever the details of the topic, would illustrate numismatic method and we are most grateful to them for the way in which they have responded. The papers are placed in the broad chronological order of the coinage discussed, beginning with the earliest issues of the Greek world and continuing to the close of the Middle Ages. The approaches which they illustrate include coining technology, the choice of types, the interpretation of find evidence, and the correlation of coins themselves with contemporary documents. The resulting volume does not pretend to be a complete exposition of the methods available to the student, but it is hoped that it will have a special usefulness in demonstrating current directions and techniques in the study of coinage. Perhaps the one general conclusion which the collection allows is the value of combining more than one method for the solution of a problem - a fitting tribute to the scholar to whom this volume is dedicated. The methods and technology of coin production provide one of the principal means for answering such basic questions as when and where a coinage was produced, and in what quantity. For many series the evidence of production is limited to what can be inferred from the specimens that survive. Spufford, however, is able to reconstruct from contemporary account books the organisation of Netherlandish mints during the fifteenth century. While strictly appropriate to the circumstances of late medieval coins and to mints organised along' factory' lines, incidental details, such as the provision of dies, are relevant to coining at other periods. Indeed, in the absence of mint documents, it is the study of dies that often best informs a discussion of chronology, mint attribution, and size of issue. Such were the methods for manufacturing most ancient and medieval coins that the products of particular dies can usually be identified by close comparison of the surviving specimens. From die assemblages of two Hellenistic series, one large, the other oflesser importance, M0rkholm notes how individual dies could under certain xxvi

Introduction

xxvii

circumstances be held available over relatively long periods. Such extended die-life suggests that flexibility may be required in determining the relative date of die-linked coins. Shared dies among coins of two issues in the Carolingian series are noted, too, by Lafaurie and Bernareggi, and the question is put whether such linkages, which occur through one face only, may be evidence for use in each case of a single mint. This would require a revised interpretation of the various town-names, the characteristic type of both series and otherwise taken as mint signatures. Establishing the number of dies employed in a coinage is also a more efficient means of calculating the size of issue than simply counting the coins that chance to survive. So, Berghaus suggests that a controversial imperial issue of Duisburg, though known from only eight specimens, may have been a relatively substantial coinage. This belies its supposedly commemorative function. On the other hand, recognition of die identities among coins comprising a single hoard leads Lallemand to infer the limited currency role of a particular denomination. Die studies are not, however, always appropriate, either from the poor condition of the coins or from the disproportionate labour in the collection and analysis of the material. In these cases, inferences from fabric, the facture of the uncoined blank, the composition of the metal, and the weight of the struck coin, are especially useful. Detailed study of the fabrics and especially of the metrology of the Crete/Cyrenaica bronze series under the Roman Republic and early Empire, in conjunction with other arguments, enables Buttrey not only to demonstrate the denominational system, but to suggest the relative chronology of the constituent issues. The stylistic homogeneity shown by some of the dies in this series, here not supported by shared usage, is gainsaid as evidence for centralised minting by the use of flans of a fabric distinctive to each part of this sometime joint province. Flan size is also shown by Buttrey to be on occasion a more reliable indicator of denomination than the very variable weights registered for particular issues. Variations in the weights of the several Carolingian gold coinages contained in the Ilanz hoard are considered by Bernareggi to indicate the acceptability, in certain circumstances, of a metrological range even among coins of precious metal. Scientific assays of their gold content are, however, lacking. The imprecision of non-destructive metal analysis is brought out by Metcalf in his study of the Merovingian silver coinage. Beyond. any 'economic' inferences, the method is potentially useful as a chronological control and as evidence for the use of distinguishable metal sources. How the' state' might enhance the face-value of a coinage by the use of alloyed metal and of differential weights and the effects of these on circulation are considered by Suchodolski, interpreting the Carolingian documents that also serve as the basis for Lafaurie's study. Finally, a particular class of fabric, the recognition of which is of special importance in studies of coin circulation, comprises re-used earlier issues. For Buttrey, the appearance of cut coins in the Cyrenaica provides a striking parallel with contemporary practice in the western provinces of the Roman Empire following Augustus' reform of the metropolitan coinage, while Morrisson demonstrates the methods and circumstances by which countermarked coins of the early Roman Empire were accommodated to the currency of sixth-century North Africa and Italy. Apart from its immediate physical characteristics, a coin's most easily distinguishable

xxviii

Introduction

features are its types - designs and inscriptions. The first requirement is, therefore, a correct reading. Berghaus bases his close examination of the Duisburg issue on a study of the individual dies. Types, their local associations, their continuity from earlier issues, and their imitation of the contemporary metropolitan coinage, are one of the principal means by which Buttrey is able to impose order on the apparently inchoate material that comprises the Crete/Cyrenaica coinage. In addition to confirming the evidence of circulation and metrology for the denominations and chronology of the series, analysis of the coin types distinguishes the issues intended for each half of the joint province. The typology and occasion of imitation in general is considered by Stewart in a detailed account of the influence of Scottish types on Continental coinage, and the particular example of imitation of English sterling in northern Germany is noted by Hatz. On the other hand the danger of over-refinement in assessing the smallest features of types, such as privy-marks, is emphasised by Blunt in demonstrating from contemporary mint indentures and the surviving coins that in the English medieval series such marks cannot be rigidly related, as previously proposed, to the periodic trials of the Pyx. On occasion, however, varieties and even changes of main type have undoubtedly an administrative purpose, as Carson notes for the Roman imperial series and Lafaurie and Dolley infer for the early Middle Ages. The significance of types, especially those of ancient coins, has long attracted scholarly attention and controversy. The limited perception of the Roman imperial currency by its contemporary users is argued by Crawford. After surveying the notices and appreciations of coins and other material remains given by ancient authors, he concludes that the only important element in a type was to identify the issuing authority. Price re-considers the nature of the authority whose types and inscriptions appear on the earliest Greek coins, suggesting their personal rather than official or state character. For Bernareggi, however, the significance of the legends on the Carolingian gold coins in the Ilanz hoard is straight-forwardly political, though he doubts their role as currency. Finally, the many 'signatures' that occur on a particular Norman series of the second half of the eleventh century are identified by Dumas, not as the names of moneyers, but as those of ducal officers effectively usurping the right of coinage. One of the numismatist's principal tools is the study of coin finds. Various classes of find are therefore either the object of specific studies or a point of departure for other contributions. Price emphasises the need to consider the whole archaeological context in which finds of coins are made. The value of single-finds in describing the area of a coinage's circulation is argued by Buttrey and Morrisson, the former inferring from the narrow circulation of bronze the attribution of the Crete/Cyrenaica series. The value of hoards not only in providing a chronological framework for undated issues, but also in determining the structure and articulation of given series is illustrated by Cars on from the favourable circumstance of a relatively large number of hoards from a single century. In particular, they serve to distinguish substantive coinages from the secondary, often commemorative, issues of the Roman series and to authenticate the regular coins from the irregular. Carson's reflections on possible monetary, rather than violent, reasons for depositing individual hoards provide a useful introduction to Lallemand's study

Introduction

xxix

comparing the composition of hoards concealed in Belgium during the last quarter of the fourth century AD with the occurrence of contemporary coins among Belgian site-finds. Another concentration of finds, mainly hoards, from a closely defined geographical area enables Hatz to identify trade and tribute payments as alternative reasons for the arrival of considerable numbers of English coins in Schleswig-Holstein between the eighth and fourteenth centuries. Hoards also define the problems posed by the Norman series whose legends are studied by Dumas, both by dating them and by providing an indication of the size of issue. Other hoards are remarkable for being limited largely, or exclusively, to single issues, a characteristic of periods in which the currency was subject to demonetisation or withdrawals. Lafaurie adduces the restricted composition of Carolingian hoards to explain the currency regulations fragmentarily preserved in ninth-century documents, while Dolley interprets the disappearance of periodic recoinage in England as the outcome of JElfred's monetary policy. Single hoards are the subject of papers by Bernareggi and Hendy. The Ilanz gold hoard, studied by the former, is one of those exceptional finds which has provided us with a good sample of a coinage otherwise virtually unknown. The apparently random number (786) of coins in Hendy's late Byzantine hoard from Bulgaria proves to be an independently documented sum of money concealed on an identifiable occasion, an exceptional conjunction of circumstances for a coin hoard. Similar precision regarding the date of withdrawal, as well as some of the circumstances surrounding their acquisition, is, however, obtained in respect of Archibald's so-called 'paper' hoards, although these payments, booked by a Scottish agent in the Netherlands at the end of the fifteenth century, are markedly different in composition from the few contemporary gold coin hoards known from the same area. An aid largely denied the ancient numismatist comprises contemporary documents treating of coinage, either directly, as mint-records, or more broadly, as currency ordinances and notarial instruments. The information contained in such documents not otherwise to be inferred from the coins themselves may often supplement the material evidence br be explained by it. In yet other cases, the documents describe the social and economic background against which the function and behaviour of particular series or issues can best be understood. The late medieval French documents studied by Cocks haw illustrate the universal phenomenon of forgery, both for private gain and as unlicensed issues. In each case those responsible for the irregularity were either officers or workmen of properly constituted mints, but the product of their frauds, if it survives, cannot now be distinguished from the regular coinage. The important role of mint officials in the production of' forgeries' in the Carolingian empire is also inferred by Lafaurie, partly from the surviving ordinances, but also from the move to centralised control of minting attested independently by the die-links. Similarly, his re-interpretation of the same documents in the light of hoard evidence reveals a chronological framework for the ninth-century imperial coinage. The various ways in which classes of documentary evidence can be related to the numismatic material are discussed by Bisson in a detailed analysis of the sources for coinage at Barcelona in the early thirteenth century. Not only does the written record supply a coinage now wholly lacking from the numismatic corpus, but it documents public, or market, reaction to changes in the king's coining standard.

xxx

Introduction

The wider context necessary for a proper understanding of coinage is again revealed by Spufford's study of mints in the Netherlands. Beyond the details of coin manufacture, the documents provide indications of the place of mint officers and contractors both in the royal administration and in local society. An understanding of contemporary social structure is crucial, too, to Price's discussion of the very origins of coined money. Here, however, inferences have to be made from anthropological sources rather than documents. Social organisation and documentary evidence are finally combined by Cipolla in his demonstration from late medieval Genoa of how the study of coinage by itself must provide a partial account only of economic transactions.

1 Thoughts on the beginnings of coinage M. J. PRICE

It was probably almost twenty years ago that Phi lip Grierson turned my thoughts to the

problems involved in the first appearance of coined metal. Since then his pen has been responsible for many important papers in various numismatic fields; but for the ancient historian the publication of his Creighton Lecture l deserves particular attention. It brings together many different strands of evidence to concentrate on the origins of money, reminding us that history is in the first instance about people, a fact which is rarely evident in numismatic studies. In that lecture the question of the origin of coinage had to be passed over in favour of the broader, more basic problems of money itself; and I would like to take the opportunity offered by this volume of essays to explore this subject further, in deep admiration of Phi lip Grierson's contributions to many spheres of numismatic study. The broad picture of the development of the first coinage is well attested, although evidence for accurate chronology is sti11lacking. 2 The earliest context in which coins have been found is in the foundations of the archaic temple of Artemis at Ephesus, finally constructed, after three earlier 'phases', during the reign of Croesus of Lydia, 560-546/5 BC. The first structure built on the site consisted of two rectangular bases linked by a cross wall, and it is one of these, filled with votive offerings, which provides the earliest context for coins. The sides of that base apparently burst, and a retaining wall was thrown around the whole complex making a single rectangular platform for the second stage of the building. This was then extended to form the third stage, and the platform of the so-called' Croesus' temple was finally thrown over the whole to make the fourth stage. The only room for disagreement is in the absolute chronology accorded each stage of building, and here scholars have been vague. Hogarth 3 dismissed the evidence of the numismatist B. V. Head, and put the first building shortly before 700 BC. Gjerstadt,4 in answer to those who thought that all the structures were merely foundations of a single building, underlined the correctness of the archaeological evidence for the phases, but on no real evidence gave each stage a 25-year existence. Jacobsthal and Robinson,5 on I UP

2

M. J. PRICE

the more pragmatic grounds of the objects discovered in the original Basis deposit, decided that a date c. 600 BC must be about right for its construction, even though it had to be admitted that there were some probable sixth-century pieces present. 6 Neither stressed the remarkable similarity of objects found inside and outside the first structure; and their date too may have been influenced by the need to provide sufficient time for three buildings to rise and fall before the construction of the platform of the Croesus temple. In the deposits as reconstructed by Hogarth and Head, some of the latest coins were found in the earliest context (pI. I, nos. 1-4,8-10, depict some of the types found in the Artemisium). An electrum coin inscribed with the name Walwel was present in the original basis, from the same reverse punch die as one found outside. 7 This small coin was struck by only half the obverse die, which in its complete state depicted two lion heads facing inwards with the inscription down the centre (pI. I, no. 2). Stylistically it comes at the end of a long series with a single lion's head, uninscribed, which has been attributed to Sardes (pI. I, no. 7).8 Typologically it resembles the coins attributed to Croesus of the mid sixth century (pI. I, nos. 11-12); and, if the latter are rightly attributed,9 it is difficult indeed to see how the Walwel issue could be far removed from the time of Croesus. In the filling of the other base, also of the first building period, was found a coin with stag forepart type (pI. I, no. 4) which shows close affinities to the famous issue ofPhanes (pI. I, nos. 5-6),10 and in particular the manner in which the head and dappled coat of the stag have been engraved suggests the possibility that the coin found in the base is a fraction of that same issue. The exact date of the Phanes coins has not been established; but the reverse punches have been engraved with patterns, and this together with the presence of the inscription would suggest a date well after the striking of the first coins. Again it is difficult to see how such well-developed types could have been engraved 50 or more years before the time of Croesus. A sixth-century date seems to be necessary for this coin also. The commonest coins found throughout the four building periods are pieces with lion's head and the fractions with a type described as a lion's paw (pI. I, no. 8). These are as common in the Basis deposit as elsewhere,l1 and since the limited number of dies from which they were struck demands a rather short period of issue, it would be surprising indeed to find them apparently as fresh and little used as ever during the building of the fourth structure, if that took place 50 or more years after the construction of the first. It should be noted that there is no evidence of any architectural member from above the foundation level to be connected with the three earlier phases. It is possible, of course, that the buildings were constructed of wood; but the fact is that we know nothing of them except that the foundations existed, and that objects of a not dissimilar nature were enclosed with each phase. A further important fact to note is that the Artemisium was built on wet sand, and we know that the earliest builders experienced difficulties in raising the structure, since Pliny 12 tells us that Theodoros ofSamos was brought in - presumably after local builders had failed - to lay the foundations on charcoal and fleeces to prevent them from sinking. It should be suggested that the four building phases were in fact tentative steps towards a final structure. We know that the original basis collapsed, and

Thoughts on the beginnings of coinage

3

~

• 6

5

10

9

8

7

4

3

2

12

II

13

Plate

I

4

M. J. PRICE

in the marshy conditions this could have happened only a short time after its construction. The second phase might be attributed to Theodoros as a trial and the pot hoard discovered in this context, found at the bottom of an earth filling, may be a second foundation deposit complementing that found in the original basisP The third phase would then be an extension of that trial to form the foundations of the temple proper. At this point another of the traditions becomes meaningful. We are told 14 that the original scheme of the architect Chersiphron was enlarged and completed by his son; and this tradition could reflect the extension of the third phase to the final platform that had begun at least during Croesus' reign. 15 If this interpretation were correct, the original basis may have been constructed less than a lifetime before the temple attributed to Croesus was started. There may never have been three separate sanctuaries constructed before the final temple was erected; and a date as late as c. 575 could be given to the earliest Basis deposit. If that were so, the recent attempt to push back the origin of coinage into the early seventh century BC must be rejected,16 and even the 'traditional' date of c. 640, based on a late seventh-century date for the earliest deposits of the Artemisium at Ephesus, may prove to be too early. Or Weidauer's study of the dies of these early electrum coins shows how few were employed for any particular series, and at the same time her work emphasises how difficult it is to place the various series in a sequence. 17 It seems rather that there were several places of minting, each with fairly short series, rather than one or two mints producing a long sequence over a long period of time. Whatever the minting arrangements, if the Walwel coin and the fraction possibly associated with the Phanes stater are to be placed before 600 BC, there is very little coinage extant which may be attributed to the years immediately preceding Croesus. On the other hand, there is very little sign of development in technique and fabric between the earliest electrum pieces (e.g. pI. I, nos. 1-10) and coinage attributed to Croesus (e.g. pI. I, nos. 11-12), less indeed than in any hundred-year period of Greek coinage; and this alone makes a date before 650 BC extremely suspect. Changing types and stylised fabric continue to be a feature of the electrum coinage of Cyzicus, Mytilene, and Phocaea; but such aspects may have developed from the tradition of early electrum, and the distinction between the early electrum and silver coinage must be made. Apart from the Artemisium deposits, the only terminus ante quem is the earliest appearance in literature of a probable coin term,18 when Alcaeus was given 2,000 staters for military expenses early in the sixth century. All in all, a date in the last quarter of the seventh century, rather than in the third quarter or earlier, for the appearance of the earliest coins seems very probable. Silver dumps, flattened but not struck with a type, were found in the deposits from the Artemisium;19 but neither there, nor in the four main hoards of related electrum coins,20 were silver coins present. It appears that in the area of Lydia and Ionia, where the electrum coinage originated, no silver coinage was minted until the time of Croesus. 21 For at least 50 years electrum coins alone were struck, even though, as the Artemisium deposits show, ingots of silver, or indeed of gold, could be used beside the coins. This phenomenon requires some explanation, particularly when viewed against later practice. In the years after Croesus the minting of electrum was considerably restricted, and by the end of the sixth century the use of silver as a coin medium had virtually supplanted

Thoughts on the beginnings of coinage

5

electrum. The earliest electrum coins are found in western Asia Minor, in the general area in which they must have been minted. Their alloy, as far as has been ascertained,22 is extremely variable, although the weights are accurate, and the coins must therefore have been used at a fixed, generally accepted value. This in turn must have been no lower than the highest point of the range of their intrinsic value, with the result that most of the coins must have been valued above their intrinsic value. This overvaluation in itself would account for their remaining within a close radius of their place of origin, since no foreign merchant would accept an overvalued nugget of metal which would not retain its value abroad. In this the early electrum differs from the later sixth-century silver coinage, of which it is true to say that most coins are more commonly found outside the area of origin than in its vicinity. On the other hand, the hoards and the deposits from Ephesus show that within the Lydian and Ionian sphere there must have been some travelling of coin to account for the remarkable variety of types found together, which can hardly have all been the products of one mint. 23 It is normally assumed that the stamp of the punch and the resulting type acted as a guarantee for the coin; but if so, such a guarantee could only hold within the limited area controlled by the issuing authority. Coins of varying intrinsic value would require some more general guarantee of value if they were to be accepted in transactions outside the immediate area of the mint, and this would be more probably effected through some convention accepted amongst traders of the ratio of electrum to gold, than through a stamp impressed on the coin. It is possible that this idea of a guarantee is an assumption based on a preconceived notion of the function of coinage, which in turn is based on later practices. Differences between the early electrum and silver coinages have already been mentioned with regard to types and circulation, and there is also a difference in the denominations minted. Whereas sixth- and early fifth-century silver is normally of stater size, with few small fractions, much of the early electrum has survived in minute and impractical fractions, down to the ninety-sixth of a stater. Some allowance must be made for the chances of survival of small silver and large electrum; but to those handling the coins, this difference is very marked, and, added to the other differences noted, suggests the possibility that the early coins of electrum and of silver were not intended to perform exactly the same functions. It has been suggested from the time of Aristotle onwards that coins were intended to facilitate trade by providing ready weighed and guaranteed pieces of metal. This is often thought of as retail trade, trade in small quantities for daily necessities; but Dr Kraay24 has pointed out that even the smallest electrum piece must have had a fair value in terms of silver, about two Attic obols, and would thus provide a whole day's pay rather than the exchange for a loaf of bread. The idea of a retail trade grows out of the existence of a generally accepted medium practical for small transactions; and conversely payment for service through a single medium is only practical when retail trade in commodities of very small value is established. Such retail trade as is demanded of an urban society could not have taken place through the medium of minute shavings of electrum, silver, or gold. It was as a result of the progress of the economy towards such a system that

6

M. J. PRICE

bronze coinage was instituted during the second half of the fifth century, and that very small denominations in silver began to be commonly minted. If the above observations are correct, it is clear that the theory proposed by R. M. Cook,25 and now widely accepted, that coinage was to provide payments for mercenaries, does not fit the facts as we now know them. Not only is it clear that the localised circulation would require that such mercenaries settled in local communities, but it is also doubtful whether mercenaries would happily have accepted overvalued lumps of electrum as payment. Or Kraay proposed a more complicated view to take account of the facts, that governments were seeking by means of these objects to standardise payments to their own servants, so that in return they could standardise taxes, fines, and similar payments to the state. 26 This appears to obviate the necessity to view the electrum coins as supplying a retail trade; but in the first instance the coins must in the main have been put into circulation as payments. The theory assumes that payment for service was acceptable, and this in turn would suggest at least in part that coins were suitable for retail trade. Fines and taxes would still represent only a small part of the coins' function. The theory also rests on the assumption that the state or monarch was the issuer of the object. Although with later coinage this is clearly so, the nature of the early electrum coinage does not preclude the possibility that private individuals might have been responsible for their issue. 27 Again, the difference with later silver coinage must be emphasised. Electrum coins were issued in numerous small series of many different types, very few of which were later adopted as city types or are obviously religious symbols - whereas the silver coinage forms into recognisable issues of particular city mints. None of the electrum coins from before the time ofCroesus appears to have been inscribed with the name of a city.28 The coins of Walwel found in the excavations at Ephesus were once thought to be the issue of King Alyattes of Sardes; but a new variety is linked by the use of the same reverse punch to the Walwel issues, but is inscribed Kalil,29 a name which has no connection with the royal line of Lydia. Another variety with facing boar heads has other letters which again do not form arecognisable royal name. 30 There are two other inscriptions, one not legible, the other the famous Phanes issue (pI. I, no. 5), which includes the surprising addition EMI 1:HMA which would be entirely superfluous if the type was a royal or state symbol, but which is perfectly acceptable if the type was derived directly from the seal of an individual. Once in circulation some coins were repunched with countermarks, mainly of a linear type, but the function of which may be related to the main type. 31 These countermarks at least must be private marks, and it must be admitted that the evidence could point to the coins themselves being the issues of private individuals not of states. Monarchs as individuals no doubt struck coins, such as the lion's head issues of 'Sardes'; but any interpretation of the function of the earliest coins should take into account the possibility that their issue was not the monopoly of the state. How much less can an individual, albeit as influential person such as a banker, 'guarantee' the circulation of a coin at a value above the intrinsic value of the metal. There can be little doubt that the origin of the coin type in general terms lies in seals, whether they are the marks of cities or of persons. The seal was a mark of ownership,

Thoughts on the beginnings of coinage

7

and in acting as a signature it identified the authority of its owner. In the same way a coin type gave the object an official mark of origin, not necessarily to guarantee its circulation, but like a seal to identify its source. At this stage in the economy there is no reason to presuppose an extension of the function of the seal merely because the types were applied to nuggets of precious metal. It is also reasonable to assume that at this stage in the economy payment in metals for service was not normal. Metals only provide livelihood when exchanged, and, as stated above, daily requirements call for a retail trade in minute quantities of metal. We can gain some idea of the practice of employment from literature;32 and it would seem that employees were normally given board and lodging in return for service, and 'payment' was received at the end of service by way of a bonus, which could presumably, but not necessarily, be given in metals. This is very different from the idea of payment for service, such as the dikastikon and the ekklesiastikon of fifth-century Athens, since the bonus was more in the form of a parting gift. The arrangement did not require that livelihood should depend on payments of money, and such a system worked perfectly well without the existence of coins for over 1,000 years, even though metals were well recognised as a form of amassing wealth or of giving valuations. 33 The servant must have been very much tied to his master, and democracy could hardly flourish without the freedom that a system such as that of coinage brought to society. The introduction of coinage could and did bring about a change in the previous system; but there is no reason to believe that the change had already taken place before the appearance of the first electrum coins. Coins could well have provided a means for standardising bonus payments, or gifts of any sort. In the first instance coins must have been regarded as payments in the broadest sense, since only in this way could they have been put into circulation. But there is this difference; that, as bonus payments, the coins are far more akin to gifts (or medals) than to coins as we know them. On a theoretical level, others such as Edouard Will 34 have stressed the importance of the gift in the early Greek economy, and the personal nature of early electrum coins seems to require a specific function of this sort. The donor could be the state, or a monarch, or indeed a private individual. While it might be realised that the recipient could use the metal to acquire other objects or to make any form of payments, he could equally keep it as indicative of wealth. This, after all, is in the nature of a bonus. The coin type as the seal of origin represents the source of the bonus, the personal authority of the issuer. The coin, accurately weighed according to a recognised weight system, would be perfectly acceptable as a gift or bonus, even if made of electrum and not of gold or silver. On the other hand it is notable that for the first 50 or so years only this alloy was struck into coin, and we may suppose that the attraction of coining electrum lay in the profit that the use of such an alloy brought to the issuer. The place of such objects in the economy would grow as the practice of electrum bonus payments and gifts of coin became more widely adopted and as the coins circulated in other transactions. By the time Croesus of Sardes had brought within his kingdom the cities of the western seaboard such as Ephesus, the economy was ripe for the reform which brought gold and silver coinage into existence for the first time. It is then that the view

8

M. J. PRICE

of coinage as a medium for standardising payments to the state becomes attractive. With the existence of coins it would become nonnal to standardise in tenns of coins; but even so, during the later sixth and early fifth centuries the evidence would suggest that much of the silver coined at the main mints went abroad, either through mercenaries or in the fonn of trade; and in these cases it seems unlikely that the issues were made specifically for use within the state. With the increasing use of coin in payments of all sorts, the fifth and fourth centuries saw many cities overvaluing coin against bullion, so that valuable silver, 'With certain notable exceptions, tended to stay at home. At the same time overvalued copper was introduced to provide a practical medium for a daily retail trade. The problems involved in the introduction of coinage are very far from being solved; but the suggestion that the early electrum coins represent gifts or bonus payments made first in the last quarter of the seventh century BC is an attempt to fonnulate a picture that is in accordance with the facts. The basic questions of data and attribution still require to be studied in detail, and it is necessary before further progress can be made to place the earliest coins in their social and historical context, so that their function can be conceived in tenns of the people who first produced them.

NOTES

P. Grierson, The origins of money, London 1977· 2 For the most recent discussions see L. Breglia, 'Il materiale proveniente dalla base centrale dell' Artemision di Efeseo e le monete di Lidia', Annali delf Istituto Italiano di Numismatica XVIII/XIX 1971/1972 [19741, 9-23; c. M. Kraay, Archaic and classical Greek coins (hereafter ACGC), London 1976, 20--30; T. Hackens, 'Chronique numismatique I. Les monnaies grecques les plus anciennes (VIle-VIe s. av. J.-c.)', L' Antiquite Classique XLVI 1977, 205-218 at 205-213; L. Weidauer, Probleme der fruhen Elektronpriigung (hereafter Probleme), Typos I, Fribourg 1975. 3 D. G. Hogarth, Excavations at Ephesus. The archaic Artemisia (hereafter Ephesus), London 1908, 239-246; compare B. V. Head, 'The coins', in the same volume, 74-93 at 91-92 (present disposition of the coins: Arkeoloji Miizesi, Istanbul). 4 E. Gjerstad, 'Studies in. archaic Greek chronology 11. Ephesus', Annals of Archaeology and Anthropology issued by the Liverpool Institute of Archaeology XXIV 1937, 15-34· I

5 P. Jacobsthal, 'The date of the Ephesian foundation deposit', JHSLXXI 1951,85-95; and E. S. G. Robinson, 'Coins from the Ephesian Artemision reconsidered' (hereafter 'Artemision '), in the same volume, 156-167. Comparing the 2,000 unstratified objects with the 1,000 of the Basis deposit, both agree (Robinson, 'The date of the earliest coins', NCS XVI 1956, 1-8 at 3) that 'many resemble basis types so closely that they can with certainty be dated to the same period'. 6 Emphasised by R. M. Cook, 'Ionia and Greece, 800--600 BC', JHS LXVI 1946,67-98 at 90 n. 190. 7 Head, Ephesus, 83, no. 43 (Basis); 85, no. 72 (unstratified); Weidauer, Probleme, 26-27, nos. 99 and I 12 (Istanbul: Arkeoloji Miizesi). 8 A. R. Bellinger, 'Electrum coins from Gordion', Essays in Greek coinage presented to Stanley Robinson (ed. C. M. Kraay, G. K. Jenkins), Oxford 1968, 10--15 at 13; compare Weidauer, Probleme, 60. 9 A good deal of what follows is based on the assumption that the earliest gold and silver pieces, with heads of lion and bull, were

Thoughts on the beginnings of coinage

IQ

I I

12

13

14 15 16

17

indeed struck under Croesus 560-546/5 BC. The coinage of these types continued to be struck under the Persians to the last quarter of the sixth century, and there are no grounds for placing their introduction much before the middle of the century. Kraay, ACGC, 23; P. R. Franke, R. Schmi tt, 'Phaneos - phanos emi sema', Ch iron IV 1974, 1-4. Head, Ephesus, 82-85. Basis deposits, 14 examples; later phases, 22 examples. Pliny, Natural History xxxvi 14. The burnt layer found in the excavations (Hogarth, Ephesus, 32 and 239) adds a graphic touch to the story. Early sixth-century sherds were found below the burnt layer (Hogarth, Ephesus, 32). Head, Ephesus, 74-75. E. S. G. Robinson, 'The date of the earliest coins', N0 XVI 1956, I -8 at 3, would prefer the coins of the pot hoard to be earlier; but the context does not allow this. Differences in engraving technique, given the novelty of engraving coin dies, could equally be due to geographical as to chronological considerations. Vitruvius vii 159; compare Strabo xiv 1-22. Herodotus i 92. The insistence of Weidauer, Probleme, 72-76, that the temple of Artemis destroyed by the Cimmerians must be on the site of the Croesus temple is without foundation. Compare Robinson, n. 13 above, 7-8. On the contrary, the destruction of that sanctuary in 626 BC could well have resulted in the move to a new site; and if so, the destruction may even provide a terminus post quem for the Basis deposit. Weidauer, Probleme, 43-57. T. R. Volk reminds me that the idea of a progression from unstruck dump to struck coin with striated surface (pI. I, no. I) to struck coin with type is an unnecessary rationalisation of the situation. There is no need to view the latter two steps as separate phases of development. All three clearly circulated together; and coins with striated surfaces and coins with recognisable types may well have been made in different places at the same time.

9

18 Alcaeusfrag. D II; L. Breglia, 'Gli stateri di Alceo', Quaderni Ticinesi [dz1 Numismatica e Antichita Classiche III 1974, 7-12. An inscription from Chios of comparable date mentions staters: A selection of Greek historical inscriptions to the end of the fifth century BC (ed. R. Meiggs, D. Lewis), Oxford 1969, 14, no. 8, B3. 19 Together with the unstruck dumps of electrum: Robinson,' Artemision', 166--167, nos. I and 48. 20 An inventory of Greek coin hoards (ed. M. Thompson, O. M0rkholm, C. M. Kraay), New York 1973, 155, no. II54 (pot hoard, Ephesus); 155-156, no. II58 (Samos); 156, no. II61 (Hellespont); 158, no. 1176 (Gordion). 21 M. Price, N. Waggoner, Archaic Greek coinage. The Asyut hoard, London 1975, 122; C. M. Kraay, 'The Asyut hoard: some comments on chronology', Ne XVII 1977, 189-198 at 197. This view may not be subscribed to by H. A. Cahn, 'Asiut. Kritische Bemerkungen zu einer Schatzfundpublikation', Schweizer Numismatische Rundschau LVI 1977, 279-287 at 281-286. 22 S. Bolin, State and currency in the Roman Empire to 300 AD, Stockholm 1958, 38-40. For an important discussion of early silver and electrum coinage from the metallurgical point of view see J. Dayton, Minerals, metals, glazing, and man, London 1978, 103- 11 5. 23 The thirteen main varieties are listed by Kraay, ACGC, 22. Of these only four may be related to later city coinage of Ionia. 24 Kraay, ACGC, 318 n. 2. 25 R. M. Cook, 'Speculation on the origins of coinage', Historia VII 1958,257-262. 26 C. M. Kraay, 'Hoards, small change and the origin of coinage', JHS LXXXIV 1964, 76--9 1 ; ACGC, 317-324. Philip Grierson makes a similar observation for the early gold Germanic and Anglo-Saxon coinages, but he also underlines the importance of the gift in the economy in 'La fonction sociale de la monnaie en Angleterre aux VIIe-VIlIe siecles', Moneta e scambi nelfalto medioevo (Spoleto, 1-27 April 1960), SSAMvIII 1961, 34 1-362 at 357-359 (reprinted in Dark Age numismatics as article xi).

10

M. J. PRICE

27 Kraay recognises this possibility in ACGC, 23, but would limit it to rulers and tyrants. With the subjugation of much of this area by the Lydian empire, one wonders whether a rich Lydian such as Pythios (Herodotus vii 27), whose wealth at a slightly later date was given as 4,000,000 darics, might not have been on an equal footing, as far as coining was concerned, with the tyrants of the Ionian cities. 28 With the possible exception of the stater of 'Phocaea' with seal type, of which a forty-eighth was present in the Basis deposit: Head, Ephesus, 87, no. 87. L. H. Jeffery, Local scripts of archaic Greece, Oxford Monographs on Classical Archaeology, Oxford 1961, 341, is very doubtful about the letter form. 29 M. Thompson, 'Some noteworthy Greek accessions',ANS-MNxII 1966, 1-18at 1-4; C. M. Kraay, 'Report of the Acting Keeper

30

31 32

33

34

of the Heberden Coin Room', Report of the Visitors [of the] Ashrnolean Museum 1968, 39-50at43-44; Weidauer, Problerne, 46-47· Weidauer, Problerne, 21, no. 56 (London: British Museum). For a detailed study of the Phanes issue, but publishing a coin of which the authenticity has been doubted, see P. R. Franke, R. Schmitt, n. 10 above. E.g. A. R. Bellinger, n. 8 above, I I: examples from the Gordion hoard. E.g. Homer, Odyssey xviii 350--361 and Iliad xxi 450--452. Compare the method of paying mercenaries in the Hellenistic period: G. T. Griffith, Mercenaries in the Hellenistic world, Cambridge 1935,264-316. See the formula evolved to describe a rich man: Homer, Iliad vi 48; x 379; xi 133; Odyssey xxi 10. E. Will, 'Refiexions et hypotheses sur les origines du monnayage', RN" XVII 1955, 5-23 at 18-21.

KEY TO ILLUSTRATIONS

Plate I I Electrum 1stater, type1ess (Fitzwilliam Museum, Cambridge: SNG IV, no. 4835). 2 Electrum 1 stater, two lion heads facing inwards, Walwel, as Head, Ephesus, pI. ii, no. 71 (Arkeoloji Miizesi, Istanbul). 3 Electrum 1 stater, similar (British Museum: BMC Lydia, p. 3, no. 16). 4 Electrum stater, forepart of stag, as Head, Ephesus, pI. ii, no. 74 (Istanbul). 5 Electrum stater, stag, Phanes (British Museum: BMC /onia, p. 47, no. I). 6 Electrum! stater, similar (British Museum). 7 Electrum 1stater, lion's head (Fitzwilliam Museum: SNG IV, no. 4837). 8 Electrum stater, lion's paw, as Head, Ephesus, pI. ii, no. 56 (Istanbul). 9 Electrum 1 stater, forepart of goat, as Head, Ephesus, pI. i, no. 14 (Istanbul). 10 Electrum! stater, two cocks, Ephesus, pI. i, no. 20 (Istanbul). I I Gold stater (croeseid), foreparts of lion and bull (Fitzwilliam Museum: McClean Collection, no. 8635). 12 Silver stater, similar (Fitzwilliam Museum: general collection). 13 Electrum 1 stater, lion's head, showing punch marks on obverse and reverse (Fitzwilliam Museum: SNG IV, no. 4838).

n

n

2 The life of obverse dies in the Hellenistic period OTTO M0RKHOLM

A new and important numismatic method which has been developed since the end of the last century is to try to identify the individual dies from which a series of coins was struck by comparing as many as possible of the surviving pieces. Obverse and reverse dies deteriorate at different rates, and they break and are replaced at different times. It is therefore possible to establish sequences of die links for specimens which share a common die, and the order of the dies within the sequence may be determined by careful study of the progressive wear of a given die as it is exhibited through successive strikings. In this way many series have been classified and the relative chronology of issues established, providing a detailed understanding of how certain mints operated. 1 In applying this method it is usually taken for granted that if two coins can be shown to be the product of the same die, they were struck at about the same time and in the same place. What, however, is meant by 'the same time'? Surely not necessarily 'within one day' or 'within one week'? Most scholars would probably agree that even a month may be too briefa span for many issues. But what then? Several months, a year, several years, even decades? And how are we to explain coins made for the same ruler or for a league of cities which bear different mint signatures on the reverse but were struck from a common obverse die? In this paper, dedicated to a scholar with a special interest in the methqdology of numismatics, I shall try to answer the first of these questions: what is meant by 'the same time'? For this I shall examine the life of some obverse dies from the Hellenistic period. The material consists of a number of dated issues which are already fully published or are shortly to be the subject of detailed studies. Conclusions derived from an examination of these issues may be helpful in estimating the likely working life of obverse dies employed for undated issues, both from Hellenistic times and from other periods. The magisterial study of Margaret Thompson of the New Style coinage of Athens enables us to examine the operation of this important mint during the second century and the early years of the first century BC. 2 The great advantage of this series is that, within each year, the coins were usually dated to the month in which they were produced. 11

12

OTTO M0RKHOLM

New Style tetradrachm (Fitzwilliam Museum, Cambridge)

This was indicated by a letter of the Greek alphabet (A-M = 1-12,with N added for an intercalary thirteenth month) placed on the amphora of the reverse type. The life of these obverse dies can therefore be determined within narrow limits. The individual obverse dies used for the tetradrachm series in general lasted for no more than three or four months, though for periods in which relatively few coins were produced the same dies were used over four or five months. 3 To date about a thousand obverse dies have been identified, of which only 41 can be shown to have lasted for seven or more months. 4 Of the 41 dies, 23 were employed over seven months: four over eight months,6 eight over nine months,7 three over ten months,S two over a twelve-month period 9 and one over a period of fifteen months. lo This last die (no. 418) was used during months 7, 8 and 9 of one year and again in month 9 of the following year. Margaret Thompson's lists indicate that for the first of these two years there was no coinage after month 9 and in the following year coining appears to have begun in month 3. It follows that during the fifteen-month period over which obverse die no. 418 was used it may well have lain idle for five consecutive months. Similarly, obverse die Thompson no. 1231 was used only during months 1 and 12 of the same year; since these are the only two months for which coins are known, it is possible that coining was suspended during the intervening ten months. The case of the second obverse die employed over twelve months (Thompson no. 1169) is different. Here the die is recorded for months 10 and II of the first year, and months 8 and 9 of the following year. The mint, however, was certainly active in the interval, so that the eight-month gap in the history of this die may well be more apparent than real and, in due course, bridged by new material. Obverse die Thompson no. 714 was used continuously in each of nine successive months. All the obverse dies mentioned above are exceptional in having a life of seven or more months. They comprise only 4 % or so of the total number of obverse dies, though it must be borne in mind that new material may add to this figure. In the case of the New Style drachms, the obverse dies lasted much longer. Of 107, fourteen, or about 13%, were used over seven or more months, including one over 32 months and another over at least 25. 11 Production of this denomination was, however, more sporadic and fewer coins were actually struck. Margaret Thompson is therefore

The life of obverse dies

13

surely right in maintaining that 'the life span of a representative obverse die is ... in inverse relationship to the production rate of the issue'.12 In order to compare a mint such as Athens, which was characterised by continuous and generally quite prolific production, with the practice of lesser mints, I shall consider some statistics from three Ptolemaic mints in Cyprus active during the second century BC. The material derives partly from the Paphos hoard 13 and partly from my own unpublished notes on these issues. Coining at the mints of Salam is, Citium and Paphos began during the reign of Pto lemy V (204-180 BC); for a period of about 80 years, down to year 5 of Pto lemy X Alexander I (= 110/109 BC), the date of the latest Cypriot coins in the Paphos hoard, the material is abundant. During this period the three mints together used some 390 obverse dies to strike tetradrachms. The' coins are dated by the regnal years of successive kings, not by months as at Athens. Instances of obverse dies being employed over two years are not infrequent,14 but in most cases they were presumably in use at the end of one year and were simply carried over to the beginning of the next, so that in fact the life of these dies may have spanned only a few months. There are, however, several cases of obverse dies lasting longer. These may be listed as in Table I. Of the 390 obverse dies employed at the three Cypriot mints, 27 (about 7 %) have lives extending over three or more regnal years. Some of these long-lived obverse dies occur in periods when only one pair of dies was in use at the mint at anyone time. Examples are Table I, nos. I, 4, 5, 10, 12, 13, I7, 23 and 24. Die no. 24 is of special interest because it was used over no fewer than five consecutive regnal years; this, so far as I know, is the longest recorded life for an obverse die in continuous use, at least for the Hellenistic period. 1D Dies no. 25-27 seem to have lain unused for several years at a time. This is quite certain in the case of die no. 27, since coins were not struck at Paphos for the years 33-35 of Ptolemy VI. Die no. 22 is a special, and rather awkward, case. The amount of wear to this obverse die revealed on coins dated to year 45 of Ptolemy VIII suggests that the reverse die was in fact used in a later year. On coins dated year 47 the obverse die appears new and fresh, while on coins of year 48 the impression of the obverse shows conspicuous traces of wear. The die is in this same damaged state when it is combined with the reverse dated 45. This reverse die must still have been available at the mint in year 48 when by some unexplained mistake it was put into use again. 20 Although this case is certainly exceptional, it provides a warning against necessarily taking dates on coins at their face value. Similar instances of the use of an obverse die over three or four years are to be found in the dated issues of other Hellenistic mints. For example, the issues of Sidon for Alexander the Great, a dated Ptolemaic series which I have recently attributed to Aradus, and that mint's autonomous series from the second and first centuries BCY These, however, add nothing to the picture derived from the more comprehensive Cypriot material discussed above. Depending on the scale of a mint's production, I would draw the following conclusions. At large and important mints, such as Athens, coin production was continuous and usually involved the simultaneous use of several pairs of dies. Under these circumstances

14

15 Table 0,,",_ die no.

I.

Use of obverse dies over several regnal years

IDterva1iDRlgDal~

Mint

"'OS

from fim

Regnal yean (corresponding datl: BC iD ilaliC$)

U~.

,

Ptolcmy VI

PtolemyVI PtokmyVI Pto1cm.yVm

• " "

173/2

171./1

1,2/1

fJI/JO

150/49

147/6

146/5

145/4

11413

Salamis

'" "

"

"

Cilium

PtolemyV

19[/90

"

" If}ll/89

'7 189/8

Cilium

PtoiemyV

'9 187/6

18()/J

'"

185/4

Citium

PtolmlY VIII

'7 144/3

143/"

"

142/1

44 127/6

U6/J

125/4

"

"

117/ 16

116/'J

112/11

1I1/IQ

Citium

PtolemyVm

Citium

PtolemyVrn

Cilium

Ptolemy IX PtolemyX

Paphos

PtoiemyVI

Paphos

Ptolemy VIII

IIB/17

,

Ptolemy V

13

Salamis

PtolemyV

14

Salamis

Ploiemy VIII

IS

Citium

Ptolemy VIII

16

Citium

PtolemyVIII

,.

54

,

,

,

177/6

17615

"

141./1

141/40

/43/2 Salamis

,.

I10/9 7 175/4

5

,.

'5 191

5

W

JO '7

/fJlJ

189/ 8

'4 IS21I

180/79

"

44 IJ7/6 49

JIB/17

"

116/15

186/5

"

,

184/3

"

183/2

1l5/14

"3/I l

Irl/Il

143/Z

142 / 1

PtolcmyVI

17

Salamis

PtolemyIX PtolemyV

"

Cilium

Ptolemy IX

'9

Paph05

129/8 47 12 4/3

122/1

Pto)emyX

PtoiemyVlII

"

[45/4 Salamis

PtolemyVI

Salami!

Ptolemy VI

Salamis

PtolemyVIII

45

126/5 31 '34/3

'7 144/3

"

, "

'7

1"/4,

157/6

"

147/6 Ptolemy VIII

23

,." '7

I'3phos

Ptolemy VIII

Citium

Ptolemy VI

Salamis

PtolemyVI

PaphOll

PtolcmyVJII Ptolemy VI

,., "

154/3 144/3

'7

"

150/49

SaLamis

"

47 124/3

,.

132/1

"

149/8

'5

JO

148/7

"

["/4

[57/6 lJ 138/7 152 / t

J4

.. "

123/2

IJI/jO J5 '47/6

"

146/5

,.

"

150/49

[32/1

"

151/50

"

150/49

"

146/5

NOICS to Table I: The regnal years carried by Ihe reverse dies are given with the corresponding date BC. Where an ob... en;e die is combined with a reverse in the Dame ofa second king, the year n[the latter's reign is shown below lhecorrespolKliog dat\: BC, c.g. DO. 8: obven:e combined with reverscs of both Pto1emy VIIl and Ptolemy IX, each dated to yeaN corresponding to 117/16 BC. I u. Svoronos u 219. DO. 1]).4a.. pL xlv, DO. 10 (London: British MUSl:WD); h. SvomnOllllo, no. 1336a, pI. xlv, uo. 12 (Hague: Koniuldijk Kabioet van Munten, Penningen en Gesocdcn Stenen): c. Svoronos 120, no. 13370%, pi, x1y, no. 13 (Hague); aU coios wrongly attributed to Ptolemy V.

1 a. SYoronoa 141. no. 1461y (Berlio: Lilbbecke Coli.); h. Paphos 30 obY. die no. 11; SYoronos 241, no. 1463,0 (Berlin: Staatl.k:he Museen); no. 14631i (VieDlla: Kuosthistorisches Museum); 153, no, 1533r (London); c. SYoronos 153. no. IW! (London); the last two coins wrongly IIltributed La Ploh:my VIII. 3 a-.:. Paphos 30, obY. die 00. 13. 4 I'l. Unpublished (llltll.nbul: Arkeoloji Miizcsi); h. Syoronos 111, no. 1362.:1, pI. xlvi, no. 11 (Athens: Etlmikon Nomismatikon MOUill:ion); c. Poenaru Bordell'· 113. no, 6 (Buc:han:~t: InstilUWl de Arheologie). 5 I'l. UnpublUhed (Winlenhur: StadLische Miinzkllbineu); b. SYoronos 111, no. 1365<11, pi. xlvi, no. 16 (paris: Bibliotheque Nationale); ,'. SYoronos 111, no. 1369a, pI. xlvi, no, 10 (Hague), 6 a. Svoronoo 156. no. 157sa. pI. IY, DO. 3 (AIeJ;lIndria: Musee GrU::o-romain); /)....c. Pap/tw 53, obY. die no. 14. 7 a, c. Paplros 56, ODv. die DD. 50; h. Syoronos 159, no. l60sa, pi. ly, 30 (Paris trade). S a-.:. Pophos 5S, ohY. die no. 6S. 9 a-c. PophO$ 5()-60, ob'V. die no. 75. 10 I'l. SYoronos liS, no, 13loa, pi. xliY, no. 9 (Athens); b. Pocnalll Bordea 113, no. 3 (Bucharest); c. SYorooos 113, no. 131111 (Berlin); no. 1313" (Berlin); coins described at 0 and c wrongly attribuled to Ptolemy V. 11 a. SYOrODOS 138, 00.1#1«, pl. xlix, no. 15 (Vienna), wrongly attributed to Ptolemy VI; h'. u.apublished (New York: American Nwnismatic Society). 11 n. SYOl"OOOS 110, no. 1339<1, pI. xly, no. 15 (London); b. SYoronos 110, no. 134QCl. pI. xly, no. 16 (London). 13 n. Syoronos 110, no. 1348a:, pI. xlv, no. 13 (Athens); h. SYoronos 119 no. 1314CI. pI. .\ly, no. 1 (Berlin), wrongly attrihuled to Ptolemy V. 14 a-h. Pnphos 33. obY. die no. 57. IS a-h. Paphor 56, obY. die no. "11. 16 a-h. Papilor 57-53, ob\I. die no. 65. 17 a. SYoronos 110, no. 1343<:1, pi. xlY, no. 19 (Athens); h. SvoronOI110, no. 134sa, pl. xlv, 00. 20 (Copenhagen: Nationalmuscet); c. SYoronos 110. no. 1346a.. pi . .\ly. no. 11 (Paris); d. uopublished (Brussels: Bibliolhb::(ue Royale). 18 o-d. Papilos S8-"I9. obv. die no. 7:. (Pll) 19 a. PophO$ 74, ob'V. die 00. "I; h. SYoronos 24t). no. 1"1011%, pi. lii, no. 1 (Berlin); c. SNG (GB) ml? v pl. lxii, no. 3443 (now London); d. unpublished (Vicuna). (Pll) le) a, c. P6phO$ 19. ob'V. die 00. 14; b. unpublisbed (Berlin). (1'1. 1) 11 a. Uopublished (Paris); h. Paphos 19. obY. die no. 15, allributed to Ptolem.y VI. (PI. 3) 11 a-c. Papilos 33-34, obY. die no. 61. h. also uopublished (New York). (PI. 3) 13 a-c. Papilos 75. obY. die 00. 8; a. also unpublished (New York); c. BI!iO unpublished (Paris), (PI. 3) 24 0-£. Papilos 51. obv. die 00. 10. o. also unpublished (Oxrord: Asbmoleao MUill:IIJD); h. Blso unpublished (New York); c. also unpublished (London); d. BI!iO unpuhlished (Paris). (Pl. 3) lS a. Paphos 19. obY. die no. 13; unpublished (London); b. unpublished (Berlin); c. SYoronos ]53. 154UL, pI. liii. no. 11 (Albens); unpublished (New York). (PI. 4) 16 a-h. Pophos 31-31. obv. die no. 4' (Pl.4) 17 a. Unpublished (VieoDB); h. SNG Copenhagen'· xL, pI. xx, no. 610; c. SNG Copenhagl.'lo XL, pI. xx, no. 611; d. SNG Copenbagl.'lO XL. pI. xx, no. 611 (1'1.4)

16

OTTO M0RKHOLM

the life of an obverse die was inevitably rather short, perhaps three to five months. A working life of more than seven months was exceptional, while a die with nine months' continuous use (Thompson no. 714) constitutes the longest recorded period. On the other hand, smaller mints, such as those established by the Ptolemies on Cyprus, though active at some point in most years, could meet the coining requirements of some years from a single pair of dies. In this class of mint an obverse die might be used for three, four or even, as in one case, five years in a row. The few instances of the use of obverse dies over a period of more than five years (M0rkholm nos. 25-27) are presumably to be explained by there being no coinage struck in some of the intervening years. Clearly the working life of coin dies must stand in inverse relation to a mint's production rate. If mints were inactive over prolonged periods dies may have been available over a much longer span of time than the five years suggested above. I know of no evidence from the Hellenistic period, but there is some possibly relevant evidence from Roman imperial times. From a group of Roman medallions struck for Marcus Aurelius, P. R. Franke has identified a reverse die used in AD 167/8 that had been used six years earlier in 161/2. 22 Further, this reverse had already been used for Hadrian between 132 and 138; this gives a working life of about 30 years. But the minting of medallions was sporadic, and there is no evidence for the use of that particular die during the intervening years. Similarly, H. von Aulock has noted a reverse die from the city of Dalisandus in Lycaonia 23 and several reverses from Codrula in Pisidia24 which were used over periods of between 30 and 80 years. In the case of Dalisandus, where the break is 80 years, von Aulock supposes that the mint simply did not function between AD 166, when this die was used for an issue for Marcus Aurelius, and some time during the reign of Philip I, when it was again used for an issue for Phi lip 11 as Caesar (244-7). The 30to 40-year gaps in the use of dies at Codrula must also result from only occasional minting. A similar case from the early modern period noted by Phi lip Grierson is the combination of an obverse die for a dicken of Basel with dated reverses from both 1499 and 1535. 25 Long-lived dies, however, are most unusual, and their longevity must be explained by their use for coinages that were produced only occasionally and in small quantities. As far as normal coin production is concerned, the examples of two classes of Hellenistic mint discussed in this paper may perhaps provide a basis for estimating the life of obverse dies elsewhere and in other periods. NOTES 1

The first student to suggest obverse die links as a means of relating and then ordering different coin issues was F. lmhoof-Blumer, •Die Miinzen Akarnaniens', NZ x 1878, I - I 80 at 56 and 69. The method was developed and applied systematically in the preparation of the Berlin corpus; see H. Gaebler, Die antiken Munzen von Makedonia und Paionia, Die antiken Miinzen

Nord-Griechenlands(ed. Elmhoof-Blumer) Berlin 1906, vii. It was also used by K. Regling in his monograph Terina, Winckelmannsprogramm XLVI, Berlin 1906, and by E. T. Newell in his studies of the coinage of Alexander, the first results of which were published in Reattribution of certain tetradrachms of Alexander the Great, New York 1912 (reprinted from American

Ill,

The life of obverse dies

2

3 4

5

6 7 8 9 10 II

Journal of Numismatics XLV 191 I, 1-10, 37-45,113-125, 194-200; XLVI 1912,22-24, 37-49, 109-116). The new method was vigorously discussed in Germany: Regling's work was criticised severely by H. von Fritze and H. Gaebler in their review, 'Terina', Nomisma I 1907, 14-22, while Gaebler's own use of the method was thought' by some to be too meticulous and too expensive in terms both of working time and of costs, see H. von Fritze, 'Nochmals das Corpus Nummorum', Nomisma 11 1908, 36-41; and F.Imhoof-Blumer, 'Bericht iiber das griechische Miinzwerk', Congres International de Numismatique, Brussels 1910, 1-3 at 2. Margaret Thompson, The New Style silver coinage of Athens (hereafter Thompson), Numismatic Studies X, New York 196 I. The impact of this important work on questions concerning mint organisation has been regrettably small, presumably because the heated discussion of the chronology has overshadowed all other aspects, see C. M. Kraay, 'Mainland Greece and Asia Minor', A survey of numismatic research 196~1965,' I. Ancient numismatics (ed. O. M0rkholm), Copenhagen 1967,35-53 at 37-38. Thompson, 72 1. I am considering only the tetradrachms from the period when the months are invariably indicated, i.e. from the twentieth issue or 177 /6 BC on Thompson's chronology. From then until the end of the series 997 tetradrachm obverse dies have been recorded. Obverse dies Thompson nos. 335, 379, 453, 468,600,619,621,691,700,717,719,753, 773,805,809,817,818,829,899,1013,1064, 1123, 1165. Obverse dies Thompson nos. 618, 673, 771, 1122. Obverse dies Thompson nos. 500, 679, 702, 714, 84 2, 884, 885, 894. Obverse dies Thompson nos. 721, 802, 126 3. Obverse dies Thompson nos. 1169, 1231. Obverse die Thompson no. 418. Obverse dies Thompson nos. 308, 326, 356, 4 11 , 473, 626, 628, 685, 722, 733, 747, 758,

12 13

14

15

16

17

18

19

20 21

17

800, 1130. Of these no. 41 I was used in months I, 2, 6, 8, and 10 of year 162/1 BC and again in months 3, 4, and 8 of year 160/59 BC; no. 685 is recorded for the first month of the year 145/4 BC and for an illegible month of year 143/2 BC (all dates according to Thompson's chronology). Thompson, 72 I. I. Nicolaou, O. M0rkholm, A Ptolemaic coin hoard (hereafter Paphos), Paphos I, Nicosia 1976 (present disposition of coins: Cyprus Museum, Nicosia). H. Kyrieieis, 'Die Portriitmiinzen Ptolemais V. und seinen Eltern', Jahrbuch des Deutschen Archiiologischen Instituts LXXXVIII 1973, 213-246 at 229 n. 55. J. A. Svoronos, Ta nomismata tou kratous ton Ptolemaion (hereafter Svoronos), Athens 1904-1908. Gh. Poenaru Bordea, 'Monede mai putin cunoscute din timpullui Ptolemaios al V -lea emise in Cipru', Studii Clasice XI 1969, 221-224. E. S. G. Robinson, The Lockett Collection v. Lesbos - Cyrenaica, Sylloge Nummorum Graecorum (GB) Ill, London 1949. A. Kromann, O. M0rkholm, The Royal Collection of coins and medals, Danish National Museum XL. Egypt,' the Ptolemies, Sylloge Nummorum Graecorum (Denmark), Copenhagen 1977 (hereafter SNG Copenhagen). Kyrieleis (n. 14 above), 229, figs. 12-16, who first drew attention to this obverse die and rightly stressed that the evident traces of wear gradually increased during its five years of use. Nicolaou, M0rkholm, Paphos, 42. E. T. Newell, The dated Alexander coinage of Sidon and Ake, Yale Oriental Series. Researches 11, New Haven 1916, 12-13 and 16-18: obverse die Q was used for years 7- I0 and obverse die cc in the years 14, 16, and 18; O. M0rkholm, 'The Ptolemaic coins of an uncertain era', Nordisk Numismatisk Arsskrift 1975-1976, 23-58 at 27-30: obverse die A5 was used in years 49, 50, and 51; obverse die A8 in years 70, 71, and 72; and obverse die AI7 in years 79, 80, and 81; G. F. Hill, Catalogue of the Greek coins of Phoenicia [in the British Museum], London

18

OTTO M0RKHOLM

1910, xxxiii and 31-32, nos. 256-262 (Aradus years 284, 285, and 287). 22 P. R. Franke, 'Zur Verwendungsdauer romischer Medaillonstempel', Chiron v 1975, 40 7-4 10. 23 H. von Aulock, Milnzen und Stiidte Lykao-

niens, Istanbuler Mitteilungen. Beiheft XVI, Tiibingen 1976, 37. 24 H. von Aulock, Milnzen und Stiidte Pisidiens I, Istanbuler Mitteilungen. Beiheft XIX, Tiibingen 1977, 33· 25 P. Grierson, Numismatics, London 1975, 110.

19

The life of obverse dies

18a

18b

19a

18c

19c

19b

20a

18d

20b Plate

19d

20c 2

OTTO M0RKHOLM

20

24b

23c

23a

21b

21a

24a

22c

22b

22a

24c

Plate 3

24d

24e

The life of obverse dies

25a

27a

25b

25c

26a

26b

27b

27c Plate 4

21

27d

3 The Roman coinage of the Cyrenaica, first century BC to first century AD T.

v.

BUTTREY

When the Romans assumed control of the Cyrenaica in the first century BC they found a Ptolemaic monetary system in use. Gold and silver coin had to be imported, presumably from Egypt; the bronze small change was locally struck, bearing Ptolemaic types but of denominations which perhaps had nothing to do with the Egyptian system (figs. 1-3).1

2

3

Ptolemaic bronze issues

How the Romans first coped with this system is unknown, but they soon began to piece out the existing currency with their own local bronze, produced sporadically until the reign of Tiberius. These were issued in limited quantities, and were often poorly struck, from dies of execrable quality. The issues are not as yet all certainly described, their denominations are conjectural, and their chronology is a tangle, for the authorising magistrates are unidentifiable. A further difficulty lies in our uncertainty over the fluctuating relationship of the Cyrenaica to Crete, with which it was sometimes bound in a single Roman province, for several magistrates struck for both areas. The study which follows consists first of a series of Notes, observations on several of the issues from various points of view - fabric, weight and denomination; type and style; mint and find spot. While these individual Notes are often on a point of detail, taken together they then help to define the nature of small-change circulation in the Cyrenaica during this period, for they make possible a re-examination of relative and absolute chronology, the relation of earlier types to later, and the relation of types to denomination.

23

T. V. BUTTREY

24

This exercise in methodology is dedicated in affection to the colleague who would appreciate it best, and whom another has already best described as notre maitre. NOTES ON THE COINAGE OF ROMAN CYRENAICA I.

The types of Lollius

L. Lollius struck six varieties of bronze, which have been recognised as representing three denominations in each of two parallel series, the one with Greek legend, the other with Latin (Robinson, BMCCyr., nos. 3-23) (pI. 5, nos. 25-30). That they circulated in two different areas is suggested by the fact that the Greek series tends to be found in the Cyrenaica, the Latin in Crete. Differential circulation might suggest separate production, which is confirmed by the differences in manufacture: the Greek legend series is struck on bevelled flans with die position usually i L the Latin on rounded flans with die position usually i i (Robinson, BMCCyr., ccxv). It is best then to follow Robinson in attributing the Greek issue to a mint in the Cyrenaica, presumably at Cyrene itself, and the Latin issue to a mint in Crete. The obverse style of the six varieties, however, is coherent, the use of die control letters is similar in both issues, and the denominations are para1\el. This bespeaks a single conception animating the whole issue. The dies must have been cut in a single workshop, presumably that of Crete given its relatively larger importance than the Cyrenaica. The Greek dies would then have been shipped to Cyrene for striking while the Latin dies were retained in Crete. This operation bears on the identification of Lollius' types, for they too can be seen as conceived in parallel series. The most obvious case is that of the Half-unit,2 where the reverse of the Greek series is the dromedary 3 appropriate to the Cyrenaica, while that of the Latin series is the stag appropriate to Crete. This kind of balance can be seen throughout the issue. Unit. The Greek bears the head ofZeus Ammon, sceptre before (Robinson, BMCCyr., no. 19). The Latin bears a 'youthful male head', sceptre before (Robinson, BMCCyr., no. 3). The reverse of both series is the sella curu/is. The types are plainly para1\el. Identification of the Latin obverse has been doubtful: Mii1\er, Afrique 1,158, notes several unlikely possibilities, e.g. a portrait of Ptole my Apion; while Robinson, BMCCyr., ccxv, takes it to be a portrait of Antony in the guise either of Dionysus with thyrsus, or of the beardless Zeus Velchanos with sceptre. The identification with Antony is fanciful- the head bears no resemblance to him at a1\ - but a Cretan Zeus to balance the Libyan Zeus is exactly right, and the sceptre is appropriate to him, as against other gods, e.g. Apo1\o, who are not norma1\y depicted with this symbol of power (see H. G. Lidde1\, R. Scott, H. S. Jones, A Greek-English Lexicon, Oxford 1940, s. V. (T'Kij1T7POV 11.2). But which Zeus? Zeus Velchanos is not a likely candidate; his sole numismatic appearance, on coins of Phaistos, shows him without sceptre and holding a cock. 4 Nor does he appear to have been widely worshipped in Crete, or to have represented Crete in any general manner. This last point is important since the bust of the Latin Unit ought to convey the notion of Crete in general, to balance the head of Ammon for the Cyrenaica. For that, another

Roman coinage of the Cyrenaica

25

beardless Zeus lies to hand, Zeus Diktaios, the young Cretan Zeus par excellence, whose birthplace was the island. The hymn from Palaikastro addresses him as kouros and associates him with Dikte; and we know that at Dikte stood the statue of a beardless Zeus. 5 The Unit of Lollius presents the two major manifestations of Zeus for the two parts of the province: for Crete, Zeus Diktaios, for the Cyrenaica, Zeus Ammon. Half-unit. The Greek coin bears a head of Apollo with bow and quiver. The head cannot be that of Libya, in spite of a superficial similarity of hair style;6 for she was not a hunting goddess, and does not have the bow and quiver as attributes. The Latin coin bears a head of Artemis with bow and quiver. The animals of the reverses refer to the two parts of the province. The dromedary for the Cyrenaica is obvious; the stag, though an evident symbol of Artemis, here refers to Crete. 7 The horizontal pairing of Apollo and Artemis is evident, but each also functions independently in the appropriate series. Apollo had instigated the earliest Greek colonisation of the Cyrenaica, and his temple was one of the most important at Cyrene. In Crete, Artemis was a national goddess in her manifestation as Diktynna. Quarter-unit. The Greek coin bears a head of Libya with reverse symbols of fertility and commercial success - caduceus with poppy and grain ear - while the Latin piece bears the club, symbol of Heracles, and wreath, symbol of victory. For the local significance of the latter, note that the Cretans believed their Heracles Idaios to have founded the Olympic games. 8 The wreath, though everywhere a common coin type, might in this context bear that extra significance. The obverse types of the two series not only refer to the two parts of the province, but cohere to form a single mythological statement. On the Unit of each series Zeus is portrayed in an appropriate local manifestation; on the Half-units are two of his Olympian offspring; on the Quarter-units are two further children of Zeus of some local eminence but less important mythologically. The identification of Zeus Diktaios as the head of the Latin Unit ties together the typology of the entire issue.

2.

The mints of Crassus

The coins of Crassus, like those of Lollius, were struck in both Greek and Latin series, but not in paired denominations as far as is now known (pI. 6, nos. 31-35). Nor were the dies cut centrally, as is shown by the differing styles of the Latin Apollo and the Greek Libya and crude Tyche. In manufacture the two series of Units are so distinct that their weight ranges barely overlap. Robinson (BMCCyr., ccxxi) properly assigns the Latin series to Crete, the Greek to the Cyrenaica, where however Crassus used two mints, Ptolemais and Cyrene, the only official of Roman times who appears to have done so. The Tyche Unit bears the inscription IITOAEMAI. Robinson knew no inscription for the rare Libya Unit (BMCCyr., ccvii, no. 25 his b = Svoronos, Ptoiemaion, no. 1902) but an example preserved in the office of the Director of Antiquities, Cyrene, can be read: ~l}l X, to 1. and r. of head. The Libya Quarter-unit also reads KYPA. Crassus' mints, therefore, struck distinctive types, and the Libya head appears to have been thought of as appropriate to the mint of Cyrene.

26

T . V . BUTTREY

3. The types of A. Pup ius Rufus

It is worth noting that the three denominations of Pupius (pI. 6, nos. 36-38) bear obverse types appropriate historically to Cyrenaican coinage: Ammon head, ram, Libya head (figs. 5, 7, 9)· The first had appeared from early times on archaic and classical silver (fig. 4), later on pre-Ptolemaic and Ptolemaic aes, most recently on the Greek Unit of Lollius (Robinson, BMCCyr., nos. 19-20). The ram had appeared rarely on aes of independent Cyrenaica (fig. 6), and most recently on a Ptolemaic issue (Robinson, BMCCyr ., cxlvi, no. ro8a) in a module just slightly smaller than that of Pupius. The Libya head was frequent on Ptolemaic aes (fig. 8). Both the Ammon head and the ram had been cut on obverse dies, mated with reverses which bore the authorising legend. Pupius simply substituted his own reverse types of sella, subsellium and serpent.

Pre-Roman

4

6

8

A. Pupius Rufus

5

7

9

Zeus Ammon

ram

Libya

4. Cyrenaican bronze of Antony and Cleopatra After the issues noted above, that of Antony and Cleopatra astonishes by its complete lack of types: the coins bear nothing but the legends ANHljYTIAjr j jBALIAj0EAjNE (Svoronos, Ptolemaion, nos. 1899-1900). The consular da te fixes the issue to 31 BC. It was struck in two denominations, the Unit and Half-unit, of which the latter is very rare (pI. 7, nos. 39-40). The major problem with this coinage is mint attribution. Muller originally omitted it, but later included it in his Afrique Supplement as Cyrenaican, 30, no. 428a. Svoronos accepted the attribution to Cyrene,9 but Regling, reviewing Svoronos, argued that it

Roman coinage of the Cyrenaica needed to be supported by find evidence. 1o As a result of these doubts the issue is not even mentioned by Robinson in BMCCyr.; Grant gave it to Patras in the Peloponnese;l1 and I have denied that it had anything to do with the Cyrenaica. 12 But a conclusive find is now to hand. In the current excavations by the University of Pennsylvania under Prof. Donald White in the Demeter Sanctuary at Cyrene, a well-worn example of the Unit has appeared. It is also halved. The issue is not likely to have travelled much, and a halved piece is most likely to have been cut locally. These coins must be restored to Cyrene; and it then follows (as Miiller, Afrique, loco cit., long ago noted) that the curious Augustan issues of Capito and Palikanus without types of any kind are easily explained as imitative of the earlier issue of Antony and Cleopatra. The literary sources note that Antony's donation to Cleopatra included Cyrene (' Libya '), of which she, according to Plutarch, Antony 54.5, or their daughter Cleopatra, according to Dio xl 41.3, was made the queen. The coin confirms Plutarch's version. If Crete too now lay in Cleopatra's domain (Dio xl 32.5), no comparable coinage was struck for the island. 5. The denominations and types of Seato Of the four varieties signed by Seato (pI. 7, nos. 41-44), two bear the head and name of Augustus (CAESAR TR POT). The obverse of the smaller is an imitation of Augustus' reformed as, with its head of the emperor (Mattingly, BMCRE I, 29, no. 137) (figs. lo-II).13 The obverse of the larger does not imitate the imperial dupondius (Mattingly, BMCRE I, 29, no. 135) whose obverse wreath appears as Seato's reverse (figs. 12-13), but bears two heads perhaps in indication of the coin's value: if one head indicates the as, two heads should indicate the dupondius. 14 Seato's other two varieties are smaller and bear no reference to Augustus, but rather the types of the same modules as struck by Pupius, namely ram and Libya/serpent. Since none of Pupius' coins bears any reference to Augustus or his reformed coinage, his issues must be earlier than Seato's, and Seato's coins, imitations. This division in Seato's typology - with Augustus and without - has for some scholars defined two separate periods of issue, indeed two different occasions on which Seato held

10

11

12

Issues of Seato with Roman prototypes

13

28

T. V. BUTTREY

office in the province: 'It is conceivable that Scato had two periods of governorship' (Robinson, BMCCyr., ccxxiv). This notion depends in good part on the assumption that the four varieties represent only three denominations: (I) dupondius and as with Augustus, (2) as and semis without Augustus. Two types of the as would mean two separate issues. But the second' somewhat lighter' as (Robinson, BMCCyr., ccxxiii) actually weighs just half of the first in the average of examples known to me. It makes more sense to see this variety as the semis, and Scato's lightest piece as the quadrans. There are then four different denominations in Scato's coinage, and there is no need to posit different issues on that basis. Nor need there be any division on the basis of the types, which are easily explained: all are imitations. Both the dupondius and the as repeat the obverses of the reformed issues of Rome (figs. 12-13, Io-II). That the smaller pieces do not mention Augustus is owing simply to there having been no Augustan prototype. The semis was not struck at the Rome mint under Augustus; the quadrans not so early. Scato therefore copied the types of his predecessor for his semis and quadrans, and struck them on flans of the appropriate fractional weight of the as. His four varieties can be regarded as a coherent set of denominations, all of whose types were deliberately derivative, struck on one occasion. 15 6. The denominations and types of Capito and Palikanus

The relative order of the two coinages is unknown;16 since the types are identical, one borrowed from the other (pI. 7, nos. 45-48). Each official struck two varieties, A with types and legends: IMP A VG TR POT in wreath/ sella (name); and B with legends only: IMP A VG TR POT /(name). Variety A reproduces on the obverse the obverse of Augustus' reformed dupondii (figs. 14-15). The reverse of the Roman issue bore the name of the moneyer, here replaced by the local official's name and title, with the received Cyrenaican type of sella. That the dupondius type is not repeated on variety B indicates that it is of another denomination, certainly then the asY It cannot be the semis, for it weighs on average about half again as much as Scato's semis. Here averages are important to take, for the individual specimens are erratic in weight, so much so that Robinson (BMCCyr., ccxxv) concluded that variety A was struck as both dupondius and as (so that it would have been impossible to distinguish between the two denominations by type). Five specimens of Palikanus' variety A, Robinson, BMCCyr., nos. 44-48, weigh 18.14, 15.23,9.16,8.27 and 7.94 g. respectively; while two of Capi to (British Museum, London, 1946 acquisition, and Bibliotheque Nationale, Paris) weigh 12.39 and 9.00 g. These terrible variations in weight are not owing to the state of wear. Most of these pieces show average use, but the example weighing 12.39 g. is in splendid condition. Rather, this is an aspect of poor manufacturing techniques. Of the first three pieces of Palikanus in Robinson, BMCCyr., all are struck from the same obverse die, yet the first weighs almost double the third. The reason is poor control of the thickness of the flans, which vary considerably and influence the weight accordingly. (The phenomenon is not limited to these issues. Sixteen Greek Units of Lollius average 20.46 g. and 32.3 mm diameter;18 yet a thin flan piece of 32 mm in the American Numismatic

29

Roman coinage of the Cyrenaica

14

15

Issue of Capito with Roman prototype

Society, New York, weighs only 15.13 g., and a still thinner one of 33 mm in the Fitzwilliam Museum, Cambridge, weighs only 12.50 g.) This is fair warning that weight can be misleading in determining denomination, and that type and flan size should be given first consideration. We ought therefore to reject Robinson's argument (BMCCyr., ccxxv) that variety A occurred on both the dupondius and the as, the latter subsequently losing its types (= variety B) so as to be distinguishable (so BMCCyr., ccxxvii n.). Rather, variety A pertains exclusively to the dupondii, which were however manufactured with great carelessness, while variety B indicates the as.

7. The denominations of Tiberius Robinson's observation (BMCCyr., ccxxvi) that three denominations are here represented must be correct (pI. 8, nos. 49-5 I). It is true that the types of the first two are almost identical, obverse Drususjreverse Tiberius and Germanicus. But they differ in the detail that on the larger denomination the head of Drusus occurs with lituus and simpulum, which are lacking on the smaller. There is no overlap of weight or of flan size between the larger and the smaller denominations in the examples available to me (BM, ANS, Cyrene excavations). The average weight (g.) of the three denominations accords well with those calculated by Mattingly in BMCRE I, liv-lvi, for Tiberian aes from Rome and Gaul:

Dupondius

Cyrene Rome or Gaul

As

Semis

3.9 2 3. 61

I take no account of the possible differences in alloys between mints. Those of Cyrene are as yet unascertained.

30

T. V. BUTTREY

8. Halved coins at Cyrene The examples known to me are: Roman Republican. One halved as was found in the Apollonia excavations, too worn for more precise identification.19 It is exactly equivalent to the material which appears in Western Europe. Soter / Libya. One halved piece is in a private collection, from Cyrene (cf. Robinson, BMCCyr., 83, no. 48, '221-140 BC '). The halving is most unusual, but the module of the piece at 27 mm is equivalent to that of the smaller Republican asses and those of Pupius Rufus. Presumably the older Ptolemaic coin was simply swept up with the Roman for cutting. 20 A. Pupius Rufus. Three halved Units are in the Cyrene Museum, the ANS (ex-Norton)21 and a private collection, all from Cyrene. The first and third have been cut horizontally across the reverse, the arms of the sella serving as guidelines. Antony and Cleopatra. One piece is fr~m the current excavations in the Demeter Sanctuary at Cyrene, a halved Unit, mentioned above. Although the piece is typeless the cut is in the manner of the Republican halvings, on the vertical axis of the obverse. Seato. The Apollonia excavations produced one halved as of Scato with head of Augustus,22 and a second has been found at Cyrene, now in a private collection. The same collection includes yet another halved coin from Cyrene, of identical module and fabric with the other two, probably of Scato but too worn to be identified.

THE CHRONOLOGY OF THE ROMAN COINAGE OF THE CYRENAICA

The relative chronology proposed by Robinson is in general unassailable on present evidence, but his absolute chronology must be questioned. None of the coins aside from those of Antony bears a date, and the officials who signed most of the other issues are completely unknown. Scholarly argument has been based largely on speculation concerning the identity and function of the officials, as well as the historical and constitutional implications in the connection between Crete and the Cyrenaica, or want of it, indicated by a given issue. For a survey of recent opinion see Perl, who reviews the problem fully without moving it any closer to solution. 23 Robinson's arrangement of the Cyrenaican material, enlarged by the addition of the Antonian issues, is as follows :24 The Lollius issue. These are the first Roman coins of the province which can be related to the Roman monetary system (see below). They are struck from dies of better style than any which follow, and the Unit is heavier. Greek and Latin series were struck for, and circulated in, the Cyrenaica and Crete. Robinson (BMCCyr., ccxiii-ccxv) dates the issue sufficiently late to coincide with Antony's Fleet Praefect bronze, and attempts to discover a portrait of Antony on the Latin as. Both suggestions are fanciful. Lollius' date remains unknown, but if it is considerably earlier than Antony's, the issue of Crassus and Pupius which follow can also be pre-Antonian. The Crass us issue. These are oflighter weight and cruder style. Greek and Latin series

Roman coinage of the Cyrenaica were struck for the Cyrenaica and Crete. This issue must follow Lollius', given the weight standard. (Grant, FITA, 55-58, reverses the two without explanation, and without noticing the weight difference.) Crassus' issue has been assigned to c. 37-34 BC on the grounds that the crocodile of the Latin as refers to Egypt; therefore the coinage was struck under Egyptian authority; therefore only while Crete and the Cyrenaica were united under Egyptian control could Crassus have struck, that is after Antony's donations to Cleopatra (Robinson, BMCCyr., ccxxi, ccxxii; Grant, FITA, 55-56). Now the whole chronology and extent of the donations is still unclear. 25 The crocodile alone (fig. 18),

16

17

18

crocodile

here said to symbolise Egyptian authority, does not derive from Egyptian coins. It elsewhere appears on Roman coins only to signify the conquest of Egypt, namely on Octavian's aurei and denarii with the legend AEGYPTO CAPTA (Mattingly, BMCRE 1, nos. 650-655), and enchained on the aes of Nemausus (figs. 16-17). Further, it is not accompanied on Crassus' coin by any authorising legend beyond his own name; why not BAnAILLAL [KAEOTIATPAL] as on the Cyrene bronze of Antony and Cleopatra? Whatever the significance of the crocodile type, Crassus' coin omits to make precisely those claims which are made for it in the modern scholarship. On balance, since Crassus claims no higher authority, I am inclined to place his coins earlier than the donations. The A. Pupius Rufus issue. These are of lighter weight than Crassus', and were struck in a Greek series only, therefore in the Cyrenaica. Robinson, Grant and Perl all place him after Actium. In the relative chronology I include him in the pre-Actian period by inserting Antony's bronze after his issue. The evidence is ambiguous. (i) Weight and module. From Crassus through the Augustan issues the weight and module of the as steadily declined. Pupius' as best fits this scheme if it precedes rather than follows Antony's, which is closer to Seato's. On the other hand, the semis and quadrans of Pup ius seem to be very close to Seato's. (ii) Type. Capito and Palikanus ignored the as type of Seato and preferred to imitate that of Antony. If Pupius was post-Antonian, their later imitation of Antony becomes even more remote. (iii) Signature. All other post-Antonian dupondii and asses bear the name, and in some cases the image, of Augustus or (under Tiberius) members of the imperial family. Pupius' usage would be strangely independent

32

T. V. BUTTREY

and anachronistic. On balance, therefore, I prefer to place Pupius' issue pnor to Antony's. Antony and Cleopatra, dated to 3 I BC by the reference to Antony's third consulship. I place the issue after those signed only with the magistrates' names, and before those which refer to Augustus. Denarii of Scarpus, struck first for Antony (fig. 19), then with quinarii for Octavian, 3 I BC (figs. 20, 22). These issues are not included in Robinson, BMCCyr., though struck

19

20

21

22

Issues of Scarpus with prototype of Octavian

in Cyrene, because they are Roman military issues. They are conveniently collected in Grueber, BMCRR n, 583-58626 = Crawford, RRC, no. 546. 27 Note that the imitation of local types which is a feature of Roman Cyrenaican coinage is found here too, in Scarpus' use of the Ammon head as obverse type. Scarpus' coinage for Antony is limited to 31 BC by the consular date and Scarp us' change of allegiance in September; his coinage for Octavian presumably followed at once. Note the close stylistic connection in the Zeus Ammon heads of the two series (Grueber, BMCRR rn, pI. cxxii, 7-ro), the work of the same engraver. So Crawford, RRC, no. 546. In Grueber, BMCRR n, 582-586, and Mattingly, BMCRE1, cxxvi, II I-I 12, Scarp us' coinage for Octavian/Augustus is dated from 30 to 27 BC for two reasons which no longer hold: the evident imitation of Octavian's type of Victory on globe (Mattingly, BMCRE I, nos. 602-604) (fig. 21) and the variant with A VGVSTVS which could date no earlier than 27 BC. But some of the IMP CAESAR and CAESAR DIVI F issues ofOctavian, from which the Victory derives, are now plausibly viewed as pre-Actian by Crawford ;28 and the A VGVSTVS piece of Scarpus appears to be imaginary (Mattingly, BMCRE I, cxxvi). Indeed Dio li 9. I suggests that Scarp us abandoned his troops to Cornelius Gallus very soon after his defection from Antony. The Scato issue. I have argued above that there was only one issue, struck after 23 B.C given Augustus' TR POT. :rhe style of the series is somewhat better than that of Capito and Palikanus, and the weights are less erratic. Grant (FITA, 138) places Scato after the other two on the basis of a supposed weight reduction. This, however, depends on reading his largest piece as a sestertius, the next as a dupondius, contrary to the meaning of the types (see above).

Roman coinage of the Cyrenaica

33

The Capito and Palikanus issues in the name of Augustus. Priority between the two issues cannot be established, and neither can be more precisely dated than subsequent to the reform of the aes of Augustus (itself of uncertain date) whose dupondius they imitate. The issue in the name of Tiberius. From the reference to Tiberius Gemellus and Germanicus, the issue must date post AD 19. 29

DENOMINATIONS AND DENOMINATION-TYPES OF THE ROMAN COINAGE OF THE CYRENAICA

The weights and types of these issues provide the evidence that they formed a single system, based on contemporary Roman coinages. As long ago as Muller (Afrique I, 169-170, Table v) Roman denominations were imposed on these coins, but without any supporting argumentation. Robinson (BMCCyr., ccxxvii) also provides Roman values for them, disagreeing with Muller as to what they might be. (Grant, FITA, 138, follows Robinson, save for the case ofScato where he follows Muller.) The treatment of the coins of Lollius reveals how the two positions arose. (I) Muller says, 'Les pesees ... correspondent assez bien au poids du dupondius romain de ce temps ... A l'epoque ou la Cyrenaique filt faite provide romaine, l'as etait du poids d'une demi-once c.il.d. de 13, 5 grammes' (169). The heaviest of Lollius' Units known to Muller weighed 27.4 g.; the semuncial as weighed a theoretical 13.6 g., i.e. one half of the Roman ounce; the Unit was therefore a semuncial dupondius. (2) Robinson (BMCCyr., ccxiii) takes Lollius' Unit to be, rather, a sestertius, on the argument that its weights correspond to that denomination in Antony's Fleet Praefect coinage. Neither of these positions can be accepted. (I) While Muller's point might theoretically be correct, in fact the coinage of the semuncial as ceased at Rome after 84 BC, according to the chronology of Crawford, RRC, nos. 354-355. The later military issues in bronze, of which Sextus Pompey's and Octavian's were struck in very considerable numbers, were light uncial asses, not semuncial; so too the local Gallic aes ofOctavian. 30 The dupondius did not exist at all as a denomination in the first century BC until the Triumviral period when one issue was struck, so small that today only a single specimen survives.31 If the Unit of Lollius was intended to conform to Roman Republican monetary usage, it cannot have been a dupondius. (2) Similarly, Antony's Fleet Praefect coinage was at variance with normal usage, in that it included a curious range of denominations including a tripondius, and was struck to two standards, quartuncial and octuncia1. 32 Not only are these pieces idiosyncratic in the extreme, but their locus of production is quite unknown. To my knowledge no example of this variegated coinage has ever been attested in a Cyrenaican find, and there is no evidence that they ever had anything to do with the province. Robinson's association of this material with Lollius' coinage (itself of uncertain date) is entirely arbitrary and unsupported by any evidence. The association of Lollius' Unit with an imaginary Republican dupondius and with the Fleet Praefect coinage has contorted the whole picture of the Cyrenaican denominations. Nonetheless, that our aes was struck in terms of Roman denominations is certainly correct. A coherent account of the whole range from Lollius to Tiberius can

34

T. V. BUTTREY

be constructed if the Cyrenaican monetary circulation of the first centuries BC and AD is laid out as it really was: see Table 2 below, which includes average weights of the specimens known to me, and suggested denominations. Since the latest issues are the clearest, and since later issues tend to imitate earlier ones, the overall shape of these coinages will emerge most easily if they are taken in reverse order. Tiberius: As indicated above the denominations dupondius, as and semis are certain from the Roman and Gallic analogues of denomination and weight. The camel of the smallest denomination must have been copied from the only piece on which the type had earlier appeared, the Apollo/camel variety of Lollius. I will argue below that Lollius' piece must have been a semis, so that the imitation under Tiberius is of denominational, not simply antiquarian significance. Capito and Palikanus: The heavier piece in each series imitates the dupondius of Augustus' reformed coinage and approaches its theoretical weight of 13.7 g. The lighter piece is typeless, but it must be the as and therein imitates the common issue of Antony and Cleopatra, which is of that denomination. Seato: The largest piece with its two heads is in obverse type analogous to the Gallic dupondius (i.e. the revalued asses of Copia and Vienna), and to the reformed dupondius in reverse type and in weight; the second piece with head of Augustus similarly imitates the obverse of the Augustan reformed as. It follows that the two smallest pieces, whose types had already appeared on coins of Pupius, must on the basis of weight be identified as semis and quadrans. Antony and Cleopatra: The larger, and much the commoner, of the two modules cannot be evaluated by type - it has none. But the average weight of normally well-circulated specimens approaches that of the semuncial as, theoretically 13.6 g. The heaviest individual specimen attested is Svoronos, Ptolemaion, no. 18996' at 15.40 g.; next come two die-linked pieces in the British Museum at 12.94 g. each. The smaller denomination then can only be the semis, of relatively low weight and narrow flan by comparison with the later issue of Scato; but only three weights are to hand (Svoronos, Ptolemaion, nos. 1900 et, P; BM 1952 acquisition). The weight of the Unit, its imitation on the asses of Capito and Palikanus, and the fact that an example has been found halved, all tie it into the Roman system and identify it as a semuncial as. A. Pupius Rufus: The Unit here too approaches the theoretical weight of the semuncial as; the heaviest example known to me, Robinson, BMCCyr., no. 28, weighs 13.08 g.33 The reverse type of sella was taken over by Pupius from the Unit of Lollius, also to be regarded as an as. The two smaller denominations bear the same obverse types - ram and Libya - as the later Scato pieces ofthe same weight and module, and must represent the same denominations, viz. semis and quadrans. That is, imitation of specific types, already seen to be so frequent at Cyrene, has a denominational function quite apart from those types associated with the reform of Augustus. Pupius himself had already borrowed both types, the Libya of the quadrans from Crassus, the ram of the semis not from Crassus (who depicted Apollo) but from a Ptolemaic prototype (see above). Nonetheless, there may be more continuity here than strikes the eye. Robinson (BMCCyr., clxviii, ccxxxiv) believed that the ram refers to, and indeed may represent, Zeus Ammon. But the other

Roman coinage of the Cyrenaica Table

2 39

Average weight (g.)

Weight (g.) Obverse type

35

Average diameter (mm)

Diameter (mm)

Figure no. (pis. S-8)

Denomination (number of examples)

Uncial standard L. Lollius Greek Latin Greek Latin Greek Latin

Zeus Ammon Zeus Diktaios Apollo Artemis Libya wreath

as semis quadrans

20.46 (16) \.. 20.84(2S) 21'SI (9) J 11.60 (8)} 10.67(4S) 10"47 (37) 4.3 6 (4) } 4·33(13) 4.3 1 (9)

32.3 (19) \.. 32.3 (9) 27.2 (10) \ 27. 6 (3S) { 17·0 (4) 16.8 (9) J

J

32.3(28) 27·S(4S) 16·9(13)

2S 26 27 28 29 30

Semuncial standard Crassus Greek

Libya Tyche Latin crocodile Latin Apollo Greek Libya A. Pupius Rufus Greek Zeus Ammon Greek ram Greek Libya Antony and Cleopatra Greek Greek

as'0

l

28·4 29·9 26·3 20.8 IS·8

I

31 32 33 34 3S

semis quadrans

IS·29 (2) 13.78 (2) 11.68 (12) 7. 0 3 (S) 3.0 3 (6)

as semis quadrans

11.67 (II) S·12 (7) 2.61 (I)

28.1 (13) 21.9 (8) 16·S (I)

36 37 38

as semis

10·97 (14) 4.66 (3)

26.1 (7) 19·5 (2)

39 40

J

12·39(16)

(3) (3) . (6) (4) (6)

J

27·7(12)

Reform of Augustus Scato Latin

Augustus and Agrippa Augustus ram Libya

Latin Latin Latin Capito wreath Latin Latin Palikanus Latin wreath Latin Tiberius Greek Drusus with symbols Greek Drusus Greek camel

dupondius

13.9 1 (4)

29·9 (S)

41

as semis quadrans

10.21 (S) S·14 (3) 2·34 (3)

26.1 (S) 21.3 (3) IS·9 (3)

42 43 44

dupondius as

12-40 (1)41 9.02 (S)42

26·8 (2) 24. 2 (2)

4S 46

dupondius as

II.7S (S) 7. 00 (2)

27·9 (S) 22·9 (2)

47 48

dupondius

14. 19 (4)

28.0 (4)

49

as semis

9· II (4) 3.9 2 (2)

24.6 (4) 20·9 (2)

So SI

T. V. BUTTREY

ram god, Apollo Carneius, was well known at Cyrene, where his head figures on many coins dating from the fourth century BC onwards (Robinson, BMCCyr., ccxl, p. 16, no. 68). The semisses of Lollius and Crassus had borne a head of Apollo who is, to be sure, not horned as Carneius; but it might be that Pupius redesigned that denomination with a view to its continuing to refer to Apollo, but in another form equally well known at Cyrene, Carneius, via a symbol, the ram, as appropriate to him as it was to Zeus Ammon. Crassus: The individual Units, with their unhelpful types, vary considerably in weight and flan size, owing to particularly poor manufacturing control, but the average of all our specimens is that of circulated semuncial asses. The average Half-unit is a bit heavy at 7.03 g. for the semuncial semis, but note that e.g. BMCCyr., no. 25 his, at 7.85 g. is struck on an unusually thick flan. 34 The average Quarter-unit falls just under the theoretical semuncial weight for the quadrans, and bears the head of Libya which we have already seen on the quadrantes of Pupius and Scato. Crassus' three denominations are therefore the semuncial as, semis and quadrans. Finally, the issue of Lollius: Plainly the six types represent three denominations, falling into corresponding Greek and Latin series. The Greek types are all imitated later as denomination-types: Apollo and Libya on the semis and quadrans of Crassus, Zeus Ammon and Libya on the as and quadrans of Pupius. Lollius' weights however are distinctly heavier than theirs. His Unit, given its type, ought to be the as, and cannot be the dupondius (see above); therefore his is an uncial issue, probably not struck up to full theoretical weight, but in that regard very like the asses of Sextus Pompey and Octavian. Given the regular inclination which we have now seen at Cyrene to imitate denomination-types, it can only be that Lollius' denominations were the same as those of Crassus' and Pupius' later and lighter issues, just as at Rome during the Republic the bronze denomination-types remained constant through various diminutions in the weight standard. A further confirmation of the adherence in the Cyrenaica to Roman weights and denominations is found in the phenomenon of the halved coins. Halved Roman asses have been found in large quantities in Italy and Sicily, Gaul and Germany.35 There were two instances of halving, involving material of different weight and module, and separated by roughly half a century. (I) In the first instance, Roman sextantal, uncial, and to a certain extent semuncial asses were divided in half, cut along the obverse vertical axis between the two parts of the Janus head, or the two heads of the late Republican provincial bronzes of Octavian. This instance is to be dated to the 20S BC and associated with Augustus' reform, which introduced an as ofless than semuncial weight and bearing one head as type. The older Janus head asses must therefore have been retariffed as dupondii, and some subsequently halved so as to provide asses again, each now with one head. Such halves are found by the hundreds all over the West. (2) The second instance occurred under Tiberius, when reformed asses of Augustus, a certain number of Tiberian asses, and early Imperial asses of the Gallic mints as well, were halved. This occurrence seems to have been virtually confined to the valley of the Rhine, and to have had the purpose of providing small change. 36 The evidence now available from the Cyrenaica adds this area to those already known

Roman coinage of the Cyrenaica

37

to have experienced halving. Here, as elsewhere, halving was carried out on the spot, as is proved by the halves oflocal coins. Two pieces disturb the scene somewhat. A halved semis of Pupius was obtained by Norton in Cyrene and is now in the American Numismatic Society.37 There is no explanation for it, although rare halved Republican semisses are known. 3B A quartered as of Pupius is in the Fitzwilliam Museum, Cambridge. Quarters are particularly uncommon everywhere, and should probably be considered freaks rather than evidence of a deliberate monetary innovation. The last two pieces aside, the picture is consistent, equivalent to the two phases of the Western halving: (I) Roman Republican to Antony and Cleopatra, (2) Seato. It is the evidence of the halving of phase (2) which is unexpected, for these halves are attested in the West almost exdusively in the Rhine valley. Here, as in the West, the Augustan asses are perceptibly smaller than those which preceded them, so that it would have been easy to distinguish between them when halving was carried out - i.e., as in Germany, the Republican and the Augustan asses ought to have been halved on two different occasions. There seem to be only three possible explanations, each of which has its difficulties: A. The asses of Seato were cut, presumably accidentally, in the 20S BC when the older Republican and local asses were revalued as dupondii and halved. The difficulty is that the reformed as of Augustus and its imitation by Seato were deliberately designed so as to demonstrate denomination at once: one head = one as. It would bespeak a particularly thoughtless process to halve the very coins which were already defined as equivalent to the halved old as. B. The asses of Seato were independently cut later, under Augustus. The difficulty lies in discovering the occasion and the purpose. C. The asses of Seato were cut under Tiberius, as in Germany. The difficulty is the hardly plausible association of the Cyrenaica with Germany in this enterprise, and the fact that the other Augustan issues, of Ca pi to and Palikanus, have not been found halved. Whatever the solution, the examples from Cyrene not only reveal that halving in the Western European mode was carried on in Africa as well, but gratifyingly confirm thereby that a Roman, rather than a local, monetary system had been in use in the Cyrenaica in the late Republic. The weight standard was first uncial, in the case of Lollius, semuncial thereafter. The denominations of as, semis and quadrans were regular. With the reform of Augustus at Rome, the dupondius was introduced at Cyrene as well. Both dupondius and as were struck under Augustus and Tiberius. The highly erratic weights of the individual specimens do not obscure the fact that taken together they reflect a deliberate adherence to a standard equivalent to that being struck at Rome.

NOTES I

The Soter /Libya issues had been struck in various modules, tending toward the smaller as time went on (E. S. G. Robinson, Catalogue of the Greek coins of Cyrenaica [in the

British Museum] (hereafter BMCCyr.), London 1927, 80-89; but ultimately without the clear distinctions of size, and therefore of denomination, which character-

T. V. BUTTREY

ised the Egyptians' own coinage. In addition, in the late second/early first centuries BC the mint of Cyrene struck a mass of small bronze coins measuring 10-15 mm, of the types Ammon/eagle and Ammon/lsis head-dress. These are not included in Robinson, BMCCyr., and were given to Egypt by J. Svoronos, Ta nomismata tou kratous ton Ptolemaion (hereafter Ptolemaion), Athens 1904, nos. 1646, 1652, 1655-1656, 17 15-1717, 17 20 - 17 24, 173 21733, 1845; but they are as a group unlike the Egyptian in crudity of style and fabric. More important, they are found in the Cyrenaica in considerable quantity: (I) at Apollonia, 42 pieces of these issues were identified, including four minor varieties not in Svoronos (T. V. Buttrey, 'The coins' (hereafter Apol/onia), Apol/onia, The port of Cyrene. Excavations by the University of Michigan, 1965-1967 (ed. J. H. Humphrey), Tripoli 1976, 345-346); (2) at Cyrene, Norton recovered 171 pieces, including examples of three varieties not in Svoronos (Buttrey, Apol/onia, 366--367); (3) in the current excavations of the Demeter Sanctuary at Cyrene, 184 examples have so far been identified; (4) E. S. G. Robinson, 'Greek coins found in the Cyrenaica', NCS IV 1944, 105- I 13, notes over 40 examples; and so on. These all must have been struck at Cyrene, as I have argued at greater length in Appol/onia, 337-338. But when the reattribution is made, nothing remains of this module in the late Ptolemaic coinage of Egypt proper. Svoronos gives only three other bronze issues measuring as little as 15 mm which he can date as late as the reigns of Ptolemy VI and VIII, in the second century BC, and all are very uncommon (nos. 1408, 1428, 1639). There is no evident coordination between the large module bronze which continued to be struck in Egypt, and the small, crude bronzes produced in great quantity at Cyrene in the second/first centuries BC. It is possible, then, that on the level of small change the monetary systems of the two areas were not commensurate. 2 For the denominations of Lollius see below. 3 The dromedary has one hump, the camel it

4

5 6

7

8 9

IO

I I

12

13

14

15

has two. Or else the other way around. On this difficult question, L. Muller, Numismatique de fancienne Afrique I-Ill and supplement (hereafter Afrique), Copenhagen 1860-1874; Ill, 191, annotation to p. 159. W. Wroth, Catalogue of the Greek coins of Crete and the Aegean islands [in the British Museum], London 1886,63, nos. 18-19. A. B. Cook, Zeus I, Cambridge 1914, 15, 65 2 . So suggests Anne E. Chapman, 'Some first century B.C. Bronze coins of Knossos', NC7 VIII 1968, 13-26 at 18, not seeing the bow and quiver. But they can be made out on Robinson, BMCCyr., pI. xlii, no. 2. For the stag in Crete in classical times see 'Kreta', Paulys Real-encyclopiidie der classischen Altertumswissenschaft 2 (ed. G. Wissowa, W. Kroll) xl-ii, Stuttgart 1924, 1741, 1742-1743. Pausanias v 7, 6-8, I; v 13, 8. And went further in his catalogue, adding to the Cyrene listings the tetradrachm with busts of Antony and Cleopatra and her legend ... 0EA NE!lTEPA as on the bronze (Ptolemaion, nos. 1897-1898). But he withdrew this suggestion in text, at vrry, and reassigned the tetradrachm to Syrian Antioch. K. Regling, [review of] Svoronos, Ptolemaion, ZjN xxv 1906, 344-399 at 397. M. Grant, From imperium to auctoritas (hereafter FITA), Cambridge 1946, 64-65. T. V. Buttrey, 'Thea neotera on coins of Antony and Cleopatra', ANS-MN VI 1954, 95- 109. H. Mattingly, Coins of the Roman Empire in the British Museum I. Augustus to Vitellius (hereafter BMCRE I), London 1923. T. V. Buttrey, 'Halved coins, the Augustan reform, and Horace, Odes I. 3', American Journal of Archaeology LXXVI 1972, 31-48. Grant (F/TA, 138) concurs, but he approaches the question from a different angle, and concludes that Scato's coins (or only the two larger denominations?) were struck at Cnossus. But the as (= Grant 'dupondius'; see below) appears to occur only in Cyrenaica: for one example from Apollonia see T. V. Buttrey, Apol/onia, 346, no. 54; for a second from Cyrene, Buttrey,

Roman coinage of the Cyrenaica

16

17

18

19 20

21 22 23

24

Apollonia, 367, no. N74; a third is in a private collection, from Cyrene; for a fourth, J. Boardman, J. Hayes, Excavations at Tochra 1963-1965, The archaic deposits II, London 1973, 123. Robinson's order, Palikanus - Capito, is apparently arbitrary; see BMCCyr., ccxxv. Grant (FITA, 135) accepts Palikanus' coins as the earlier' by their higher weight and closer imitation of the Roman models'. But Capito's dupondius is of marginally better style (Robinson, BMCCyr., pI. xliv, no. 5, as against nos. I -3), and heavier weight. See Table 2. As noted above, the typeless variety B must have been imitated from the typeless issue of Antony and Cleopatra, which suggests the continuing circulation of those coins under Augustus. Robinson, BMCCyr., 116, nos. 19-20; BMCCyr., cciv-ccv, nos. 18 bis a to 20b, 2od-20f, 20h; BM, 1936 acquisition; G. Macdonald, Catalogue of Greek coins in the Hunterian Collection, University of Glasgow Ill, Glasgow 1905, 576, no. 66; ANS, 2 pieces. Buttrey, Apollonia, 346, no. 56. In the same way, several halved Mamertine bronzes were found at Morgantina, of the same module as Republican asses and presumably cut along with them, see T. V. Buttrey, n. 14 above, 40 n. 57. Buttrey, Apollonia, 367, no. N72. Buttrey, Apollonia, 346, no. 54. G. Perl, 'Die romischen Provinzbeamten in Cyrenae und Creta zur Zeit der Republik', Klio LII 1970, 3 I 9-354; 'Nachtriige', Klio LIII 1971,369-379. The first two issues catalogued by Robinson appear to be exclusively Cretan in origin. (I) Roma/bee. These uncommon pieces are struck on an old-fashioned dumpy flan, quite unlike any of the properly Cyrenaican issues (pI. 5, no. 23). Although one find in the Cyrenaica is attested (R. C. Bond, J. M. Swales, 'Surface finds of coins from the city of Euesperides', Libya Antiqua II 1965, 101, no. 137), the coinage is undoubtedly Cretan for the types can be paralleled there. See B. V. Head, A guide to the principal coins of the Greeks, London 1932,

83,

39

VII.B. 10, a tetradrachm of Gortyna c. 69 with types Roma (i.e. helmeted female head)/ Artemis Ephesia with bee symbol. (2) The Licinius issue. These are struck on a thin flan of small module and low weight, again not relating to any of the issues from the Cyrenaica (pI. 5, no. 24). Although the legends of the two faces point to both parts of the joint province-KPHTA, AIBYH-two pieces with provenience come from Crete (Robinson, BMCCyr., I 13, no. 2; cciii 2 bis a), while none of the Cyrenaican excavations known to me has produced any. 25 For a resume of the problem as far as Crete is concerned, see Chapman n. 6 above, 13-'4· 26 H. A. Grueber, Coins of the Roman Republic in the British Museum (hereafter BMCRR) I-Ill, London 1910. 27 M. H. Crawford, Roman republican coinage (hereafter RRC), Cambridge 1974. 28 M. H. Crawford, [Review of] K. Kraft, Zur Munzpragung des Augustus, JRS LXIV 1974, 246- 247. 29 I omit several suggested additions to the list of Roman Cyrenaican issues. Grant introduces an entirely new coinage, struck by P. Cosconius, otherwise unknown (FITA, 260-261). The attribution depends on identifying the reverse type, a curious plant, as the silphium. To my eye it bears no resemblance. Alfoldi (who believes Pupius, Scato, Capito and Palikanus to have been praefecti class is) would add the aes of Clovius and Oppius (Crawford, RRC, nos 476, 550) and the aurei and denarii of Q. Cornuficius (Crawford, RRC, no. 509) (,Commandants de la flotte romaine stationee it Cyrene sous Pompee, Cesar et Octavien', Melanges d'archeologie, d'epigraphie et d'histoire offerts a Jerome Carcopino, Paris 1966, 25-43). There are two major arguments, neither of which is tenable. (I) The style of these issues is said to be related, and that of Oppius very close to the coins ofCornuficius. The three obverse types of the last include a head of Jupiter Ammon, historicaiJy the major type of the coinages of Cyrene. Cornuficius' mint would then have been at Cyrene, and so too the mint of the two aes BC

40

T. V. BUTTREY

issues. But even allowing the stylistic associations, Cornuficius himself was never anywhere near Cyrene as far as the sources indicate, and his obverse types also include a bust of Africa, never found at Cyrene, and a head of Tanit, historically the major type of the coinages of Carthage. Clearly the obverses taken together refer to his province and cannot be assumed to be equivalent to mintmarks. (2) Alfoldi's chronology places Clovius' coinage in 46 BC, then Cornuficius' and Oppius', in 41-40 BC. Lollius' style, he believes, could then derive from Oppius' (p. 30), and the rest of the Roman coinages of Cyrene follow. In other words, Clovius would stand earliest in the Roman coinages of the Cyrenaica. But in stylistic justification of his attribution of Cornuficius' coinage and the aes said to be related to it Alfoldi looks for 'ne ... pas une simple Moneta castrensis temporaire, mais un hotel des monnaies bien installe, avec une longue tradition artistique. Or, precisement, l'outillage, I'experience et le personnel necessaires it un resultat aussi accompli existaient it Cyrene' (p. 29). This is not the case at all; there had been no mint in operation at Cyrene for at least 50 years prior to the date which Alfoldi would assign to Clovius, that is since the land was bequeathed to the Romans on the death of Ptolemy Apion. And the last coinage struck under Ptolemaic authority consisted ofiarge quantities of unutterable junk (Robinson, BMCCyr., 86-89, and the small bronzes discussed in n. I above). The 'long artistic tradition' had been dead for a couple of centuries. There is then no reason to assign either the aes issues of Clovius and Oppius or Cornuficius' coinage to Cyrene. 30 Buttrey, n. 14 above, 34-37: the average of a number of Sextus' Janus-Pompey asses is 21.52 g; of Octavian's DIVOS IVLlVS asses, 20.80 g.; of asses of Copia and Vienna, 19.13 and 19.43 g .. respectively. The one exception is the rare semuncial as of Atratinus, Crawford, RRC, no. 530 = Grueber, BMCRR 11, 501 (14.2 g.). 31 T. V. Buttrey, 'The unique "as" ofCn. Piso Frugi, an unrecognized semuncial dupondius', Studi Oliveriani XI 1963,3-10. I accept

32

33

34

35 36

37 38 39

40

the revised date suggested by Crawford, RRC, no. 547. GrueberinBMCRRII, 511-512 n.,doubted that there actually had been two standards in use. But their distinction and identification can be supported statistically; see T. V. Buttrey, 'Studies in the coinage of Marc Antony' (diss. Princeton 1953), Appendix I. The weight of Robinson, BMCCyr., 118, no. 30, '7.95' g., should be corrected to 11.54 g. The weight given in Robinson, BMCCyr., '8-48' g., is in error. Several of the diameters reported for Crassus in BMCCyr. should be corrected, viz., no. 25,0.87 inch; no. 25 his, 0.79 inch; no. 26,0.59 inch. Buttrey, n. 14 above. T. R. Volk now informs me of one halved reformed as certainly found in Italy (see G. B. Frescura, 'Monete romane trovate a Lagole nel 1950-51', Archivio Storico di Belluno, Feltre, e Cadore XXII 1951,74-75), now in the Museo della Magnifica Comunitit di Cadore, Pieve di Cadore. Additionally a small number in northern Italian collections may well have been found locally - Aosta (M useo Archeologico), 3; Benevagienna (Museo Civico), I. Buttrey, Apollonia, 367, no. N73· Buttrey, n. 14 above, 33 n. 16. The examples collected for this Table are from Robinson, BMCCyr., including both those in the introduction and in the catalogue proper; subsequent BM acquisitions, coins from the Hunterian and Fitzwilliam Museums, the ANS, a few pieces from the curren t Cyrene exca va tions or in the Cyrene Museum, and a private collection assembled in Cyrene; for Antony and Cleopatra, the examples listed by Svoronos, Ptolemaion. I have not usually included the weights given by Muller, since some might be duplications of the other material, and in any case the state of preservation is not ascertainable. Not only are the three as issues from the three mints divergent in average weight, their weight ranges do not even overlap save for a single heavy crocodile (Latin) piece: Libya (Greek), 15.32-15.25 g.; Tyche

Roman coinage of the Cyrenaica (Greek), 14.30--13.25 g.; crocodile (Latin), 15.08,13.10--9.08 g. Consequently the average as weight and diameter given here are informational only, but it is the case that in weight the denomination does approach the semuncial standard. 41 The British Museum specimen. A second piece in Paris weighs only 9.00 g., but is holed.

42 Here only I have averaged in Muller's weights (Staatliche Munzsammlung, Munich, and Nationalmuseet, Copenhagen) with the others (Paris, ANS, private collection). The range is still enormous, I 1.80--5.57 g. The ANS piece at I 1.32 g. is struck on a very thick flan, fully 3 mm thick.

KEY TO ILLUSTRATIONS

Text figures I Ptolemaic, mint of Cyrene, 3rd/2nd c. BC, as Svoronos, Ptolemaion, no. 872 (Fitzwilliam Museum, Cambridge: Leake Collection). 2 Ptolemaic, mint of Cyrene, 2nd/1st c. BC, as Svoronos, Ptolemaion, no. 1655 (Fitzwilliam Museum: Leake Collection). 3 Ptolemaic, mint of Cyrene 2nd/1st c. BC, as Svoronos, Ptolemaion, no. 1845 (Fitzwilliam Museum: Leake Collection). 4 Cyrene, 5th/4th c. BC, rev. silver tetradrachm (Fitzwilliam Museum: McClean Collection, no. 993 6). 5 Pup ius Rufus, obv. bronze Unit (British Museum: BMCCyr., pI. xliii, no. 2). 6 Ptolemaic, mint ofCyrene, (?), 3rd/2nd c. BC, obv. bronze issue (Nationalmuseet, Copenhagen: SNG XL, no. 456). 7 Pup ius Rufus, obv. bronze half-unit (British Museum: BMCCyr., pI. xliii, no. 4). 8 Ptolemaic, mint of Cyrene, rev. bronze Unit (Fitzwilliam Museum: Weber gift). 9 Pup ius Rufus, obv. bronze Quarter-unit (Nationalmuseet, Copenhagen: SNG XLI, no. 1317). 10 Augustus (Rome), obv. copper as of C. Asinius Gallus (Fitzwilliam Museum: general collection). I I Scato, obv. bronze as (Fitzwilliam Museum: general collection). 12 Augustus (Rome), obv. orichaIcum dupondius of C. Asinius Gallus (Fitzwilliam Museum: general collection). 13 Scato, rev. bronze dupondius (British Museum: BMCCyr., pI. xliii, no. 7). 14 Augustus (Rome), orichaIcum dupondius of Cn.Piso Cn.f. (Fitzwilliam Museum: general collection). 15 Capito, obv. bronze dupondius (British Museum: BMCCyr., pI. xliv, no. 2). 16 Augustus (Rome), rev. silver denarius dated 28 BC (Fitzwilliam Museum: Hart gift). 17 Nemausus, rev. bronze as (Fitzwilliam Museum: Hart gift). 18 Crassus, obv. bronze as (Fitzwilliam Museum: general collection). 19 Scarpus for Antony, silver denarius (Fitzwilliam Museum: general collection). 20 Scarpus for Octavian, silver denarius (Staatliche Munzsammlung, Munich). 21 Octavian (Rome), rev. silver denarius (Medagliere Capitolino, Rome). 22 Scarpus for Octavian, silver quinarius (Fitzwilliam Museum: general collection).

Plate 5 23 Crete, bronze Unit (British Museum: BMCCyr., pI. xxxix, no. 4). 24 Licinius (Crete), bronze issue (British Museum: BMCCyr., pI. xxxix, no. 5).

42

T. V. BUTTREY

(For nos. 25-51 see Table 2) 25 British Museum: BMCCyr., pI. xli, no. 7. 26 British Museum: BMCCyr., pI. xxxix, no. 27 British Museum. 28 British Museum. 29 British Museum. 30 British Museum: BMCCyr., pI. xli, no. 3.

10.

Plate 6 31 British Museum. 32 Kunsthistorisches Museum, Vienna: BMCCyr., pI. xlii, no. 33 British Museum: BMCCyr., pI. xlii, no. 8. 34 British Museum: BMCCyr., pI. xlii, no. 9. 35 British Museum: BMCCyr., pI. xlii, no. 12. 36 British Museum: BMCCyr., pI. xliii, no. 3. 37 British Museum: BMCCyr., pI. xliii, no. 4. 38 British Museum: BMCCyr., pI. xliii, no. 6.

10.

Plate 7 39 British Museum. 40 British Museum. 41 British Museum: BMCCyr., pI. xliii, no. 7. 42 British Museum: BMCCyr., pI. xliii, no. 8. 43 British Museum: BMCCyr., pI. xliii, no. 9. 44 British Museum: BMCCyr., pI. xliii, no. 10. 45 Bibliotheque Nationale, Paris: BMCCyr., pI. xliv, no. 5. 46 Bibliotheque Nationale, Paris: BMCCyr., pI. xliv, no. 6. 47 British Museum: BMCCyr., pI. xliv, no. I. 48 British Museum. Plate 8 49 British Museum: BMCCyr., pI. xliv, no. 7. 50 British Museum: BMCCyr., pI. xliv, no. 9. 51 British Museum: BMCCyr., pI. xliv, no. 10.

I am grateful to the Curators of the collections cited for providing casts and photographs.

43

Roman coinage of the Cyrenaica

24

29

30

Plate 5

44

T. V. BUTTREY

31

32

33

35

38

Plate 6

Roman coinage of the Cyrenaica

45

40

44

43

46

Plate 7

T. V. BUTTREY

49

50

Plate 8

51

4 Roman imperial coin types and the formation of public opinion M. H. eRA WFORD

The authority and personality of the Roman emperor and his government were mediated to the subject population in a variety of ways. Perhaps the most important was the complex of stories circulating probably even among relativdy humble people, many of them about contact with the emperor or his representatives. 1 Of course such contact itself, when it occurred, also mediated the authority and personality of the emperor and his government; the imperial cult, too, had an important role to play, as did the erection of figured monuments and imperial buildings and the distribution of imperial largesse. Within the empire as a whole, the plebs in Rome naturally had privileged access to the benefits of imperial rule and saw far more of the emperor and his works. On the nature and importance within this context of the role of the imperial coinage widely differing views are expressed; at one extreme there is the view that the emperor himself paid particular attention to the choice of types for his coinage in order to draw attention to his virtues and his successes and that these types had a major impact on the population of the Roman Empire, at the other extreme the view that only a minor department of government was involved and that the pictorial types of the imperial coinage were little noticed and often misunderstood. 2 I begin by looking at the extent to which historians in antiquity made use of buildings, statues, inscriptions, archaeological discoveries and coins as historical evidence and then consider the impact which was made by all of these and by coins· in particular on the public as a whole. My conclusions are perhaps valid only for the educated classes, who created most of such records as we have which are relevant to this investigation - but it is unlikely that the impact of a coin on an illiterate peasant was greater. Thucydides, as we might expect, uses an antiquarian's method of research to aid historical enquiry.3 He attempts to use the evidence provided by the clearance of the graves on Delos in order to establish the identity of the primitive inhabitants of the island. 4 He is aware in his discussion of Mycenae of how little can be deduced from architectural remains about the size or importance of a city." The dedicatory inscription

47

M. H. eRA WFORD

of Pausanias on the tripod at Delphi, erected after the battle of Plataea, is admitted as evidence for the improper ambitions of that wayward monarch. 6 The altar of Apollo and a stele on the Acropolis are used as evidence for the family of Peisistratus. 7 Similarly, Aristotle cites an inscribed discus at Olympia in his discussion of Lycurgus. 8 Polybius, whose testimony is unwisely rejected by Livy, records the size and composition of Hannibal's army for the invasion of Italy on the basis of the bronze inscription set up by Hannibal in the temple of Hera Lacinia at Croton, an inscription which Polybius himself sought out and inspected. 9 And in his critique of Timaeus, Polybius' mounting indignation reaches its peak with the remark that Timaeus went to the trouble of collecting inscriptions from hidden and inconspicuous parts of the buildings; and still he was inaccurate. lo Diodorus justifies his belief in the revival of Sicily after Timoleon by an appeal to the number and grandeur of the buildings erected. l l When Cicero was trying to reconstruct the composition of the Commission of Ten sent to help L. Mummius with the settlement of Greece after the fall of Corinth in 146 BC, he accepted Atticus' identification of A. Postumius Albinus as a member of the Commission on the basis of a statue at the Isthmus. 12 Suetonius adduces statues of Vespasian's father, dedicated 'to an honest tax-collector', in his account ofVespasian's family.13 And Plutarch occasionally (unlike Suetonius) uses statues as a guide to the appearance of the subjects of his Lives, in order to be able to argue from their appearance to their character. 14 For the use of coins as historical evidence we have only, so far as I know, the example of the Historia Augusta, where they are used rather to lend verisimilitude to historical fiction. But their use in this way implies the possibility of their use in history. In the most explicit example, the author and his friends are debating whether or not one Firmus was ever emperor. The opinion is advanced that he wore the purple, struck coins and was called Augustus. Severus Archontius even produces coins of Firmus to prove the point. Needless to say, none now survive and there is no reason to suppose that they were ever more than a figment of the imagination of the author of the Historia Augusta. 15 The technique is a favourite one of his; Odenathus, Trebellianus, Victoria, all are endowed with coins and the title of Augustus or Augusta. 16 None deserves a place in history. It is nonetheless interesting to note that the assumption lying behind the argument is that the issuing of coinage is a mark of sovereignty, an assumption to which I return below. It is also interesting that the modern discovery of an emperor from his coins, Uranius Antoninus, regarded by Michael Grant as one of the chief services of numismatics to history, should share its methodology with historical fictionY It is apparent, then, that the use of monuments of one sort or another, including coins, as historical evidence was known to the historians of antiquity. It is also apparent that the people responsible for all these monuments, buildings, statues, inscriptions and coins, attached some importance to them. One might presume this in any case from the expense normally involved in their production, but explicit evidence may also be found. For Theopompus, inscriptions were so important to the community responsible for their erection that a plausible accusation to level against Athens could be that of forging the inscription relative to the peace of Callias. 18 In the private sphere, three Thebans of

Roman coin types and public opinion

49

the second rank went to the length of erecting a monument bearing an inscription explaining that it had really been they, not Epaminondas, who had won the Battle of Leuctra. 19 For a later period, Polybius records a remarkable debate among the Aetolians besieging the city of Medion. They were on the point of reaching the annual change over of generals and they believed that they were about to capture the city. A full-scale debate ensued - who should have the right to have his name inscribed on the trophy, the general who had conducted the siege or the general who would within the next few days receive its submission? The Aetolians reached the equitable decision that both should be inscribed on the trophy. Unfortunately, the Illyrians appeared on the scene, the Aetolians were soundly defeated and the siege was raised. The citizens of Medion gleefully inscribed the names of both defeated generals on the trophy which they erected. 20 For the Roman period, Dio records as a matter of some importance the decision of Tiberius to inscribe the temple of Castor and Pollux with the name of Drusus as well as with his own.21 And Vespasian, established as the ruler of the Roman state, took steps to replace the inscribed records of that state destroyed in the fire on the Capitol in AD 69. 22 Again, in the private sphere, Trimalchio's long discussion of what sort of tomb he wanted ended with the quotation of the inscription it was to bear, not the least important part of which was the statement that he never listened to a philosopher. 23 There is also much evidence for the importance attached to imperial statues. An unsuccessful attempt at a prosecution for maiestas under Tiberius involved the charge that the defendant had sold a statue of Augustus along with some gardens. Tiberius' reply is well known, that his father had not been deified in order that the honour should be the ruin of his fellow-citizens. 24 But later history belied these good intentions and the relevant title in the Digest states that wanton damage to, melting down or sale of consecrated statues of the emperor were all illegal. The emperors cared for their statues. 25 For the only explicit evidence that the emperors cared for the types of their coinage we have to go to the fourth-century treatise, de rebus bellicis. The author of this tract, in the course of his suggestions for the better government of the Empire, offers some new ideas for coin types. As preserved, his designs are sadly unimaginative, but no matter. They would hardly have been offered unless the author believed that the subject was of interest to an emperor. 26 This belief is widely shared by modern scholars who are under no obligation to believe that the emperor cared for everything; and the fact that much trouble seems to have been taken over the imperial coinage is adduced in support of their notion. It can hardly be denied that great care was indeed lavished on the imperial coinage, in both design and execution, and I should not wish to reject the belief of the author of the de rebus bellicis as wholly unfounded. But explanations other than direct imperial interest may be invoked for the high quality of much of the imperial coinage and a consideration of other evidence suggests that the author of the de rebus bellicis was expressing a hope rather than documenting an actuality. Some monuments certainly made an impact on those who saw them, even if they were not historians in search of source material. The author of the speech concerning Neaira preserved in the Demosthenic corpus would hardly have appealed to his listeners'

50

M. H. eRA WFORD

knowledge of a painting in the Painted Stoa to prove the presence of the Plataeans at the Battle of Marathon unless he and they had been reasonably familiar with itY The paintings carried in Roman triumphs to portray the course of the war are recorded in an abundance of sources, not only historical and antiquarian. 28 They are included, for instance, in the derisive account given by Cicero of the triumph which Piso claimed was his for the asking. Herodian remarks on the impression made on the population as a whole and on himself by pictures of the emperor. And right at the end of classical antiquity the author of the panegyric to Maximian and Constantine talks of the painting in the palace at Aquileia depicting in effect the mission of Constantine, a painting about which the author seems to have heard in ordinary conversation. 29 When Demosthenes was attempting to stir the Athenians to action against Philip II of Macedon, he appealed to his audience's knowledge of the buildings of Athens and the marvels of the temples and the offerings therein, the achievements of the great days of the fifth century, as well as to his audience's knowledge of the deeds of Miltiades, Aristides and the rest. 30 Statues also excited attention. The Elder Cato of course rebuked those who regarded the honour of a statue as important. 31 Cicero made effective use of the statues in the Forum in the Sixth Philippic, inviting his audience to compare those of L. Antonius and of Q. Marcius Tremulus, victor over the Hernici in 306 BC. 32 The thirty-first and thirty-seventh orations of Dio Chrysostom are full of references to statues. As for inscriptions, in his defence of M. Fonteius, Cicero appeals to the antiquity of his client's family, an antiquity documented and known from the inscribed records of its achievements. The younger Pliny records an inscription which he has himself seen, that of Pallas on his tomb along the Via Tiburtina. 33 But what of coins? The ancients certainly examined their coins carefully, if only to make sure that they were not false. They also had the habit of scratching inscriptions on coins (fig. 1).34 Many of these appear to be gibberish, meaningless collections ofletters,

KOPArrnN

in faint graffito before head

but those which are intelligible often suggest that the coins were intended as presents, whether to gods or to other human beings. It would be surprising if no-one ever noticed what type or legend was on a coin and it is possible that conscious selection sometimes took place of a type suitable as a present or to accompany a burial. An as of Domitian with Fortuna as reverse type was found in the mast step of the Roman ship from

Roman coin types and public opinion

51

Blackfriars. 35 But the range of things which were regularly noticed, or at any rate noticed and recorded, is curiously limited. Before looking at these records, we must make two important distinctions. References to coin types may be of two kinds, which I should like to call direct and indirect. If a man takes a coin out of his purse, looks at the type and then records it, that is a direct reference, even if it comes to us by way of another author. But a chronicler may also record of a ruler that he struck a certain type, without ever seeing any example of the coin in question, but deriving his information from the records of government. I regard a reference to a coin type of this kind as indirect. The second distinction which I wish to make is this. The types and legends of an ancient coin normally had two functions, first to identify the authority responsible for the coin, second to proclaim the message, if any, which that authority wished to put out. The first function is clearly the more important. There is good evidence that in the Greco-Roman world the striking of coins was universally regarded as a mark of sovereignty. 36 As I have noted above, this universal belief appears to lie behind the use, or abuse, of numismatic evidence by the author of the Historia Augusta. But we can go further than that. In the course of the celebrated debate between Agrippa and Maecenas, staged by Dio in 29 BC, Maecenas assumes that Rome can regulate the coinages of the cities of the Empire. And a curious story told by Dio of Vitellius, that he made no attempt to eliminate the coinage of Nero, Galba and Otho, carries the same implication, that the striking of coinage and the possession of sovereign power were linked. 37 Thus Vespasian began to strike coinage soon after his proclamation; alleged coins of Perennis were regarded as evidence of a plot to seize the throne; Severus allowed coins to be struck for Albinus after making him Caesar.38 And it is because the striking of new coinage was linked with the refoundation of Abonouteichos as lonopolis that Lucian records the fact; his actual description of the coins is phantasmagorical. 39 In considering references to coin types we must therefore distinguish carefully between references to types which identify the issuing authority and references to those which convey some further message. Some of the most striking references to coin types conveying some message which we possess seem to be indirect. This comes out particularly clearly in the case ofDio's record of the placing of Caesar's head on the coinage; the measure is recorded by Dio along with the other measures voted by the senate towards the end of Caesar's life. 40 There is no evidence that Dio or his sources ever saw the coins in question (fig. 2). Perhaps the most dramatic coin types of antiquity are those of the celebrated issue of Brutus, with the head of the Liberator on the obverse and the cap of Liberty between two daggers with the legend EID MAR on the reverse (fig. 3). The production of this issue is duly recorded by Dio: 41 'Brutus dealt with these affairs (before going to Asia to meet Cassius), and on the coins which he struck he placed his portrait and the pileus with two daggers, proclaiming thus and by the legend that he had freed his country with Cassius.' The description is tolerably accurate, but again there is no reason whatever to suppose that Dio or his sources ever saw one of these coins. They were not produced in large quantities and are of extreme rarity now. Furthermore the description of the

M. H. eRA WFORD

2 JULIUS CAESAR

3

4

BRUTUS

AUGUSTUS

5

6

NERO

CONSTANTlNE

legend, with the involvement of Cassius, is not such as one would derive from an inspection of the coins, which bear no reference to Cassius. I think we may believe that Dio's record of the coin does not derive from autopsy by anyone, but from a chronicle of the activities of Brutus. The same is true, I think, of a record in Suetonius of a coin type of Augustus. In 45-44 BC, while the future Augustus was at Apollonia waiting for Caesar, his fortune was told by Theogenes on the basis of his horoscope. In due course, Suetonius continues, he made his horoscope public and struck a silver coin with as its type the sign of the zodiac Capricorn, under which he was born (fig. 4). Augustus was, of course, not born under Capricorn, but under Libra, but the reason for the error of Suetonius (and of Manilius) need not detain US. 42 The linking of the remark about Augustus' coin type with the story of his horoscope and its publication suggests to me that the record of the coin type comes from the memoirs of Augustus or an associate, not from observation. There is no way of telling whether the brief reference by Suetonius to the coin type of Nero showing him playing the lyre derives from observation (fig. 5).43 Only with the description of the consecration coinage of Constantine in the Vita Constantini of Eusebius am I reasonably sure of the relevance of observation (fig. 6).44 While we cannot conclude from the passages we have been considering that the Greeks and Romans often noticed the programmatic coin types with which they were confronted, one inference is possible. If chroniclers thought it worthwhile to record what rulers put on their coins, we may reasonably infer that the rulers themselves and the educated classes as a whole attached at least some importance to the subject. The belief of the author of the de rebus bellicis, which we have already considered, is thus to a certain extent vindicated. What did the Greeks and Romans notice about the coins which they handled? The

Roman coin types and public opinion

53

7

8

9

quadrigatus

bigatus

victoriatus

10

11

'double'

cistophorus

most straightforward procedure for naming a coin is to adopt the name of the denomination, to treat it in fact as purely and simply a piece of money. Both drachma and denarius are names of this kind. But both in the Greek and in the Roman world coins were often named after their types. The Roman examples which come most readily to mind are quadrigati, bigati, victoriati, each named after a type, Jupiter in a quadriga, Victory in a biga and Victory crowning a trophy respectively (figs. 7-9).45 (It is presumably not an accident that in each case the legend ROMA is associated with the reverse type.) Tetrarchic nummi were perhaps called' doubles' because they bore two figures on the reverse (fig. 10).46 Greek examples are equally ready to hand. One of the commonest coins of the later Hellenistic world was the cistophoros, named from the cista mystica which it bore (fig. I I). Similarly, the New Style silver coinage of Athens was distinguished from the Old Style by the olive-wreath which it bore on the reverse, occasioning the name stephanephoros (fig. I2). The late Lycian issues were known as citharephoroi, from the lyre which formed the reverse type (fig. 13).47 Coin types in the Greek world were also noticed without giving their names to the coins in question. The type of the Persian daric (fig. 14) provided Agesilaus with the occasion for a bitter joke at the expense of the Athenians and Thebans bribed by Persia to stir up trouble in Greece, that he was driven from Asia by 30,000 archers. 48 The basic types of the Republican bronze coinage, a head of a deity on the obverse and a prow on the reverse (figs. 15-16), similarly found their way into the imagination

54

M. H. eRA WFORD

stephanephorus

13

14

citharephorus

•archer'

16

as

quadrans

of the users of the coinage. The Roman equivalent of the slightly curious English expression, heads or tails, was capita aut navia. 49 Two further comments may be made. The erroneous belief that these coins were the oldest produced by Rome made their types the most discussed of all antiquity and provided the stimulus for increasingly wild antiquarian and etymological speculation. 50 But despite this interest, Pliny, true to form, asserted wrongly that while the reverse type of the as was a prow, that of the triens and quadrans, two of the fractions of the as, was a raft. 51 He was misled by the poetic expression quadrans ratitus which he had read in Lucilius and generated a clear example of the preference for literary speculation rather than visual investigation. The Greeks and Romans, then, noticed at least some coin types which had no necessary connection with the issuing authority. But all the coin types so noticed were very common indeed (despite which they were still sometimes mistakenly read or reported) and were almost devoid of any kind of programmatic content. The only hypothesis which could justify the belief that programmatic coin types were noticed is unverifiable and improbable. It could be argued that their interest was limited to the period of their appearance, and that they were not recorded precisely because of their ephemeral significance. But, if this is so, it is surprising that Cicero's numerous references to ephemeral political concerns of the late Republic do not include a reference to coin types. Cicero knew that Pompeius struck coinage at Apollonia in 49 BC, but he does not record the types of this or any other issue. 52 In any case, by far the most widely noticed feature of ancient coins was that element

Roman coin types and public opinion

55

in the types which indicated the issuing authority. This appears most clearly in the case of the Roman imperial coinage, on which the head of the emperor is the most significant feature. The best-known example is probably the one from the New Testament, where Jesus examines the image and title of the emperor on a denarius and pronounces the dictum, 'give to Caesar what is of Caesar'.53 The same concern with the head of the emperor occurs also in Talmudic texts, where it is stated that the coins used for the redemption of the second tithe must have the imperial image intact. 54 The colloquial expression domini for coins, surely deriving from observation of the imperial image, is also worth noticing in this context. 55 The phrase Kaisaros nomisma was used by Epictetus, 'neither a banker nor a shopkeeper may reject the coinage of the emperor, but ifsomeone produces it, whether he wishes or not, he must give over what is being sold for it'.56 We need not doubt that the identifying feature of the coinage of the emperor was the image of the emperor. An imperial law of AD 343 made the point explicitly, 'all the solidi, on which our images appear and for which there is universal veneration, must be treated as of the same value ... '57 Of course, the corollary of the fact that the emperor's head on the Roman imperial coinage symbolised his authority was the fate which this head suffered when the emperor portrayed fel1. 58 The defacement of reliefs and inscriptions which included a representation or a mention of an emperor who suffered damnatio memoriae is a wen-known phenomenon. It did not escape notice in antiquity. On Domitian's death the senate decreed that his inscriptions were everywhere to be destroyed and his memory to be obliterated. 59 As early as the fourth century BC, Lycurgus told a story (probably apocryphal) about the melting down of a statue of Hipparchus, son of Charmus, sentenced to death in absentia for treason. 60 Coins of the Roman Empire, surely because of the presence on them of the imperial portrait, did not always escape the fate of the man responsible for their production. Dio records that after the death of Caligula the senate decreed that an the bronze coinage which bore his image should be melted down. An analogous political judgment is implied in Statius' derisive reference to an as Gaianus. 61 Some thrifty souls in the Rhine army camps who seem to have wished to

17 CARACALLA

and

18 GET A

NERO

M. H. eRA WFORD

express their disapproval of Ca ligula without going to the lengths demanded by the decree of the senate contented themselves with a chisel blow to deface the emperor's image. Their handiwork survives to be inspected. 62 Caracalla had the coinage of Geta melted down.63 A political judgement is also implied in another story of Epictetus. A man was offered a coin of Nero and rejected it in favour of one of Trajan, despite the fact that the coin of Nero was of greater intrinsic value. 64 There is a curious Jewish example of the same attitude; because of the outrages committed by Hadrian against the Jews in and after AD 132, some Talmudic texts held that the coins bearing his image should not be used and even went so far as to reverse their normal rule, saying that they could be used when so worn that the image could not be recognised. 65 The evidence of coinages other than the Roman imperial confirms the view that the only really important element in the typology was that which served to identify the issuing authority. Many Greek coins were named after the city which issued them, Kyzikenoi and so on. But one can go further: the norm for Greek coinages was for the type of the coinage of a city to be the current badge of that city. When Sestos decided to resume bronze coinage in the second century BC, the twin reasons given are that the sema of the city (in this case a seated Demeter) may be current and that the city may derive profit from the coinage (fig. 19).66 A city's sema was known independently of her coinage; it was often used in the fourth and third centuries BC to decorate an inscription relating to the city or to one of its members. 67 In the field of coinage, the p%i (colts) of Corinth

19

20

21

Sestos

Corinth

Aegina

22

23

24

Athens

Athens

Arab imitation

Roman coin types and public opinion

57

(fig. 20), the chelonai (tortoises) of Aegina (fig. 21) are familiar, and the glaukes (owls) of Athens (fig. 22) are the most familiar of all. The importance of the sema of Athens is apparent not only from the extent to which it was copied (figs. 23-24), but also from the comment of Xenophon that the coins of Athens were in demand everywhere and even carried a premium outside Athens. 68 Their preponderance in the Aegean at the turn of the fifth and fourth centuries is attested by a story of Plutarch. Gylippus was entrusted with the money being sent back to Sparta by Lysander. The money was placed in bags and sealed with a note inside each bag saying how much there was. But in the course of the journey home Gylippus unpicked the stitching at the bottom of the bags, removed some of the money and stitched the bags up again. The ephors were at a loss to account for the discrepancy until Gylippus' servant informed them that there were many owls asleep under Gylippus' roof-tiles. 69 Further evidence that the only really important thing about a coin was whether it issued from a respectable authority or not may be drawn from the habit of naming coins after people. The Lydian Kroiseioi (fig. 25) and Persian Darics (fig. 26) belong to this class, as do the Carian Maussoleia (fig. 27) and the Macedonian Philippeioi (fig. 28), the great gold coinage which found its way into Hellenistic literature as the gold coinage par excellence and made its way from there through early Roman comedy into Augustan poetry; coins were likewise named for Alexander and his successors. The whole ofSulla's coinage in the east was named Lucullion (e.g. fig. 29), largely, I think, because Lucullus was remembered as the collector of the levies of money which went to provide the metal for the coinage. 7o Three oddities are coinages apparently named from women, the Philistideion, the Berenikeion and the Demareteion. The Philistideion is simply mentioned by Hesychius and is presumably the portrait issue struck by Hieron 11 (fig. 30). The Berenikeion is mentioned only in passing by Pollux in the middle of a list and presumably refers to the coinage of Berenice 11, consort of Ptolemy III Euergetes and ruler of Cyrenaica in her own right, a coinage which bears her portrait and inscription (fig. 31).71 The Demareteion is a curious and instructive case. By analogy with other coins named after people, it should be a common coin. But it cannot now be identified with certainty. Diodorus, Pollux and Hesychius tell contradictory, etiological stories about its origin, which are patently invented. Again by analogy, it should either be named, like the Lucullion, because Demarete provided the wherewithal to produce it, or because it bore what was believed to be her portrait. 72 In my view it is the common fifth/fourth-century tetradrachm of Syracuse, with the head of Arethusa mistaken for that of Demarete (fig. 32). It seems, then, that there is little evidence for official interest in coin types and even less evidence that in the Greco-Roman world coin types which may be called programmatic had much impact, although other aspects of coins and coin types were noticed. 73 This fact is probably even more significant than the fact that the situation with respect to coin types differs markedly from the situation with respect to other monuments. But the facts require at least some attempt at explanation. Why did the inhabitants of the Roman Empire not notice the programmatic element in the ever-changing coin types with which they were bombarded?

58

M. H. CRAWFORD

25

27

26

28

29 ~.--......

.

.....i~~~ .: . .~r:/, .. :'

' ~~'

J'

f

.",.,/

I

I

~ ' .'j.."" :i"I "~(i) . t}: I. ',

.. t

I

'~:-=J/- .

30

31

32

First of all, surely, because those who were educated enough systematically to make sense of the reverse types of the imperial coinage had much better ways of finding out about the emperor and his activities. It is also relevant precisely that the types were ever-changing. The vast majority of the inhabitants of the Roman Empire, if they saw . coins at all, saw new issues as a tiny part of a mass of issues covering a century or more. It is not surprising that it was the head of the emperor which was noticed - and doubtless alone symbolised for many people the fact that they belonged to the Roman Empire. Another factor must have been sheer size. A picture carried in a triumph is both more striking and easier to see than the type of a coin an inch or less across.

Roman coin types and public opinion

59

One problem remains, to find an explanation for the diversity, imaginativeness and often great beauty of Roman imperial coin types. The rulers of the Roman Empire were on the whole intelligent men and I find it hard to believe that with so much else on their hands they, or indeed their senior advisers, devoted day-to-day attention to the devising and designing of types of which almost no-one took any notice. The reason, I think, is a combination of accident and human nature. The accident is that Rome, when it adopted the idea of coinage, was governed by an intensely competitive oligarchy. The Republican oligarchy entrusted the production of its coinage, as it was bound to do, to an annually changing magistracy; the rest followed. We saw earlier that the norm for Greek coin types was the badge of the city and all that the Roman oligarchs did with the coinage of their city, when they abandoned the public types with which it began, was to place their own private badges on it. Questions about the reception of this coinage by those who saw it probably did not arise. The victor in the civil wars eventually excluded all but his own badges from the coinage and from this point or from one very soon after it the imperial coinage was, I think, largely carried on by an independent artistic tradition. The Romans were technically capable of the mass-production of coin dies, but they do not seem to have attempted this until very late, no doubt for the same reasons which inhibited technological advance in other fields. Instead they employed artists, engravers, to cut the dies. It would not be surprising if this group experimented and innovated, bringing their skill to bear in the creation of artistically satisfying types, which represented their ideas of the moment about the state and the ruler for whom they worked. In other words, the model I propose is a largely ascending one; of course, an emperor may have issued a general directive to ensure that his coinage represented his 'personality' and the model may thus far be a descending one; but the pattern for the most part is surely one of a mint doing its best for its patron. The most conspicuous results of this effort are of course the great medallions of the later Empire. The officials of the mint doubtless knew how to portray an emperor without much prompting,just as a governor of Asia, Avidius Quietus, writing to Aezani, knew very well how to portray Hadrian. 74 One can easily understand the obvious fact that the coinage of some emperors reflects their reigns quite accurately, without postulating a massive effort to mould public opinion, directed from above and doomed to failure. One can also easily understand that the coinage of some emperors only reflects their reigns in a very odd way: thus the coinage of Vespasian, at firs\ sight innovatory, with its enormous range of changing types, in fact borrows almost every type from the Republic or earlier emperors.75 The mint is presumably following a general directive to make the coinage as like earlier coinage as possible; the result is as far removed from anything resembling propaganda as one can imagine. Not the least of an emperor's functions was that of patron of the arts; conspicuous consumption in this field was expected of him. But no-one has suggested, I think, that all emperors personally inspected and passed the designs for all the buildings and statues erected on their behalf. There is no reason to suppose that they did this for the coinage either.

60

M. H. eRA WFORD NOTES

The following additional abbreviations are employed: ANRW ILLRP BMCRR

McClean OGIS RE RIC RRC SNG

Aufstieg und Niedergang der romischen Welt: Principat (ed. H. Temporini, etc.), Berlin/New York 1972A. Degrassi, Inscriptiones latinae liberae rei publicae, Florence 1957-63. H. A. Grueber, Coins of the Roman Republic in the British Museum, London 19 10. S. W. Grose, Catalogue of the McClean Collection of Greek coins [in the Fitzwilliam Museum], 3 vols., Cambridge 1923-9. W. Dittenberger, Orientis graeci inscriptiones selectae, Leipzig 1903--05. Paulys Real-encyclopiidie der classischen. Altf'rtumswissenschaft (ed. G. Wissowa, etc.), Stuttgart, Munich 1894The Roman imperial coinage (ed. H. Mattingly, etc.), London 1923M. H. Crawford, Roman republican coinage, Cambridge 1974 (pubiished 1975). Sylloge Nummorum Graecorum [GB], London 1931-

See, provisionally, F. G. B. Millar, The emperor in the Roman world, London 1977, 3-4· 2 The central expositions of the first view in recent times have been by H. Mattingly, The emperor and his clients, Todd Memorial Lecture II, Sydney 1948; and C. H. V. Sutherland, Coinage in Roman imperial policy, London 1951 (compare the attempt to defend the approach by the latter in the President's Annual Address, N0 XI 1951, Proceedings, 6-20 at 13-19). An incisive critique of this approach is that of A. H. M. Jones, 'Numismatics and history', Essays in Roman coinage presented to Harold Mattingly (ed. R. A. G. Carson, C. H. V. Sutherland), London 1956, 13-33 at 15 (reprinted, with a brief bibliographical mise-au-point by myself, in A. H. M. Jones, The Roman economy (ed. P. A. Brunt), Oxford 1974,62-81). Sutherland replied in 'The intelligibility of Roman imperial coin types', JRS 2GLIx 1959, 46-55, whence the brief exposition in The emperor and the coinage, London 1976, 96-101; the approach is essentially sterile, since it ignores much of the available evidence. 3 See A. D. Momigliano, 'Ancient history and the antiquarian' , Journal ofthe Warburg and Courtauld Institutes XIII 1950,285-315 at 287-288 (reprinted in A. D. Momigliano, Contributo alia storia degli studi classici, I

Storia e Letteratura XLVII, Rome 1955, 67-106 at 100; Studies in historiography, London 1966, 1-39 at 4); see also R. Weiss, 'The study of ancient numismatics during the Renaissance', NO VIII 1968, 177-187. 4 Thucydides i 8, I; for attempts to relate the evidence of Thucydides to the archaeological evidence see R. M. Cook, 'Thucydides as archaeologist', Annual of the British School at Athens L 1955, 266-270; c. R. Long, , Greeks, Carians and the purification of Delos', American Journal of Archaeology LXII 1958, 297-306; J. Boardman, 'Sickles and strigils', JHS XCI 1971, 136-137; compare Plutarch, Solon 10,4-5, for an alleged allusion to burial habits by Solon and his opponents in the argument over the Athenian claim to Salamis. 5 Thucydides i 10, 1-3; Cook, n. 4 above. 6 Thucydides i 132, 2; compare [Demosthenes]lix 97. 7 Thucydides vi 54, 6-55, 2, with A. W. Gomme, K. J. Dover, A. Andrewes, A historical commentary on Thucydides IV, Oxford 1970, 317-337 at 330-334. 8 Plutarch, Lycurgus I = Aristotle (ed. V. Rose, Leipzig 1886),frag. 533. 9 Polybius iii 33, 18; 56,4; Livy xxi 38, 2. 10 Polybius xii I I, 2. I I Diodorus xvi 83. 12 Cicero, ad Atticum xiii 32, 3; on the whole episode see E. Badian, • Cicero and the

Roman coin types and public opinion commission of 146 BC', Hommages a M. Renard(ed. J. Bibauw), Collection Latomus Cl, Brussels 1969, 54-65; E. Rawson, 'Cicero the historian and Cicero the antiquarian', JRS LXII 1972, 33-45 at 40. 13 Suetonius, Vespasian I; compare Vespasian 12 for the actual tombs of the Vespasii. See in general A. Stein, Romische Inschriften in der antiken Literatur, Prague 1931; R. Chevallier, Epigraphie et litterature aRome, Faenza 1972, ch. 2. 14 A. E. Wardman, 'Description of personal appearance in Plutarch and Suetonius: the use of statues as evidence', Classical Quarterly2 XVII 1967, 414-420, with earlier bibliography. 15 Scriptores Historiae Augustae (hereafter SHA), Firmus 2, I. 16 SHA, The 30 pretenders 26,3; 31, 3; The two Gallieni 12, I for Odenathus; K. Menadier argues that the type alleged is that of the coinage of the age of Julian (' Die Miinzen und das Miinzwesen bei den Scriptores Historiae Augustae', ZjNXXXI 1914, 1-144 at 55). For other numismatic figments of the imagination of the author of the Historia Augusta see Alexander Severus 25, 9; Antoninus Diadumenianus 2, 6. 17 M. Grant, Roman history from coins, Cambridge 1958, 58. 18 F. Jacoby, Die Fragmente der griechischen Historiker, Berlin 1923- , fr. 153-154, compare Plutarch, Omon 13, 4-5. Note also H. Volkmann, 'Die Inschriften im Geschichtswerk des Herodot', Convivium, Beitriige zur Altertumswissenschaft. Konrat Ziegler dem Lehrer und Freunde . .. zum 70. Geburtstag, Stuttgart 1954, 41-65. 19 M. N. Tod, A selection of Greek historical inscriptions 11, Oxford 1948, 92--94, no. 130. 20 Polybius ii 2, 8-1 I and 4, 1-2. 21 Dio Iv 27, 4. 22 Suetonius, Vespasian 8; compare Demosthenes xxii 69-78. 23 Petronius, Satyricon 71, 12. 24 Tacitus, Annals i 73, 2. 25 Digesta Iustiniani xlviii 4. 4, 1-6; see also Suetonius, Tiberius 58; Philostratus, Apollonius of Tyana i 15. It remains true, of course, that people did not always care whether a statue was really of the person

61

named on its base, Cicero, ad Atticum vi I. 17; A. E. Wardman (see n. 14 above); H. Blanck, Wiederverwendung alter Statuen als Ehrendenkmiiler bei Griechen und Romern 2, Rome 1969, Studia Archaeologica XI. See also K. Scott, 'The significance of statues in precious metals in emperor worship', Transactions and Proceedings of the American Philological Associatian LXII 1931, 101- 12 3. 26 Anonymous, de rebus bellicis 3, 4; see E. A. Thompson, A Roman reformer and inventor, Oxford 1952, 36-37. Eusebius, Vita Constantini 4,15, also assumes that the emperor chose coin types (see n. 44 below). 27 [Demosthenesl lix 94; on this text see A. M. Snodgrass, Arms and armour of the Greeks, London 1967,94; on the fame of the Stoa Poikile see R. E. Wycherley, 'The painted stoa', Phoenix VII 1953, 20--35; C. M. Robertson, Greek art, Cambridge 1975, 24 2. 28 W. Ehlers, 'Triumphus', RE VIIA, 493-5I1 at 503. 29 Cicero, in Pisonem 60; Herodian iv 8, 1-2; v, 5, 7; Panegyrici latini 7 (6), 6. 30 Demosthenes iii 25. 31 Plutarch, Cato Maior 19. 32 Cicero, Philippics vi 12; compare vii 16. 33 Cicero, pro Fonteio 41; Pliny, Letters vii 29; compare viii 6. 34 J. Friedlaender, 'Eingeritzte Inschriften auf Miinzen', ZjN III 1876, 44-46 (expanding 'Miinzen mit eingeritzten Aufschriften', Berliner Bliitter fur Munz- Siegel- und Wappen-kundev 1868, 146-147); G. F. Hill, A handbook of Greek and Roman coins, London 1899, 197; BMCRR 1,30; ILLRP I, 88, no. I14 - d'Ailly Collection no. 921 (Bibliotheque Nationale, Paris); ILLRP I, 127, no. 191; C. WCl on RRC 1,419-436, no. 408 (Staatliche M useen, Berlin); C. M. Kraay, Archaic and classical Greek coins, London 1976, 16-17; C. T. Seltman, The temple coins of Olympia, Cambridge 1921, 10, no. 218 A (reprinted from Nomisma VIII 1913, 23-65; IX 1914, 1-33; XI 1921, 1-39). 35 P. R. V. Marsden, A ship of the Roman period from Blackfriars, London 1967. 36 See Herodotus iv 166. 2; I Maccabees 15,6.

62

37

38 39

40 41 42

43 44

45 46

47

48 49

50

M. H. eRA WFORD

OGIS I, 370, no. 229 at line 55 (in a treaty between Smyrna and the incolae of Magnesia, the legal tender of Smyrna is imposed in Magnesia). Dio lii 30, 9, with my remarks in 'Finance, coinage and money from the Severans to Constantine', ANRWII-2, 560-593 at 561; Dio Ixiv 6, I. Tacitus, Histories ii 82; Herodian i 9, 7; ii 15, 5· Lucian, Alexander 58; in view of Alexander's hostility to the city of Amastris, it is ironical that dies for the coinage of Abonouteichoslonopolis were probably always made in Amastris (U. Westermark, 'Amastris-Abonoteichos', Numismatica Stockholmiensia I 1975/ 197 6 [197 8], 7-8 at 7). Dio xliv 4, 4· Dio xlvii 25, 3. Suetonius, Augustus 94; Manilius ii 497; K. Kraft, 'Zum Capricorn auf den Miinzen des Augustus', Jahrbuch fur Numismatik und Geldgeschichte XVII 1967, 17-27 at 17 for this error (reprinted in Gesammelte Aufsiitze zur antiken Geldgeschichte und Numismatik I (ed. H. Castritius, D. Kienast), Darmstadt 1978, 262-272). Suetonius, Nero 25. Eusebius, Vita Constantiniiv 73; veiled head of Constantine on the obverse; Constantine going up to heaven in a chariot on the reverse. Compare iii 47 for the mother of Constantine on gold coins; iv 15 for Constantine looking upwards on gold coins. I suspect that with the advent of the Christian Empire coin types were more noticed, though I am not sure why. Pliny, Natural History xxxiii 46. See the texts collected in Crawford, n. 37 above, 580 n. 80. l. R. Melville lones, 'Greek coin names in -phoros', Bulletin of the Institute for Classical Studies XXI 1974, 55-74. Plutarch, Agesi/aus 15, 6; Artaxerxes 20, 4; Moralia 211b. Macrobius, Saturnalia i 7, 22; Festus, de verborum significatu at Navia; Origo gentis romanae iii 5. RRC, 7 I 8, n. 2. Roman antiquarian speculation actually invented coin types: Plutarch, Quaestiones romanae 274e-f; Poblicola

51 52

53 54

55 56 57 58

59

I I - ox, sheep, and pig as types, on the grounds that these were the source of the wealth of early Rome. Pliny, Natural History xxxiii 45. Cicero, ad familiares xiii 29, 4. Lucilius 278-281W suggests that the rich did not normally handle coins but this does not of course mean that Cicero would not have reported changes in coin types, if their impact had been politically important. Luke 20, 24; Mark 12, 17; Matthew 22, 2 I. E. Lambert, 'Les changeurs et la monnaie en Palestine du ler au I1l e siecle de l'ere vulgaire d'apres les textes talmudiques', Revue des Etudes Juives L1 1906,217-244 at 243 n. 4; LII 1906, 24-42 at 29 n. 2. The exposition of D. Sperber, Roman Palestine 200-400, Ramat-Gan 1974,69, is less lucid. Martial iv 28, 5. Arrian, Epictetus iii 3, 3. Codex Theodosianus ix 22. To show disrespect for the coinage of a living emperor was of course lese majeste, see the texts cited in nn. 24-25. Suetonius, Domitian 23; Pliny, Panegyricus

51.

60 Lycurgus, in Leocratem I 17; see H. Blanck, n. 25 above, 109-112, for other examples; Plutarch, Solon 12, 3. 61 Dio Ix 22, 3; Statius, Si/vae iv 9, 22. See W. Trillmich, Familienpropaganda der Kaiser Caligula und Claudius, Antike Miinzen und Geschnittene Steine VID, Berlin 197 8, 46 n. 99. 62 H. Chantraine, Novaesium III: Die antiken Fundmunzen der Ausgrabungen in Neuss, Limesforschungen VIII, Berlin 1968,22. For the countermarking and overstriking of coins of Elagabalus from Phoenicia and Palestine, see A. Kindler, 'The damnatio memoriae of Elagabal on city coins of the near East', Gazette Numismatique Suisse no. 1171980,3-7; on ancient erasures in general see M. Bernhart, 'Erasionen', [Festschrift] Heinrich Buchenau, Munich 1922, 1-8 (I owe this. reference to T. R. Volk); K. Regling, ZjN 1922, 166 n. 5, at 169; P. Berghaus, Festschrift F. Dorner I, Leiden 1980, 158. 63 Dio Ixxvii 12,6. See K. A. Neugebauer, 'Die Familie des Septimius Severus', Antike XII

Roman coin types and public opinion 1936, 155-172 at 162; fig. 17, a coin with facing portraits of Caracalla and Geta, the latter erased. 64 Arrian, Epictetus 5, 17. For contemporary cancellation of Nero's portrait, see fig. 18. 65 Lambert, n. 54 above, [u] 241-243. 66 OGIS I, no. 339; see the commentary of L. Robert, 'Les monetaires et un decret hellenistique de Sestos', RN6 xv 1973,43-53. Note Periplus of the Red Sea 47 for the legends of the coins of Apollodotus and Menander. 67 L. Lacroix, 'Les "blasons" des villes grecques', Etudes d' Archeologie Classique I 1955/1956[1958] (Annalesdef Est: Memoire no. 19),91-115; T. Ritti, Sigle ed emblemi sui decreti onorari greci, Atti della Accademia Nazionale dei Lincei: Memorie della Classe di Scienze Morali 8 XIV-V, Rome 19 69. 68 Xenophon, Poroi 3, 2. 69 Plutarch, Lysander 16, 2. 70 RRC, 80 n. I; compare Martial xii 57, 7 for Neronia massa. 71 Pollux ix 85. 72 Diodorus xi 26, 3; Pollux ix 86; Hesychius s. v. !!./Ylpap€T€tov; for a dedication by Gelon and his brothers see [Simonides] (ed. E. Diehl, Anthologia Lyrica Graeca 11, Leipzig 1925), no. 106. 73 Contra, e.g. H. Gesche, 'Die Reiterstatuen der Aemilier und Marcier', lahrbuch fur Numismatik und Geldgeschichte XVII 1968, 25-42; M. Manson, 'La Pietas et le sentiment de l'enfance a Rome d'apres les monnaies', RBNS ex XI 1975, 21-80; R. Fears, Princeps a dUs electus: the divine

election of the emperor as a political concept at Rome, Papers and Monographs of the American Academy in Rome XXVI, Rome 1977, 199-205; D. H. Euan-Smith, 'Obverse portrait propaganda', Quaderni Ticinesi [dll Numismatica e Antichita Classiche VI 1977, 257-269; W. Trillmich, n. 61 above, 4-5. Contrast the caution, on general grounds, of D. Mannsperger, 'ROM ET AVG. Die Selbstdarstellung des Kaisertums in der romischen Reichspragung', ANRW II-i, 919-996, doubting whether the coin types or legends of antiquity resembled a newspaper or a royal diary and pointing to the likely roles of tradition, climate of opinion and initiative in the adoption of a coin type. 74 OGIS I, no. 502; see R. Fears, n. 73 above, 279, for a clear use of rather striking - and appropriate - types being chosen without intervention by the emperor. 75 T. V. Buttrey, 'Vespasian as moneyer', NC7 XII 1972, 89-109; the suggestion that Vespasian had been moneyer under Tiberius and so acquired a numismatist's love of diverse coin types is, as Buttrey admits, fanciful. Note that G. G. Belloni, 'Monete romane e propaganda. Impostazione di una problematica complessa', Contributi del lstituto di Storia Antica [della Universita Cattolica del Sacro Cuore] IV 1976, 131-159, and'Significati storico-politici delle figurazioni e delle scritte delle monete da Augusto a Traiano', ANRWII-i, 997-1144, sees coin types as for record purposes, not for persuasion.

KEY TO ILLUSTRATIONS

(All Fitzwilliam Museum, Cambridge) Text figures I Metapontum, obv. didrachm (FM: McClean Collection, no. 924). 2 P. Sepullius Macer for Julius Caesar, denarius as RRC, 489, no. 480/5b (FM: Museum of Classical Archaeology, Gow Collection). 3 L. Plaetorius Cestianus for Brutus, denarius as RRC, 518, no. 508/3 (FM: Hart gift). 4 Augustus, denarius as BMCRE I, 56, no. 305 (FM: general collection). 5 Nero, dupondius as BMCRE 1,249, no. 256 (FM: Young gift).

M. H. eRA WFORD

6 Constantine 11, etc., for Divus Constantinus, nummus as RIC VIII, 539, no. 12 (FM: general collection). 7 Roman Republic, didrachm as RRC, 145, no. 30/1 (FM: general collection). 8 Roman Republic, denarius as RRC, 251, no. 206/1 (FM: Young gift). 9 Roman Republic, victoriatus as RRC, 192, no. 102/1 (FM: McClean gift). 10 Third tetrarchy: Galerius Maximian, nummus as RIC VI, 208, no. 676a (FM: Grierson gift). I I Ephesus, cistophoric tetradrachm (FM: SNG lv-vi, pI. lxxxv, no. 4434). 12 Athens, New Style tetradrachm (FM: SNG Iv-iv, pI. lix, no. 3206). 13 Patara (Lycia), drachm (FM: SNG lv-vii, pI ci. no. 5045). 14 Persia, 5th-4th centuries BC, gold daric as Kraay, Archaic and classical Greek coins, pI. iv, no. 82 (FM: Leake Collection). 15 Roman Republic, as as RRC, 158, no. 56/2 (FM: general collection). 16 Roman Republic, quadrans as RRC, 159, no. 56/5 (FM: general collection). 17 Stratonicea (Caria) for Caracalla and Geta, bronze issue (FM: SNG lv-vi, no. 4732); for an undefaced specimen of this issue, see SNG Copenhagen XXVI, pI. xii, no. 510. 18 Nero, dupondius, type ofD. W. MacDowell, The western coinages of Nero, Numismatic Notes and Monographs CLXI, New York 1979, 172, no. 193 (FM: general collection). 19 Sestos, bronze issue (FM: McClean Collection, no. 4173). 20 Corinth, stater (FM: McClean Collection, no. 5312). 21 Aegina, stater (FM: McClean Collection, no. 6042). 22 Athens, tetradrachm (FM: SNG Iv-iv, pI. lvi, no. 3063). 23 Athens, triobol (FM: SNG Iv-iv, pI. lvi, no. 3109). 24 Gaza (Palestine), drachm (FM: McClean Collection, no. 9547). 25 Croesus, gold stater (FM: McClean Collection, no. 8635). 26 Persia, 5th-4th centuries BC, gold daric (FM: Tremlett bequest). 27 Maussolus, tetradrachm (FM: McClean Collection, no. 8519). 28 Philip 11, gold stater (FM: SNG Iv-iii, pI. xxxvi, no. 2026). 29 Roman provincial administration, New Style tetradrachm of Athenian type (FM: SNG Iv-iv, pI. lix, no. 3231). 30 Syracuse, sixteen litrae (FM: McClean Collection, no. 29 I 2). 31 Cyrene, didrachm (FM: McClean Collection, no. 9966, as of Berenice I). 32 Syracuse, tetradrachm (FM: SNG Iv-ii, pI. xxii, no. 1245).

5 Coin hoards and Roman coinage of the third century AD R. A. G. CARSON

In his little book Numismatics, l in the chapter devoted to coin finds and hoards, Philip Grierson wrote that' hoards are for the most part less obviously interesting than coins in collections, since they often consist of hundreds of virtually identical objects ... but, quite apart from being the ultimate source of all the coins that one sees in collections, coin finds are the numismatist's most valuable single guide to classification and dating'. This statement is of general application to all series of coins, and to all periods within any given series, but there are some series and periods for which the evidence of coin hoards is more valuable than for others, and for the Roman coinage of the third century AD the evidence of coin hoards is, for a variety of reasons, particularly valuable, and the techniques by which this evidence can be exploited merit consideration. The ultimate object in the study of any series of coins is to create, from a mass of material, as fully detailed a chronological picture of the coinage as possible, so that evidence of the coinage can make its contribution to the history of the period. In such a re-creation one of the prime sources, that of contemporary historians, is conspicuously lacking in the third century, in contrast with both the earlier and later centuries of the Empire. The contemporary sources fail in the early part of the century, Dio Cassius in the reign of Severus Alexander (222-35), and Herodian only a few years later in 238, and recourse must be had to the testimony found in the epitomists or in the juridical sources, or else in documents such as papyri and inscriptions. In the third century, also, the amount of coinage which carries a reasonably overt indication of date of issue is significantly smaller than hitherto, and it becomes much less common for offices such as the consulship, whose dates are known, or the annual recording of the tribunician power, to appear in the imperial titulature on coins in anything like the fashion of coinage in the first and second centuries. The methods so far advocated for calculating the total output of coinage at any given period are neither convenient nor, seemingly, reliable, for it remains true that the coins available for study are only a sample of those that have been found, while coins that have been found form only a sample of those that have been lost, and these in turn are

66

R. A. G. CARSON

only a sample of those originally in circulation. 2 While it is true that the formation and deposition of hoards are affected by factors other than quantity and availability of coins, it is the case that hoards of Roman coinage of the third century are much more numerous, at least in Britain and in other western areas of the Empire where there is a long tradition of satisfactory recording of hoards, than hoards of coins of the first two centuries AD. The evidence of coin hoards of the third century is therefore of particular value, for only when a significant number of hoards are of a similar pattern can we safely regard them as representative of the circulating medium. Given the availability of a satisfactorily large number of hoards, it becomes possible from the study of them, and the careful comparison of one hoard with another, to arrange in chronological order whole classes of coins whose precise relation to each other would otherwise be uncertain. In the third century the only hoards which are of significant use in respect of establishing the pattern of the coinage are those of the silver denominations, first the denarius, and later the antoninianus. It is only the hoards of silver that provide a frequency both of occurrence and quantity that can be utilised for this purpose. Gold was indeed also hoarded, but the occurrence of such hoards is so rare that their evidence can make little contribution to this particular problem. The same unfortunately is true of hoards of bronze coins, and for this there appear to be two basic reasons. In the first place it is not surprising that fewer hoards of bronze coins than of precious metal coins were formed in antiquity, both because coins in bronze were token and oflittle intrinsic value, and because any appreciable sum made up of such coins would be cumbersome and weighty. In the second place, hoards of bronze coins discovered in modern times, as they fall outside the purview of treasure trove regulations, are less likely to be reported and, thus, recorded; furthermore, as bronze coins in general are less valuable in commercial terms less effort is made to market them, a process which might bring the existence of such a hoard to light. In the third century, therefore, it is in reality only the hoards of either denarii or antoniniani that exist in sufficient numbers to enable consistency of pattern to be checked, and which present quantities of each type sufficient to provide a basis for statistical analysis. Even here, where there is a considerable number of large hoards, the totality of the potential evidence is not available to be used; for it is the unfortunate truth that from the many hoards which are known covering any specified period within the century, the proportion which is recorded with the sufficiency of detail to make the reconstruction of the hoard assured is usually small. For instance, in the one detailed exercise of this type which I have carried out, for the coinage of Severus Alexander from AD 222 to AD 235, a rigorous search of publications yielded only twelve hoards recorded with adequate detail, 3 out of a total of some 46. Of the types of evidence provided by such hoards one of the most obviously useful is that of statistics, which has at least two major applications. An analysis of the numbers of each reverse type in a given coinage points to a numerical distinction between reverses which furnish its staple constituents, the substantive issues, and reverses which, distinctive in their types as they are, represent a special issue. Of the denarii of Severus Alexander, those with reverse types which are associated with special events such as the

Hoards and Roman coinage of the third century assumption of a consulship, repairs to the Colosseum and the Nymphaeum, the re-dedication of the temple of Jupiter Ultor, exist as very rare examples in collections, but never appear in hoards; and where those with a Liberalitas reverse, recording the largesse dispensed on such special occasions, do appear in hoards, their quantity is sufficiently less than that of ordinary reverses to confirm them as components of a special, restricted, and short-lived issue. For instance, of the denarii of the special issue associated with the accession of Severus Alexander, the reverse with the processional quadriga is completely absent from the great Reka-Devnia hoard, 4 and the accompanying Liberalitas reverse is represented by 77 examples. This is to be compared with such quantities as 148, 60, 176, and 108 in the same hoard for the reverses of the substantive issue, marked as early by TRP in the inscription and thus falling in the same period as the Liberalitas reverse of the special issue. 5 That this is not a chance survival is confirmed by the statistics of other hoards which, being numerically much smaller than Reka-Devnia, show the reverses of this substantive issue represented by single figures, but the special issue Liberalitas reverse represented by, at best, one coin, and more often not represented at all. The incidence of the various reverses of any group of coins in a hoard does not provide a prime criterion for establishing the pattern of issues, but it does perform a useful function in this context. The re-creation of the composition of an issue, and the pattern of issues, depends in the first place on the premise that in the first half of the century the coinage of the mint of Rome was produced by a six-officina system, though it is only on some issues of Phi lip I and his family (244--<) that the coins are marked by officina numerals, either Roman or Greek. The reconstruction of an issue is achieved by building round such dated reverses as there are, reverses which can be associated through a shared form of titulature, similarity of portrait and content of reverse type. It is at this point that the statistics provided by coin hoards can be used to check the correctness of the proposed issue. Where the figures for the six reverses of an issue are seen to be more or less of the same order of magnitude, the acceptability of the grouping of six reverses in an issue is enhanced. Again from the Reka-Devnia hoard, the figures for the six types assigned to issue four of the coinage of Severus Alexander are 40,47,25,40,18,33, and, when the data from the further I I hoards analysed is added, the totals for the types of this issue are 56, 60, 35, 45, 28, 38.6 The content of hoards which span the coinage of a whole reign can be used in such ways, but even where a hoard covers only part of a reign, its content of types can be added to a cumulative total for the issues which it does include. Hoards, however, whose content includes the coinage of only part of a reign, have a particular use in establishing the pattern and sequence of issues, especially when the content includes dated coins. Such hoards, when analysed, provide at least a reasonable indication that certain undated types fall within specific date brackets, and there fortunate chance has provided a number of hoards which terminate at successive points in a reign, a whole range of undated types can be attributed to specific periods. This was found to be the case with the coinage of Julia Mamaea, mother of Severus Alexander, none of whose coins carries an overt indication of date. Four of the recorded

Table ). Incidence oJ reverse types oJ Julia Mamaea in Jour British hoards Hoard

,7

Hoard

226 (?)

Terminating

AD

Severus Alexander

!RP v

(?)

IVNO CONSERVATRIX

AD

28

229

LIB£RALIT AS AVG IlII

IVNO CONSERVATRIX

Hoard 3' AD

230

TRP VIlli

I¥NO CONSERVATIUX

Hoard 4 10 AD

232

TRP XII

fYNO CONSERVATRlX

VENVS GENETRIX

VENVS GENETRIX

VESTA

(palladium)

VEST A

VESTA (patera)

VENVS VICTRIX

(palladium)

VESTA

(palera)

VENVS VICTRIX

VENERI I-r:UCI

F[LlCITAS PVBUCA

standing

FELlCIT AS PVBLlCA

seated

FECVND AVGVSTAE

standing

FECVND AVGVST AE

seated

IVNO AVGVSTAE

R. A. G. CA RSON

70

hoards do not include the whole range of the coinage of Sever us Alexander and Mamaea, but terminate within the reign at points marked by the latest dated coin of Severus Alexander in each hoard. The coins of Mamaea present in each of the four hoards can be tabulated as in Table 3. Analysis points to the issue of five reverses by about AD 226 (viz. Juno Conservatrix, Venus Genetrix, Vesta (palladium), Vesta (patera), Venus Victrix), the addition of one further reverse by AD 229 (Veneri FeJici), of yet another by AD 230 (Felicitas Publica (standing», and by AD 233 a further four reverses (Felicitas Publica (seated), Fecund Augustae (seated, standing), Juno Augustae). This is as far as analysis of the hoards can advance the allocation of reverses of Mamaea within the reign of Severus Alexander, and the more detailed attribution to specific issues within the periods must depend on the use of other criteria. The data assembled from a large number of hoards covering a particular area of coinage has one further function of considerable utility. The true picture of the content of any coinage is usually much obscured in standard listing by the lack of distinction between entries which represent the real .coinage material and entries which, on investigation, represent either at best hybrids or inaccurately reported coins or even forgeries. The comparison of data from a number of hoards with standard lists readily distinguishes these latter categories by their absence from hoards, and, by their removal, leaves the true coinage material for detailed investigation. This point can be illustrated from the coinage of Claudius 11 (268-70). For antoniniani of the mint of Rome, RIC lists 100 reverse types of which about half are recorded with two forms of obverse inscriptions, and thus a material of some 150 coin types. l l The material thus presented is inchoate and not meaningful, but an analysis based on the changing forms of obverse inscription, coupled with the observable absence initially of officina letters in a six reverse-type issue, and then the presence of marks of a I 2-officina system in three stages, produces a pattern

IMP C CLA VDlVS AVG

IMP C CLA VDlVS AVG

IMP CLA VDlVS AVG

IMP CLA VDlVS A VG

rev.

rev.

rev.

rev.

SALVS AVG

GENIVS AVG

GENIVS AVG

SECVRlT AVG

~I Synopsis of issues for Claudius II (Fitzwilliam Museum, Cambridge)

Hoards and Roman coinage of the third century

71

of four successive issues involving only 42 coin types, something less than a third of the total suggested by the RIC entry. The validity of this pattern was borne out by the statistics derived from the Hollingbourne hoard 12 which contained 432 antoniniani of Claudius 11 of the mint of Rome. The existence of the pattern of this coinage is no novelty, for its essentials were worked out almost a century ago by Markl,13 though his account obscures the picture by listing every coin of Claudius 11 of which he could find a record. From his lists it is clear that the majority of the recorded coins, which cannot be substantiated from comparison with hoard material, consist of hybrid coins mostly with reverses of Gallienus, defective coins, misreported coins, and dubious coins with no authority other than the lists of Banduri and Tanini.l 4 It is only in the latter half of the third century that the signatures of the mints begin to appear on the coins, usually in the form of their initial letters, but this is by no means as universal or consistent a practice as it becomes after Diocletian's reform. The attribution of antoniniani to mints has, therefore, to depend on the use of other criteria such as the known number of officinae in use at a mint, the form of officina marking either by Roman numerals, initials of Latin ordinals, or Greek numerals, forms of obverse inscription, form of obverse portrait, or, less objectively and securely, style of portrait. Hoards can provide some assistance. From an analysis of hoard material groups of antoniniani can be segregated with reasonable certainty, representing the product of various mints, and the find spots of hoards can make some contribution to the identification of the mints of the several groups. It is probably a statement of the obvious, but in general terms the varying quantities of each group of coins have a bearing on the location of mints in relation to the find spot of a hoard, the mint of the most numerous group being nearest to the find spot, with a downward progression to the most distant mint having the smallest representation. For a hoard with a find spot in Britain, for instance, the mints of the groups in descending order of magnitude are likely to be situated in Gaul, in Italy, in the Balkans, and in the eastern provinces, though there may, of course, at times be political or other circumstances which distort the general rule. Where mints are located in relative proximity to each other in contrast with the distance from either to the find spot, the difference in frequency of the identifiable groups is likely to be so slight as not to be of real help in identifying the mints. To take the instance of the hoard from Hollingbourne in Kent already mentioned,15 the statistics of the coins of the Gallic emperors, Marius, Victorinus, and the Tetrici, attributed to two mints identified by Elmer as Cologne and Trier, are: Emperor Marius (268) Victorinus (268-70) Tetricus I and 11 (270--3)

Cologne

Trier

9 536 568

31 7 212

11

For Victorinus and the Tetrici Cologne shows some preponderance, though not overwhelming, over Trier, despite the fact that the putative mints are almost equidistant

72

R. A. G. CARSON

from the find spot. The mint of one group identified as Cologne by Ehner had earlier been thought to be Lyons, but, if the above statistics have any significance, they support Cologne as the mint rather than the more distant Lyons. I6 The statistics of groups of antoniniani also shed some light on historical and monetary circumstances of a period. In the large Gloucester hoard I7 the statistics of the main western mint groups are as follows:

Emperor

Lyons

Milan

Ticinum

Rome

Siscia

Other mints

Aurelian (270--5) Tacitus (275-6) Florian (276) Probus (276-82)

41 2,772 148 4,57 8

948

37 2 385 5 762

34 1 487 29 780

735 34 9 37 2

302 11

9 44

In the coinage of Aurelian the very small number of coins from the mint of Lyons reflects the short space of time between the recapture of Gaul from Tetricus in 274 and Aurelian's death in August of the following year. The quantity of coins of the north Italian mints of Milan and Ticinum is of remarkable dimensions. The mint of Milan is positively identified by the addition of its initial letter before the mint was moved to nearby Ticinum, which adds its initial letter to the officina letter and the formula XX of Aurelian's reformed antoninianus in the exergue. The quantity of coins, 948 for Milan and 372 for Ticinum, offers a check on the suggested date for the reform of early 274, a date to which the number of pre-reform coins of Milan compared with that of the post-reform coins of Ticinum lends support. The relatively small number of 341 coins of Aurelian from the mint of Rome compared with 1,320 from the combined mints of Milan and Ticinum, as well as 735 coins from the Balkan mint of Siscia, requires an explanation. The pronounced activity of Siscia in this reign is in marked contrast with what is shown by the statistics of the following reigns tabulated above, where the product of Siscia is less than a tenth of that of Rome for Tacitus and Florian, and less than half for that of Probus. At first sight, the unexpectedly large quantity of coins from SisciaI8 compared with those from Rome for Aurelian prompted the suspicion that some of the traditional attributions to Siscia might be erroneous, especially since for some issues there is a co-incidence at the two mints of officina pattern and marking, but no secure grounds for varying the attributions could be discovered. Since the reduction in the Rome coinage appears on the evidence for the following reigns to be peculiar to Aurelian, the explanation that suggests itself is that the reduction is linked with a specific event, the temporary closure of the Rome mint after its first issue for Aurelian, a closure which may