Buddhism and Gandhara: An Archaeology of Museum Collections 9781138896819, 9781351252768

Citation preview

Buddhism and Gandhara Gandhara is a name central to Buddhist heritage and iconography. It is the ancient name of a region in present-day Pakistan, bounded on the west by the Hindu Kush mountain range and to the north by the foothills of the Himalayas. ‘Gandhara’ is also the term given to this region’s sculptural and architectural features between the first and sixth centuries CE. This book re-examines the archaeological material excavated in the region in the nineteenth and early twentieth centuries and traces the link between archaeological work, histories of museum collections and related interpretations by art historians. The essays in the volume underscore the diverse cultural traditions of Gandhara – from a variety of sources and perspectives on language, ethnicity and material culture (including classical accounts, Chinese writings, coins and Sanskrit epics) – as well as interrogate the grand narrative of Hellenism of which Gandhara has been a part. The book explores the making of collections of what came to be described as Gandhara art and reviews the Buddhist artistic tradition through notions of mobility and dynamic networks of transmission. Wide ranging and rigorous, this volume will appeal to scholars and researchers of early South Asian history, archaeology, religion (especially Budd­hist studies), art history and museums. Himanshu Prabha Ray is affiliated with Ludwig Maximilian University Munich, Germany, and is recipient of the Anneliese Maier Research Award of the Humboldt Foundation. She is former Chairperson of the National Monuments Authority, Ministry of Culture, Government of India and former Professor, Centre for Historical Studies, Jawaharlal Nehru University, New Delhi, India. She is Member of the Governing Board, The Oxford Centre for Hindu Studies. Her recent books include The Archaeology of Sacred Spaces: The Temple in Western India, 2nd Century BCE – 8th Century CE (with Susan Verma Mishra, 2017); The Return of the Buddha: Ancient Symbols for a New Nation (2014); and The Archaeology of Seafaring in Ancient South Asia (2003). Among her earlier works are The Winds of Change: Buddhism and the Maritime Links of Early South Asia (1994) and Monastery and Guild: Commerce under the Satavahanas (1986), in addition to the edited volumes Bridging The Gulf: Maritime Cultural Heritage of the Western Indian Ocean (2016); Indian World Heritage Sites in Context (2014); and Satish Chandra and Himanshu Prabha Ray (eds.), The Sea, Identity and History: From the Bay of Bengal to the South China Sea (2013). Her latest book is entitled Archaeology and Buddhism in South Asia (2017). Her research interests

include maritime history and archaeology of the Indian Ocean, the history of archaeology in South and Southeast Asia and the archaeology of religion in Asia.

Archaeology and Religion in South Asia Series Editor: Himanshu Prabha Ray

Ludwig Maximilian University Munich, Germany; former Chairperson of the National Monuments Authority, Ministry of Culture, Government of India and former Professor, Centre for Historical Studies, Jawaharlal Nehru University, New Delhi, India

Editorial Board: Gavin Flood, Former Academic Director, Oxford Centre for Hindu Studies; Jessica Frazier, Academic Administrator, Oxford Centre for Hindu Studies; Julia Shaw, Institute of Archaeology, University College, London; Shailendra Bhandare, Ashmolean Museum, Oxford; Devangana Desai, Asiatic Society, Mumbai; and Vidula Jaiswal, Jnana Pravaha, Varanasi, former Professor, Banaras Hindu University This series, in association with the Oxford Centre for Hindu Studies, reflects on the complex relationship between religion and society through new perspectives and advances in archaeology. It looks at this critical interface to provide alternative understandings of communities, beliefs, cultural systems, sacred sites, ritual practices, food habits, dietary modifications, power and agents of political legitimisation. The books in the series underline the importance of archaeological evidence in the production of knowledge of the past. They also emphasise that a systematic study of religion requires engagement with a diverse range of sources such as inscriptions, iconography, numismatics and architectural remains. Books in this Series Women and Monastic Buddhism in Early South Asia Garima Kaushik Archaeology and Religion in Early Northwest India History, Theory, Practice Daniel Michon Negotiating Cultural Identity Landscapes in Early Medieval South Asian History Edited by Himanshu Prabha Ray For a full list of titles in this series, please visit www.routledge.com/ Archaeology-and-Religion-in-South-Asia/book-series/AR

Buddhism and Gandhara An Archaeology of Museum Collections Edited by Himanshu Prabha Ray

First published 2018 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 711 Third Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2018 selection and editorial matter, Himanshu Prabha Ray; individual chapters, the contributors The right of Himanshu Prabha Ray to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. Maps not to scale. The international boundaries, coastlines, denominations, and other information shown in any map in this work do not necessarily imply any judgement concerning the legal status of any territory or the endorsement or acceptance of such information. For current boundaries, readers may refer to the Survey of India maps. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record has been requested for this book ISBN: 978-1-138-89681-9 (hbk) ISBN: 978-1-351-25276-8 (ebk) Typeset in Sabon by Apex CoVantage, LLC

Contents

List of figuresvii List of tablesx Notes on contributorsxi Forewordxiv JENS-UWE HARTMANN

Acknowledgementsxvii Introduction

1

HIMANSHU PRABHA RAY

1 Greek or Indian? The Questions of Menander and onomastic patterns in early Gandhāra

33

STEFAN BAUMS

2 “Tis all here. A treasure locked”: unlocking the wonder house of the Chinese Buddhist travelogues

47

MAX DEEG

3 Numismatics of ‘the Other’: investigating coinage and ‘Greekness’ at Taxila

70

SHAILENDRA BHANDARE

4 Region through text: representation of Gandhāra in the Mahābhārata TANNI MOITRA

104

vi  Contents 5 Charles Masson: a footloose antiquarian in Afghanistan and the building up of numismatic collections in the museums in India and England

130

SANJAY GARG

6 The collection of Gandharan art in the residence of the Malakand Political Agent, Khyber Pakhtunkhwa Province, Pakistan

154

PIA BRANCACCIO

7 Vajīrasthāna/Bazira and beyond: foundation and current status of the archaeological work in Swat

173

LUCA M. OLIVIERI

8 The beginning and development of Gandhāran collections in German public museums

213

BRITTA SCHNEIDER

9 Decoding Gandharan art: the making of museum collections in India

232

HIMANSHU PRABHA RAY

Index261

Figures

0.1

Boddhisattva from Shahbazgarhi collected by Alfred Foucher from the Afghanistan-Pakistan frontier in 1896–1897 and now in the Musée Guimet, Paris 4 0.2 Sites mentioned in the Periplus Maris Erythraei12 0.3 Wheel-shaped stupa excavated at Sanghol SGL-11, Punjab 16 0.4 Wheel-shaped stupa at SGL-5 unearthed during exavations at Sanghol 17 3.1 Punch-marked coin with ‘Taxila’ mark on reverse 72 3.2 Uniface die-struck copper coin attributed to Taxila 73 3.3 An ‘elephant and lion’ coin of Taxila 73 3.4 ‘Taxila’ coin with arched hill flanked by Brahmi letter ‘Go’ 92 3.5 ‘Taxila’ coin with lion and swastika; the reverse is blank 93 3.6 Indo-Greek copper coins of Apollodotos I counterstruck by ‘Taxila’-type devices 94 3.7 Indo-Greek copper coins of Apollodotos I counterstruck by ‘Taxila’-type devices 95 3.8 ‘Lion × elephant’ type coin, counterstruck on Apollodotos I 95 3.9 ‘Taxila’ coin with elaborate lotus standard 97 3.10 ‘Taxila’ coin with symbolic configurations on both sides; the stylized ‘lotus standard’ symbol is seen besides the arched hill on the obverse 97 3.11 ‘Elephant and horse’ coin with stylized lotus standard in front of the elephant 98 6.1 View of the garden with sculptures set in the wall, Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan 155

viii  Figures 6.2

View of the Malakand Pass, Khyber Pakhtunkhwa Province, Pakistan 6.3 Sculpture of a Bodhisattva, Gandharan art (1st–3rd c. CE), Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan 6.4 Sundial built with reused Gandharan sculptures, Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan 6.5 Architectural element from a Buddhist stupa, Gandharan art (1st–3rd c. CE), Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan 6.6 Fragment of a relief depicting the Buddha’s Parinirvana, Gandharan art (1st–3rd c. CE), Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan 6.7 Headless sculpture of a seated Buddha, Gandharan art (1st–3rd c. CE), Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan 6.8 View of Palai, Shahkot Pass, Khyber Pakhtunkhwa Province, Pakistan 6.9 Seated Buddhas from Malakand (MK 8, MK 9, MK no number), Gandharan Art (1st–3rd c. CE), Swat Museum, Saidu Sharif, Khyber Pakhtunkhwa Province, Pakistan 6.10 Close-up of a Sundial built with reused Gandharan sculptures viewed from above, Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan 7.1 The new Swat Museum: front view 7.2 Map of Swat 7.3 Trench BKG 11 seen from the northwest 7.4a BKG 11/4–5: orthophoto 7.4b BKG 11/4–5: plan with residential units 7.5 BKG 11, Sacred Building B: detail of the niche of the cultural court 7.6 BKG 11, Units B and D: axonometric restitution 7.7 Temple B: the small stele of Hariti BKG 3636 (w. 16.2 cm, h. 24.4 cm, t. 5.1 cm) 7.8 BKG 11, Unit D: the Buddhist shrine

157

158 159

160

160

161 167

168

169 175 177 181 190 192 192 194 195

Figures ix 7.9

BKG 11, Unit D: the non-Buddhist miniature stele BKG 2304 (h. 17.2 cm, w. 8.5 cm, t. 2.8 cm) 7.10 BKG 11, Temple K 7.11 BKG 11, Temple K: axonometric restitution 7.12 Gumbat-Balo Kale: a view from the west 7.13 Gumbat-Balo Kale: restitution of the terrace seen from the east 7.14 Amluk-dara: the stupa terrace 7.15 Amluk-dara: axonometric restitution of the stupa terrace 7.16a Amluk-dara: panel AKD 95 in its stratigraphic context 7.16b Amluk-dara: panel AKD 95: the Birth of Siddhartha (w. 36.5 cm, h. 19.0 cm; t. 4.5 cm) 8.1 Examples of Greek art in India 8.2 Gandhāra exhibition (1926), room III towards room II, Museum für Völkerkunde, Berlin 8.3 Gandhāra exhibition (1926), room II, wall 2, Museum für Völkerkunde, Berlin 9.1 Sign in South Asia Gallery Asian Art Museum, San Francisco 9.2 Indian Museum Kolkata no. G16 A23286 9.3 Indian Museum Kolkata no. 2377 A 23244 Rishyasringa Jataka from Yusufzai 9.4 Heras Institute collection no. 30.37 9.5 Heras Institute collection no. 30.43 9.6 The imposing building that houses the collections of Chhatrapti Shivaji Maharaj Vastu Sangrahalaya, Mumbai 9.7 CSMVS departure of prince Siddhartha 9.8 CSMVS Buddha image from Mirpur Khas 9.9 Stucco Buddha head from Sanghol in the site museum 9.10 Railing pillar from Sanghol in the site museum

196 197 197 199 199 200 200 201 202 216 222 223 233 238 238 241 242 243 245 246 249 250

Tables

1.1 1.2

Greek names in Gāndhārī inscriptions Family relationships involving bearers of Greek names in Gāndhārī inscriptions (f = father, m = mother, w = wife, s = son, u = uncle, gm = grandmother, gs = grandson, gu = great-uncle) 5.1 Charles Masson (1800–1853): a career graph 7.1–7.2 Concordance table between BKG and other excavated sites in Swat 9.1 Table of eighteen stone sculptures in the Heras Institute Museum

38

40 132 184 240

Contributors

Stefan Baums is at the Institut für Indologie und Tibetologie, LudwigMaximilians-Universität München, Germany. He was a Shinjo Ito Postdoctoral Fellow at UC Berkeley during the 2010–2011 academic year. He studied Indology, Tibetology and Linguistics at the Georg-August-Universität Göttingen, Germany; Sanskrit, Nepali and Buddhist Studies at the School of Oriental and African Studies, University of London, UK; and South Asian and Buddhist Studies at the University of Washington, USA. He finished his Doctorate from the University of Washington, USA, in 2009 for the edition and study of a first-century Gāndhārī birch-bark manuscript containing a commentary on a selection of early Buddhist verses. His research interests include Buddhist philology and epigraphy, the beginnings of written Buddhist literature, the interaction of written and oral modes of text transmission, the development of Buddhist hermeneutics and the description of Gāndhārī language and literature. Shailendra Bhandare is Assistant Keeper, Numismatics, Ashmolean Museum, Oxford University, UK. His publications include: ‘Numismatics and History: The Maurya-Gupta Interlude in the Gangetic Plain’ in Between the Empires: Society in India 300 BCE to 400 CE, 2006; ‘Not Just a Pretty Face: Interpretations of Alexander’s Numismatic Imagery in the Hellenic East’ in Memory as History: The Legacy of Alexander in Asia, 2007; and ‘From Kautilya to Kosambi and Beyond: The Quest for a Mauryan/Asokan Coinage,’ in Reimagining Aśoka: Memory and History, 2012. His research interests include Indian coinage: its historical context, utility and evolution with specific attention to the early historic, medieval (Mughal) and modern periods; and the applicability of Indian numismatics for various Indological themes such as art, iconography, epigraphy and archaeology.

xii  Contributors Pia Brancaccio is Associate Professor and Co-director of the Art History Program at Drexel University, Philadelphia, USA. She completed her doctorate in Indian Art History and Archaeology at the Universita’ degli Studi di Napoli ‘L’Orientale’, Italy. She also held research positions at the Metropolitan Museum of Art, Getty Research Institute and the Philadelphia Museum of Art. She has done extensive work on Buddhist art in ancient South Asia with a special focus on the Deccan plateau and the northwest of the Indian subcontinent. Her publications include a monograph on The Buddhist Caves at Aurangabad: Transformations in Art and Religion (2010), two edited volumes entitled Living Rock: Buddhist, Hindu and Jain Cave Temples in Western Deccan (2013) and Gandharan Buddhism: Archaeology, Art and Text with Kurt Behrendt (2006), as well as several articles in conference proceedings and academic journals (Ars Orientalis, Archives of Asian Art, East and West and South Asian Studies). Max Deeg is Professor in Buddhist and Religious Studies, School of History, Archaeology and Religion, Cardiff University, Wales, UK. His areas of research include the history of South Asian and East Asian Buddhism, the role and function of narratives in Asian religious traditions and the reception history (Rezeptionsgeschichte) of Religious Studies and the study of Asian religions. Some of his publications are a (German) translation and study of the GaosengFaxian-zhuan (Record of the Eminent Monk Faxian), a (German) translation of the Lotus Sutra and various articles and essays on the history of Buddhism in South Asia and East Asia and other Asian religions (Manichaeism, Daoism, Brahmanism). Sanjay Garg is a numismatist and economic historian of international repute and has extensive research experience. His academic contributions range from economic and monetary history and currency and coinage of South Asia to architecture, literature and archival studies and have been widely published in print and multimedia which include twenty-three books and over seventy-six research papers. He has been serving the National Archives of India since 1988 and is presently holding the post of Deputy Director. He also served as head of the Research Division of the SAARC Cultural Centre, Colombo, Sri Lanka, where he was also the Founder and Executive Editor of the SAARC Culture (annual research journal of the SAARC Cultural Centre). He is presently also the Editor of the Numismatic Digest (annual research journal of the Indian Institute of Research in Numismatic Studies, Nashik).

Contributors xiii Tanni Moitra is a visiting post-doctoral fellow at the Graduate School of Distant Worlds (GSDW), Ludwig Maximilian University, Munich, Germany. She submitted her doctoral thesis titled Extraordinary Circumstances: A Study of Āpad and Āpaddharma in the Early Indian Literary Traditions (c. 6th Century BCE to 3rd Century CE) in 2016 at the Centre for Historical Studies, Jawaharlal Nehru University, New Delhi, India. Luca Maria Olivieri is Director of the Italian Archaeological Mission in Pakistan (ISMEO). He was Director of the ACT-Field School Project in Swat from 2011 to 2016. He has been working in Swat since 1987, where he has mainly focused on the still on-going excavation project at the urban site of Bir-kot-ghwandai (Barikot), on rock-art field-research and on the archaeological map of the Swat Valley project. He has authored almost a hundred publications, including six monographs, on NW Pakistan archaeology, rock art, heritage management and archaeological methodology. His most recent work is the archival repertory/study on the history of archaeological research in Swat from 1895 to 1937, which includes several unpublished autograph letters by Sir Aurel Stein. Britta Schneider studied History of Indian Art, Indian Philology and Classical Archaeology at Freie Universität, Berlin, Germany. From there she received her master’s degree in 1994 with a thesis on “The Reclining Woman with Child in Indian Stone Sculpture”. Afterwards she trained as museum assistant at the Museum für Indische Kunst, Berlin, and worked as a guide for the Museumspädagogischer Dienst. Since 2012 she has been employed as a research assistant for the project “Buddhist Manuscripts from Gandhāra.” Schneider published brief communications on Indian games and in this regard held workshops for children and adults in Berlin and Munich. Together with Ingo Strauch she is co-editor of Hariśyenalekhapañcāṣikā, a felicitation volume in honour of Harry Falk, which was published in 2013.

Foreword

In 2013 Himanshu Prabha Ray received the prestigious Anneliese Maier Prize awarded by the Alexander von Humboldt Foundation in Germany. The main objective of this prize is fostering research collaboration between the laureate and German colleagues in related subjects. As the focus of her research connected with the award, Ray chose the theme “Cross-Cultural Dialogue: India and the Wider World in Ancient History” and decided to associate this project with the “Distant Worlds” Graduate Studies Programme at her host institution, the Ludwig Maximilian University in Munich. This programme offered itself as a suitable partner because it unites all subjects and institutions in Munich dedicated to research on ancient civilizations – from Latin Studies to Sinology, from Prehistory to the various archaeologies dealing with the ancient world, from the university to the relevant museums, from academic libraries to the Bavarian Academy of Sciences and Humanities. When it came to selecting a topic that would immediately bring Himanshu Prabha Ray into a fruitful interdisciplinary dialogue with a significant number of members of the Graduate School, Gandhāra was a natural choice. In the study of antiquity there is no other region related to India that allows – and invites – scholars of an equally wide variety of subjects to contribute to a better understanding of historical developments and processes of exchange and amalgamation, be they political, economic, religious or art historical, to name but a few. Until the end of the twentieth century, the term Gandhāra was mainly used for describing a certain art historical development that was characterized by the employment of Graeco-Roman models for expressing Indian Buddhist contents. This led to a unique fusion and resulted in works of art that must have appeared highly attractive to the eyes of early Western spectators who were conversant with the classical art of Greece and Rome, because such objects looked at the

Foreword xv same time exotic and yet surprisingly familiar. The evident relation to Greek art, at that time still viewed as the benchmark of artistic expression, procured Gandhāran art a well-disposed reception from the very beginning of its discovery. This mode of reception continues into the twenty-first century, and the fact that an equally well-esteemed religion provided the contents for these artistic expressions surely reinforces it. The peacefully meditating Buddha in his Graeco-Roman costume – a better combination is difficult to imagine. Indeed Gandhāra was a region that had witnessed and absorbed, in the wake of Alexander’s campaign, a significant amount of Greek influence. Viewed from the supposed centres of the Indian Buddhist world, however, it seemed to be located rather at the margin. This perspective has undergone a dramatic change in the last two decades. Recent manuscript finds reveal that the area was a cultural centre of its own with an enormous radiation, something like a hub – if this modern expression is appropriate here – that connected India with Central Asia and eventually with Rome and China, two of the superpowers at that time. Script appears to be an amazingly late invention in India, used only for epigraphical purposes before the first century BCE. According to radiocarbon datings, the earliest among the recently found manuscripts date to that century; if correct, this is a sensational finding. All the present-day evidence suggests that the Buddhists were among first – if not the very first – to employ the innovative medium of script for the preservation of a religious tradition. Gandhāra must have been a stronghold in this process, because all the manuscripts are local products. As in the case of the slightly later development in art, the form of a foreign model was copied for the expression of exclusively Buddhist contents. This time, however, it was the format and Aramaic script of the scrolls of the Achaemenid Empire that supplied the model. Art and literature needed a tremendous amount of political and economic support in order to flourish, and the obvious availability of both factors underlines once more the central importance of the area. In 2012 the Bavarian Academy of Sciences and Humanities in Munich initiated a long-term research project on the manuscripts from Gandhāra. This project serves as another institutional partner for Ray’s research interest. In order to provide a venue for dialogue, she envisaged a series of colloquia with the aim of both revisiting earlier research questions and generating new perspectives. So far, there have been two conferences. The first took place on March 27 and 28 of 2014 in Munich under the theme of “Rethinking the Greeks in Gandhāra”; the second was held in Delhi on April 16 and 17 of 2015

xvi  Jens-Uwe Hartmann in collaboration with the India International Centre and the National Museum Institute, both in New Delhi, and dealt with “The Making of Museum Collections, With Special Reference to Gandhāra”. The present book unites selected contributions to both conferences and arranges them under a common perspective. Gandhāra is a historical reality and at the same time a modern construct. The latter owes its genesis mostly to art and archaeology and their presentation in exhibitions and museum collections; thus, the various manifestations of the construct also reflect the dynamics of contemporary reception. Focusing on the beginnings of archaeological investigations in Gandhāra and the integration of the findings into museum collections, the book explores not only the creation of the construct but also its present-day repercussions in Asia and the West. Jens-Uwe Hartmann Professor of Indology, University of Munich, Germany

Acknowledgements

A book of this nature would not have been possible without the academic and financial support extended to me by Alexander von Humboldt Foundation of Germany. I am grateful to Professor Monika Zin for nominating me for the five-year Anneliese Maier Research Award in 2012 and to Professors Jens-Uwe Hartmann and Rolf Schneider for agreeing to host the research project in the Distant World Graduate Studies Programme of Ludwig Maximilian University. In addition, Professor Hartmann has helped guide the project in a variety of ways, including writing a preface to this volume. The two conferences organised in Munich in March 2014 and in Delhi in April 2015 were attended by scholars and researchers from several disciplines, though they were not able to contribute to the volume. Nevertheless the animated discussion that ensued has found its way into many of the chapters. I am thankful to Professors Rolf Schneider and Monika Zin, LMU; Grant Parker, Stanford University; Rachel Mairs, Reading University; Michael Alram, Vienna; Juhyung Rhi, Seoul National University; Anupa Pande, National Museum Institute; Najaf Haider, Jawaharlal Nehru University; and Daniel Michon, Claremont College. The two conferences also provided a platform to several researchers who enriched the discussions, and I would like to mention Kasper Evers, Copenhagen University; Lauren Morris, LMU; Rachel Poser, Harvard University; Lubna Sen, Abhijit Kumar; and Ujjwal Ankur from National Museum Institute and Kanika Singh, Ashoka University. India International Centre (IIC) and the National Museum Institute collaborated with me for the Delhi conference, and I would especially like to acknowledge warmly the support provided by Ms. Premola Ghose of IIC. Finally, I owe a debt to all the contributors to this volume who have invested time and energy in revising their presentations and making them available for publication, and also to the anonymous referees for

xviii  Acknowledgements their encouragement. Without the persistent support of the Routledge team, Shashank Shekhar Sinha, Aakash Chakrabarty and Rimina Mohapatra, the chapters would not have received their present elegant form in this book. It is hoped that the reader will find much of interest relating to ancient Gandhāra and also its modern construction in this volume.

Introduction Himanshu Prabha Ray

The earliest archaeological excavation in the northwestern part of the subcontinent was undertaken in 1830 at Manikyala near Rawalpindi.1 The stupa had been discovered by M.E. Elphinstone (1779–1859) in 1808 while on a diplomatic mission to Shah Shuja at Peshawar. Manikyala was an extensive Buddhist site2 located on the Grand Trunk Road in the present-day Pakistan by Jean-Baptiste Ventura (1792/3– 1858), one of the French officers in the employ of the Punjab court of Maharaja Ranjit Singh (1780–1839). Ventura decided to spend his money and time in opening the Buddhist stupa at Manikyala, which local tradition regarded as the resting place of Sikandar, or Alexander’s, horse. After his excavations, Ventura informed Ranjit Singh in a short note in Persian that the resting place of Sikandar’s horse had been discovered,3 when in fact what he found were reliquaries and relic deposits spaced at regular intervals in the central core of the stupa.4 Ventura also found perhaps the first copper coin with a Buddha image in the Manikyala stupa and since then the number of such coins has grown.5 This incident is characteristic of the diverse cultural traditions that intersected in Gandhara, the ancient name of a region in northwest Pakistan bounded on the west by the Hindu Kush mountain range and to the north by the foothills of the Himalayas. Gandhara is also the term given to this region’s sculptural and architectural features between the first and sixth centuries CE. In ancient Gandhara Buddhist religious centres were built outside the urban centres and were composed of a sacred area for public worship and a private monastic section with viharas and small devotional structures. These served the needs of at least three distinct communities: lay followers, resident monks and local and long-distance pilgrims.6 References in Greek literature to Alexander’s invasion of the East moulded European perception of the region, as the British tried to devise military strategies to establish their rule in the Punjab and

2  Himanshu Prabha Ray Afghanistan and to comprehend the complex geography and history of the northwestern part of the subcontinent in the nineteenth century.7 In contrast local knowledge associated many of the abandoned Buddhist stupa sites with Sikandar, the hero of the Persian epic poem the Shahnameh (Book of Kings) of Ferdowsi (935–1020).8 Finally the former held sway. The nature of imperial discourse current in Britain from 1860 to 1930, the period when British imperialism was at its height, affected the way in which images from Greek and Roman archaeology were invoked in academic literature. In turn, these influenced writings by popular authors, which sustained this discourse and moulded European attitudes towards the past.9 The chapters in this volume critically analyse this construction of the region of ‘Gandhara’ in the nineteenth and twentieth centuries by drawing attention to its descriptions in other sources, such as the Greek, Chinese, Sanskrit and numismatic discoveries, as well as its interlinkages with other regions. A second thrust is on the process of collecting and the extent to which early collections of sculptures from the region not only fuelled interest in the Graeco-Buddhist Gandhara School of Art, but also created it as a separate field of study. For example, the 1883 Catalogue of the Indian Museum in Calcutta by John Anderson divides the collections under the following sections: Asoka, Indo-Scythian, Gupta and Mahomedan, with the Indo-Scythian Gallery displaying sculptures from Mathura in central India and Amaravati in Andhra as well.10 In contrast, the 1930 publication on the Peshawar Museum has an additional chapter on the history and art of Gandhara, which proposes a separate and autonomous status for the art of the region as compared to other parts of the subcontinent.11 I start the discussion on these issues by tracing the beginnings of collections from what later came to be defined as the region of Gandhara.

Museum collections and the discourse on the Buddha image In 1849 the governor-general of India formally annexed the Punjab and Sikh territories and the East India Company initiated attempts at collecting antiquities from their newly acquired territories. In 1860s a makeshift museum was established in Peshawar, and in 1867 the new Lahore Museum was established to house collections from sites in the northwestern part of the subcontinent. In 1876 the sculpture of a seated Buddha with flaming shoulders was found near the modern village of Beni Hissar and donated to the Asiatic Society of Bengal and the same year transferred to the India Museum. It was the first

Introduction 3 Gandharan sculpture seen by Europeans in India.12 Regular circulars were sent out by the Punjab government exhorting officers to obtain contributions for the Lahore Central Museum. This resulted in a series of substantial collections from the Yusufzai sites in the period between 1868 and 1871, “all made with the sole view to acquiring sculpture and with little regard to the sites, or frequently even to recording the precise find-spots of the pieces themselves”.13 Both Takht-i-Bahi and Sahri Bahlol seem to have been likely sources of sculptures acquired by the Lahore Museum in 1868. From the early 1870s, Gottlieb Wilhelm Leitner (1840–1899), principal of the Lahore Government College, took an active interest in the art of Gandhara. In 1873 he was sent as the official Punjab government representative to the Vienna International Exhibition where sculptures from Gandhara were displayed. In his catalogue, Leitner described these as Graeco-Buddhistic and Indo-Bactrian. The Musée Guimet was established in Paris in 1889 to house the collections of Emile Guimet (1836–1918), son of a prominent industrialist from Lyon. The Gandhara collection of the museum was the outcome of missions by Alfred Foucher (1865–1952), a French scholar to the Peshawar district. Foucher brought back some 100 pieces, which formed the core of the collection. In May 1923 the French signed an agreement with the Amir of Afghanistan that gave them a monopoly on archaeological work in the country.14 Speaking at a lecture in the Musée Guimet in Paris, Alfred Foucher made explicit the link between museum collections and the making of the art historical discourse on the Buddha image (Figure 0.1). One of the advantages of the Musée Guimet most appreciated by its orientalist lecturers is that they are free to dispense with the oratorical precautions which they must take everywhere elsewhere . . . You could not but be aware of the fact that on crossing the threshold of this Museum you would immediately find yourself transported from Europe to Asia, and you would hardly expect me to apologize for speaking of Buddha in the home of Buddha.15 Images excavated from archaeological sites in the northwestern part of the subcontinent and housed in museums in Europe became the basis of Foucher’s thesis on the Greek origin of the Buddha image. To illustrate his views, Foucher used images from Hoti Mardan in Yusufzai, as well as the Buddha figures from Lahore Museum, which he dated to the centuries prior to the Common Era. Foucher described

4  Himanshu Prabha Ray

Figure 0.1 Boddhisattva from Shahbazgarhi collected by Alfred Foucher from the Afghanistan-Pakistan frontier in 1896–1897 and now in the Musée Guimet, Paris Source: All images in the chapter are from the personal collection of the author

the region of Gandhara as an undulating plain dotted by cities on trade routes to the Ganga Valley to the east and to the Khyber Pass in the west. Foucher then goes on to list miracles performed by the Buddha in the region, which resulted in the region acquiring sanctity of its own, especially with pilgrims from China. Coins found in stupa deposits in Gandhara showing the image of the Buddha with the Greek legend ‘Boddo’ provided further fodder to sustain Foucher’s claim.16 Joe Cribb has examined 6 Buddha-image gold coins and 109 copper coins, though more are known in existence.17 These coins were issued

Introduction 5 late in the reign of Kanishka I, but were not special issues and instead should be located firmly in the Kushan monetary system as they represented about 1.35% of Kanishka’s coinage. The coins represent three types of Buddha images accompanied by labels in Bactrian written with Greek letters identifying them as the Buddha on the gold coins and Maitreya on the copper coins.18 Thus an important link is evident between nineteenth-century archaeological work, museum collections and interpretations by art historians based on these collections. Of all the sculptural remains collected in museums, perhaps the one that attracted the most attention was the image of the Buddha himself. John and Susan Huntington have argued that Foucher’s essay started a chain of several unfounded claims with reference to early Buddhist art, one of these being the assumption that there was reticence against including figurative images of Śākyamuni Buddha in pre-Gandhara art, which Gandhara sculptors help overcome.19 This is further supported by Susan Huntington’s studies showing that “modern Western authors created categories and hierarchies of art whereby the Indic art was judged to be inferior compared with what were held to be the higher aesthetic and communicative standards of the European tradition” – aniconism being one of them.20 It has also been convincingly shown that relics were the object of veneration in early Buddhism, with a stupa containing a relic being accorded higher ritual status and that this was accepted by all nika¯ya (the term is often wrongly translated as ‘sects’). Within this framework, images of the Buddha had secondary status as compared to the relic stupa. In the majority of Gandharan monasteries in the Peshawar valley, for example, Buddha images were usually placed in a series of multiple chapels surrounding the court of the main stupa.21 Juhyung Rhi has shown that one possibility of increasing the value of a Buddha image was by installing a relic in it and that this was a practice followed in Gandhara as evident from an examination of images. Occasionally a hole has been detected in the uṣṇīśa of stone Buddha images, and it is suggested that this functioned as a receptacle for the relics. Combined together, a relic and an image would have become a more efficacious means to communicate with the worshiper: an image could be enlivened by a relic, and a relic could take a more concrete communicable form through an image.22 Clearly images and art objects from Gandhara cannot be discussed in isolation and need to factor in developments in other parts of the subcontinent.

6  Himanshu Prabha Ray

Defining the region The region of Gandhara, along with that of Kamboja, is mentioned as one of the sixteen mahājanapadas, or kingdoms, in early Buddhist literature located on the uttarāpatha, or northern route, though a precise definition of the region is not provided in the texts. Gandhara also occurs as one of the rich satrapies of the Achaemenid Empire (550–330 BCE). Art historians23 define the region based on the style of art that developed and make a distinction between the core area of the art that flourished in the Peshawar valley, bounded to the north and west by foothills, to the east by the Indus River and to the south by flatlands, though in the nineteenth century it was used for a much larger area encompassing eastern Afghanistan, Punjab, the Swat valley and Kashmir. More recently the term Greater Gandhara has been coined that covers neighbouring regions such as Swat and other river valleys to the north, the area around Taxila to the east and the eastern edge of Afghanistan to the west.24 The term nagara, or town, first occurs in Sanskrit literature around the middle of the first millennium BCE, and the grammarian Panini (generally ascribed to the fifth century BCE) describes towns in the vicinity of Taksasila, or Taxila. Taxila is also one of the most extensively excavated historical sites in South Asia. Cunningham visited it and reported on its widespread ruins extending about 5 km from north to south and 4 km east to west. John Marshall excavated it from 1913 to 1934, lured as he was by its Greek association.25 Mortimer Wheeler selected Taxila as the site of his Training School of Archaeology, because most of all, “it lies at the foot of the Himalaya, in a terrain sufficiently reminiscent of Greece”.26 In addition he also did one season’s work there in 1944–1945. In the last several decades, the Pakistan Department of Archaeology has undertaken archaeological excavations at the site. The beginnings of settlement in the Taxila Valley date to the fourth millennium BCE when Neolithic villages were first located in the area. The importance of Taxila Valley lay in its strategic location on north–south and east–west routes. Settlements such as at Sarai Khola and Hathial, as well as numerous archaeological sites dated to the late second to first millennium BCE, termed Gandhara Grave Culture, indicate that urban development in the Taxila Valley was a continuous process.27 By the sixth century BCE (i.e. the time of the Achaemenids), the first fortification of Taxila is in evidence and habitation spread on both banks of the Tamra rivulet. Excavations at Charsada have confirmed its pre-Achaemenid beginnings

Introduction 7 and its prominence as a major urban centre. Archaeological surveys indicate a hierarchy of sites with sixty-one sites being less than 5 hectares. It is also evident that regions with a high concentration of Buddhist and Early Historic sites continue to be the centres of dense populations on account of the fertility of the soil and water resources.28 The major expansion at Taxila, as indicated by the settlement at Sirkap, which John Marshall termed the Greek city, is dated from the first century BCE to the first half of the second century CE. A reappraisal of the archaeological data indicates that domestic structures at Sirkap combined both residences and workshop spaces. The distribution of private and public religious loci is complex and occurs throughout the city, and although some of the religious structures were Buddhist, others were not.29 In contrast to sites further north of the Indus, the presence of Northern Black Polished Ware and silver bar punch-marked coins from Taxila indicates connections with the Ganga Valley material culture from the beginning of settlement at the site.30 It is equally significant that in addition to its links with Ganga Valley sites, the archaeological evidence indicates interaction between Taxila and sites in central India, such as at Adam, district Nagpur. An inscription from Mathura recording a gift of a pillar by a monk named Jivaka who was originally from Uddiyana or the Swat valley provides another example of mobility around 300 CE. In the year 77, in the fourth month of summer, on the fourth day in the monastery of Maharaja Rajatiraja Devaputra Huviska, the gift of the monk Jivaka, the Odiyanaka (native of Uddiyana), consisting in pillar base 25. May welfare and happiness of all sentient beings prevail; To the community of the four quarters.31 A report published in the 19 March 2008, edition of Dawn newspaper from Islamabad featured the article a ‘Rare Mathura Style Buddha Sculpture Found in Pakistan’. The red sandstone sculpture from Mathura depicts the Buddha seated cross-legged on a throne with both soles of the feet marked with dharmachakra symbols.32 It was found by a team from the Taxila Institute of Asian Civilizations, Quaid-i-Azam University, Islamabad, at the site of Badalpur situated about 9 km northwest of Taxila Museum. Another Mathura-style sculpture of the Buddha was discovered earlier from the surface of the Bhari Dheri site in Taxila Valley. The excavators suggested that the images were perhaps donated to the monasteries at Taxila and Badalpur by pilgrims travelling from north India. Lohuizen-de Leeuw has also discussed the

8  Himanshu Prabha Ray widespread finds of Buddha images and the cultural interactions that these indicated between Taxila and Mathura.33 Ksatrapas and Mahaksatrapas reigning in Mathura extended Saka administration and religious patronage from the northwestern borderlands to an important cultural and commercial centre in the Ganga Yamuna doab by the late first century BCE. Kharoṣṭhī inscriptions on a lion capital from Mathura provide keys to understanding links between various Saka families that ruled Taxila and Mathura.34 The inscription commemorates the installation of Buddhist relics and donation of a stupa and monastery to a Sarvastivadin monastic community by the chief queen of Mahaksatrapa Rajula. The Mathura lion capital inscriptions thus indicate kinship connections between the family of Mathura Ksatrapas and Saka lineages of the northwest.35 Thus, interactions between the region of Gandhara and other centres in the subcontinent are in evidence, raising issues of interactions, mobility and exchange.

Mobility across the Hindu Kush The Behistun inscription of King Darius I (519 BCE) includes Bactria in the list of twenty-three satrapies of his kingdom. In 515 BCE Darius extended Persian control southeast of the Indus, creating the new satrapy of Hindush, roughly corresponding to modern Sindh in southern Pakistan. The foundation inscription of Darius for Persepolis mentions gold from Bactria and teak from Gandhara. Indians served as mercenaries under Darius, and there are references to a colony of Indians in the fifth century BCE near the old Sumerian city of Nippur. At Persepolis itself peoples from twenty-three regions are mentioned and workers include Greeks, Indians, Egyptians, Babylonians, etc. An Indian woman, Busasa, is said to have maintained an inn at Kish. There are references to Persian kings maintaining several translators for comprehending the many languages of the regions within their empire. Herodotus, in the book Histories, names Gandhara as a source of tax collections for King Darius and to the logioi present at the Persian court, and there are references to Persians speaking Greek and vice versa.36 Well-maintained road networks connected Achaemenid centres in Iran to the provinces of Bactria, Gandhara and Sindh, and these regions regularly paid tribute to the Persian ruler. In contrast to the Persian inscriptions, the region of Gandhara figures prominently in early Sanskrit and Pali sources, as well as the Sanskrit epic, the Mahābhārata. The Sanskrit grammarian Panini, for example, is said to come from

Introduction 9 the town of Śalātura in the vicinity of Taxila (near modern Peshawar) and while discussing the peculiarities of the language of the northwest contrasts it with that prevalent in the east. Thus for him the distinction is between the language of the northwest and the east, whereas other grammarians like Yaska and Patanjali adopt a different perspective in their threefold division of northwest, central or Aryavarta and eastern language peculiarities. However, Panini does not see the northwest as the centre of the Sanskrit-speaking universe but adopts a neutral stance while highlighting the peculiarities of regional speech. In contrast, Patanjali writing in the second century BCE emphasizes the linguistic superiority of Aryavarta and refers to the northwest as inhabited by yavanas, Greeks and Sakas.37 Numismatic data indicate that trade networks across the Hindu Kush pre-date the Hellenistic Greeks in the region and that Taxila was already a major centre at that time, with evidence of a distinctive silver coinage. In Taxila and Gandhara the currency was based on oblong stamped silver bars weighing 1 satmana (100 rattis, circa 11.2 grammes) together with fractional denominations that had a circular outline and a slightly scyphate shape. A geometric symbol of fairly constant and characteristic form was typical and was stamped twice on the large bars and once on the fractional denominations. Coins of this series are known from a number of finds encompassing a large area from the margin of the Afghan plateau in the west to Taxila in the east.38 Another site where a pre-Mauryan coin was found is Mathura. The silver punch-marked coin had nine distinct punches with the central being a pentagon enclosing a sphere. The coin itself was concave or scyphate and weighed 6.7 gm. The compositional analysis showed that the coin was 97.17% silver with minor alloying of gold (0.715), copper (0.88%) and lead (1.19%), whereas the Taxila punch-marked coins contained a higher percentage of copper shown to be between 13% and 25%.39 In the context of Gujarat, the first-century CE text the Periplus Maris Erythraei refers to Syrastrene or Surashtra, Eirinon or the Rann of Kutch beyond which lies Barake, or the Gulf of Kutch (section 40). The region is described as “very fertile” and “in the area there are still preserved to this very day signs of Alexander’s expedition, ancient shrines and the foundations of encampments and huge wells (section 41).” Somewhat later, the author refers to “old drachmas engraved with the inscriptions, in Greek letters, of Apollodotus and Menander, rulers who came after Alexander” being found in the market of Barygaza (section 47).40 How does one explain the presence of coins with Greek writing in Gujarat in the early centuries CE? How

10  Himanshu Prabha Ray did Taxila and other centres in northwest India connect with those in Gujarat and western India or those in the Ganga Valley? This is an issue that is discussed in the next section as the chapter traces the wider linkages of the northwestern part of the subcontinent. Sailing down the Indus and early contacts with the Persian Gulf First-hand information on parts of the Indian Ocean became available to the Greeks after Alexander’s campaigns to the east. Nearchus was commissioned to sail along the Indus by Alexander. Another participant in the voyage was Onesicritus, though the relationship between the two is not quite clear. Alexander believed that the ocean was relatively close, because the ‘Indian Gulf’ (Arabian Sea) formed but one stretch of water with the Persian Gulf, and the Hyracanian Sea (Caspian Sea) with the Indian Gulf. From the Persian Gulf our fleet shall sail around to Libya, as far as the Pillars of Heracles (Straits of Gibraltar). (Anabasis Alexandri, V. 26.1–3) The boats for the voyage were those of the Greek type which had been transported by carts in sections, the shorter ones in two pieces and the triacontoroi in three (Anabasis Alexandri, V.8.5). Further details are available in Strabo (Geography, 16.1.11) who states, citing Aristobulus, that the boats were built in Phoenicia and Cyprus with bolts and could be taken apart. Others were made in Babylon from the cypress trees in the groves and the parks. There is no unanimity in the sources on the total strength of the fleet. Arrianus mentions 2,000 vessels, whereas Diodorus reduces the figure by half (i.e. to 1,000). In the Indica, Arrianus refers to a figure of 800, and these were manned by seafaring communities, such as the Phoenicians, Cypriots, Carians and Egyptians who served in the army.41 The account by Arrianus of the coastal voyage does not provide ethnographic details of the route travelled, nor does it contain any useful sailing instructions and glosses over local maritime activity encountered in the region. There are nevertheless several indications for the existence of earlier sailing traditions and maritime contacts. For example, at the fishing village called Mosarna, a pilot, a Gedrosian called Hydraces, sailed with the flotilla and guided it (Indica, 27.1–2). This reliance on local pilots for guidance across the Indian Ocean was a regular feature, and there are several references to the practice in

Introduction 11 Greek as well as later accounts. The Portuguese portolanos, for example, refer to the river Hab, the extreme western branch of the Indus delta, as “rio dos pilotos” – the name originating from the fact that it was here that the foreign vessels found fishermen capable of guiding them into the Indus delta.42 Arrianus refers to variations in stellar configurations as noticed during the journey: Of the stars they had seen hitherto in the sky, some were completely hidden, others showed themselves low down towards the earth; those which had never set before were now observed both setting and at once rising again. (Indica, 25.6) He discusses the natural phenomenon of shortening shadows, an observation that he could not have made himself on account of his limited experience along the Makran coast, but which he may have acquired from those who travelled farther south to Sri Lanka. Though trading activities are not referred to directly, the reference to the availability of cinnamon on the Arabian coast (Indica, 32.7) implies the existence of sea-borne trade between Arabia and India. This is further confirmed by a statement in Theophrastus (Enquiry into Plants, ix.7) that fragrant plants are partly from India, whence they are sent by sea. Within this background of early Greek sources, the Periplus Maris Erythraei provides the earliest evidence of collating information for the benefit of Greek merchants and traders on the route between Egypt, India and East Africa (Figure 0.2). It was written in koine Greek, a dialect easily understood by merchants and traders. Although the language of the text is coherent and correct, the style and grammar are mediocre. Unlike other writings of the period, it shows no interest in the history and literature of the region. What is significant for our purpose is that it contains graphic accounts of coasts to be avoided as well as the major landfalls. For example, to set a course along the coast of Arabia is altogether risky, since the region offers poor anchorage, is foul with rocky stretches, cannot be approached because of cliffs, and is fearsome in every respect. (section 20) No information is available on the author of the Periplus, nor are there any indications within the text for dating it. There have

12  Himanshu Prabha Ray

Figure 0.2 Sites mentioned in the Periplus Maris Erythraei

nevertheless been several attempts to date it either on the basis of historical events mentioned therein or through a comparative analysis with other texts.43 The Periplus also differs substantially from other contemporary literary works, such as the Historia Naturalis of Pliny, Strabo’s Geography (written and revised between 25 BCE and CE 23) and the Geography by Claudius Ptolemy written in CE 150. The Periplus Maris Erythraei discusses the difficult entry into the Gulf of Khambat, which led to Barygaza or Bharuch, for vessels coming from seaward, because the mouth of the river on which Barygaza stands is hard to find. It is low and not visible even from close quarters; besides there are shoals in the river and hence the mouth is difficult to navigate. To help vessels overcome these difficulties, local fishermen in the king’s service came out into the entrance of the gulf in local craft called trappaga and kotymba and guided incoming watercraft.44 The Periplus mentions Barabrike in the Indus delta as one of the sites on the west coast of India which imported goods such as silver plate, glass vessels and a little wine, all of which are represented at Taxila and Begram. In return there are references to the export of lapis lazuli, which came from Badakshan, and Chinese skins and yarn. Taxila was

Introduction 13 thus an active participant in the maritime exchange network. Excavations at Sirkap, Taxila, yielded a number of western objects such as a bronze statuette of Harpokrates, repoussé silver emblema representing Silenus or Satyr, bronze saucepans with handles ending in rams’ heads, a silver spoon, at least two engraved gems, a small quantity of glass and the neck of an earthenware amphora. The cache from Begram comprised more than 150 glass objects, including one mosaic vessel and eight ribbed bowls produced either by casting or blowing. Some of these vessels such as the ribbed bowls, some of the plain brown pieces and the facet-cut vessels are generally attributed to the first to second centuries CE. The mosaic glass, the gilded glass and the cut glass with open-work decoration date somewhat later to the late second or early third century. There are others with which one cannot find parallels in the Roman Empire, though they are chemically similar to objects made in Egypt. In the Persian Gulf ed-Dur, a large archaeological site on the west coast of the Oman peninsula has provided crucial data for this maritime connection.45 The site has been identified with Omana mentioned in the Periplus as exporting pearls, purple clothing, wine, dates, gold and slaves to India.46 Ed-Dur was on the shipment route to Charax at the head of the Persian Gulf and from there via the overland route to Palmyra.47 Further support for this hypothesis comes from the finds of fifteen Characene coins at ed-Dur and the fact that 40% of the diagnostic pottery at the site has Characene analogies. The site dates to the first century CE, and 147 objects of glass recovered from archaeological excavations reached the site between 25 BCE and CE 75. The numerous excavation seasons indicate that only a limited number of beach-rock buildings existed there, and a majority of the living quarters consisted of palm frond dwellings. The dead were buried in individual stone graves, although large tombs with a shaft entrance for multiple burials existed as well and were found all over the site. In addition the sun god Shamash was worshipped in a temple as evident from an eight-line inscription in Aramaic on a stone basin found next to the temple. The site participated in an extensive network of trade and exchange as indicated by northeast Arabian coin finds from the site, as well as foreign issues. The latter group comprises four Roman coins (one of Augustus and three of Tiberius), eastern Mediterranean coins (three Nabataean coins of Aretas IV c. 9 BCE–CE 40, Gaza), southern Mesopotamia (eleven Characene coins), south Arabia (two coins from Hadhramawt), Persian (Parthian and Persis) and five Indian coins dated to the first century BCE to the first century CE.

14  Himanshu Prabha Ray Indian coins include copper karshapanas of the Ujjain type and those of Agnimitra, Abhiraka and Bhumaka.48 Chemical analyses have confirmed the Roman origin of the glass found at ed-Dur. Forty per cent of the samples were made by casting and 57% of the cast vessels were ribbed bowls, and 43% were made of mosaic glass. Of the 59% samples made by inflation, 70.1% were of blown glass. It is significant that a majority of the vessels were used for eating and 48.4% were flat or shallow vessels. Twenty-six per cent were small bottles, jars and boxes, perhaps for storing and/ or transporting cosmetic or medicinal preparations.49 This analysis of the shapes of the vessel challenges the parameters of the earlier debate centred on luxuries or prestige items required by the state. Further evidence of the Persian Gulf–Gujarat–Gandhara connection comes from finds of dedicatory and burial inscriptions written in Greek as well as Greek letters on pottery from Bahrain. These reinforce the identification of the island with Tylos, which was a stopping point for people as well as a homeland for others who understood and used the Greek language, even though their Semitic names suggest that they were not Greeks.50 This brief overview of the interconnections of what came to be defined as the region of Gandhara in the nineteenth and early twentieth centuries has underscored the fallacy of studying Gandhara as an isolated self-contained unit. Also significant is the marginalization of archaeological research in Gandhara after 1920. The discovery of third millennium BCE Harappan sites in the Indus valley in the 1920s transformed the antiquity of civilization in the region and shifted the focus of archaeological research to the protohistoric period. It is significant that after the partition of the subcontinent in 1947, foreign teams continued their quest for the Greeks and Gandhara proper west of the river Indus in Pakistan; much less attention has been paid to the archaeological context of the religious architecture and the Punjab plains,51 with the exception of work in the Swat valley by the Italian Archaeological Mission (Olivieri in this volume). Royal patronage and expansion of Buddhism Alexander Cunningham assigned a majority of sculptures that he collected from the region in the 1860s to the first two centuries of Kushan rule and thus established a link between the dynasty and art.52 The chronology of Gandharan sculptures has since been reworked. The excavations at Butkara I in Swat have yielded a fairly substantial group of reliefs with scenes from the life of Buddha that are now

Introduction 15 ascribed to the end of the first century BCE or the first half of the first century CE, both on the basis of archaeological evidence and on stylistic grounds. In some of the reliefs, the Buddha is only symbolically represented, thereby challenging Foucher’s focus on the image of the Buddha in Gandhara.53 Buddhism is said to have been introduced in the region by the Mauryan ruler Aśoka as evident from his rock edicts found at Shahbazgarhi and Mansehra dated to the third century BCE. Further corroborating evidence comes from the Milindapañha, or Questions of King Milinda, a text dated to the second century BCE that records a philosophical dialogue between the Indo-Greek king Menander and the Buddhist monk Nagasena. The figure of Menander who ruled from c. 155 to 130 BCE presents the duality of the Indo-Greeks. In the Buddhist tradition, Menander is said to have converted after a conversation with Nagasena, and the Pali text, the Milindpañha dated to 150 to 100 BCE, presents the dialogue between the two. In contrast coins of the ruler show him as a Hellenistic ruler with figures of Athena on the reverse with the legend ‘Menander, the saviour king’. The material record of Buddhism in the form of stupas, images and structural remains, however, appears from the first century BCE onwards. It is in the first two centuries of the Common Era that Buddhism proliferates, with patronage under the Kushans being credited for this expansion as will be discussed in the next section. It is also around this time that Buddhism expanded beyond the Indian subcontinent to establish a foothold in Iran and China.54 The first Kushan (CE 40–360) ruler who the Chinese writings credit with conquering lands south of the Hindu Kush as far as the northwest frontier region and Kashmir is identified on coins and the Rabatak inscription as Kujula Kadphises, and the beginnings of Kushan rule date to CE 40.55 The Rabatak inscription records that under Kanishka, Kushan rule extended along the Ganga Valley as far east as Patna to Campa, or modern Bhagalpur in Bihar. In c. CE 233, the Sasanians took control of Bactria and soon captured Gandhara from Vasishka (c. CE 246–67) and Taxila from Vasudeva II (c. 280–320). The muchdiminished Kushan rule continued into the fourth century CE, though the last few kings, such as Mahi, Shaka and Kipunadha, are known mainly from their coins. Did political unity create cultural unity? Can we identify homogeneity in material culture in north India in the first three centuries CE of Kushan rule? More than two dozen stupas are known to have been built on the wheel-shaped pattern in South Asia, and their distribution ranges from sites in Gandhara or northwest India to those in the upper Yamuna

16  Himanshu Prabha Ray basin, with the cluster being densest around the mouths of the Krishna and Godavari rivers.56 The dharmacakra pattern appears around the first century CE. The spokes of the wheel vary from eight at Shah-ji-kiDheri to sixteen at Dharmarajika at Taxila to eight at Sanghol 2 (SGL11), to eight plus eight at the Jain stupa of Kankali Tila, to twelve plus twenty-four plus thirty-two at Sanghol 1 (SGL-5) (Figures 0.3 and 0.4). Thus the wheel-shaped pattern is confined in north India to structures lying northwest of Mathura. Was the use of the dharmacakra pattern symbolic or a mere construction device? Snodgrass urges that the wheel-shaped stupa was symbolic.57 On the other hand, the introduction of the wheel-shaped plan for the larger stupas at Nagarjunakonda is explained as being motivated by considerations of structural stability and economy of building material, and several structural features determined the adoption of the plan. The wheel-shaped pattern was used mainly for the construction of large-sized stupas. It is suggested that the use of a wheel-shaped pattern may have been essential to withstand the pressure of building high drums, such as those depicted in Amaravati reliefs. This is further corroborated by the study of the Dharmarajika stupa at Taxila where spokes are observable at the top of the extant height of the stupa, which reaches 13.5 metres above the ground.58

Figure 0.3 Wheel-shaped stupa excavated at Sanghol SGL-11, Punjab

Introduction 17

Figure 0.4 Wheel-shaped stupa at SGL-5 unearthed during excavations at Sanghol

Kuwayama also suggests that builders in the northwest used a square podium to impart height to the stupa, as evident from the Dharmarajika example, a practice also adopted at Sanghol. Most stupas at Taxila were faced with kanjur ashlar, or chiselled blocks of local porous stone that required a covering of lime plaster and stucco. There was a distinction in building techniques, as evident in the Buddhist monuments from Swat and Yusufzai when compared to those from Taxila.59 These regional differences are important when assessing the stupas in other parts of the subcontinent. Contrary to received wisdom that credits the Kushans with the proliferation of the Mathura school of art at Mathura,60 recent research has shown that Mathura was the seat of a fully developed school of art and architecture from the second century BCE onwards (i.e. at least 250 years before the reign of Kanishka). Stone sculpture representing a variety of themes was produced and the multi-religious affiliation of these needs to be emphasized. These included anthropomorphic representations of Jinas, Yaksa and Yaksi images; ayagapattas; architectural fragments; railing pillars; and so on. The absence of surviving monumental architecture at Mathura dating to this early period has led to

18  Himanshu Prabha Ray the neglect of the art traditions of this early period, which continued unabated well into the sixth century CE.61 Clearly it is imperative to delink transformations in material culture from political changes. These results are further substantiated by the archaeological excavations at Sonkh, located 22 km southwest of Mathura city. Detailed documentation dates changes in the cultural deposit at Sonkh from the second century BCE to the second century CE and underscores the fallacy in attributing social and cultural transformation to political dynasties. This is an issue that we take up for discussion in the next section. Locating Gandhara in the multicultural milieu of early India A much-neglected theme that continues to be ignored is the multicultural milieu of Gandhara, where Buddhism was one of many religions practised. However, most of the sacred architecture that survives in the region is Buddhist in character, and this may not be a mere coincidence, as we have referred to the selective collections of sculptures based on their Hellenistic/Greek affinities. The Guide Book to Peshawar Museum, for example, refers to multi-armed figures of Shiva in its collection, which bear a close resemblance to those on the coins of Huvishka,62 and this raises the question of worship to the deity in early Punjab. The worship of Shiva in the form of linga was well established in the centuries just prior to the Common Era. Three cylindrical-shaped objects in a row on a pedestal have been represented on a variety of silver punch-marked coins of Imperial series (third century BCE), and the obverse of the square copper specimens of Taxila (third-second century BCE) bears a linga on a pedestal (indicated by a wavy line). Shiva is also shown on the Kuninda coinage of the late second century BCE. The obverse of the copper coins bears a standing deity, holding a trident with battle axe on the shaft in the right hand, with a star/crescent behind his shoulder. The depiction of images of Shiva continues, and the figure holding a trident and club in his hands which appears on the obverse of a square copper coin of Saka King Maues is identified as Shiva,63 because of a copper seal found at Sirkap (Taxila) where a figure similarly carrying a trident in his left hand and a club in the right hand bears a Brahmi inscription Shiva rakshitasa and Tridusa Vibhumitrasa in Kharoshti script of the first century CE; the bull below the left leg of the main figure is also significant.64 The coins of Kanishka I are of special importance because they depict a wide range of deities in varied iconographic forms, but Shiva

Introduction 19 continues to occupy an important position, and the coins show further advancement in the iconography of Shiva not known from contemporary sculptures. Shiva on the coins of Kanishka I always stands frontally in dvibhanga (two folds) with his face turned to right. He stands with one of his legs extended slightly forward to suggest mobility. He wears rudraksha armlets, bracelets and anklets of beads. Shiva, whether two-armed or four-armed, is always depicted one-faced. Trident and water vessel continued to be the most favoured attributes, with the water vessel shown in outstretched hands, usually the lower-right hand turned downwards so as to suggest the pouring act. Besides, the damru or drum, antelope, vajra or thunderbolt and ankusha or goad are also frequently represented. Did Shiva have a larger following and a broader base in the northwestern part of the subcontinent as compared to other deities, or was the use of the Shiva image a matter of personal choice for the Kushan rulers? We do know that Kanishka I was instrumental in bringing about several changes of crucial significance, such as the use of Bactrian instead of Greek both on his coins and inscriptions and the setting up of sanctuaries, one of them being at Surkh Kotal where images of Kushan kings were enshrined. The Rabatak inscription found not far from Surkh Kotal in 1993 is crucial for an understanding of several aspects of Kushan chronology, but more significantly for this chapter, Kushan notions of kingship as well. The inscription refers to Kanishka as having obtained kingship from Nana and from all the gods, but does not mention Oesho or Shiva who appears prominently on Kanishka’s coins. Lines 4–7 of the Rabatak inscription give a list of the chief cities of north India, which Kanishka controlled, and four of the five names can be identified as those of Saketa, Kausambi, Pataliputra and Champa. The major part of the inscription, however, concerns the foundation of a sanctuary, or bagolaggo (devakula), perhaps at Rabatak itself, which seems to have been an extensive site. Lines 9 and 10 name the divinities to be worshipped in the temple and include Umma in this list. Umma, or in Bactrian Ommo, appears on a coin of Huvishka along with Oesho or Shiva. An interlinear inscription written above the names of Zoroastrian deities contains names of Hindu gods, such as Mahasena and Vishakha.65 Another site that has provided valuable evidence for the presence of a Shiva temple is that of Kashmir Smast in the Mardan district of Pakistan. The natural caves in a limestone mountain were enlarged in several phases, and several Hindu caves have been found here with wooden architectural interiors. The settlement of the site can, like many others in the northwest, be traced back to the first through third

20  Himanshu Prabha Ray centuries CE and continued until the eighth to ninth centuries CE.66 Thus no simple answers account for the popularity of Shiva in Gandhara and the Punjab or the adoption of Shiva images on coins. It nevertheless raises the issue of contextualising Buddhist sculptures within a larger cultural milieu. Re-engaging with Gandhara: the Gandhari–Prakrit manuscripts The language of Gandhara has come to be known as Gāndhārī, one of the regional dialects of the Prakrit, which was written in the Kharoṣṭhī script. The earliest written records from the area are the multilingual rock and pillar inscriptions of Aśoka, mentioned earlier, employing the Kharoṣṭhī script, as well as Brāhmī and Greek. Some of the earliest of the second-century CE manuscripts are written in Gāndhārī using the Kharoṣṭhī script. This script was, however, gradually abandoned for Brāhmī, and the written language developed towards a standardized Sanskrit, with an intermediate stage of semi-colloquial Sanskrit that has come to be known as Buddhist Hybrid Sanskrit. The Greek script is represented by one leather manuscript written in Bactrian. In 1994 the British Library Oriental and India Office Collections acquired a collection of twenty-nine fragments of manuscripts written on birch bark scrolls in the Gāndhārī language (a dialect of Prakrit) and in the Kharoṣṭhī script. They were found inside a clay pot, also bearing an inscription in the same language, recording its dedication to the teachers of the Dharmaguptaka School, suggesting that the manuscripts belonged to the library of a monastery of this school. The Dharmaguptakas are known to have played a prominent role in the dissemination of Buddhism to central Asia and China.67 Although some of the texts, including the examples noted earlier, can be directly identified with or at least related to texts extant in other languages such as Pali, Sanskrit or Chinese, the majority of them, including many of the avadānas and the commentaries, appear to have no parallels in the previously known Buddhist literatures. Between 1993 and 1995, in the middle of the political destabilization caused by the Taliban and shortly after their destruction of the Bamiyan Buddhas, a treasure trove of Buddhist manuscript fragments on birch bark and palm leaf was found by inhabitants of the Bamiyan valley. The manuscripts provide valuable insights into the genres of literature that were important for the Buddhist community of the area. Numbering more than 5,000 pieces in total, the fragments entered the

Introduction 21 international antiquities market and were soon dispersed into the private collections of Martin Schøyen in Norway and of Ikuo Hirayama and Genshu Hayashidera in Japan. The manuscripts of the Schøyen collection are dated from the second to eighth century CE. Among the discovered manuscripts, the oldest stratum dates to the third century CE and includes the Mahayana texts Bhadrakalpikasūtra and Bodhisattvapiṭakasūtra, both in Gāndhārī written in Kharoṣṭhī script; the Aṣṭasāhasrikā Prajñāpāramitā in Buddhist Hybrid Sanskrit written in Kushan-Brahmi script; the Śrīmālādevīsiṃhanādanirdeśasūt ra and the Mahāvastu both in Sanskrit written in Gupta-Brahmi script; and the Larger Sukhāvatīvyūha in Gilgit-Bāmiyān type I script, which is a sixth- to seventh-century variation of the Gupta-Brahmi script.68 Two of the manuscripts contain various basic rules for the Buddhist monastic order. The texts are shown to belong to the MahāsāṃghikaLokottaravādin school, possibly a further indication of the regional origin of the collection as a whole, though this is an issue that needs further research. Outline of the book The foregoing discussion underscores the need to reassess the archaeological material ‘excavated’ in the nineteenth and early twentieth centuries. This edited volume has two objectives, as discussed at the beginning. The first section includes chapters that underscore the diverse cultural traditions of the region of Gandhara from a variety of sources and perspectives, such as Greek cultural appropriation (Stefan Baums), Chinese writings (Max Deeg), local coins (Shailendra Bhandare) and Sanskrit epics (i.e. the Mahābhārata) (Tanni Moitra). Baums discusses the patterns and mechanisms of Greek cultural appropriation in Gandhara through a survey of the personal names, titles and dynastic relationships in Gandharan primary sources (coins, seals, inscriptions, manuscripts) from the second century BCE to the third century CE and compares these to onomastic practices in other parts of the Hellenistic world. The chapter by Deeg considers the role and function of the socalled Chinese pilgrimage records in the reconstruction of Buddhist history in India. The chapter argues that these records were part of the nineteenth- and early twentieth-century activities of collecting and constructing historical knowledge and data, which enabled early archaeologists in India like Alexander Cunningham and Aurel Stein to identify and locate Buddhist sites on the ground. The chapter, as an example or case study, focuses on Xuanzang’s description of Swat and

22  Himanshu Prabha Ray the way it influenced the historical view of the archaeological research in the region. As discussed by Shailendra Bhandare, the binary of ‘Greek’ and ‘non-Greek’ has not only dominated the discourse about the spread of Hellenism to regions across the Hindukush in art history, but also prompted visible numismatic articulations like incorporating inscriptions in Indian scripts and languages on Indo-Greek coinage. Chronologically, this is adjudged to have taken place sometime during the late second century BCE, under the reigns of Greek rulers like Pantaleon, Agathocles and Eucratides. The finds of the non-Greek coins in the excavations at Taxila gave a further boost to this ‘narrative of the conquest’ as articulated in the interpretative writing of John Marshall who adjudged Taxila to have become a Greek city. Methodologically, Bhandare deploys new numismatic evidence to show how the nonGreek coins fit into the circulatory landscape of Gandhara, though the contours of this circulation system require further research. Moitra explores the socio-cultural and political representation of Gandhara in Sanskrit texts with a specific focus on the epic the Mahābhārata. It is often suggested by historians that the northwestern region of the subcontinent is represented as ‘foreign’ or ‘other’ from the perspective of the ‘indigenous’ Brahmanical society of Madhyadeśa. The Mahābhārata provides a valuable resource for study of these issues, because Gandhara does not merely appear as a reference but plays an active role in the central narrative of the epic. Gandhara is integral to the Mahābhārata because two of its central characters, Gāndhārī and Śakuni, belong to this region. A political connection through matrimony is established between Gandhara and Kuru by the marriage of Dhṛtarāṣtra, the Kuru king, to Gāndhārī, the princess of Gandhara. Interestingly, Gāndhārī is considered the most befitting to enter the Kuru lineage as the queen of Hastināpura in north India because of her religious preferences and inclination for Brahmanical norms. She is shown as the worshipper of Shiva and the follower and upholder of varṇāśramadharma (referring especially to a person’s responsibility regarding class [varṇa] and stage of life [āśrama]); Śakuni, Gāndhārī’s wicked and crafty brother, on the other hand, is shown to have impeccable knowledge of dharma which he manipulates to bring about the battle of Kurukṣetra. The second section interrogates the grand narrative of Hellenism of which Gandhara has been a part by highlighting the process of collecting. It incorporates chapters that specifically deal with the making of collections of what came to be described as ‘Gandharan art’. Archival research on the history of these collections is an under-researched

Introduction 23 theme, as the focus has generally been on iconography and the relationship between text and image. The first chapter in this section on the coin collections of Charles Masson (1800–1853) addresses this lacuna by examining archival resources. Between 1833 and 1838 Masson explored or excavated nearly fifty monuments in the vicinity of Peshawar and Kabul and amassed a staggering amount of antiquities. These included a large number of coins, of which the British Museum received a substantial share. Other recipients of Masson’s coin collection were the Asiatic Society, Calcutta [Kolkata], Ashmolean Museum, Oxford and the Fitzwilliam Museum, Cambridge. The story of Masson’s antique-collecting escapades in Afghanistan during the years preceding the First Anglo-Afghan War (1839–1842) and the subsequent dispersal of his fabulous collection of antiquities, which included seals, beads, ingots, weights, ornaments, plaques, sculptures, reliquaries (caskets, boxes and bowls), arrowheads, discs, amulets, buttons, etc., together with an estimated 60,000 coins, in various museums of India and England lie scattered in the official archives of the British East India Company as well in his private correspondence. Based on these archival sources, Sanjay Garg presents a coherent narrative of the coins discovered and collected by Masson and how his discoveries galvanized the efforts of his contemporary, James Prinsep (1799–1840) in deciphering the ancient Indian scripts and helped museums in India and England build an impressive numismatic collection from this part of the world. Other collections discussed include that of the Political House at Malakand, Khyber Pakhtunkhwa Province, Pakistan, by Pia Brancaccio. The Gandharan pieces collected in Swat in the early 1930s remain cemented in the walls of the House Garden. Original correspondence between Captain Cobb and the director of the ASI Frontier Circle sheds light on the genesis of the collection. The chapter concludes with reflections on how modern practices of collecting and exhibiting Gandharan art are still somewhat informed by an approach inherited from the past. Of the archaeological excavations continuing at Gandharan sites in Pakistan is that of the Italian Mission to the Swat valley, as discussed by Luca Olivieri. The site of Barikot (Bīr-koṭ-ghwaṇḍai) has been a project of the Italian Archaeological Mission since 1978. Excavations, interrupted in 2006, were resumed in 2011 and are currently in progress. Barikot was basically a fortified outpost founded by the Indo-Greeks towards the end of the second century BCE, mantained with its military features by the Saka-Parthians as a fortified cantonment, that, after being demilitarized, was turned

24  Himanshu Prabha Ray into a well-organized cluster of residential units in Kushan’s time. In 2012–2013 three major cultic areas were documented in phases dated to the mid-third century CE. These cultic areas represent three different typologies: 1) a Buddhist shrine located in an open courtyard of a private residential unit; 2) a sacred building characterized by an open courtyard with benches and niches with votive sculptures in situ and a large structure featured by a raised paved platform; and 3) a sacred precinct characterized by an open courtyard with a Buddhist shrine and a sacred building (distyle in antis, with naos and a rear room, with a deposit of luxury objects). Whereas the Buddhist shrines perfectly match coeval Gandharan examples, the other cultic buildings apparently find their antecedents in Central Asian models. Berlin, Munich, Stuttgart – all three cities have museums with sculpture from Gandhara, but the history and size of these collections differ considerably, as evident in the chapter by Britta Schneider. With the progressing western rediscovery of Gandhara which started in the first half of the nineteenth century onwards, artefacts from that region were put on display in Europe. In the same period, the former royal collections were transformed into public exhibitions. Museums were built and collections started growing mostly due to travellers and explorers who donated or sold their private treasuries collected during their journeys. In the beginning, Gandharan sculptures were part of ethnological collections. Some still are, and only the Indian Department of the Asian Art Museum, Berlin, has become totally independent from ethnology. The final chapter by Himanshu Prabha Ray draws on collections of Gandharan sculptures in museums in India along two lines of enquiry: one, the nature and size of collections in some of the major museums of the country, such the Indian Museum, Kolkata, founded in 1814 and with the largest collection of 1,602 Gandharan objects and two, the National Museum, New Delhi, which was inaugurated on 15 August 1949, two years after Indian independence, has 688 objects. In contrast to the Indian Museum’s collection, which was complete before 1927, the National Museum continued to add pieces until 1987. Other sizable collections include those in the Government Museum, Chandigarh, and the Chhatrapati Shivaji Maharaj Vastu Sangrahalaya, Mumbai, though the history of the collection is unique in each case. How are these differences to be understood or contextualized? The focus on ‘collecting’ rather than ‘collections’ provides insights into the changing nature of engagement between the region of Gandhara and the history of the subcontinent.

Introduction 25 It is important to underline the premise that Buddhist artistic tradition looks very different if Gandhara is considered a unit unto itself, on the one hand, or as a part of a broader swathe of Indus valley civilizations, on the other. Thus the volume emphasizes the need to unbundle the intertwined strands of nineteenth- and twentiethcentury scholarship vis-à-vis language, ethnicity and material culture so as to foster interdisciplinary research on mobility, connectivity and the wider networks of transmission.

Notes 1 The first recorded discovery of a Buddhist stupa dates to 1798 when Colin Mackenzie discovered the remains of a stupa at Amaravati in Andhra, Colin Mackenzie, ‘Extracts of a Journal,’ Asiatick Researches, 9, 1807: 272–278. In 1800 a local doctor excavated a stupa at Vaisali in Bihar. J. Stephenson, Excursions to the Ruins and Site of an Ancient City Near Bakhra, Journal of the Asiatic Society of Bengal, 4, 1835: 128–138. 2 The remains at the site consisted of a great stupa, south of the modern village, fourteen smaller buildings of the same kind, fifteen monasteries and many isolated massive stone walls. 3 J.-M. Lafont, Conducting Excavations and Collecting Coins in Maharaja Ranjit Singh’s Kingdom (1822–1839), in Himanshu Prabha Ray edited, Coins in India: Power and Communication, Marg Publications, Mumbai, 2006: 98–107. 4 Elizabeth Errington, Reliquaries in the British Museum, in David Jongeward, Elizabeth Errington, Richard Salomon and Stefan Baums edited, Gandharan Buddhist Reliquaries, University of Washington Press, Seattle and London, 2012: 119–126. 5 Joe Cribb, Kanishka’s Buddha Image Coins Revisited, Silk Road Art and Archaeology, 6, 1999–2000: 151–189. 6 Kurt A. Behrendt, The Buddhist Architecture of Gandhara, Brill, Leiden, 2004: 27. 7 Daniel Michon, Archaeology and Religion in Early Northwest India: History, Theory, Practice, Routledge India, New Delhi, 2014. 8 Himanshu Prabha Ray and Daniel T. Potts edited, Memory as History: The Legacy of Alexander in Asia, Aryan Books International, New Delhi, 2007. 9 Phiroze Vasunia, Alexander and Asia: Droysen and Grote, in Himanshu Prabha Ray and Daniel T. Potts edited, Memory as History: The Legacy of Alexander in Asia, Aryan Books International, New Delhi, 2007: 89–104. 10 John Anderson, Catalogue and Handbook of the Archaeological Collections in the Indian Museum, Part I, Asoka and Indo-Scythian Galleries, Trustees, Calcutta, 1883. 11 H. Hargreaves, Handbook to the Sculptures in the Peshawar Museum, Government of India, Calcutta, 1930: 7: ‘The school of Grandhara, on the other hand, though later in date is not a natural continuation of the Ancient Indian School but exhibits clear evidence of Hellenistic influence, displays a greater mastery over technical difficulties and introduces

26  Himanshu Prabha Ray new and foreign motifs.’ Nevertheless the author concedes that it is Buddhist in content, and this provides strong links with other parts of the subcontinent. 12 Elizabeth Errington, The Western Discovery of the Art of Gandhara and the Finds of Jamalgarhi, Unpublished Ph.D. thesis, School of Oriental and African Studies, London, 1987: 48. 13 Errington, The Western Discovery of the Art of Gandhara: 127. 14 Annabel Walker, Aurel Stein: Pioneer of the Silk Road, John Murray, London, 1995: 245–248. 15 Alfred Foucher, The Greek Origin of the Image of Buddha, in Alfred Foucher, The Beginnings of Buddhist Art and other Essays in Indian and Central-Asian Archaeology, Paul Geuthner, Paris, 1917: 111. 16 Alfred Foucher, The Greek Origin of the Image of Buddha, in Alfred Foucher, The Beginnings of Buddhist Art and other Essays in Indian and Central-Asian Archaeology, Paul Geuthner, Paris, 1917: 128. 17 Joe Cribb, Kanishka’s Buddha Image Coins Revisited, Silk Road Art and Archaeology, 6, 1999–2000: 151–189. 18 Joe Cribb, Kanishka’s Buddha Image: 151. 19 John C. Huntington, The Origin of the Buddha Image: Early Image Traditions and the Concept of Buddhadarśanapunyā, in A. K. Narain edited, Studies in Buddhist Art of South Asia, Kanak, New Delhi, 1985: 23–58. 20 Susan L. Huntington, Buddhist Art Through a Modern Lens: A Case of a Mistaken Scholarly Trajectory, in Julia A. B. Hegewald edited, In the Shadow of the Golden Age: Art And Identity in Asia from Gandhara to the Modern Age, E. B. Verlag, Berlin, 2014: 81 [79–114]. 21 Juhyung Rhi, Images, Relics, and Jewels: The Assimilation of Images in the Buddhist Relic Cult of Gandhāra: Or Vice Versa, Artibus Asiae, 65.2, 2005: 169–211. 22 Juhyung Rhi, Images, Relics, and Jewels: The Assimilation of Images in the Buddhist Relic Cult of Gandhāra: Or Vice Versa, Artibus Asiae, 65.2, 2005: 203. 23 W. Zwalf, A Catalogue of the Gandhara Sculpture in the British Museum, British Museum Press, London, 1996, volume I: 12–13. Kurt A. Behrendt, The Buddhist Architecture of Gandhara, Brill, Leiden, 2004: 12. 24 Richard Salomon, Ancient Buddhist Scrolls from Gandhāra, The British Library, London, 1999: 3. 25 John Marshall, Taxila, 3 volumes, Cambridge University Press, Cambridge, 1951: Preface. 26 R.E.M. Wheeler, Still Digging, E. P. Dutton & Co. Inc., New York, 1956: 189. 27 A.H. Dani, The Historic City of Taxila, UNESCO, Paris, 1986: 81. 28 Jonathan Mark Kenoyer, New Perspectives on the Mauryan and Kushana Periods, in Patrick Olivelle edited, Between the Empires: Society in India 300 BCE to 400 CE, Oxford University Press, Oxford, 2006: 40–46. 29 R. Coningham and Briece R. Edwards, Space and Society at Sirkap, Taxila: A Reexamination of Urban Form and Meaning, Ancient Pakistan, XII, 1997–1998: 63. 30 F.R. Allchin, The Archaeology of Early Historic South Asia, Cambridge University Press, Cambridge, 1995: 131. 31 H. Lueders and Klaus Ludwig Janert, Mathura Inscriptions, Vanden hoeck & Ruprecht, Goettingen, 1961: 68, no. 62.

Introduction 27 32 www.qau.edu.pk/new-discoveries-at-badalpur-buddhist-monastery/ accessed on 13 July 2015. 33 J.E. Van Lohuizen-de Leeuw, Gandhara and Mathura: Their Cultural Relationship, in Pratapaditya Pal edited, Aspects of Indian Art, E. J. Brill, Leiden, 1972: 27–43. 34 F.W. Thomas, The Inscription on the Mathura Lion Capital, Epigraphia Indica, 9, 1907–1908: 135–147. 35 Richard Salomon, An Inscribed Silver Buddhist Reliquary of the Time of King Kharaosta and Prince Indravarman, Journal of the American Oriental Society, 116.3, 1996: 418–452. 36 Thomas McEvilley, The Shape of Ancient Thought, Allworth Communications, New York, 2002: 9–15. 37 Madhav M. Deshpande, Sanskrit and Prakrit: Sociolinguistic Issues, Motilal Banarsidass, Delhi, 1993: 80–81. 38 M. Mitchiner, India: Minute Silver Coins of the Early Mauryan Empire, East and West, 33. 1–4, December, 1983: 113: Specimens from Afghanistan include the Chaman Hazouri hoard, coins from Mir Zakah and those from Bhir mound, Taxila (EHC Walsh, Punch-marked Coins from Taxila, MASI, no. 59, 1939). Further to the east the kingdoms of Kasi and Kosala used the same weight standard, but the denominations used by them were the half satamana (50 rattis) and the quarter sattamana (25 rattis) . . . At a more southerly latitude lay Avanti on the Narmada valley and a different weight standard was followed in this region where silver was tariffed according to the karsapana of 32 rattis (circa 3.6 grammes). Coins of this region were always stamped with a single symbol (p. 114). The coinage of Magadha like that of Avanti was struck to the karsapana weight standard, but like its other neighbours, Kasi and Kosala, the Magadhan coinage was composed of several punch marks. After a period of evolution, Magadha was to adopt the characteristic 5-punch coin design that was to remain standard (p. 115). 39 Nupam Mahajan and R. Balasubramaniam, Scanning Electron Microscopy Study of an Ancient Silver Punch-Marked Coin with Central Pentagonal Mark, Numismatic Digest, 21–22, 1997–1998: 153–167. 40 Lionel Casson, The Periplus Maris Erythraei, Princeton University Press, Princeton, 1989: 75–81. 41 K. Karttunen, India and the Hellenistic World, Finnish Oriental Society, Helsinki, 1997: 42. 42 M. Kervran, Multiple Ports at the Mouth of the River Indus, in Himanshu Prabha Ray, ed., Archaeology of Seafaring: The Indian Ocean in the Ancient Period, ICHR Monograph Series I, New Delhi, 1999: 70–153. 43 L. Casson, Periplus Maris Erythraei, Princeton University Press, Princeton, 1989. 44 L. Casson, Periplus Maris Erythraei: section 43–44. 45 In the first century CE ed-Dur has provided data for fishing and pearling activity, with fish accounting for the largest portion of the faunal inventory followed by shellfish. A bell-shaped lead weight identical to the stone weights used by pearl divers was found in a house. The recovery of large pearl oyster shells in graves at ed-Dur also suggests that the pearl was a highly valued commodity. D.T. Potts, The Roman Relationship with the Persicus Sinus from the Rise of Spasinou Charax (127 BCE) to the Reign of Shapur II (CE 309–379), in Susan E. Alcock

28  Himanshu Prabha Ray edited, The Early Roman Empire in the East, Oxbow Monograph 95, Oxford, 1997: 92–93. 46 Periplus Maris Erythraei: section 36. 47 David Whitehouse, Ancient Glass from ed-Dur, Arabian Archaeology and Epigraphy, 11.1, May, 2000: 87–128. 48 E. Haerinck, International Contacts in the Southern Persian Gulf in the Late 1st Century BC/1st Century AD: Numismatic Evidence from ed-Dur, Iranica Antiqua, 33, 1998: 293–295. 49 Ibid, p. 119. 50 P.L. Gatier, P. Lombard and K.M. Al-Sindi, Greek Inscriptions from Bahrain, Arabian Archaeology and Epigraphy, 13.2, November, 2002: 223–233. 51 Daniel Michon, Archaeology and Religion in Early Northwest Inddia: History, Theory, Practice, Routledge, London, New York, New Delhi, 2015: 64–65. 52 Cunningham Reports vol. II (1871), vol. V (1875), vol. XIV (1882) and vol. XIX (1885). 53 Maurizio Taddei, Narrative Art between India and the Hellenistic World, Transcultural Studies, 1, 2015: 13. 54 Richard Salomon, Ancient Buddhist Scrolls from Gandhara: The British Library Kharoṣṭhī Fragments, The British Library, London, 1999: 5–6. 55 Elizabeth Errington and Vesta Sarkkhosh Curtis edited, From Persepolis to the Punjab, The British Museum Press, London, 2007: 67–71. 56 Soshin Kuwayama, A Hidden Import from Imperial Rome Manifest in Stupas, in R. Allchin, B. Allchin, N. Kreitman and E. Errington edited, Gandharan Art in Context, The Ancient India and Iran Trust, Cambridge, 1997: 119–120. 57 Adrian Snodgrass, The Symbolism of the Stupa, Southeast Asia Program, Cornell, 1988: 78–79. 58 Kuwayama, A Hidden Import from Imperial Rome Manifest in Stupas: 148–149. 59 Shoshin Kuwayama, Kanjur Ashlar and Diaper Masonry: Two Building Phases in Taxila of the First Century AD, in Doris M. Srinivasan edited, On the Cusp of An Era: Art in the Pre-Kusana World, Brill, Leiden and Boston, 2007: 201–202. 60 See for example, Shashi Asthana, Kushana Art of Mathura, in S.P. Gupta edited, Kushana Sculptures from Sanghol (1st–2nd Century AD): A Recent Discovery, National Museum, New Delhi, 1985: 25. 61 Sonya Rhie Quintanilla, Emergence of the Stone Sculptural Tradition at Mathura: Mid-Second Century BC – First Century AD, Ph.D. thesis, Harvard University, 1999. 62 Guide Book to Peshawar Museum: 57. 63 P. Gardner, Coins of Greek and Scythian Kings of Bactria and India, British Museum, London, 1886: 71, pl. XVII, 3. 64 John Marshall, Annual Report, Archaeological Survey of India, 1914– 1915: 35, pl. XXIV, 50, Seal No. 2. 65 Nicholas Sims-William and Joe Cribb, A New Bactrian Inscription of Kanishka the Great, Silk Road Art and Archaeology, Kamakura: Journal of the Institute of Silk Road Studies, 4, 1995–1996: 75–127.

Introduction 29 66 M. Nasim Khan, Exploration and Excavation of the Earliest Sivaite Monastic Establishment at Kashmir Smast (A Preliminary Report), Ancient Pakistan, 14, 2001: 218–272. 67 Richard Salomon, Ancient Buddhist Scrolls from Gandhara: The British Library Kharoṣṭhī Fragments, The British Library, London, 1999: 10. 68 Kazunobu Matsuda, On the Importance of the Buddhist Manuscripts in the Schøyen Collection, in Jens Braarvig and Fredrik Liland edited, Traces of Gandhāran Buddhism: An Exhibition of Ancient Buddhist Manuscripts in the Schøyen Collection, Hermes Publishing House, Oslo in collaboration with Amarin Printing and Publishing, Bangkok, 2010: XXVIII.

References Allchin, F.R. 1995. The Archaeology of Early Historic South Asia. Cambridge: Cambridge University Press. Anderson, J. 1883. Catalogue and Handbook of the Archaeological Collections in the Indian Museum, Part I: Asoka and Indo-Scythian Galleries. Calcutta: Trustees. Asthana, S. 1985. Kushana Art of Mathura, in S.P. Gupta (ed.), Kushana Sculptures from Sanghol (1st‑2nd century AD): A Recent Discovery. New Delhi: National Museum, pp. 25–40. Behrendt, K.A. 2004. The Buddhist Architecture of Gandhara. Leiden: Brill. Casson, L. 1989. The Periplus Maris Erythraei. Princeton: Princeton University Press. Coningham, Robin A.E., and B.R. Edwards. 1997. Space and Society at Sirkap, Taxila: A Reexamination of Urban Form and Meaning. Ancient Pakistan XII.8: 47–76. Cribb, J. 1999–2000. Kanishka’s Buddha Image Coins Revisited. Silk Road Art and Archaeology 6: 151–189. Cunningham, A. 1871–1885. Archaeological Survey of India, Reports, vol. II, Calcutta: Office of the Superintendent of Government Printing, 1871; Archaeological Survey of India, Reports for 1872–73, vol. V, Calcutta: Office of the Superintendent of Government Printing, 1875, Report of a Tour in the Punjab in 1878–79, vol. XIV, Calcutta: Office of the Superintendent of Government Printing, 1882; and Report of a Tour through Behar, Central India, Peshawar and Yusufzai 1881–82, vol. XIX, Calcutta: Office of the Superintendent of Government Printing, 1885. Dani, A.H. 1986. The Historic City of Taxila. Paris: UNESCO. Daniel, M. 2015. Archaeology and Religion in Early Northwest Inddia: History, Theory, Practice. London, New York and New Delhi: Routledge. Deshpande, M.M. 1993. Sanskrit and Prakrit: Sociolinguistic Issues. Delhi: Motilal Banarsidass. Errington, E. 1987. The Western Discovery of the Art of Gandhara and the Finds of Jamalgarhi. Unpublished Ph.D. thesis. London: School of Oriental and African Studies.

30  Himanshu Prabha Ray Errington, E. 2012. Reliquaries in the British Museum, in D. Jongeward, E. Errington, R. Salomon, and S. Baums (eds.), Gandharan Buddhist Reliquaries. Seattle and London: University of Washington Press, pp. 119–126. Errington, E., and V.S. Curtis (eds.), 2007. From Persepolis to the Punjab. London: British Museum Press. Foucher, A. 1917. The Beginnings of Buddhist Art and Other Essays in Indian and Central-Asian Archaeology. Paris: Paul Geuthner. Gardner, P. 1886. Coins of Greek and Scythian Kings of Bactria and India. London: British Museum. Gatier, P.L., P. Lombard, and K.M. Al-Sindi. 2002. Greek Inscriptions from Bahrain. Arabian Archaeology and Epigraphy 13.2, November: 223–233. Haerinck, E. 1998. International Contacts in the Southern Persian Gulf in the Late 1st Century BC/1st Century AD: Numismatic Evidence from ed-Dur. Iranica Antiqua 33: 273–302. Hargreaves, H. 1930. Handbook to the Sculptures in the Peshawar Museum. Calcutta: Government of India. Huntington, J.C. 1985. The Origin of the Buddha Image: Early Image Traditions and the Concept of Buddhadarśanapunyā, in A. K. Narain (ed.), Studies in Buddhist Art of South Asia. New Delhi: Kanak Publications, pp. 23–58. Huntington, S.L. 2014. Buddhist Art Through a Modern Lens: A Case of a Mistaken Scholarly Trajectory, in J.A.B. Hegewald (ed.), The Shadow of the Golden Age: Art And Identity in Asia from Gandhara to the Modern Age. Berlin: E. B. Verlag, pp. 79–114. Karttunen, K. 1997. India and the Hellenistic World. Helsinki: Finnish Oriental Society. Kenoyer, J.M. 2006. New Perspectives on the Mauryan and Kushana Periods, in P. Olivelle (ed.), Between the Empires: Society in India 300 BCE to 400 CE. Oxford: Oxford University Press, pp. 33–50. Kervran, M. 1999. Multiple Ports at the Mouth of the River Indus, in H.P. Ray (ed.), Archaeology of Seafaring: The Indian Ocean in the Ancient Period. New Delhi: ICHR Monograph Series I, pp. 70–153. Kuwayama, S. 1997. A Hidden Import from Imperial Rome Manifest in Stupas, in F.R. Allchin, B. Allchin, N. Kreitman, and E. Errington (eds.), Gandharan Art in Context. Cambridge: Ancient India and Iran Trust, pp. 119–149. Kuwayama, S. 2007. Kanjur Ashlar and Diaper Masonry: Two Building Phases in Taxila of the First Century AD, in D.M. Srinivasan (ed.), On the Cusp of an Era: Art in the Pre-Kusana World. Leiden and Boston: Brill, pp. 201–232. Lafont, J.-M. 2006. Conducting Excavations and Collecting Coins in Maharaja Ranjit Singh’s Kingdom (1822–1839), in H.P. Ray (ed.), Coins in India: Power and Communication. Mumbai: Marg Publications, pp. 98–107. Lueders, H. 1912. A List of Brahmi Inscriptions. Calcutta: Government Printing Press.

Introduction 31 Mackenzie, C. 1807. Extracts of a Journal. Asiatick Researches 9.1807: 272–278. Marshall, J. 1914–1915. Annual Report, Archaeological Survey of India. Calcutta: Government of India. Marshall, J. 1951. Taxila, 3 vols. Cambridge: Cambridge University Press. Matsuda, K. 2010. On the Importance of the Buddhist Manuscripts in the Schøyen Collection, in J. Braarvig and F. Liland (eds.), Traces of Gandhāran Buddhism: An Exhibition of Ancient Buddhist Manuscripts in the Schøyen Collection. Oslo: Hermes Publishing House, in collaboration with Bangkok: Amarin Printing and Publishing, pp. XXVIII–XXIX. McEvilley, T. 2002. The Shape of Ancient Thought. New York: Allworth Communications. Michon, D. 2014. Archaeology and Religion in Early Northwest India: History, Theory, Practice. New Delhi: Routledge India. Mitchiner, M. 1983. India: Minute Silver Coins of the Early Mauryan Empire. East and West 33.1–4, December: 113–123. Nasim Khan, M. 2001. Exploration and Excavation of the Earliest Sivaite Monastic Establishment at Kashmir Smast (A Preliminary Report). Ancient Pakistan 14: 218–272. Nupam, M., and R. Balasubramaniam. 1997–1998. Scanning Electron Microscopy Study of an Ancient Silver Punch-Marked Coin with Central Pentagonal Mark. Numismatic Digest 21–22: 153–167. Potts, D.T. 1997. The Roman Relationship with the Persicus Sinus from the Rise of Spasinou Charax (127 BC) to the Reign of Shapur II (AD 309–379), in S.E. Alcock (ed.), The Early Roman Empire in the East. Oxford: Oxbow Monograph 95, pp. 89–107. Quintanilla, S.R. 1999. Emergence of the Stone Sculptural Tradition at Mathura: Mid-Second Century BC – First Century AD. Ph.D. thesis. Harvard University. Ray, H.P., and D.T. Potts, eds. 2007. Memory as History: The Legacy of Alexander in Asia. New Delhi: Aryan Books International. Rhi, J. 2005. Images, Relics, and Jewels: The Assimilation of Images in the Buddhist Relic Cult of Gandhāra: Or Vice Versa. Artibus Asiae 65.2: 169–211. Salomon, R. 1996. An Inscribed Silver Buddhist Reliquary of the Time of King Kharaosta and Prince Indravarman. Journal of the American Oriental Society 116.3: 418–452. Salomon, R. 1999. Ancient Buddhist Scrolls from Gandhara: The British Library Kharoṣṭhī Fragments. London: British Library. Sims-William, N., and J. Cribb. 1995–1996. A New Bactrian Inscription of Kanishka the Great, in Silk Road Art and Archaeology. Kamakura: Journal of the Institute of Silk Road Studies, vol. 4, pp. 75–127. Snodgrass, A. 1988. The Symbolism of the Stupa. Cornell: Southeast Asia Program. Stephenson, J. 1835. Excursions to the Ruins and Site of an Ancient City near Bakhra. Journal of the Asiatic Society of Bengal 4: 128–138.

32  Himanshu Prabha Ray Taddei, M. 2015. Narrative Art Between India and the Hellenistic World. Transcultural Studies 1. http://heiup.uni-heidelberg.de/journals/index.php/ transcultural/article/view/22215/16737/, accessed on 25 August 2016. Thomas, F.W. 1907–1908. The Inscription on the Mathura Lion Capital. Epigraphia Indica 9: 135–147. Van Lohuizen-de Leeuw, J.E. 1972. Gandhara and Mathura: Their Cultural Relationship, in P. Pal (ed.), Aspects of Indian Art. Leiden: E. J. Brill, pp. 27–43. Vasunia, P. 2007. Alexander and Asia: Droysen and Grote, in H.P. Ray, and D.T. Potts (eds.), Memory as History: The Legacy of Alexander in Asia. New Delhi: Aryan Books International, pp. 89–102. Walker, A. 1995. Aurel Stein: Pioneer of the Silk Road. London: John Murray. Walsh, E.H.C. 1939. Punch-marked Coins from Taxila, Memoirs Archaeological Survey of India, no. 59. New Delhi: Archaeological Survey of India. Wheeler, R.E.M. 1956. Still Digging. New York: E. P. Dutton & Co. Inc. Whitehouse, D. 2000. Ancient Glass from ed-Dur. Arabian Archaeology and Epigraphy 11.1, May: 87–128. Zwalf, W. 1996. A Catalogue of the Gandhara Sculpture in the British Museum, 2 vols. London: British Museum Press.

1 Greek or Indian? The Questions of Menander and onomastic patterns in early Gandhāra Stefan Baums This chapter reconsiders the relationship between Hellenistic society and Buddhism in Bactria and Gandhāra between the third century BCE and the second century CE.1 It does so with a dual focus: the question of the literary antecedents, authorship and audience of the Milindapañha/Nǎxiān bǐqiū jīng 那先比丘經 will be raised first; then it introduces new epigraphic evidence for the state of Hellenistic society and Buddhism in the period under consideration and presents a comprehensive analysis of the relevant onomastic data in the inscriptions; and finally, it concludes by bringing this epigraphic evidence to bear on the questions raised about the Milindapañha and its place in society.2 In briefest historical outline,3 Menander was a Bactrian Greek king who reigned around 150 BCE and conquered parts of India (up to Mathurā and, temporarily, Pāṭaliputra), leaving behind a very large coin issue, brief references in the works of the Alexander historians and – unique among Greek rulers – a literary echo as the interlocutor of the monk Nāgasena in the Buddhist scholastic dialogue preserved to us in Chinese translation as the Nǎxiān bǐqiū jīng 那先比丘經 (‘Sūtra of the Monk Nāgasena’) and in Pali translation as the Milindapañha (‘Questions of Menander’). The extant Chinese translation was prepared in the fourth century CE on the basis of an earlier translation of the third century CE and goes back to an Indian original in a language other than Pali, and possibly Gāndhārī.4 The Pali translation consists of an old core (pages 1–89 of Trenckner’s edition) corresponding to the Chinese translation (though itself incorporating several younger elements) and at least three later textual layers (pages 90–420 of Trenckner’s edition) that had been added to it by the time the Pali commentaries were composed in the fifth century CE.5 The dialogue of Menander and Nāgasena takes place over a period of two days. On the first day, Menander drives his chariot to visit Nāgasena in his

34  Stefan Baums assembly, challenges him with a series of questions on the nature of the world and leaves on horseback, thoroughly convinced by Nāgasena on all points. On the second day, Nāgasena visits Menander in his palace and answers a further series of questions (occasionally interjecting questions of his own). The text ends on the morning of the third day when Menander and Nāgasena meet one last time to assure each other of their respect.6 In an early article on Hellenism in Bactria and India, the Scottish historian W. W. Tarn (1902: 272–274, following Takakusu Junjirō 高楠 順次郎) suggested that the dialogue between Menander and Nāgasena may have been based on an earlier dialogue between the Buddha and a king Nanda (a version of which is preserved in a Chinese avadāna collection) and that the legend, reported by Plutarch, of the division of Menander’s ashes into eight parts likewise appears to be based on the division of the Buddha’s relics into eight parts. Concerning the historical king Menander’s attitude to Buddhism, Tarn considers it likely that he was favorably disposed to the Indian religion for political expediency, although we have no positive evidence for any more active support (the title δίκαιος = dhamika, Skt. dhārmika, and the eight-spoked wheel on Menander’s coinage can refer to general Indian notions of just kingship and universal rule).7 Tarn further suggested that the author of the dialogue may have been attracted to Menander simply because he was the most powerful and famous of the IndoGreek rulers. (I will return to the larger question of why a Greek ruler was chosen at all.) Meanwhile, after Albrecht Weber’s early proposal that the Milindapañha could be a direct Indian response to the Socratic dialogues, Indological opinion had come to prefer the Upaniṣadic dialogues and other inner-Indian models as its most likely literary antecedents.8 But Tarn’s thoughts on this matter developed in the opposite direction, and in an excursus in his groundbreaking history of the Greeks in Bactria and India,9 he presented an elaborate theory that the familiarity with Greek customs in the old core of the Milindapañha presupposed an original written by a Greek in the Greek language soon after Menander’s time; this hypothetical original was in turn based on the legend of Alexander asking ten questions of Indian gymnosophists and the hypothetical Greek Ur-Milindapañha itself travelled to the West and there inspired an unknown Greek author to compose the core of the dialogue between King Ptolemy II and the seventy-two Jewish elders that is preserved in Aristeas’ letters to Philocrates. For Tarn, his theory of a Greek Ur-Milindapañha formed part of a larger postulate of a body of lost Bactrian Greek literature, for which he could

Greek or Indian? 35 only adduce the evidence of one unusual city name ‘Iomousa’, which he explains as a nickname derived from the beginning of a hymn Ἰὼ Μοῦσα ‘Hail, O Muse’, and another slightly less unusual city name, ‘Euthymedeia’ (Εὐθυμέδεια), which he takes as a nickname plucked from a composition in hexameter verse (in which this name would scan). Soon after the publication of Tarn’s book, Jan Gonda devoted an entire article10 to justified criticism of Tarn’s arguments for a Greek Ur-Milindapañha, favouring, like most of his fellow Indologists,11 an Indian literary origin (and in particular Buddhist canonical dialogues like that between the Buddha and King Ajātaśatru in the Śrāmaṇyaphalasūtra), although Gonda cautiously adds that although he does not believe that Tarn’s specific arguments carry weight, he cannot entirely rule out a Greek prototype for the Milindapañha. A. K. Narain in Appendix I of his 1957 book on the Indo-Greeks similarly argues against that part of Tarn’s argument which hinges on a supposed underlying Greek distinction between the Indian variant word forms Yona and Yonaka (both meaning ‘Greek’ or ‘Western foreigner’ in general), but he does not address the larger question of a Bactrian Greek literature and a possible origin of the Milindapañha in it. Tarn died in 1957, just one year before the first of an unbroken string of discoveries of Greek inscriptions from Bactria that support his theory of a living Greek literary tradition in this country in the third and second centuries BCE. This string of discoveries was foreshadowed by the single Greek phrase, διὰ Παλαμήδου, ‘on behalf of Palamḗdēs’, at the bottom of a Bactrian-language inscription discovered at Surkh Kotal and first published in 1954.12 In 1958 this was followed by the discovery at Kandahar (Alexandria in Arachosia) of a bilingual Greek-Aramaic summary version of the Buddhist edicts of Emperor Aśoka, and in 1964 by the further discovery in the same city of a faithful Greek translation of Aśoka’s Rock Edicts XII and XIII,13 both dated to the third century BCE. One year later, French excavations started in Ai Khanum (maybe Alexandria on the Oxus), and between 1965 and 1978 these brought to light a total of four Greek stone inscriptions, ca. thirty financial records inscribed on pottery and the remains of two literary manuscripts. Of special significance for our purposes is an extract from the Delphic Maxims inscribed around the year 300 BCE by a certain Kléarkhos in the heroon of Kinéas),14 a papyrus manuscript containing a Greek philosophical treatise dating from ca. 250 BCE and a parchment manuscript preserving fragments of a Greek drama dating from ca. 200 BCE.15 A later find at Kandahar (first published in 1979) was

36  Stefan Baums the statue of a hunting dog with a Greek verse inscription by its owner, the son of Aristō̃naks, expressing his gratitude for having been saved from the attack of a wild animal (third century BCE).16 Most recently, an altar of Hestía was discovered in Kuliab (ca. 100 km northeast of Ai Khanum) with an epigram recording its dedication by a certain Hēliódotos to King Euthúdēmos and his son, the ‘glorious conqueror’ (of Gandhāra) Dēmḗtrios.17 The historical reference assigns this inscription to the beginning of the second century BCE. In addition to these literary monuments, over the years a sizeable number of non-literary Greek inscriptions, graffiti and papyri have been found at various sites in Bactria.18 Although the inscriptions mentioned so far, whether imported or produced locally (as predicted by Tarn), are purely Greek in form and content, one last important Greek inscription recently discovered, probably in Kandahar, has an author with an Indian name.19 In this funerary stele, Sōphutos (probably = Subhūti), son of Naratos (probably = Nārada), recounts in hexametric verse how, deprived of his family fortune but well trained in archery and the arts, he went abroad, earned a new fortune and returned in triumph to his hometown. The great importance of the stele of Subhūti lies in the fact that it attests at an early date, probably the second century BCE, the existence of thoroughly Hellenized Indian expatriates, considering the Greek towns of Bactria their home and engaging in the production of Greek literary art. The inverse situation is presented by the well-known pillar of Hēliódōros in Vidiśā, dating from ca. 110 BCE.20 In the inscription on this pillar Hēliódōros (Heliodora), son of Díōn (Diya) from Taxila and ambassador of the Bactrian king Antialkídas (Aṃtalikita), records its dedication to the supreme god Vāsudeva (i.e., Viṣṇu) and pays his respects to the local ruler Kāśīputra Bhāgabhadra. Just as in the early second century BCE some members of the Indian upper class had settled in Bactria and adopted Hellenistic culture, by the late second century BCE Bactrian Greeks had settled in India and engaged in Indian religious practices. Apart from the coin legends and Hēliódōros’ pillar, nine Indianlanguage inscriptions produced by bearers of Greek names were known to Tarn.21 The two earliest of these, occurring in dated inscriptions, are Theódotos (Theuduta), the meridarch, dedicating a Buddhist reliquary in Swat in the early first century BCE (CKI22 32) and Theódōros (Thaïdora), son of Datia, donating a pond using a Buddhist formula in (probably) the year 73/72 BCE (CKI 57). The other seven inscriptions are undated, but five of them are likely to belong to the

Greek or Indian? 37 late first century BCE or early first century CE: the seal ring of Theodámas (Theudama) from Bajaur (CKI 34), the seal ring of Dēmḗtrios (Timitra) from Vidiśā, the seal ring of Deínippos (Denipa) from Taxila (CKI 106) and two silver cups of Theódōros (Theutara), son of Theōrós (?) (Thavara), from Taxila (CKI 88, 89). The remaining two inscriptions belong to the Kuṣāṇa period: the training weight of the wrestler Ménandros (Miṇaṃdra, i.e., Menander) from an unknown findspot (CKI 143) and the dedication of a Buddha image by Sophḗ (Sapha) from Jamalgarhi (CKI 118). One Kuṣāṇa-period inscription not mentioned by Tarn is the Zeda well dedication of Hipeadhia (or, in Falk’s (2009b) new reading, Hiperecaa) whose hybrid name may contain the Greek element Hippías or Hippo- (CKI 148). Likewise not mentioned by Tarn (because they were discovered and published just one year before his 1938 history) are two further inscriptions: the relic dedication of Macayemaṇa from Puṣkalāvatī honouring a king Avakhajhada whose name appears to be a hybrid of Greek Eukhē and Iranian -zāða ‘son’ (CKI 178), and the important and controversial Shinkot Casket from Bajaur (CKI 176). The latter contains a series of inscriptions recording the donation and reestablishment of relics of the Buddha, and the oldest of these inscriptions (if genuine) contains an incompletely preserved dating by regnal year of King Menander, which would make it the earliest (mid-second century BCE) Indian inscription after the Aśokan edicts (mid-third century BCE). Falk (2005)23 has raised doubts about the authenticity of this particular inscription on the Shinkot Casket, but because part of his argument hinges on the unusual formulation of a separate passage on the casket, and because the apparent correction of the date from the eighth to the fourteenth day of the month (both customary days for relic installations) appears authentic, I am still inclined to accept this reference to Menander on the Shinkot Casket as genuine.24 The years since 1938 have brought to light thirteen more Gāndhārī inscriptions by bearers of Greek names. In order of publication, these are (1) the Bajaur relic dedication of Śatruleka mentioning his son Ménandros (Menaṃdra) and dated to the year 19/20 CE (CKI 257); (2) the relic dedication of Saṯaṣaka, son of Hermaĩos (Hirmaa), reinstalled by Aprakhaka, son of Hēlióphilos (Heliuphila), and dated to the year 98/99 CE (CKI 328); (3) a possibly spurious relic dedication from Hadda dating to 19/18 BCE and mentioning a Hermaĩos (Hirmaa), son of Mahomava, and another Hermaĩos (Hirmaa), son of Soṇakṣita, among the donors (CKI 455); (4) the ownership inscription of Theodámas (Theudama) on a silver vessel from a first-century CE Central Asian hoard (CKI 727); (5) the relic dedication of Helaüta (see

38  Stefan Baums later), son of Dēmḗtrios (Demetria), dated to 63/64 CE (CKI 564); (6) the dedicatory inscription on a silver vessel of the meridarch Kalliphō̃n (Kaliphoṇa) from the Mohmand Agency dating from the second or early first century BCE (CKI 552); and (7–13) seven seals, sealings and tokens from the Aman ur Rahman collection belonging to Eukratídēs (Evukratita, CKI 917), to a son of Ísandros (Isaṃdra, CKI 972), to a son of Zēnóphilos (Zenupila, CKI 987), to Theuta (see later, CKI 969), to Theodámas (Theudama, CKI 978), to Dionusódōros (Dinisidora, CKI 1000) and to Deinokrátēs (Denukrata, CKI 1030).25 See Table 1.1 for an alphabetical overview of all Greek names attested in Gāndhārī inscriptions currently known. Excluded both in Tarn’s list and in my expanded list are the numerous Greek names of Gandhāran rulers attested on their coinage; the names of Macedonian months, used in parallel with the Indian system Table 1.1  Greek names in Gāndhārī inscriptions Name

Inscription

Aṃtikini, Aṃtekine (Antígonos) Aṃtiyoka, Aṃtiyoga (Antíokhos) Alikasudara (Aléksandros) Avakha (Eukhḗ) Isaṃdra (Ísandros) Evukratita (Eukratídēs) Kaliphoṇa (Kalliphō̃n) Turamaya (Ptolemaĩos) Thavara (Theōrós?) Theutara, Thaïdora, (*Theu)sora (Theódōros) Theudama (Theodámas)

CKI 13, 27 (Aśoka, rock edict) CKI 2, 13, 16, 27 (Aśoka, rock edict) CKI 13, 27 (Aśoka, rock edict) CKI 178 (relic donation) CKI 972 (sealing) CKI 917 (token) CKI 552 (phial) CKI 13, 27 (Aśoka, rock edict) CKI 88, 89 (silver cups) CKI 57 (pond donation), 88, 89 (silver cups), 955 (token) CKI 34 (seal), 727 (silver vessel), 978 (sealing) CKI 32 (reliquary) CKI 1000 (seal) CKI 106 (seal) CKI 1030 (seal) CKI 564 (relic donation) CKI 13, 27 (Aśoka, rock edict) CKI 143 (weight), 176 (reliquary), 257 (reliquary) CKI 118 (image donation) CKI 328 (reliquary), 455 (gold leaf) CKI 328 (reliquary) CKI 987 (token)

Theuduta (Theódotos) Dinisidora (Dionusódōros) Denipa (Deínippos) Denukrata (Deinokrátēs) Demetria (Dēmḗtrios) Maka (Mágas) Menaṃdra, Minaṃdra, Minedra (Ménandros) Sapha (Sophḗ) Hirmaa (Hermaĩos) Heliuphila (Hēlióphilos) Zenupila (Zēnóphilos) Source: Prepared by author

Greek or Indian? 39 of months in Gāndhārī inscriptions; and the three Hellenistic administrative titles attested in Gandhāra: stratega (stratēgós), meridarkha (meridárkhēs) and aṇaṃkaya (anagkaĩos). In particular, the ubiquitous and prestigious Greek names of rulers on coins will have served as an important vector for the adoption of Greek names in Gandhāra in general. Considering only the epigraphic data, several observations can be made on the mechanisms whereby these Greek names were acculturated in their Gandhāran context. First of all, three of the inscriptions carry their names in parallel in Kharoṣṭhī and in Greek script. Although the token CKI 955 ((*Theu)sorasa, θεοδ(*ωρου)) and the sealing CKI 978 ((*The)uda(*masa), Θε⟨*ο⟩δ(*α)μ(*ου)) contain just the names, a much more interesting case is presented by the inscription of Kalliphō̃ n (CKI 552). Like the stele of Subhūti and the pillar of Hēliódōros, it illustrates a particular type of interaction between Hellenistic and Indian religious practices. The silver vessel in question belongs to a hoard of nine plates and bowls, three of which are inscribed: Kalliphō̃ n’s inscription is given in Greek (Καλλιφων μεριδαρχης ευξαμενος ανεθηκεν τωι Χαοσει ‘Kalliphō̃ n, making a vow, dedicated [this] to Kháos’) as well as in Gāndhārī (Kaliphoṇena meridarkhena praṭiśunita nirakaṭe Boasa), a second inscription is in Greek only and a third in Gāndhārī only. The inscriptions reveal that the whole set of vessels was dedicated in a Greek ritual in the sanctuary of a deity, but the deity in question was Indian rather than Greek. Its name is given as Boa in the Gāndhārī version of Kalliphō̃ n’s inscription (interpreted by Falk 2009a as Bhava, a primordial form of Śiva) and translated as Kháos into Greek in an instance of interpretational Graeca.26 A second type of acculturation relates to those cases where the same person had both an Indian and a Greek name, two of which are attested in our corpus. Denukrata from the seal CKI 1030 was according to the inscription (Denukratasa Sagharakṣidasa) also known as Sagharakṣida, and Indravarma’s biscript sealing CKI 1035 (Iṃdravarmasa strategasa, Αλεξανδρου στρατηγου) tells us that he had the Greek name Aléksandros.27 One suspects that in the case of Iṃdravarma, his choice of Greek name was influenced by the assonance of Iṃdra- with -andros and by the semantic affinity of -varma, ‘protection’, and Aleks-, ‘defending’. It is worth noting that the practice of double names in Indian and Greek was employed not just by rulers but, in the case of Saṃgharakṣida, even by somebody who appears to have been a Buddhist and possibly a monastic.

40  Stefan Baums The final step in the Gandhāran intermingling of Greek and Indian onomastic systems is those names that appear to combine Greek and Indian elements. Both Helaüta in the relic dedication CKI 564 and Theuta on the seal CKI 969 appear to share the same final element -uta, which is likely to correspond to the common Sanskrit name element -gupta ‘protected by (a deity)’. Here it is combined with two different Greek first elements, namely Hēlio- (the sun god) and Theo- (god). The creation of such bilingual portmanteau names was facilitated by the shared inheritance in India and Greece of the Indo-European aristocratic custom of compound names.28 In the last part of this chapter, I would like to use data from the inscriptional corpus described earlier to examine what, if anything, the existence of Greek names in India tells us about the cultural identification of their bearers. The corpus contains ten bearers of Greek names and two bearers of Greek–non-Greek hybrid names that tell us something about their family relationships as part of identifying themselves or sharing the merit of a religious donation, summarized in Table 1.2.

Table 1.2 Family relationships involving bearers of Greek names in Gāndhārī inscriptions (f = father, m = mother, w = wife, s = son, u = uncle, gm = grandmother, gs = grandson, gu = great-uncle) Century

Greek name

Family relations

CKI no.

2 BCE 1 BCE

Díōn (f) Datia (f)

– 57

1 BCE – 1 CE 1 CE

Hēliódōros (Vidiśā) Theódōros (Kaldarra) Theódōros (Taxila) Ménandros

88, 89 257

2 CE

Hermaĩos (Hadda) Hermaĩos (Hadda) Theodámas (Central Asia) Helaüta Hiperecaa (Zeda)

Theōrós? (f) Subhūtikā (gm) Vijayamitra (gu) Śatruleka (f) Davili (m) Indrasena (u) Indrasena (b) Mahomava (f) Soṇakṣita (f) Budhala (s) Dēmḗtrios (f) Vasativaca (w) Leaka Sacaloka (s) Saṯaṣaka (s) Muṃji (gs) Aprakhaka (s) Raja (s)

564 148

Hermaĩos unclear

Hēlióphilos Zēnóphilos

Source: Prepared by author

455 455 727

328 328 987

Greek or Indian? 41 The surprising result of this survey is that only one or two of these Greek-named persons had relatives that themselves bore Greek names, namely Hēliódōros, son of Díōn, and possibly Theódōros, son of Theōrós. (In addition, hybrid-named Helaüta has a Greeknamed father Dēmḗtrios.) In all other cases, the family members of the bearers of Greek names have Indian or foreign (including Scythian and Parthian) names. A particularly telling case in point is that of Ménandros, a scion of the well-known royal family of Apraca in Bajaur: whereas Ménandros himself has been given the name of the most famous foreign ruler of Gandhāra (living two centuries before him), half of his relatives have adopted Indian names (Subhūtikā, Vijayamitra, Indrasena) and the other half retain their foreign names (Śatruleka, Davili).29 We have to conclude from this that although the use of Greek names in Gandhāra bespeaks the continuing historical memory and prestige of the Indo-Greek rulers (especially Menander), we can deduce nothing from it about the ethnicity or cultural self-identification of their bearers (unless we have positive evidence that the whole family bore Greek names), and we can make no more than guesses about the amount of Hellenism entering into the complex and eclectic cultural makeup of these foreign rulers of the Indian northwest that chose to bestow Greek names on some of their offspring. At the end of this chapter I would like to return to the Questions of Menander, a product of precisely the complex cultural landscape of the first centuries BCE and CE that is exemplified in the epigraphic evidence. Are we able to formulate a clearer idea of its authorship and intended audience than before? Much of the century-long discussion of this question has operated with monolithic entities – Greek and Indian – for the one and the other: either the Questions of Menander were taken as a Greek text co-opted by Indian redactors for Buddhist proselytization or as an Indian text with Menander substituted for a generic Indian ruler to appeal to a Greek audience. As we have seen, however, throughout the epigraphic record we have evidence of Indians adopting Hellenistic culture in the Greek city-states of Bactria (Subhūti), Greeks settling in India and practising Indian religion (Hēliódōros) and a community in the borderlands of Gandhāra producing bilingual records of donations to a god that offers an Indian and a Greek interpretation (Bhava and Kháos) at the same time. This cultural melding process reached its culmination when people who were by origin neither Greek nor Indian began to settle in and rule Gandhāra and freely adopted elements of either culture as they found them on the ground.

42  Stefan Baums One recent analysis of the Questions of Menander arrives at nuanced results that appear commensurate with this cultural background. Vasil'kov (1993) argues that the Questions owe their overall literary form to an Indian culture of verbal challenge and contest (as elaborated by the Dutch Indologists Heesterman and Kuiper) that left its imprint on Buddhist as well as non-Buddhist literature. Under this interpretation, King Menander challenges Nāgasena in Nāgasena’s assembly during their first encounter, is defeated and as a consequence has to leave the site of the verbal contest on horseback rather than on his chariot. On the second day, Nāgasena returns the challenge when in turn he visits Menander’s royal assembly, and although the Buddhist context demanded that Menander be the one to receive answers to his doubts, Vasil'kov considers it significant that only on this second day are some questions also posed by Nāgasena to Menander.30 Menander’s repeated defeat and the acknowledgement of the superiority of the victor conclude the traditional Indian scheme of the verbal contest, which contrasts sharply with the impartial pursuit of truth that is the object of a Socratic dialogue. Although the literary form of the Questions is thus purely Indian (with a Buddhist overlay of older patterns), Vasil'kov argues – to my mind convincingly – that the microstructure of the argument, with its innumerable appeals to natural phenomena and cultural objects (some of the latter specifically Greek), engages with Greek habits of debate that would have been prevalent in the Greek literary culture of Bactria (which, as we have seen, has been richly attested by recent archaeological discoveries) and were still current among the intended audience of the Questions. Applying Vasil’kov’s conclusions to the epigraphic data surveyed in this chapter, one may therefore propose that the author of the Questions of Menander in the form in which we have them was steeped in the age-old Indian tradition of verbal debate and its corresponding literary form; that the author of the Questions (like the authors of other Buddhist dialogues before him) modified this literary form to emphasize the superiority of the knowledge of the Buddhist sage Nāgasena; and that he intended the text for the conversion of an audience that was neither Indian nor Greek, but part of the cosmopolitan melting pot of Gandhāra that was Indianized enough for the literary form of the Questions to appeal to it, Hellenized enough to be persuaded by its Greek style of argumentation and worldly enough to identify with the figure of the most famous foreign ruler of Gandhāra as he undergoes conversion to Buddhism.

Greek or Indian? 43

Notes 1 A first version of this chapter was presented at the 2011 meeting of the American Academy of Religion in a panel on ‘Greco-Roman Culture and Buddhism’ organized by Mariko Walter. The argument was updated and expanded for a presentation at the workshop ‘Rethinking the Greeks in Gandhāra’ at the Ludwig-Maximilians-Universität München (27–28 March 2014). 2 There is a noticeable absence of Greek names in the large number of Gāndhārī birch-bark manuscripts that have recently come to light (Baums 2014), in contrast to the contemporary Indo-Scythian rulers that figure in the Gāndhārī avadāna collections (Salomon 1999: 141– 151). The only potential trace of a Greek cultural influence on Gāndhārī literature was noted by Brough 1962: 207–208 who suggested that in Khotan Dharmapada verse 97 dharmatraka contains a multilingual pun on Sanskrit tarka ‘thought’ and Greek τροχός ‘wheel’. If correct, this would imply that the intended audience of Gāndhārī literature included some with knowledge of Greek. Another noticeable absence is that of Greek names among the large cache of Gāndhārī administrative documents from Niya (cf. the onomasticon in Padwa 2007: 309–333), in spite of the well-known Hellenistic motifs in the seals of these documents. The only potential Classical name among the Gāndhārī sources from Central Asia is Tita in the Miran wall painting CKI 443 (if, in fact, it is a rendering of Latin Titus). 3 G. Fussman, L’Indo-Grec Ménandre ou Paul Demiéville revisité, Journal asiatique, 281, 1993: 63–66. 4 Oskar von Hinüber, A Handbook of Pāli Literature, Walter de Gruyter, Berlin, 1996: 83. 5 Von Hinüber, Handbook: 83–86. 6 Y. Vasil'kov, Did East and West Really Meet in Milinda’s Questions?, Культурология, 1, 1993: 67, von Hinüber, Handbook: 83–84. 7 W.W. Tarn, The Greeks in Bactria & India, University Press, Cambridge, 1938: 262–264. 8 M. Winternitz A History of Indian Literature, Vol. II: Buddhist Literature and Jaina Literature, University of Calcutta, Calcutta, 1933: 176. 9 Tarn, The Greeks: 414–436. 10 J. Gonda, Tarn’s Hypothesis on the Origin of the Milindapañha, Mnemosyne, 2, 1949: 44–62. 11 More recently Fussman, L’Indo-Grec Ménandre, 1993. 12 P. Bernard, Langue et épigraphie grecques dans l’Asie Centrale à l’époque hellénistique, in Jan A. Todd, Dora Komini-Dialeti and Despina Hatzivassiliou (eds.), Greek Archaeology without Frontiers, “Open Science” Lecture Series, Athens, 2002: 86–92. 13 Bernard, Langue et épigraphie: 94–103. 14 Bernard, Langue et épigraphie: 75–78. This inscription is the point of departure for the general discusssion of Hellenism in Bactria in Parker 2007. 15 Bernard, Langue et épigraphie: 81. 16 Bernard, Langue et épigraphie: 103.

44  Stefan Baums 17 Paul Bernard, Georges-Jean Pinault and Georges Rougemont, Deux nouvelles inscriptions grecques de l’Asie centrale, Journal des savants, 2004: 333–356. 18 Cf. Claude Rapin, Fouilles d’Aï Khanoum, VIII: La trésorie du palais hellénistique d’Aï Khanoum: L’apogée et la chute du royaume grec de Bactriane, De Boccard, Paris, 1992: 387–392 and Bernard, Langue et épigraphie, for overviews. 19 Paul Bernard, Georges-Jean Pinault and Georges Rougemont, Deux nouvelles inscriptions grecques de l’Asie centrale, Journal des savants, 2004: 227–332. 20 Tarn, The Greeks: 313–314, 388. 21 Tarn, The Greeks: 389–392. 22 CKI = Corpus of Kharoṣṭhī Inscriptions (Part II of Baums and Glass 2002 –). 23 Harry Falk, The Introduction of Stūpa-Worship in Bajaur, in Osmund Bopearachchi and Marie-Françoise Boussac (eds.), Afghanistani: ancien carrefour entre l’Est et l’Ouest: actes du colloque international organisé par Christian Landes & Osmund Bopearachchi au Musée archéologique HenriPrades-Lattes du 5 au 7 mai 2003, Brepols, Turnhout, 2005: 347–358. 24 Stefan Baums, Catalog and Revised Texts and Translations of Gandharan Reliquary Inscriptions, David Jongeward, Elizabeth Errington, Richard Salomon and Stefan Baums, Gandharan Buddhist Reliquaries, Seattle: Early Buddhist Manuscripts Project, 2012: 202–203. 25 A fourteenth newly discovered inscription on a stone slab from Takht-ibahi appears to record a donation by a certain Bhadraśila, son of Iphaṇa (Nadiem 1989; revised reading by Andrew Glass, CKI 596). The name of the father sounds un-Indian and bears a certain resemblance to Greek names in Iphi-, but the similarity is not strong enough to include it in our list of Greek names. 26 A fourth inscription, the sealing CKI 940, likewise appears to preserve parallel names in Kharoṣṭhī (/// ṇasa) and in Greek script (/// εθρανου), but both versions are too fragmentary to say whether a Greek name was intended. There is no match for the substring εθραν in the online version of the Lexicon of Greek Personal Names (http:// www.lgpn.ox.ac.uk). The inverse type of biscript inscription presents Indian or local names both in Kharoṣṭhī and in Greek script. Examples include Budha(*dha)trasa / Βοδαζατρασα (CKI 104), Madeasa / (*Μ)οδειου (CKI 933), Śatralakasa / Σατρολαιου (CKI 943), a ma ? / δαβο (CKI 962), Naṃdasa / Νανδου (as well as Brāhmī Ṇadasa, CKI 977) and Mitriśamaputrasa / Μιτασαμαπατασα (as well as Brāhmī Mitraśamapūtrasa, CKI 1077). The examples including Brāhmī can be further compared with the biscript Kharoṣṭhī Brāhmī seal of Indravarma (CKI 364), giving his name in Kharoṣṭhī (Iṃdravarmasa iśparasa) and his title in Brāhmī (Avajarajasa). 27 Aman ur Rahman and Harry Falk, Seals, Sealings and Tokens from Gandhāra, Reichert Verlag, Wiesbaden, 2011: 179 read στρατεγου, but the expected η is clear on their photograph. 28 As mentioned earlier, Hiperecaa in the well donation CKI 148 may be a third such portmanteau name containing a Greek first element Hippías or Hippo-, but here the identity of the second element remains elusive and does not appear to be Indian.

Greek or Indian? 45 29 Compare with this the epigraphic record of Ai Khanum in Bactria, which consists mostly of Greek names, but also contains one name in Aramaic and one in unclear script (Parker 2007: 175–176). 30 Yaroslav Vasil'kov, Did East and West Really Meet in Milinda’s Questions?, Культурология, 1, 1993: 64–77.

References Baums, S. 2012. Catalog and Revised Texts and Translations of Gandharan Reliquary Inscriptions, in D. Jongeward, E. Errington, R. Salomon, and S. Baums (eds.), Gandharan Buddhist Reliquaries. Seattle: Early Buddhist Manuscripts Project, pp. 200–251. Baums, S. 2014. Gandhāran Scrolls: Rediscovering an Ancient Manuscript Type, in J.B. Quenzer, D. Bondarev, and J.-U. Sobisch (eds.), Manuscript Cultures: Mapping the Field. Berlin: De Gruyter, pp. 183–225. Baums, S., and A. Glass. 2002. Catalog of Gāndhārī Texts. https://gandhari. org/catalog/ accessed on 3 March 2017. Bernard, P. 2002. Langue et épigraphie grecques dans l’Asie Centrale à l’époque hellénistique, in J.A. Todd, D. Komini-Dialeti, and D. Hatzivassiliou (eds.), Greek Archaeology Without Frontiers. Athens: “Open Science” Lecture Series, pp. 75–108. Bernard, P., G.-J. Pinault, and G. Rougemont. 2004. Deux nouvelles inscriptions grecques de l’Asie central. Journal des savants 227–356. Brough, J. 1962. The Gāndhārī Dharmapada. London: Oxford University Press. Falk, H. 2005. The Introduction of Stūpa-Worship in Bajaur, in O. Bopearachchi, and M.-F. Boussac (eds.), Afghanistan: ancien carrefour entre l’Est et l’Ouest : actes du colloque international organisé par Christian Landes & Osmund Bopearachchi au Musée archéologique Henri-Prades-Lattes du 5 au 7 mai 2003. Turnhout: Brepols, pp. 347–358. Falk, H. 2009a. Greek Style Dedications to an Indian God in Gandhara. IndoAsiatische Zeitschrift 13: 25–42. Falk, H. 2009b. The Pious Donation of Wells in Gandhara, in G. Mevissen, and A. Banerji (eds.), Prajñādhara: Essays on Asian, History, Epigraphy and Culture in Honour of Gouriswar Bhattacharya. New Delhi: Kaveri Books, pp. 23–36. Fussman, G. 1993. L’Indo-Grec Ménandre ou Paul Demiéville revisité. Journal asiatique 281: 61–138. Gonda, J. 1949. Tarn’s Hypothesis on the Origin of the Milindapañha. Mnemosyne 2: 44–62. Hinüber, O. von. 1996. A Handbook of Pāli Literature. Berlin: Walter de Gruyter. Nadiem, I.H. 1989. A Fresh Kharoshthi Inscription from Takht-i-Bahi. Journal of Central Asia 12: 209–216. Narain, A.K. 1957. The Indo-Greeks. Oxford: Clarendon Press.

46  Stefan Baums Padwa, M. 2007. An Archaic Fabric: Culture and Landscape in an Early Inner Asian Oasis (3rd – 4th Century C.E. Niya). Unpublished Ph.D. dissertation. Harvard University. Parker, G. 2007. Hellenism in an Afghan Context, in H.P. Ray and D.T. Potts (eds.), Memory as History: The Legacy of Alexander in Asia. New Delhi: Aryan Books International, pp. 170–191. Rahman, A., and H. Falk. 2011. Seals, Sealings and Tokens from Gandhāra. Wiesbaden: Reichert Verlag. Rapin, C. 1992. Fouilles d’Aï Khanoum, VIII: La trésorie du palais hellénistique d’Aï Khanoum : L’apogée et la chute du royaume grec de Bactriane. Paris: De Boccard. Salomon, R. 1999. Ancient Buddhist Scrolls from Gandhāra: The British Library Kharoṣṭhī Fragments. Seattle: University of Washington Press. Tarn, W.W. 1902. Notes on Hellenism in Bactria and India. Journal of Hellenic Studies 22: 268–293. Tarn, W.W. 1938. The Greeks in Bactria & India. Cambridge: University Press. Tarn, W.W. 1951. The Greeks in Bactria & India, 2nd edition. Cambridge: University Press. Trenckner, V. 1880. The Milindapañho: Being Dialogues Between King Milinda and the Buddhist Sage Nāgasena. London: Williams and Norgate. Vasil'kov, Y. 1993. Did East and West Really Meet in Milinda’s Questions? Культурология 1: 64–77. Winternitz, M. 1933. A History of Indian Literature, Vol. II: Buddhist Literature and Jaina Literature. Calcutta: University of Calcutta.

2 “Tis all here. A treasure locked” Unlocking the wonder house of the Chinese Buddhist travelogues Max Deeg [The lama] had heard of the travels of the Chinese pilgrims, Fo-Hian and Hwen-Thiang, and was anxious to know if there was any translation of their record. He drew in his breath as he turned helplessly over the pages of Beal and Stanislas Julien. Tis all here. A treasure locked.1

This is a crucial passage from Rudyard Kipling’s novel Kim in which the Tibetan lama encounters the mine of information, the Chinese travel records, which will become the clue for his search for the “Fountain of Wisdom”, the spring created by the bodhisattva during the archery competition in Kapilavastu at the spot where his arrow hit the ground after having penetrated several targets. Stanley Abe has aptly interpreted the colonialist tenets of the Lahore Museum episode,2 the Wonder House or “Ajaib-Gher”, but what I want to discuss today is another aspect of nineteenth-century and early twentieth-century investigations in the Indian past. My title obviously implies a parallel between museums as places of collecting artefacts – the Lahore Museum whose curator Kipling’s father was – and texts – the Chinese “pilgrims’ ” records – used as quarries for collecting information for all kinds of purposes. Tapati Guha-Thakurta has, while also using Kim as a point of reference, claimed “twin histories of museums and archaeology in nineteenthcentury India” in the sense of “collecting, conserving, classifying”.3 I would, in the case of the investigation of the Indian past, argue for an intrinsic connection between the two collecting activities insofar as the collection of textual information was – in the earliest period of Indian archaeology at least used to identify historical sites of importance almost exclusively Buddhist – in order to collect, conserve and classify artefacts. The connection between collecting data, knowledge

48  Max Deeg as it were, and artefacts is, again, clearly depicted by Kipling who has the museum’s curator say to the lama: “Welcome then, O lama from Tibet. Here be the images, and I am here . . . to gather knowledge. Come to my office awhile.”4 What this episode signalizes – and I could go on analyzing Kim with respect to the perceptions of museal artefacts and texts presented and represented in the novel – is, I would claim, the intimate interrelation between collecting information and data from textual sources and retrieving and collecting objects of material culture. The “images” assembled in the Lahore Museum – and in others, one may add – are clearly the result of collecting activities, and their finding and identification, partly at least, are the result of consulting with texts. And these texts are not coming from the direct contextual vicinity of the sites from which the collected artefacts are originating, Indian canonical texts as it were, because of their overall non- or not-any-more-existence. The textual sources Kipling is referring to are, of course, the travelogues of Chinese monks who came to and studied in India between the early fifth and ninth centuries of the Common Era. These texts, often somewhat misleadingly called “pilgrim records”,5 are linked with names such as Faxian 法顯, Song Yun 宋雲 and Huisheng 惠生, Xuanzang 玄奘, Yijing 義淨, Huichao 惠超 and Wukong 武空. The texts themselves can be considered collections – collections of knowledge about Buddhist sacred places and the monuments and narratives linked to them in India. In a way, what is reflected in the passage from Kim discussed earlier is reminding us of the collecting activities, the products of which the German sociologist and historian Jürgen Osterhammel has called one of the emerging elements of nineteenth-century public and urban civilization, “Treasuries of Knowledge and Memories”, and described them as follows: “Archives, libraries, museums, and other collections might be called treasuries of memories.” Osterhammel also reminds us that “[The] boundaries between their various subcategories developed only gradually”. He emphasizes: “Treasuries of memory preserved the past as a virtual present. Yet the cultural past remains dead if it is nothing but treasured. Only in the act of appropriation, comprehension, and sometimes reenactment does it come alive.”6 Collections and recollections go hand in hand. One also could refer briefly to Michel Foucault’s concept of “archaeology of knowledge”, which he develops from a discussion of two different forms of history: a traditional one and a new one. This does not mean that I ascribe completely to either of them; I would rather insist that in practice both are needed to get the discipline of engaging with

“Tis all here. A treasure locked” 49 the past and creating from the process of the archaeology of knowledge a contextualized collection of knowledge. I thus deliberately “misuse” and reinterpret Foucault’s following description of a dynamic between the textual and the archaeological “realm”: [H]istory, in its traditional form, undertook to “memorize” the monuments of the past, transform them into documents, and lend speech to those traces which, in themselves, are often not verbal, or which say in silence something other than what they actually say; in our time, history is that which transforms documents into monuments. In that area where, in the past, history deciphered the traces left by men, it now deploys a mass of elements that have to be grouped, made relevant, placed in relation to one another to form totalities. There was a time when archaeology, as a discipline devoted to silent monuments, inert traces, objects without context, and things left by the past, aspired to the condition of history, and attained meaning only through the restitution of a historical discourse; it may be said, to play on words a little, that in our time history aspires to the condition of archaeology, to the intrinsic description of the monument.7 It is well known that South Asian archaeologists – and not only those, of course – have in the nineteenth and twentieth centuries8 exploited, and still do exploit, these texts as collections of information which may be applied for the interpretation of their sites, objects and data in a partly successful, partly very unproductive way. Successful insofar as quite a lot of sites could only be identified with the help of these records – unproductive in the sense of having used these texts out of context. One could say that, in a way, archaeologists sometimes reached the contextualization and interpretation of their own objects by the de-contextualization of the records they used. This is, of course, also partly true through the fact that the texts were conceived as almost timeless and “frozen” documentation of a past difficult to penetrate9 without asking the now usual questions of genre, cultural setting, motivation, intentionality, etc. If they delivered in the late nineteenth century mostly translations from Asian texts, useful pieces of information then, these were broken from the quarry of the text; if they did not correspond or even disagreed with other evidence, they were “corrected”, ignored or declared to be representing a wrong or distorted view without asking for the reasons of the disagreement and the possible background and motivation on the record’s side.10

50  Max Deeg In comparison to the described approach I suggest a different one, and I would like to draw here some parallels between texts – in my case the Chinese Buddhist travelogues – and collections of material cultures in the hope that some of the concerns and problems which I feel have not been addressed with regard to the texts become clearer through such a comparison. I do not think that keeping the two discourses, the archaeological and the textual, separate will do either side any good. We should neither play out each position against each other, nor, of course, be as naïve on the respective other side as Sir Mortimer Wheeler who, in his Foreword to Abu Imam’s biography of Alexander Cunningham, remarks on the two Chinese travellers Faxian and Xuanzang and their use by the “father” of South-Asia archaeology: “Cunningham wisely used the tools most ready to his hands, and these tools were the recorded travels of two highly intelligent Buddhists of the 5th and 7th centuries.”11 The question is how “wisely” Cunningham, as the curator of collections of artefacts, really used the tools of the curators of the relevant texts (i.e. the translators of the Chinese travelogues). We know now that the application of Cunningham’s method did not always lead to convincing results, and I would suggest that this had to do with a lack of communication between himself and the texts – starting with the fact that he had to rely on first Abel Remusat’s and Stanislas Julien’s French translations and later on Beal’s English translations of his two main sources, Xuanzang and Faxian. Over time the uncritical use of the Chinese records – and other textual material – of the nineteenth and early twentieth centuries has been critically assessed, but unfortunately this has not led to a new entangled discourse on a broader scale. It seems, with exceptions, of course, that textualists and archaeologists are going separate ways. The problem I see in the relationship of the two collections is a theoretical divorce and, resulting from it, at least a partial breakdown of communication. By theoretical divorce I mean an isolationist approach to the study of Indian history and the history of Buddhism in India in which textualists and archaeologists – in a way reifying themselves as a separate group by the use of a particular ideological rhetoric – often do not cooperate with each other in an effective way. This is reflected, for example, in the following simplistic statement by the archaeologist Robin Coningham: Traditionally knowledge of the history and nature of Buddhism has come from a combination of two sources, ancient texts and modern devotional practice. Archaeology has seldom been utilised in this process, apart from the largely unscientific clearing

“Tis all here. A treasure locked” 51 of ‘Buddhist’ monuments . . . These views [i.e.: the prevalence of textual and anthropological studies, MD] are so widely held that when such clearing has revealed material evidence which conflicts with these sources, it is commonly interpreted as a local aberration or degenerative practice.12 Coningham’s statement is naïve because one can turn the last statement into its opposite, that textual sources are still often appropriated for the interpretation of archaeological data out of context. Dilip Chakrabarti’s opinion is more balanced in this respect and may be taken as a plea for a closer cooperation between textual and archaeological scholars: “There are two good reasons why a historical study of ancient India cannot realize its full potential on the basis of textual sources alone.”13 But Chakrabarti’s statement also reflects the “struggle” over supremacy of interpretation of the past which is sometimes fought by the proponents of the two “collections”, textual and “objectual”.14 Despite all this criticism about the predominance of the study of texts in Buddhist studies over archaeology which seems to dovetail with a postcolonial critique of a colonial “archaeology of knowledge”, one thing seems obvious: meaningful interpretation of Buddhist history in India (and elsewhere) can neither be done without the texts – if they are available at all – nor without archaeology. What is needed for combining the two collections, the one of material culture and the one of textual data, in a balanced way for the benefit of valuable and solid conclusions and the production of knowledge is what generally would make the world as a whole a better place: dialogue and communication. In a way, what I am arguing for is such a dialogue between curators of artefacts and curators of texts. I suggest that bringing archaeologists, curators and textualists into dialogue will improve our understanding of artefacts and texts in a common context and will enhance the adequate presentation of collections in museums, as well as the representation and interpretation of history in general. Because the regional focus of this volume is Gandhara, I discuss a concrete example from what has been called Greater Gandhara, from Swat, the old Udyāna (sometimes reconstructed as Uḍḍiyāna). Swat is actually an excellent example for the communication between textual sources and archaeology because of the wealth of archaeological sites and artefacts preserved,15 but also because of the fact that in addition to the two main texts mentioned before, Faxian’s and Xuanzang’s,16 the region is described by two other Chinese travellers: by Song Yun

52  Max Deeg and Huisheng from the beginning of the sixth century, and by Huichao from the first half of the eighth century. Swat attracted the attention of historians and archaeologists in the nineteenth century because it was “Alexander territory” (i.e. an area where Alexander had moved through on his Indian campaign). The region was, however, difficult to access because it was barred from the Peshawar plains by a mountain range. Alexander Cunningham in his The Ancient Geography of India (1871: 81ff.) presents a collection of information retrieved from the Chinese travelogues without giving much archaeological data: he only presents data from one collection, not being able to match it with data from the other. In his report Cunningham expresses his regret about the inaccessibility of the region (i.e. the inability to collect more specimens from the region): [T]he broad and fertile valley of the Suwât River is known to be rich in ancient remains, which will only lessen both in number and value as successive years pass over them. During the past year Dr. Leitner procured some specimens of Buddhist sculpture from Suwât; and lately I have been fortunate enough to obtain two inscriptions from northern Yusufzai, one from Suwât and one from Bajâwar. These few trophies, which have been wrested with difficulties from the forbidden territory of a bigoted people, are sufficient to prove that both Suwât and Bajâwar must possess many ancient remains of the same style and date as those of Takht-i-Bahi and Jamâl-garhi to the south of the mountains. (1875: 1) His frustration obviously resulted from the discrepancy between the availability of data from one “collection”, the textual one, and the inability of linking these data with the other, the geo-archaeological one. This essential step in this direction would be reserved for another colonial “collector”. The next direct report on the Swat valley, this time collected from direct experience in the region, is Major Harold Arthur Deane’s (at that time C.S.I., Political Agent Dir and Swat) 1896 essay in the Journal of the Royal Asiatic Society. Deane follows the method laid out by Cunningham insofar as he almost slavishly follows Xuanzang’s record: As the Swat valley, and neighbourhood, which constitute the principal portion of the old province of Udyana, have hitherto been inaccessible for archaeological research, the following rough notes (. . .) may induce others better qualified to devote some attention

“Tis all here. A treasure locked” 53 to this interesting neighbourhood. They are principally connected with the travels of the Chinese pilgrim Huan Tsiang, as given in Beal’s “Buddhist Records of the Western World.”17 How absolute and uncritical Deane’s dependence is on Xuanzang’s record is shown by the next sentence in his article: The Pilgrim says (. . .) that there were 1400 old Sangharamas on the banks of the river Su-po-fa-sutu [sic!], the present Swat river. This was probably not an exaggeration, as ruins are now found all through the country. Unfortunately, however, the majority lie in Upper Swat, which is at present closed to Europeans.18 Despite the fact that other Chinese travellers like Faxian, Song Yun19 and Huichao left information about the region, Xuanzang’s position as the most reliable and comprehensive informant had already been established to a degree that Deane could just ignore these other sources. We also see the typical treatment of the text: either Xuanzang’s data are correct and differences are interpreted away, or the information in Xuanzang’s Datang-xiyu-ji 大唐西域記 (“Records of the Western Regions”) – as has been observed, for instance, in the case of Cunningham – is tweaked to corroborate with the topographical or archaeological situation. The next major step in the exploration of and collection of data in the Swat valley, again following Xuanzang’s “footsteps” in almost exactly the same way as Cunningham, is Aurel Stein.20 Different from Deane – to whom one of Stein’s monographs (1929) on the region is dedicated – Stein, as usual without almost no hesitation and doubt, identified most of the monuments mentioned in the Chinese travelogues in situ. In Swat Stein’s “Orientalist” soul becomes visible because, despite the fact that he mostly still relied on the Chinese travelogues and particularly on Xuanzang,21 he confessed and claimed: What drew my eyes so eagerly towards Swat was not merely the fame that this ancient region, the ancient Uḍḍiyāna, had enjoyed in Buddhist culture, nor the traces that early worship and culture were known to have left in numerous as yet unsurveyed ruins. Nor was it only the wish to find myself back on the tracks of those old Buddhist pilgrims who travelled from China to the sacred sites of Swat, and whose footsteps I have had the good fortune to follow in the course of my expeditions through the desert wastes of Innermost Asia and across the high ranges of the Pamirs and

54  Max Deeg Hindukush. May the sacred spirit of old Hsüan-tsang, the most famous of those pilgrims and my adopted “Chinese patron saint”, forgive the confession: what attracted me to Swat far more than such pious memories was the wish to trace the scenes of that arduous campaign of Alexander which brought the great conqueror from the foot of the snowy Hindukush to the Indus, on his way to the triumphant invasion of the Panjab.22 There is, in my view, a slight shift in the approach of Stein’s successor in the region, Giuseppe Tucci, giving the archaeological evidence more weight than the Chinese records.23 Although identification of the sites mentioned in the Chinese travelogues still plays a considerable role in the 1958 report, in his 1977 paper on Swat Tucci is, as Stein before him, more interested in Alexander’s campaign in the region and other issues. If the positivist “reading” of the travelogues discussed so far is inadequate, how then should these sources be read and used? My suggestion is to read both textual and archaeological evidence in relation to other textual sources (epigraphic, numismatic), as scant as it may be, and to take into account the evidence in situ and chronological considerations. We then may come to sounder results with “softer” conclusions than either by ignoring the records completely or by overusing them in relation to identification of sites and individual artefacts. I would like to demonstrate the suggested approach by two case studies in the context of Swat. They both make use of narratives in the Chinese travelogues which are connected to places in the Swat valley. One of them is the story of the Buddha’s relics brought back to Swat by King Uttarasena, and the other example is on the location – or rather localization – of jātakas or previous birth stories in the Swat valley. The first example brings evidence from the textual collections, the Chinese travelogues and epigraphy, into dialogue with material collections, and the second example then exemplifies the possible consequences for the historical interpretation of data found in the textual corpus in case of silence of the other collection. My first example is restricted to Xuanzang’s report and is not found in any other source. It is the story of the king of Udyāna, Uttarasena, the son of an escaped survivor of King Virūḍhaka’s massacre of the Śākyas in Kapilavastu and a local nāginī. Uttarasena, we are told, receives direct permission from the Buddha on the latter’s return from subduing the nāga Apalāla in the upper Swāt valley to collect his share of the Buddha’s relics after his parinirvāṇa.24 Uttarasena’s stūpa of the Buddha’s relics is, according to Xuanzang, sixty or seventy Chinese

“Tis all here. A treasure locked” 55 miles (ca. 30–35 km) southwest of the capital of Udyāna (i.e. from today’s Mingora). Although this story is not found in any other Buddhist source, there is a possibility to create links between the narrative reported by Xuanzang and other information that can be collected from Swat. There is, first of all, the importance of relic veneration in the region, clearly shown by the existence of stūpas and (inscribed) reliquaries and art historical evidence of such a cult from the region from an earlier period. Xuanzang’s narrative could be read in a very general way as a post ex facto legitimization of the flourishing and vibrant relic cult in the valley. But the most striking point is that a possible connection can be made with concrete data coming from archaeology and epigraphy in the form of the well-known Senavarman inscription on a Buddhist reliquary. The document originally was in a private collection, and its provenance is not clear but there are indications that it may have been found in Swat.25 The Kharoṣṭhī inscription, written in Gāndhārī on a gold plate, presents us with a relatively complete lineage of the Iṣmaho royal family of Oḍi, first edited, translated and commented on by Bailey (1980), then by Fussman (1982), Salomon (1986) and von Hinüber (2003). The link between the Uttarasena legend in the Xiyu-ji and the inscription is an early king in the dynastic lineage called Utaraseṇa (Skt. Uttarasena).26 The second important and more recent evidence of a link between an actual historical dynasty in the region and Xuanzang’s story is the interpretation of the name Iṣmaho as being a Gāndhārī form of Skt. Īkṣvāku occurring in a Gāndhārī text fragment and in the Senavarman inscription.27 This shows that the Oḍi dynasty obviously claimed the same solar descendent – Īkṣvāku – as attributed to the Śākya clan in Buddhist literature. What the Senavarman inscription indicates is that there was indeed a relic claimed to be the Buddha’s, and it may well be, as Richard Salomon assumes – and I think he has a very good point in the light of all the evidence – that the caitya referred to in the inscription and the Uttarasena stūpa in Xuanzang’s Datang-xiyu-ji (“Records of the Western Regions”) here were identical or at least connected by the enshrined relics. If the dynastic lineage derived by Salomon28 from the Senavarman inscription is correct, the line of Uttarasena/Utaraṣeṇa was continued by his son Vasusena/Vasuṣeṇa but then moved to another competing side line: Bhadrasena – Ajitasena – Senavarman (Fussman 1982: 18; more assertively Salomon 1986: 287 and 289). Although we do not know the political events behind such a possible bifurcation of the

56  Max Deeg dynastic lineage, this seems to be supported and reflected by the narrative of the usurpation of power through Uttarasena’s father from the previous king of Udyāna in Xuanzang’s story. With the identification of Iṣmaho as Īkṣvāku, it is clear that there was no individual and historical “founder” Iṣmaho.29 It also seems evident that the dynastic line of Senavarman rather claimed a mythical solar origin.30 In the Senavarman inscription, however, there is no hint to a claimed Śākya ancestry. One possible explanation for this is that the legend in the Xuanzang’s “Records”, for unknown reasons and hundreds of years later, preserves a foundation narrative of the rivaling dynastic lineages of Uttarasena and Vasusena which may have interpreted the original, rather general, Īkṣvāku ancestry as having a more specific Buddhist Śākyan origin and further added the mythical element of nāga-descent of the royal house. What can be concluded from the evidence so far is that the dynastic foundation legend was in use between the period of a flourishing Buddhist culture in the Kuṣaṇa period and Xuanzang. The story was well established at the time of the redaction of the Mūlasarvāstivādavinaya: in the narrative of the Buddha’s travel to the Northwest during which he subjugated the nāga Apalāla in northern Swat in the Bhaiṣajyavastu of this Vinaya, we find the following brief allusion to the story when the Buddha travels back to Magadha after the conversion of Apalāla and place names are listed as a kind of itinerary: [dhānyapure uttarasenasya rājño mātā satyeṣu pratiṣṭhā]pitā (“At Dhānyapura the mother of king Uttarasena became convinced of the [four noble] truths.”).31 Yijing’s Chinese clearly supports Tucci’s reconstituted text (T.1448.40c.26f.): 世尊復到稻穀樓閣城。於此城中 化勝軍王母，令住四諦已。 (“The World-Honoured one then arrived in the city [called] “Rice Tower”. In this city [he] converted the mother of king “Supreme Army”32 [and] made her rest in the four [noble] truths.”).33 The existence of an assumed original relic of the Buddha in the region is clearly demonstrated by the Senavarman inscription in which this relic is elaborately eulogized. This relic obviously was considered an important element of the dynastic legitimation strategy, as another Gāndhārī-Kharoṣṭhī inscription by Senavarman’s father, Ajitasena, on a reliquary demonstrates the kings of Oḍi enshrined – thus copying on a minor scale Aśoka’s distribution of the relics – parts of the relics into stūpas in their realm, in the specific case into the great stupa (mahathuba) of Tira.34 The site of the stūpa the Senavarman inscription refers to was identified by Stein35 – and before him by Deane36 who obviously did not

“Tis all here. A treasure locked” 57 see the monument but had it described “by natives” – with the Shingardar stūpa in the village of the same name, near Ghalegai (Deane: Ghaligai; Stein: Ghalagai) northeast of Barikot (Stein: Bir-Kot) and more than 16 km southeast of Mingora. This identification was questioned by Tucci,37 who instead proposed Nawekalai (Naway Kalay), 500 m north of Kota, a village ca. 30 km (25 km according to Google Earth) southwest of Mingora, which according to him was abundant with archaeological remains.38 Both identifications bear their specific problems: Shingardar is far too close to Mingora to match Xuanzang’s distance of 60 Chinese miles. As far as I know no stūpa has been found around Nawekalai, although such an important relic must have been enshrined in a substantial one, even if Xuanzang’s height of more than 60 feet may be exaggerated. The Uttarasena stūpa may well have been even situated more to the west of Nawekalai, somewhere near Chakdara or Bhatkela. The origin of the Senavarman inscription – if the caitya referred to in the document is in fact linked to the Uttarasena stūpa in the Xiyu-ji – could have helped to clarify the issue, but because it is not known, the conclusions drawn by both Stein and Tucci must remain speculative. My second example deals with a cluster of self-sacrifice jātakas, stories of previous existences of the Buddha, located in the Gandharan plain and Udyāna.39 Here the textual basis is broader and allows some historical reconstruction insofar as the earlier records by Faxian and Song Yun/Huisheng can be read against what Xuanzang reports in and for the first half of the seventh century, whereas Huichao almost a century later only states that Buddhism was flourishing in the area without making reference to specific legends and sites. As in the Gandharan plain, the Chinese travellers locate places and stupas linked to former existences of the Buddha as narrated in the jatakas with Udyāna or Swat. Suheduo 宿呵多 (lower part of the Swat valley) Faxian, the earliest of the Chinese travellers, only reports of a stūpa dedicated to the place where the bodhisattva in his previous existence as Śivi (or Śibi) sacrificed himself in form of a dove.40 For the Gandharan plain (Jiantuowei 犍陀衛/Gandhāvatī) he locates the jatakas of Śibi’s eye sacrifice, of the bodhisattva’s sacrifice of his head (Zhuchashiluo 竺剎尸羅/Takṣaśilā) and of the hungry tigress (vyāghrī).41 Song Yun, travelling in Swat and Gandhara at the beginning of the sixth century, adds the narrative of Viśvantara (Biluo 鞞羅) to the legends of Śibi (?),42 of the hungry tigress, of the bodhisattva’s donation of his skin, the bodhisattva’s eye sacrifice and the sacrifice of his flesh to rescue a pigeon (T.2092.1019c.18ff.).

58  Max Deeg By Xuanzang’s time the jātaka sites have increased considerably: he records the place of the self-sacrifice of Kṣāntivādin close to the capital of Udyāna, the places of the bodhisattva’s self-sacrifice for a half-stanza of the dharma and of the prince Sarvadatta (or Sarvadada) in the mountains south of the capital, the place where the bodhisattva breaks his bones and uses his skin in order to write half a verse of a sutra, the places of the Śibi/Śivika-dove jātaka, the self-sacrifice of the bodhisattva as the serpent Sarvauṣadhi, of the spring miracle performed by the bodhisattva as a peacock king and the place of the Maitrībala jātaka according to which the bodhisattva fed five yakṣas with his blood. Because jātakas are, in principle, not fixed in terms of location, they can be linked quite easily with particular locations. In some cases there are elements in the narrative which may have led to such a localization, and I would like to demonstrate this in the case of the story of Kṣāntivādin (Pāli Khantivādī, Chin. Chandiboli 羼提波梨, Renruxianren 忍辱仙人, etc.), the story of the enduring ascetic who is cut into pieces by King Kalibhū (Pāli Kalābu, Chin. Geli 歌利, Jiali 迦利, etc.). Xuanzang gives a brief account of this jātaka in the third fascicle of his “Record”: Four or five miles to the east of the city of Mengjieli43 there is a big stūpa, [where] miraculous phenomena are very frequent; it is here, when the Buddha in former times was the saint “Enduring Humiliation”,44 [where] king Jieli45 (. . .) cut off [his] limbs.46 Some versions of the story may have facilitated its relocation in the mountainous regions in general and more particularly in Swat. In the Pali version Kṣāntivādin, Pāli Khantivādī hails from Takṣaśilā, Pali Takkasilā (no.313, Khantivādī-Jātaka; Fausbøll 1963: 39ff.), and in the Mahāvastu (Sénart: 3, 357ff.; Jones: 3, 355ff.) he originates from Uttarakuru so that the link with a northwestern or northern region would have been deducible from these texts. In the Sengjialuocha-suo-ji-jing 僧伽羅剎所集經, *Saṅgharakṣa-samuccayasūtra (?), translated by Saṅghabhadra/Sengjiafadeng 僧伽跋澄 (fl. after 380), the king goes on a hunting tour deep in the mountains (shenshan 深山; T.194.118a.25ff.). In the Xianyu-jing 賢愚經, “Sūtra of the Wise and the Fool”, translated by Huijue 慧覺 (fl. around 445) and others, although the story is clearly located in Benares (Boluonai 波羅奈), the king together with his harem undertakes a pleasure trip to the mountains (shanlin 山林) (T.202.359c.21ff.).

“Tis all here. A treasure locked” 59 It is, of course, difficult to decide what triggered or justified the final translocation of the story from Benares to a very individual place in Swat. I think, however, that the version in the Mūlasarvāstivādavinaya gives another clue for the location being exactly in the region of Udyāna: there the saint Kṣāntivādin settles down in a garden that belonged to his father (pituḥ santakam udyānaṃ gataḥ; Gnoli 1978: 8, line 2), and it is there where he is mutilated by his younger brother, King Kalibhū. A similar point can be made for the Pali version where the ascetic is invited to the royal garden (rājuyyāne) in which the whole story of mutilation unfolds. In the light of the “folk-etymological” interpretation of the place name Uḍḍiyāna in its “regularized” form, Udyāna as “garden”,47 it seems totally plausible that the appellative udyāna was purposely interpreted as a place name and the jātaka more easily translocated to Swat. Another indirect clue for a localization of this legend and others in the wider region of the northwest is given by the set of stories of the Buddha’s, that is, the bodhisattva’s, self-sacrifices alluded to in the Rāṣṭrapālaparipṛcchā.48 The first couple of the jātakas this text refers to include not only the one of Kṣāntivādin but also others located in the wider area, like the one of the hungry tigress (Vyāghrījātaka), the one of King Śibi donating parts of his body to save the life of a dove, etc.; the cluster of stories – also found in the anonymous Pusa-benxing-jing 菩薩本行經 (T.155.119a.26ff.) from the time before the Eastern Jin 東晉 (317–420) – seems to indicate their common location in the northwest of India. Being remarkably absent in the oldest translation of the sūtra by Dharmarakṣa/Zhu Fahu 竺法 護 (fl. ca. 265–310), the Deguang-taizi-jing 德光太子經 (T.170), it is found in the Skt. version, the Tibetan and the later Chinese translation Huguo-pusa-hui 護國菩薩會 (T.310.461b.25ff.) by Jñānagupta/ She’najueduo 闍那崛多 (523–600), who himself hailed from Gandhara, and an even later one by Dānapāla/Shihu 施護 (fl.982–1017), who came directly from Udyāna. Jñānagupta’s translation lists, in this sequence, Visvantara, Kṣāntivādin, Śyāma, the story of the bodhisattva using his skin as parchment for writing a half-stanza of the dharma, Vyāghrī, Śibi-dove, etc.49 A similar series of narratives is alluded to in the Dasheng-bensheng-xindi-guan-jing 大乘本 生心地觀經/*Mahāyāna-jātaka-cittabhūmi-samādhi-sūtra translated by Prajña/Banruo 般若 (744–ca. 810), who hailed from Gandhāra (T.159.249a.13ff.): Viśvantara, Vyāghrī, Śibi-pigeon, Mṛga-Jātaka and the self-sacrifice for a half-stanza of the dharma; in this text these jātakas are explicitly located in the Snow Mountains, xueshan 雪山.

60  Max Deeg This textual situation and the fact that the early Chinese travellers like Faxian and Song Yun visited only some of the sites related to jātakas whereas Xuanzang refers to a greater number of them seems to indicate a transfer or localization of some of the stories to the wider region between the third and the fifth centuries. The almost complete lack of clear textual,50 archaeological or art historical51 evidence – as far as I am aware – of the presence of these self-sacrificial jātakas in the region seems to suggest the following scenario: in the late Kuṣaṇa period events from former existences of the Buddha were localized, probably a little bit later than in the Gandharan plains and possibly triggered by them, in the Swat region. This may have been connected to an old localization of the legend of the Buddha’s visit and subjugation of the nāga Apalāla in the area and the vibrant relic cult during the first centuries of the Common Era.52 This time scale is also supported to some extent by the prominence of selfsacrificial motifs (dehadāna; Yaldiz: “Aufopferungslegenden”)53 in the rock art along the Karakorum highway54 and in the art of the Northern Silk Road55 and the fact that the bulk of Chinese translations of jātaka collections stems from the period between the late third and the early sixth centuries. This reading of the historical development would also explain why the most prominent Buddhist site in the Swat valley, Butkara, is not prominently mentioned and linked to specific narratives in the travellers’ accounts56 because it flourished before these developments and was, as an older sacred place, not “free” to have a story located at it. At the beginning these localizations of the previous existences of the bodhisattva in the region seem to have been rather restricted. Faxian’s witness of only one site in the wider region of the Swat valley seems to reflect this. But the localization of jātakas, especially those of selfsacrificial content, increased constantly until the seventh century when Xuanzang visited the region. Briefly afterwards the meaning ascribed to the sites – possibly with the political decline of the Kuṣaṇa and the lack of patronage – seems to have fallen into oblivion again. Nothing is known and preserved – at least not recognizably – of these localized jātakas in terms of material culture after Xuanzang’s visit; the silence of the early eighth-century traveller Huichao may be taken as a witness of this changed situation and can be explained by the rise of esoteric Buddhism in the region when a different iconographic programme is expressed in form of the Hindu Śāhi rock images studied by Anna Filigenzi57 and other scholars. In historical terms and with regard to content and context, the Xuanzang report sits well in the middle of the thriving period of assumed

“Tis all here. A treasure locked” 61 decline of Buddhism after the Kuṣaṇa period58 and before the stronger evidence of esoteric Buddhism in the region,59 already indicated by Xuanzang’s somewhat dismissive remark on the use of magical spells in the region and then attested to by Huichao at the beginning of the eighth century. Collections – museal, archaeological and textual – are doomed to be static, if not silent, if they are not allowed to communicate with each other. Hopefully the examples from the Swat and Gandhara were able to demonstrate how such a contextualization, in this specific case coming from the textual side, can lead to the reconstruction of possible historical scenarios which can and should be explored further and may enable understanding and be able to enrich the presentation and exposition of other collections.

Notes 1 Rudyard Kipling, Kim, Doubleday, Page & Company, New York, Toronto, 1921: 13. 2 Stanley Abe, Inside the Wonder House: Buddhist Art and the West, in Donald S. Lopez edited, Curators of the Buddha: The Study of Buddhism under Colonialism, The University of Chicago Press, Chicago, London, 1995: 64ff. 3 Tapati Guha-Thakurta, Monuments, Objects, Histories : Institutions of Art in Colonial and Postcolonial India, Columbia University Press, New York, 2004: 43ff. 4 Rudyard Kipling, Kim, New York, Toronto: Doubleday, Page and Company, 1921: 11. 5 See Max Deeg, When Peregrinus Is Not Pilgrim: The Chinese “Pilgrims” Records – A Revision of Literary Genre and its Context, in Christoph Cueppers and Max Deeg edited, Dharmayātra – Buddhist Pilgrimage in Time and Space, Lumbinī International Research Institute, Lumbinī, 2014: 65–95. 6 Jürgen Osterhammel, The Transformation of the World. A Global History of the Nineteenth Century, Princeton University Press, Princeton, 2014: 32. 7 Foucault 2002: 7f. 8 Various historical overviews and critical analyses of early modern colonial Indian archaeology have been published in the last few decades of which I only want to point out D.K. Chakrabarti, History of Indian Archaeology: The Beginning to 1947, Munshiram Manoharlal Publishers, Delhi, 1995; T. Guha-Thakurta, Monuments, Objects, Histories, 2004; Singh, The Discovery of Ancient India: Early Archaeologists and the Beginnings of Archaeology, Permanent Black, New Delhi, 2004; and H.P. Ray, Archaeology and Empire: Buddhist Monuments in Monsoon Asia, Indian Economic and Social History Review, 45.3, 2008: 517–449 and H.P. Ray, The Return of the Buddha: Ancient Symbols for a New Nation, Routledge, New Delhi, 2014.

62  Max Deeg 9 An example of this “timeless” interpretation of the Chinese records is Cunningham’s use of them (Janice Leoshko, 2003, Sacred Traces: British Explorations of Buddhism in South Asia, Aldershot, Burlington: Ashgate: 43), especially of Xuanzang in combination with the Greek records about Alexander’s failed conquest of northwest India, with both sources allowing Cunningham to apply them like a historical hermeneutical “bracket” of the regions covered by them. 10 See Leoshko, 2003, Sacred Traces: 42ff. 11 Abu Imam, 1966, Sir Alexander Cunningham and the Beginnings of Indian Archaeology, Asiatic Society of Pakistan, Dacca: VI. 12 Robin Coningham, The Archaeology of Buddhism, in Timothy Insoll edited, Archaeology and World Religion, Routledge, London and New York, 2001: 61–95. 13 Dilip Chakrabarti, India – An Archaeological History. Palaeolithic Beginnings to Early Historic Foundations, (Second Edition), Oxford University Press, New Delhi, 2015: 1. 14 In this Orientalist scheme of interpretation, the textualists are often seen as the representatives of the colonial “Orientalists” without really defining where the archaeologists like Cunningham and Fergusson are to be positioned other than that the former gave into the predominance of the text; see e.g. Guha-Thakurta, Monuments, Objects, Histories, 2004: 35f. 15 See e.g. Luca Maria Olivieri and Massimo Vidale, Archaeology and Settlement History in a Test Area of the Swat Valley: Preliminary Report on the AMSV Project (1st Phase), East and West: 56.1–3, 2006: 73–150. 16 The valley is still of interest to Tibetan Buddhists because it is the home place of the Tantric master Padmasaṃbhava. 17 H.A. Deane, Note on Udyāna and Gandhāra, Journal of the Royal Asiatic Society, 28.4, 1896: 655–677. 18 Deane, Note on Udyāna and Gandhāra, 1896. 19 Song Yun’s report is only referred to in the context of Kafiristan (Deane 1896: 662) about which Xuanzang has little to report. 20 An overview of Stein’s activities in the region is given in Rienjang, Wannaporn, Aurel Stein’s Work in the North-West Frontier Province, Pakistan, in Helen Wang edited, Sir Aurel Stein, Colleagues and Collections, The British Museum, London, 2012: 1–10. Stein’s first survey of a part of the region was published in 1905. In the first volume of Serindia, Detailed Report of Explorations in Central Asia and Westernmost China, Volume I, Text, Oxford University Press, Clarendon, 1921 (1921: 1ff.) he discusses Swāt at some length. 21 Aurel Stein, On Alexander’s Track to the Indus: Personal Narrative of Explorations on the North-West Frontier of India Carried out under the Orders of H.M Indian Government, Macmillan and Co. Limited, London, 1929: 14f. In Stein, Serindia, he even discusses the Chinese sources into two chapters, “Early Chinese Pilgrims to Udyāna” (p.5ff.) and “Udyāna in Chinese Records of T’ang Time” (p. 14ff.), were are dedicated almost exclusively to Xuanzang. 22 Stein, On Alexander’s Track to the Indus, 1929: 1f. 23 This is already indicated in Guiseppe Tucci, Travels of Tibetan Pilgrims in the Swat Valley, The Greater India Society (Greater India Studies, no. 2),

“Tis all here. A treasure locked” 63 Calcutta, 1940: 1: “our only informants are the Chinese pilgrims, but the description which they have left of the place is not very much detailed”. 24 For a detailed discussion see Deeg, 2011: 194ff. 25 Richard Salomon, The Inscription of Senavarma, King of Oḍi, Indo Iranian Journal, 29, 1986: 261. Stefan Baums informed me that the piece has been lost recently. 26 See the dynastic family tree in Salomon, The Inscription of Senavarma, 1986: 289. 27 Richard Salomon and Stefan Baums, Sanskrit Īkṣvāku, Pāli Okkāka, and Gāndhārī Iṣmaho, The Journal of the Pali Text Society, 39, 2007: 201–227. 28 Salomon, The Inscription of Senavarma, 1986: 288f. 29 As assumed by Gerard Fussman, Documents épigraphiques kouchans (III): L’inscription Kharoṣṭḥī de Senavarma, roi d’Oḍi: une nouvelle lecture, Bulletin de l’École Francaise de la Extreme Orient, 71: 7 1982, but relativized in Fussman 1986: 7, as “ancêtre mythique”; Salomon leaves out the name completely, and Oskar von Hinüber, Beiträge zur Erklärung der Senavarma-Inschrift. Stuttgart: Franz Steiner Verlag (Akademie der Wissenschaften und der Literatur, Abhandlungen der geistes- und sozialwissenschaftlichen Klasse, Jahrgang 2003, Nr.1: 33), puts it in parentheses. 30 I cannot help but point out a possible continuity between the Īkṣvāku kings of Udyāna/Oḍi and the solar kingship and symbolism of the late Turki Śāhi rulers in the valley: see Anna Filigenzi, Sūrya, the Solar Kingship and the Turki Śāhis: New Acquisitions on the Cultural History of Swat, East & West, 56.1.3, 2006: 195–203; Post-Gandharan, non-Gandharan: An Archaeological Inquiry into a Still Nameless Period, in Michael Alram, Deborah Klimburg-Salter, Minoru Inaba, and Matthias Pfisterer edited, Coins, Art and Chronology II: The First Millenium C.E. in the IndoIranian Borderlands, Verlag der Österreichischen Akademie der Wissenschaften, Wien, 2010a: 407–427; & Post-Gandharan Swat, Late Buddhist Rock Sculptures and the Turki Śāhis’ Religious Centres, Journal of Asian Civilizations, 34.1, 2011: 193–201. 31 Guiseppe Tucci, Preliminary Report on an Archaeological Survey in Swat, East & West, 9, 1958: 327a., note 24, restoring from the edited text: dhānyapu[ram anuprāptaḥ; dhānyapure senarājaḥ paramasatyeṣu prati] ṣṭhāpitaḥ. (Bagchi 1, 9, line 4). 32 Shengjun 勝軍 normally is the translation of Prasenajit, the contemporary of the Buddha and king of Kosala. But here it is clearly used for Uttarasena (sheng 勝 in the meaning of “excellent, utmost”). 33 Jean Przyluski, Le Nord-Ouest de l’Inde dans le Vinaya des Mūla Sarvāstivādin et les textes apparentés, Journal Asiatique, (Novembre‑Décembre), 1914: 513. 34 Fussman, Documents épigraphiques kouchans., 1986: 9, who tentatively identifies this place name with Tirah southwest of the Khyber Pass. 35 Stein, On Alexander’s Track to the Indus, 1829: 49ff. 36 Deane, Note on Udyāna and Gandhāra, 1896: 660. 37 Tucci, Preliminary Report on an Archaeological Survey in Swat., 1958: 299f.; see also Filigenzi, Sūrya, the Solar Kingship and the Turki Śāhis., 2006: 197 & Post-Gandharan Swat., 2011: 199.

64  Max Deeg 38 Tucci, Preliminary Report on an Archaeological Survey in Swat., 1958: 300ff.; see also Salomon, The Inscription of Senavarma., 1986: 289. 39 For the localization of jātakas due to their general “placeless-ness” – which is not completely true outside of the Pāli collection – see Appleton, A Place for the Bodhisatta., 2007. 40 Deeg, Das Gaoseng-Faxian-zhuan, 2005: 226f. and 521 (German translation). 41 Deeg, Das Gaoseng-Faxian-zhuan, 2005: 228ff. and 521f. (German translation). 42 I am not fully convinced that the place of saduo 薩埵, Skt. (mahā)sattva, refers to the famous story of the bodhisattva feeding his body to the tigress – as suggested by Chavannes 1903: 407, note 8, and maintained by Wang 1984: 228 – which, according to Song Yun himself, was eight days’ journey to the southeast of the capital and therefore probably beyond the mountain range in the Gandharan plain east of Takṣaśilā and therefore at some distance from the place which Faxian and Xuanzang propose for the jātaka of the tigress. As a whole, Song Yun’s location of the different jātakas seems to be a bit confused. 43 The area of modern Mingora. 44 Renru(-xian) 忍辱(仙): (ṛṣi) Kṣāntivādin. 45 羯利 / *kɨat-lih, Skt. *Kali. 46 T.882b.24ff. 瞢揭釐城東四五里有大窣堵波，極多靈瑞，是佛在昔作忍辱 仙，於此為羯利王(唐言鬪諍。舊云哥利，訛也)割截支體。 47 E.g. in Xuanzang’s biography by Huili (T.2053.230b.14): 唐言苑，昔阿輸 迦王之苑也。 (“In the language of the Tang [this means] “garden”, [and] in the past this was the garden of king Aśoka.”). 48 See Daniel Boucher, Bodhisattvas of the Forest and the Formation of the Mahāyāna: A Study and Translation of the Rāṣṭrapālaparipṛcchā-sūtra, University of Hawai’i Press, Honolulu, 2008: 26ff. 49 For the translation of the Skt. see Daniel Boucher, Bodhisattvas of the Forest and the Formation of the Mahāyāna, 2008: 131f., Skt. text in Finot 1901: 21f.; cp. also Dānapāla’s translation, T.321.5a.24ff. 50 The only local reference in the few previous birth stories (pūrvayoga) in Gāndhārī, edited by Timothy Lenz, A New Version of the Gāndhārī Dharmapada and a Collection of Previous Birth Stories: British Library Kharoṣṭhī Fragments 16 + 25. Seattle, London, University of Washington Press, Gāndhāran Buddhist Texts: Vol. 3), 2002, is that one of them, reflecting a more regional color, is located in Takṣaśilā (no.5, p. 182ff.), and another is the story of Viśvantara; the latter, however, contains no location. 51 For a comparative discussion of the respective stories (Kṣāntivādin, Sudāna, Viśvantara, Vyāghrī, Śibi-pigeon and Maitrībala) and their art historical representation see Dieter Schlingloff, Ajanta: Handbuch der Malereien / Handbook of Paintings – Erzählende Wandmalereien / Narrative WallPaintings, 3 vols., Otto Harrassowitz, Wiesbaden, 2000 (English version: Delhi 2013). 52 See the huge number of reliquaries from Swat listed in the appendix of David Jongeward, Elizabeth Errington, Richard Salomon and Stefan Baums, Gandharan Buddhist Reliquaries, University of Washington Press, Seattle and London (Gandharan Studies I), 2012: 254ff.

“Tis all here. A treasure locked” 65 53 On this type of legends see generally Reiko Ohnuma, Head, Eyes, Flesh and Blood: Giving Away the Body in Indian Buddhist Literature, Columbia University Press, New York, Chichester, 2007. 54 Jason Neelis, La Vieille Route Reconsidered: Alternative Paths for Early Transmission of Buddhism Beyond the Borderlands of South Asia, Bulletin of the Asia Institute, 16, 2006: 143–164; on the Jātaka representations (Śibi, Vyāghrī) see Volker Thewalt, Jataka-Darstellungen bei Chilas und Shatial am Indus, in Peter Snoy edited, Ethnologie und Geschichte – Festschrift für Karl Jettmar, Steiner, Wiesbaden (Beiträge der Südasienforschung, Bd. 86) 1983: 622–634. 55 See Marianne Yaldiz, Archäologie und Kunstgeschichte Chinesisch Zentralasiens (Xinjiang), E.J.Brill, Leiden and New York (Handbuch der Oirentalistik VII.3.2), 1987: 45ff. 56 On Butkara see, beside the detailed excavation reports by Facenna; an overview is given by Behrendt 2004: 47ff. and 99ff. The earliest period of the site (Butkara I) is dated to the second century BC: see Jongewald, Errington, Salomon, Baums, Gandharan Buddhist Reliquaries, 2012: 8 & 27. Some scholars suggest the identification of the so-called Tuoluo monastery (Tuoluo-si 陀羅寺) mentioned only by Song Yun as being situated north of the city (of Mingora?) with Butkara, but I do not find this very convincing. 57 Anna Filigenzi, A Vajrayanic Theme in the Rock Sculpture of Swat, in Giovanni Verardi, and Silvio Vita edited, Buddhist Asia: Papers from the First Conference of Buddhist Studies Held in Naples in May 2001, Italian School of East Asian Studies, Kyoto, 2003: 37–55, 2010a, 2011. 58 Filigenzi, The Shahi Period: A Reappraisal of Archaeological and Art Historical Sources, in Michael Alram, Deborah Klimburg-Salter, Minoru Inaba and Matthias Pfisterer edited, Coins, Art and Chronology II: The First Millenium C.E. in the Indo-Iranian Borderlands, Verlag der Österreichischen Akademie der Wissenschaften, Wien, 2010b: 407–427. 59 On archaeological evidence see Filigenzi, A Vajrayanic Theme in the Rock Sculpture of Swat, 2003.

References Abe, S. 1995. Inside the Wonder House: Buddhist Art and the West, in D.S. Lopez (ed.), Curators of the Buddha: The Study of Buddhism Under Colonialism. Chicago and London: University of Chicago Press, pp. 63–106. Appleton, N. 2007. A Place for the Bodhisatta: The Local and the Universal in Jataka Stories. Acta Orientalia 8.1: 109–122. Bailey, H.W. 1980. A Kharoṣṭhī Inscription of Seṇavarma, King of Oḍi. Journal of the Royal Asiatic Society 112.1: 21–29. Behrendt, K. 2004. Buddhist Architecture of Gandhāra. Leiden and Boston: Brill (Handbook of Oriental Studies 2.17). Boucher, D. 2008. Bodhisattvas of the Forest and the Formation of the Mahāyāna: A Study and Translation of the Rāṣṭrapālaparipṛcchā-sūtra. Honolulu: University of Hawai’i Press. Callieri, P. 2006. Buddhist Presence in the Urban Settlements of Swāt, Second Century BC to Fourth Century CE, in P. Brancaccio, and K. Behrendt (eds.),

66  Max Deeg Gandhāran Buddhism: Archaeology, Art, Texts. Vancouver and Toronto: UBC Press, pp. 60–82. Chakrabarti, Dilip K. 1995. History of Indian Archaeology: The Beginning to 1947. Delhi: Munshiram Manoharlal Publishers. Chakrabarti, Dilip K. 2015. India – an Archaeological History: Palaeolithic Beginnings to Early Historic Foundations, 2nd edition. New Delhi: Oxford University Press. Chavannes, É. 1903. Voyage de Song Yun dans l’Udyāna et le Gandhāra (518– 522 p.C.). Bulletin de l’École Franςaise de l’Extrême-Orient 3: 379–441. Coningham, R. 2001. The Archaeology of Buddhism, in T. Insoll (ed.), Archaeology and World Religion. London and New York: Routledge, pp. 61–95. Cunningham, A. 1871. The Ancient Geography of India, I: The Buddhist Period Including the Campaigns of Alexander, and the Travels of HwenThsang. London: Trübner and Co. Cunningham, A. 1875. Archaeological Survey of India: Report for the Year 1872–73. Calcutta: Office of the Superintendent of Government Printing. Deane, H.A. 1896. Note on Udyāna and Gandhāra. Journal of the Royal Asiatic Society 28.4: 655–677. Deeg, M. 2005. Das Gaoseng-Faxian-zhuan als religionsgeschichtliche Quelle: Der älteste Bericht eines chinesischen buddhistischen Pilgermönchs über seine Reise nach Indien mit Übersetzung des Textes. Wiesbaden: Otto Harrassowitz (Studies in Oriental Religions 52). Deeg, M. 2012. Secular Buddhist Lineages – The Śākyas and Royal Descendencies in Local Buddhist Legitimation Strategies. Religions of South Asia 5.1‑2 (guest editors: Simon Brodbeck, James Hegarty): 189–207. Deeg, M. 2014. When Peregrinus Is Not Pilgrim: The Chinese “Pilgrims” Records – A Revision of Literary Genre and its Context, in C. Cueppers, and M. Deeg (eds.), Dharmayātra – Buddhist Pilgrimage in Time and Space. Lumbinī: Lumbinī International Research Institute, pp. 65–95. Dharma Fellowship of His Holiness the Gyalwa Karmapa (anonymous, without year). Uddiyana Until the Eighth Century: A Short Historical Overview, www.dharmafellowship.org/library/essays/uddiyana.htm/ accessed on 16 March 2015. Fausbøll, V., ed. 1963. The Jātaka, Together with Its Commentary: Being Tales of the Anterior Births of Gotama Buddha, vol. 3. London: Pali Text Society, Luzac and Co. Ltd. Filigenzi, A. 2003. A Vajrayanic Theme in the Rock Sculpture of Swat, in G. Verardi, and S. Vita (eds.), Buddhist Asia: Papers from the First Conference of Buddhist Studies Held in Naples in May 2001. Kyoto: Italian School of East Asian Studies, pp. 37–55. Filigenzi, A. 2006. Sūrya, the Solar Kingship and the Turki Śāhis: New Acquisitions on the Cultural History of Swat. East and West 56.1–3: 195–203. Filigenzi, A. 2010a. Post-Gandharan, non-Gandharan: An Archaeological Inquiry into a Still Nameless Period, in M. Alram, D. Klimburg-Salter, M. Inaba, and M. Pfisterer (eds.), Coins, Art and Chronology II: The First

“Tis all here. A treasure locked” 67 Millenium C.E. in the Indo-Iranian Borderlands. Wien: Verlag der Österreichischen Akademie der Wissenschaften, pp. 389–406. Filigenzi, A. 2010b. The Shahi Period: A Reappraisal of Archaeological and Art Historical Sources, in M. Alram, D. Klimburg-Salter, M. Inaba, and M. Pfisterer (eds.), Coins, Art and Chronology II: The First Millenium C.E. in the Indo-Iranian Borderlands. Wien: Verlag der Österreichischen Akademie der Wissenschaften, pp. 407–427. Filigenzi, A. 2011. Post-Gandharan Swat: Late Buddhist Rock Sculptures and the Turki Śāhis’ Religious Centres. Journal of Asian Civilizations 34.1: 193–210. Foucault, M. 1989. The Archaeology of Knowledge. London and New York: Routledge (French original: L’archéologie du savoir, Paris: Edition Gallimard, 1969). Fussman, G. 1982. Documents épigraphiques kouchans (III): L’inscription Kharoṣṭḥī de Senavarma, roi d’Oḍi: une nouvelle lecture. Bulletin de l’École Franςaise de l’Extrême-Orient 71: 1–46. Fussman, G. 1986. Documents épigraphiques kouchans (IV): Ajitasena, père de Senavarma. Bulletin de l’École Franςaise de l’Extrême-Orient 75: 1–14. Guha-Thakurta, T. 2004. Monuments, Objects, Histories: Institutions of Art in Colonial and Postcolonial India. New York: Columbia University Press. Hinüber, O. von. 2003. Beiträge zur Erklärung der Senavarma-Inschrift. Stuttgart: Franz Steiner Verlag (Akademie der Wissenschaften und der Literatur, Mainz, Abhandlungen der geistes- und sozialwissenschaftlichen Klasse, Jahrgang 2003, Nr.1). Imam, A. 1966. Sir Alexander Cunningham and the Beginnings of Indian Archaeology. Dacca: Asiatic Society of Pakistan. Jones, J.J. 1956. The Mahāvastu (Translated From Buddhist Sanskrit), vol. 3. London: Luzac and Co. Ltd. Jongeward, D., E. Errington, R. Salomon, and S. Baums. 2012. Gandharan Buddhist Reliquaries. Seattle and London: University of Washington Press (Gandharan Studies I). Kipling, R. 1921. Kim. New York and Toronto: Doubleday, Page and Co. Lenz, T. 2002. A New Version of the Gāndhārī Dharmapada and a Collection of Previous Birth Stories: British Library Kharoṣṭhī Fragments 16 + 25. Seattle and London: University of Washington Press (Gandhāran Buddhist Texts, vol. 3). Leoshko, J. 2003. Sacred Traces: British Explorations of Buddhism in South Asia. Aldershot and Burlington: Ashgate. Neelis, J. 2006. La Vieille Route Reconsidered: Alternative Paths for Early Transmission of Buddhism Beyond the Borderlands of South Asia. Bulletin of the Asia Institute 16: 143–164. Neelis, J. 2011. Early Buddhist Transmission and Trade Networks: Mobility and Exchange Within and Beyond the Northwestern Borderlands of South Asia. Leiden and Boston: Brill (Dynamics in the History of Religions, vol. 2).

68  Max Deeg Ohnuma, R. 2007. Head, Eyes, Flesh and Blood: Giving Away the Body in Indian Buddhist Literature. New York and Chichester: Columbia University Press. Olivieri, L.M. 2008. Fallstudie Swat: Die Stadt Barikot und ihr Umfeld, in Kunst- und Ausstellungshalle der Bundesrepublik Deutschland GmbH (ed.), Gandhara: Das buddhistische Erbe Pakistans: Legenden, Klöster und Paradiese. Mainz: Verlag Philipp von Zabern, pp. 294–297. Olivieri, L.M., and M. Vidale. 2006. Archaeology and Settlement History in a Test Area of the Swat Valley: Preliminary Report on the AMSV Project (1st Phase). East and West 56.1–3: 73–150. Osterhammel, J. 2014. The Transformation of the World: A Global History of the Nineteenth Century. Princeton and Oxford: Princeton University Press. Przyluski, J. 1914. Le Nord-Ouest de l’Inde dans le Vinaya des MūlaSarvāstivādin et les textes apparentés. Journal Asiatique Novembre‑Décembre: 493–568. Ray, H.P. 2008. Archaeology and Empire: Buddhist Monuments in Monsoon Asia. Indian Economic & Social History Review 45.3: 517–449. Ray, H.P. 2014. The Return of the Buddha: Ancient Symbols for a New Nation. London–New York–New Delhi: Routledge. Rienjang, W. 2012. Aurel Stein’s Work in the North-West Frontier Province, Pakistan, in H. Wang (ed.), Sir Aurel Stein, Colleagues and Collections. London: The British Museum, pp. 1–10. Salomon, R. 1986. The Inscription of Senavarma, King of Oḍi. Indo-Iranian Journal 29: 261–293. Salomon, R., and S. Baums. 2007. Sanskrit Īkṣvāku, Pāli Okkāka, and Gāndhārī Iṣmaho. The Journal of the Pāli Text Society 39: 201–227. Schlingloff, D. 2000. Ajanta: Handbuch der Malereien/Handbook of Paintings – Erzählende Wandmalereien/Narrative Wall-Paintings, 3 vols. Wiesbaden: Otto Harrassowitz (English version Delhi 2013). Senart, É. 1897. Le Mahāvastu, vol. 3. Paris: Imprimerie Nationale. Singh, U. 2004. The Discovery of Ancient India: Early Archaeologists and the Beginnings of Archaeology. New Delhi: Permanent Black. Stein, M.A. 1905. Report of Archaeological Survey Work in the North-West Frontier Province and Baluchistan for the Period from January 2nd, 1904, to March 31st, 1905. Peshawar: Government Press, North-West Frontier Province. Stein, M.A. 1921. Serindia: Detailed Report of Explorations in Central Asia and Westernmost China, Volume I: Text. Oxford: Clarendon Press. Stein, M.A. 1929. On Alexander’s Track to the Indus: Personal Narrative of Explorations on the North-West Frontier of India Carried out Under the Orders of H.M Indian Government. London: Macmillan & Co. Ltd. Stein, M.A. 1930. An Archaeological Tour in Upper Swat and Adjacent Hill Tracts. Memoirs of the Archaeological Survey of India No. 42. Calcutta: Government of India Central Publication Branch. Stein, M.A. 1942. From Swat to the Gorges of the Indus. Geographical Journal C.2: 49–56.

“Tis all here. A treasure locked” 69 Stein, M.A. 1944. Archaeological Notes from the Hindukush Region. Journal of the Royal Asiatic Society (pt.1 & 2): 5–23. Thewalt, V. 1983. Jataka-Darstellungen bei Chilas und Shatial am Indus, in P. Snoy (ed.), Ethnologie und Geschichte – Festschrift für Karl Jettmar. Wiesbaden: Steiner (Beiträge zur Südasienforschung, Bd. 86), pp. 622–634. Tucci, G. 1940. Travels of Tibetan Pilgrims in the Swat Valley. Calcutta: Greater India Society (Greater India Studies, no. 2). Tucci, G. 1958. Preliminary Report on an Archaeological Survey in Swat. East & West 9: 279–328. Tucci, G. (with contribution from K. Enoki and B. Brentjes). 1977. On Swat: The Dards and Connected Problems. East & West 27: 9–103. Wang Yi-t’ung. 1984. A Record of Buddhist Monasteries in Lo-Yang, by Yang Hsüan-chih. Princeton: Princeton University Press. Yaldiz, M. 1987. Archäologie und Kunstgeschichte Chinesisch-Zentralasiens (Xinjiang). Leiden and New York: E. J.Brill (Handbuch der Oirentalistik VII.3.2).

3 Numismatics of ‘the Other’ Investigating coinage and ‘Greekness’ at Taxila Shailendra Bhandare Prologue Taxila, located in the Punjab province of present-day Pakistan, has been a focus of visualizing the ‘Greek’ in Indian history for a long time. Its beginnings lie in the Alexander narrative, reconstructed from Classical texts like Arrian’s Anabasis, which refer to an Indian ruler named ‘Taxiles’ who became Alexander’s ally. Coins enter the narrative at this juncture too – Taxiles is said to have paid Alexander a tribute of ‘thousand talents of coined silver’ (Plutarch, Alexander, 59 – “So, after receiving many gifts and giving many more, at last he lavished upon him a thousand talents in coined money. This conduct greatly vexed Alexander’s friends, but it made many of the Barbarians look upon him more kindly”).1 Taxiles also owed allegiance to Alexander in his campaign farther east of the Punjab, against his enemy Porus, and it famously culminated in the battle of the Hydaspes. The credit of identifying the city of Taxila goes to Alexander Cunningham, who, working from the survey of archaeological remains and mapping them on various textual accounts in Greek and Chinese, commented that ‘extensive ruins near Shah Dheri’ were the ‘the famous Taxila of the Greeks’ and indeed, ‘the ancient capital of the Panjab’.2 In the formative years of the study of Indian numismatics, the origin of coinage in India was a highly debated aspect. Many Western numismatists saw a familiarity with Greek coinage as a precondition to have triggered the development of indigenous coins in India, and thus dated the advent of coinage in India commensurate with the Indian campaigns of Alexander. But the mention of Classical writers like Plutarch of Taxiles paying ‘coined silver’ to Alexander meant the Indians knew of coined money even before Alexander, so it ended this debate. Alexander Cunningham declared, “The Hindus were in actual possession of a real coinage at the time of Alexander’s expedition”.3 The

Numismatics of ‘the Other’ 71 numismatic analysis of Cunningham set the tone for most future investigations and interpretations, as we shall see further in this chapter. Taxila and its constituent sites, such as the ‘Bhir Mound’, Sirkap, Sirsukh, Jandiyal, and Mohra Moradu, became the subject of extensive archaeological exploration and excavation under the directorship of Sir John Marshall over nearly three decades. The reports of the excavations further cemented the ‘Greek’ element in the narrative of Taxila.4 Central to the narrative was not only the ‘similarity’ of artistic archetypes, such as pillar capitals, or motifs such as the ‘double-headed eagles’, but also an urban plan which was thought to be essentially Greek. Marshall described it as a “typically Greek chess-board pattern, with streets cutting one another at right angles and regularly aligned blocks of buildings”.5 The excavations yielded an enormous amount of numismatic data, a lot of which still remains to be properly studied and published. Coins helped Marshall put his narrative of a ‘Greek’ Taxila in further perspective as can be seen in his assessment of the data. Although he seemingly acknowledged “there is nothing typically Greek about buildings, nor are there any remains of temples, altars, public monuments or statues such as we are accustomed to associate with the Greeks” in spite of the city being laid on a Greek grid plan, he posited the numismatic evidence into the narrative to be “where Greek Art manifests itself most prominently . . . the stylistic history of which is singularly lucid and coherent”.6 The narrative constructed by Marshall had a lasting impact on the history of Taxila in particular, and the history of the Greeks in India, or the narratives of ‘Hellenism’, in general. Taxila occupied the position of a node where cultures mixed and produced hybrids. The roots of such interpretations lay in Classical Studies, with nineteenthcentury Romantic revivalist historians like Johann Gustav Droysen, the father of the term ‘Hellenismus’. In the post-colonial discourse, however, the term ‘Greek’ was problematized, particularly in the aftermath of Edward Saïd’s ‘Orientalism’. It attracted criticism from several quarters, posited between the Droysenian idea of mischkultur and hybridity and Homi Bhabha’s Location of Culture. It also attracted attention from a more incisive post-colonial inquiry into the concept of ‘identity’ in the ancient world and what its connections were with ethnicity.7 My chapter attempts to unpick the role coins played in the narrative, particularly from the viewpoint of numismatic analysis and interpretation which in turn led to the archaeological inferences which the like of Marshall drew from the data. I will also present new numismatic data to argue for a different interpretation, which in turn casts a

72  Shailendra Bhandare doubt on earlier inferences, specifically the archaeological chronology of various sites under the umbrage of Taxila. I hope the analysis and data I present here will help further deconstruct the notion of what exactly ‘Greek’ means in the context of Taxila and feed it further into the debate about Essentialism in ancient history and archaeology. The numismatic chronology of Taxila Alexander Cunningham was responsible for creating a numismatic category titled ‘coins of Taxila’ in his book Coins of Ancient India, published in 1891.8 By this time, the knowledge that the silver ‘punch-marked coins’ were the earliest of Indian coins had been well established. Cunningham described two such coins and remarked on a peculiar mark they carry on the reverse (for a similar coin, see Figure 3.1), conjecturing that it could be “the mark or sign of Taxila”. One of the reasons for this conjecture is that it also occurred on a gold coin he reportedly obtained at Taxila, now in the British Museum collection. Cunningham then described a range of “square copper pieces of single dies, and of the standard peculiar to India” (for example, see Figure 3.2) and commented on the symbols they carry. Evident in the commentary is the fact that Cunningham sees many of these symbols as Buddhist or Hindu – the chaitya and ‘Bodhi Tree’, ‘plan of a monastery with its cells’, ‘a building which may be a temple’, and so on. He then moves to ‘double-die pieces’, the salient of which are ‘Elephant and Lion types’ (Figure 3.3). He also described a number of other types in this category – most of them uninscribed, but a few carrying inscriptions indicating they were struck by Nigamas, or guilds.

Figure 3.1 Punch-marked coin with ‘Taxila’ mark on reverse Source: All images have been taken from Heberden Coin Room, Ashmolean Museum, University of Oxford, UK. Used with permission.

Numismatics of ‘the Other’ 73

Figure 3.2 Uniface die-struck copper coin attributed to Taxila

Figure 3.3 An ‘elephant and lion’ coin of Taxila

The internal chronology of these ‘local’ coins is suggested by Cunningham on the basis of a “small parcel of 143 square copper coins”, which was found in 1884 at Shah-ji-ki-dheri. He was able to examine 135 of these and managed to acquire 127 for his own collection. The breakup of the 135 was as follows: • Eighty-four Indian coins of one face only – average weight 140.8 grain • Twenty-seven Indian coins of two faces – average weight 183.3 grains

74  Shailendra Bhandare •

Twenty-four Greek (nine Pantaleon and fifteen Agathocles) – average weight 180.5 grains

Cunningham identified the single-face coins to be struck on the ‘Indian standard of 80 ratis, or 144 grains’, whereas the two-face coins, ‘both Indian and Greek’, were struck to the heavier standard of 100 ratis, or 180 grains. He then remarks, As all these coins were found together, they must have been current at the same time. But as the greater number are of the Indian standard, I infer that they must belong to the indigenous coinage prior to the Greek occupation. I infer also that the deposit must have been at an early period of the Greek rule, as shown by the presence of the two early Greek kings, Pantaleon and Agathokles. The Greek coins preserve the Indian type of the Lion, as well as the square shape.9 Cunningham’s analysis ends here, but it sets the tone for all future assessments of coins found at Taxila. He obviously regards some of the coins to be ‘Indian’ and others to be ‘Greek’. The reason why they are ‘Indian’ is conceivably because they have ‘Indian’ (Buddhist/ Hindu) symbols, a different shape – rectangular rather than round – and a weight corresponding to the ‘Indian standard’. The coins of Pantaleon and Agathocles do have inscriptions, which are in Greek and Brahmi, whereas the Indian coins do not. As such, and because they are more numerous in the parcel, Cunningham contended that they must antedate the ‘Greek’ coins. This is a very clear example of an antiquarian method of taxonomic categorization, with ‘Essentialism’ about the inherent characters of the objects being classified as a guiding principle. The ‘Indian’ coins, depending on their physical features, are thus ‘Othered’ from being ‘Greek’, thereby creating ‘Greek’ and ‘Indian’ as numismatic categories. This numismatic categorization is based on a purely Essentialist premise: ‘Greek’ is manifest when coins are inscribed, round, neatly executed with animate motifs, and struck with dies, whereas they are regarded as ‘Indian’ when they are uninscribed, square, mainly symbolic in terms of motifs, and struck with either one or two dies. This ‘taxonomic Othering’, as I call it, later emerged as a fulcrum for classifying these coins which had a huge impact on how other kinds of data, particularly the archaeological data emerging out of Marshall’s digs, was analyzed and interpreted. I would like to emphasize here that the ‘taxonomic Othering’ I have referred to is, in a sense, an analytical rather than a ‘physical’

Numismatics of ‘the Other’ 75 categorization. If we see coins for what they are physically, of course, Cunningham was right: these coins bear the traits he recognized as such in terms of shape, size, weights, motifs, and so on. However, the way he treats them analytically and puts them in these pigeonholes of ‘Greek’ and ‘Indian’ is where the ‘Othering’ comes about. In doing so, Cunningham ostensibly does not take into account aspects beyond these physical attributes – like the circulatory context of the coins or how such aspects as shape could have been dictated by far more mundane considerations such as how the coin-manufacturing technology worked, and in response to what sort of circulatory demands the coinage were subjected to, rather than whether they were ‘Greek’ or ‘Indian’ in any way. The important consideration that a ‘coinage tradition’, which emerges out of a complex interplay of cultural, economic, and political factors, could well have been behind these physical aspects is also largely overlooked by Cunningham. The major fallout of the scheme of classification and the placement of the so-called ‘Indian’ coins immediately before the advent of the Greeks was the creation of a Taxila as an ‘independent state’ between the Mauryan and the ‘Greek’ periods in the historical narrative about Taxila. Thus, Taxila was taken to be a ‘state’ under its ruler Taxiles, the adversary of Porus and ally of Alexander. Then came the Mauryan interlude which was briefly followed once again by a ‘Taxilan’ state; this state then was taken over by the Greeks during the rule of Pantaleon and Agathocles, who made it a Greek outpost. Once the Greeks took over, in Marshall’s words, “for six long centuries . . . Taxila was never again to be free from the yoke of Western conquerors”. It is worthwhile to note that the key framework within which the numismatic evidence was analyzed and interpreted was provided by texts. These were often taken ad lib and ad hoc – for example, while discussing the Mauryan presence at Taxila beyond the rule of Asoka Marshall took the mention of a stupa dedicated to his son Kunala by the Chinese traveller Xuanzang as a chronological marker, even after acknowledging that the mention was clearly ‘apocryphal’10 and notwithstanding the fact that Xuanzang was writing nearly half a millennium later. This helped to bridge the gap between Mauryan Taxila and the advent of the Greeks in the late second century BCE under Pantaleon/Agathocles as Cunningham had propounded. This numismatic chronology for Taxila came to be generally accepted – barring minor discord with respect to some details – by Classical historians and numismatists. As Ancient Indian numismatics grew as a subject, the Taxilan coins came to occupy a categorical ‘time and space’ locus. In Marshall’s excavation reports, three numismatists,

76  Shailendra Bhandare namely E.H.C. Walsh, John Allan, and R. B. Whitehead, wrote separate reports on the coin data that were recovered. Marshall himself consolidated his views upon their writings, sometimes offering his own views. Exactly why Marshall enrolled three different authors to treat numismatic evidence is not known, but it appears to be the result of some of them having specialized knowledge about or interest in certain series of coins (for instance, Walsh dealt only with punch-marked coins). The contribution of these three, and of Marshall himself, with specific reference to the coin data excavated at Taxila will be discussed in more detail later. But before that, it will be worthwhile to see how other numismatists have treated what was regarded as the ‘Indian’ coinage of Taxila. Taxila before the Greeks: numismatic assessments The most significant numismatic treatment of the Taxila coinage came at the hands of John Allan, who published the catalogue of coins of Ancient India in the British Museum in 1936.11 Allan clearly stated that “it is on the authority of Cunningham that most of these coins in the (British) Museum collection are attributed to Taxila”. He then produced a classification much like Cunningham had done, splitting the bulk into ‘inscribed’ and ‘uninscribed’. In the latter category, he identified the ‘three-arched hill with crescent’ as a symbol ‘characteristic of Taxila’ which accompanies disparate symbols on the uninscribed die-struck coins, including ‘a plan of the courtyard of a monastery with cells around and a stupa in the centre’. On the basis of the symbolic configuration, he assigned them eight varieties: Var. a through Var. e, and Vars. h through j. Chronologically, Allan assigned all these coins to the first quarter of the third century BCE because certain varieties were ‘found in one pot amid ruins of Taxila. . . (with) coins of Pantaleon and Agathocles of Taxila fabric’. This is ostensibly a reference to the find Cunningham made which we have already discussed. Implicit in this statement is Allan’s recognition that the coins of the Greek rulers were in a ‘non-Greek’ fabric, thereby emphasizing the ‘Othering’ of the physical attributes. Allan then proceeded to describe another large group of uninscribed die-struck coins ‘of similar fabric’, attributing them to ‘Taxila’ on the basis of the repertoire of symbols they share with other such coins.12 One of them is the ‘three-arched hill with crescent’ which he had adjudged to be ‘characteristic of Taxila’ while describing the previous group. These coins bear the motifs of an elephant and a lion, both of which are depicted in association with the so-called ‘Taxila’ symbols

Numismatics of ‘the Other’ 77 and classified into Vars. a through e. On certain varieties, however, the symbol is replaced with a ‘plant’, as Allan recognizes it, and the lion replaced by a ‘galloping horse with a star’ above. These constitute Vars. e and g. The horse, according to Allan, is ‘familiar on the coins of Euthydemus and some other Greek kings’ and ‘the type of galloping horse is not an Indian one, nor is the star’. Allan also considers that ‘the idea of a main type with subsidiary symbol is Greek, and not Indian’ and ‘on purely Indian coins, all the symbols are the same size’. He then concludes that Vars. a through d of this group of coins ‘should be attributed to Greek kings in Taxila’. The fact that they were also represented in the pot found by Cunningham is taken as corroborative evidence. Because coins of Vars. e and g were not in the pot, they ‘may therefore be a little later in date than the former’. Allan’s analysis exemplifies the Essentialist approach – he clearly makes his mind up on what is ‘Greek’ and what is ‘Indian’ when it comes to classificatory characteristics of the coins. The corroboration he seeks also follows the same logic – some of the motifs on the coins are ‘Greek’ in his view: the lion is Pantaleon’s ‘favourite type’, the monogram on one of the coins which looks like ‘Ā’ is the ‘initial of Agathocles’, and ‘the star and the plant (as Allan identifies it) link these coins with those bearing the name of Agathocles’. These characteristics prompt Allan to remark that these coins might be Greek issues, even though the fabric of the coins is clearly ‘Indian’ as the Essentialist definition goes – they are rectangular, differently manufactured than the Greek coins, and carry no inscriptions. William Tarn, who wrote the narrative The Greeks in Bactria and India,13 imagined a different reality. He was convinced that the conquest of Taxila by the Greeks happened not close to Pantaleon or Agathocles, but their predecessor, Demetrios.14 Described famously as the ‘Conqueror of the Indians’ by Justin, the remarkable portrait of Demetrios on his silver coins wearing a headgear fashioned as an elephant scalp, as well the motif of an elephant wearing a bell in its neck, were both taken as numismatic substantiations of this event by Tarn. He also contended that Demetrios laid the foundations of a new ‘Greek’ city at Sirkap15 and entrusted his agenda of a southward conquest to his kinsman General Apollodotos.16 Tarn does not attribute the ‘Indian’-style coins of Agathocles to Taxila; Agathocles in his view ruled in a ‘sub-kingdom that lay primarily in Iran’.17 Instead, he explains the similarity between the bilingual coins of Agathocles to the ‘Indian’-style uninscribed coin by suggesting that Agathocles might have struck his coins because he was ‘putting into circulation under his own name more coins known and acceptable to the merchants of

78  Shailendra Bhandare Taxila’. So in effect, Tarn agrees that ‘Taxila’ possessed a coinage of its own prior to, or parallel with, its purported Greek conquest, but he does not take the similarity of the inscribed Greek coins of Agathocles and Pantaleon with the so-called ‘Taxila’ coins to suggest that they manifest a political interface at Taxila. With regard to the other types of ‘Taxila’ coins, Tarn has precious little to say. He does contend that the coinage continued after Demetrios’s conquest of the city, referring to Allan’s remark of some of the varieties being ‘a little late’ than those contained in Cunningham’s pot, which was deposited concurrently in the reigns of Agathocles/Pantaleon. Tarn’s quest to substantiate the Greek rule in Taxila takes him in the direction of finding a ‘Greek’ type associated with that city, which he contends ‘ought to be discoverable on Taxila’s own coinage’.18 He identifies this to be the ‘elephant’ type which is ‘infinitely commonest’ among the various types of Taxila coins. Tarn is conceivably referring here to the uninscribed ‘lion × elephant’ coins which he takes to be a part of the Taxila coinage. Conveniently for Tarn, the coins of Apollodotos are of ‘elephant × bull’ type as well, so it substantiates not only his identification of the elephant as a ‘Greek’ type associated with Taxila, it also fits neatly with the dynastic order (Demetrios the conqueror appoints Apollodotos the governor) he suggests in his narrative reconstruction. Tarn’s view that the Agathocles/Pantaleon issues were not struck in Taxila effectively dissociated the identity of these coins as markers of a political interface between the ‘independent state of Taxila’ and ‘Taxila under the Greeks’. This interface, according to Tarn, antedated the two rulers. However, Tarn regarded the ‘lion × elephant’ type coins to be of a ‘Taxila type’ which was struck almost concomitantly with the Agathocles/Pantaleon issues but under Apollodotos, who Tarn considered to be the Greek ruler of Taxila under Demetrios. Tarn’s views regarding the Greek succession are no longer tenable, but his dissociation of the Agathocles/Pantaleon coins with Taxila, in direct contradiction to Cunningham and Allan, is worthy of note. But it did not find favour with numismatists. They were guided by the apparent similarity between the fabric and the motifs of the two series – both were struck in an ‘Indian’ fabric in being rectangular and followed an ‘Indian’ standard, and the lion on the Agathocles/Pantaleon coins was evidently similar to the lion of the ‘lion × elephant’-type coins. Michael Mitchiner, while accepting the narrative of an ‘independent Taxila state’ in the immediate aftermath of the Mauryans, suggests a somewhat different treatment.19 According to him, what was clubbed together into a single category as the local/‘Indian’ coinage of Taxila

Numismatics of ‘the Other’ 79 falls under two categories: coins with ‘thick compact’ fabric and coins with ‘thin broad’ fabric. Each of these comprises many series, but the “influence of Greek art” is evident on only one of them, according to Mitchiner. He agrees that the coinage in both fabrics ends with the Greek king Agathocles striking what could be the terminal types in both fabrics. Mitchiner further suggests that each of these fabrics must be linked to the two premier cities of Gandhara: Taxila and Puṣkalavati. The ‘thick compact’ coins are attributed to Pushkalavati, and the ‘thin broad’ coins to Taxila. He bases his surmise on numismatic typology and the ‘amount of Greek influence’ shown by the coins in each series (as if it were a quantifiable prospect!). The ‘influence’ logic is nothing but ‘taxonomic Othering’ in a slightly different garb – coins such as those of the ‘lion × elephant’ were adjudged to have ‘more Greek influence’, manifested in the fact that they were ‘struck on both sides’ and constitute a ‘single device’ appearing on each side, which are in a ‘style’ that in evolutionary terms ‘is suggestive of Greek influence’. The basis of this interpretation is that the motifs such as lion, elephant, and horse are similar in execution to the ‘succeeding Indo-Greek coins of Apollodotos I and their portrayal is “distinctly Greek”’. On the other hand, the coins of an earlier sort within the same fabric/series have a lower weight, have a multitude of symbols, and are primarily struck on one side. This makes them an ‘essentially an Indian coin type’. This is then adopted or continued by the Greek rulers when they make an appearance in Gandhara. The latest numismatic assessment of the ‘Taxila’ coinage is by Wilfried Pieper20 who ascribes its chronological placement to the ‘independent’ period between the Mauryas and the Greeks approximately between 232 BCE, the proposed date of the death of Asoka and c.190– 180 BCE when Agathocles ‘made Taxila a province of his realm’. This is in turn based on Osmund Bopearachchi’s dating of Agathocles.21 The Taxila coins thus ‘served as prototypes for Hellenized copies struck by Agathocles and Pantaleon’. He also suggests that ‘the series is so extensive and the varieties so numerous that the continued striking of these coins seems to have been tolerated by the Indo-Greek kings of Taxila for use in local trade’. However, he does not say for how long. He also doubts Bopearachchi’s contention (following Tarn) that the square elephant/lion and elephant/horse coins could be issues of Demetrios I because if the Greeks had ‘invented’ the square copper types at Taxila, ‘the absence of any legend on them would be unusual and against the rule of their coinage’. To Pieper, the technique of producing a uniface coin is ‘obviously simpler’ than to strike a coin between

80  Shailendra Bhandare two engraved dies, which is ‘more advanced’, and therefore he chooses to suggest that the uniface coins of Taxila might be earlier than the biface types. Pieper compares the symbolic programme of the Taxila coins with ‘Mauryan’-cast copper coins known from northern India and sees them as the prototypes for the animal types from Taxila. There is an ‘Indian side’ to this story as well. Babu Durga Prasad, an Indian numismatist, propounded a view that a common symbol evident on a wide range of Indian coins, identified by numismatists as ‘Three-arched Hill surmounted by a Crescent’, was in fact a ‘Mauryan royal symbol’, or Rājānka. Noted Indological scholar Dr K. P. Jayaswal was also of the same opinion – in fact it is difficult to ascertain who articulated it first between Prasad and Jayaswal. Jayaswal appears to have made the assertion in 1933 while delivering the presidential address of the Oriental Conference at Baroda, but Prasad contends that [t]he symbol was first mentioned by me to be connected with the Mauryas, as I found it on the remains of half a dozen remains of definitely known Mauryan monuments, as well as cast copper coins dug out from the Mauryan levels at different ancient sites.22 Prasad adduced evidence to this effect from the so-called ‘Mauryan’ texts like the Arthasastra, which indicated that a practice of marking various objects with such a royal mark was indeed prevalent during Mauryan times. Although the text does not give any idea as to what the mark actually looked like, the identification of the symbol as such was inferred by one of its prominent constituents, the ‘crescent’, which is Chandra in Sanskrit, and therefore conveniently linked to Chandragupta, the first Mauryan ruler. Needless to say, this is all purely conjectural; however, it gained wide acceptance and contributed subsequently to the identification of punch-marked silver coins as ‘Mauryan’ currency. Because the three-arched hill symbol predominates on a lot of the Indian-style coins of Taxila, Prasad contended that these were in fact Mauryan issues. Most noteworthy is the inclusion of the ‘lion × elephant’ types – regarded as the immediate precursors of the Pantaleon/ Agathocles coins by other numismatists – into this very broad category of Mauryan coins, because they, too, carry the arched hill symbol. By Durga Prasad’s contention, a period of independence for Taxila before the supposed Greek advent under Pantaleon/Agathocles would not appear to exist. The succession to the Greeks from the Mauryans would be direct. However, this view did not find takers in the

Numismatics of ‘the Other’ 81 historical narrative about Taxila, probably because the overwhelming analysis provided in the Taxila excavation reports seemingly goes against it. However, in the numismatic section John Allan wrote for the Taxila excavation report, he deployed the logic of interpreting the arched hill symbol as Mauryan to emphasize the Mauryan character of some of the ‘Indian’ coins. If we recap these varied numismatic perspectives, certain distinct analytical tropes emerge. First, the chronological placement of the Taxila coinage is taken by most numismatists to be a post-Mauryan phenomenon. Archaeological substantiation was sought for this (see later in this chapter), by the fact that Taxila had an independent coinage soon after the Mauryans had been a part of the discourse even before this substantiation was published. The inference that the coinage of Taxila was post-Mauryan and pre-Greek is primarily based on a typological construct, in which the discussion has mainly been to identify and assign coins to ‘Greek’ and ‘non-Greek’ types. The characteristics of these types are physical as well ideological – if the coins are square, ‘rough’, uninscribed, and symbolic in appearance, they are adjudged to be ‘non-Greek’; if they are round, neat, inscribed, and carry animate motifs, they are inferred to be of ‘Greek’ type. The next step is to identify, in individual instances, which of these have been influenced by the other and then relate such a connection to a type-based chronology. This is then mapped with the known historical wisdom to suggest where exactly the political interface lay. To most numismatists, this interface appears to lie about the time of Agathocles/Pantaleon, which then makes them deduce that this was the time Taxila ‘lost’ its independence and became a ‘Greek’ city. The historical reconstruction is thus a direct outcome of the selective ‘Othering’ of certain types of coins with respect to the bulk of coins associated with Taxila in both excavated and non-excavated contexts. Taxila’s local coinage – archaeology and numismatics John Marshall excavated various sites comprising the ancient city of Taxila for a long time. Excavations began in 1914 and continued until the late 1930s. A consolidated report was published in 1951.23 This report curiously has four analytical sections on coins written by four authors: Marshall himself, John Allan of the British Museum, R. B. Whitehead, and E.H.C. Walsh. Each presents his own analysis about the coin finds. Marshall and Allan write more from a ‘top-down’, summary perspective; Whitehead’s take is that of an antiquarian/numismatist (focusing on “significant coins”, separating out those which

82  Shailendra Bhandare he found more interesting), and Walsh writes mainly about the two hoard finds unearthed at the site of Bhir mound. A ‘historical section’ appears at the beginning of the report in which the general chronology and periodization of the site are outlined. A cursory glance at this section suffices to suggest that the general structure follows the norm we have seen in the preceding section of this chapter – in the decades following Alexander’s retreat from the Punjab, Taxila first exists as a ‘Mauryan’ province, then becomes independent for a while, and then is occupied by the Bactrian Greeks. The interpretation of coins is largely based on the stratigraphic sequencing and identifying certain layers of Bhir and Sirkap as ‘Mauryan’, which inter alia has been essentially a numismatic exercise, based mainly on the analysis of the Bhir Mound coin hoards as forwarded by Walsh. To an extent therefore, the archaeological and numismatic narratives given in the Taxila volumes feed off each other in a circular way. The four writers also present different views, particularly with respect to the local Taxila coins, which draw upon and/or differ from each other’s analysis, as well as in ways they use their forbearers’ views and conclusions. At the same time, each of them arrives at what appears to be a curious mix of interpretations, sometimes completely ignoring important factual details others have noted. Marshall devotes a number of pages to the ‘Local Copper Coinage of Taxila’.24 In the opening paragraphs of this section, he noted that the archaeology of the finds of these coins makes it certain that although they were largely issued from c. third century BCE (“it would be unsafe to infer that any of these coins were struck before the third century BCE”), the coinage was not ‘short-lived’ and did not end with the ‘Greek conquest before the middle of the second century BCE’ as Allan had postulated. Marshall’s inference is based on the numbers of these coins recovered from different levels – he found 179 coins from the so-called Mauryan and ‘autonomous’ periods of the Bhir mound settlements, but from the Sirkap cities, he recovered 502 coins, with 432 coming from the top two strata. A type-wise distribution was also telling – whereas 25 types and varieties were recovered from Sirkap, only 11 were common in the assemblage uncovered at Bhir, and none of the “noteworthy types such as the ‘Tree-in-railing’ ” were found there. According to Marshall, the Sirkap cities “did not come into existence until well after the Greek conquest”, so finding such a large number of these so-called ‘Indian’ coins at Sirkap came as a bit of a surprise. Although Marshall agrees on the ‘residuality’ of coins from an earlier period in circulation (“ancient Indian coins often remain in circulation for several generations after they had ceased to be

Numismatics of ‘the Other’ 83 minted”), he found the sheer number of these coins rather daunting to admit that their issue had been stopped after the Greek conquest. He therefore contended, “On the whole, it seems more likely that these local coins continued to be struck for some time after the Greeks, and possibly after the Saka conquest also”. This contention could have killed the narrative about the existence of an ‘independent’ Taxila based on placing these coins into the Maurya–Greek interlude as ‘Indian’ issues preceding the Greek conquest in an interpretative sense. But it did not, primarily because even though archaeology was clearly guiding Marshall in suggesting that the issue of these could have been carried on well after the Greek conquest, the art historian in him appears to have ignored these warnings, as evident from the successive paragraphs of his analysis. In his estimation of what the symbols of these uninscribed coins might mean, he seems to be guided by a preconceived notion that they are ‘Buddhist’. Marshall takes the ‘three-arched hill surmounted by a crescent’ symbol to stand for a Buddhist stupa. Types which have various ancillary symbols constituting a programme around this so-called ‘stupa’ – a ‘pillar’, a tree in railing, an ‘undulating river’, a female figure holding a flower, etc. – are also alluded to have a ‘Buddhist’ connotation. He even sees the “plan of a sanghārāma with two rows of monastic cells and a small flame-shaped object in the middle” on one coin type. In a feat of interpretative certitude, Marshall declares with regard to the meaning of these symbols, there “can hardly be much room for doubt” – the chaitya or stupa is evidently the great Dharmarajika stupa built by Asoka; the stream is the Tamra Nala, which flowed at the foot of the stupa; the pillar is an Asokan pillar; the two rows of cells and the sangharama are parts of the monastery with the central flameshaped object being a diminutive description of the stupa; and the treein-railing must be the Bodhi tree from Bodhgaya which Asoka had transplanted at Taxila. As for the female holding the flower, Marshall chooses to identify her as Maya, the Buddha’s mother, even though other numismatists like Allan had opined earlier that she could well be Lakshmi. Marshall’s essentialist arguments are quite surprising because he gives no specific evidence to his interpretations. It is interesting to see that much like the numismatists ‘Other’ the coins in an Essentialist manner, he ‘Others’ Buddhism from Hinduism and regards the symbols on the uninscribed coins as exclusively Buddhist. In this, he is probably guided by his work on the Buddhist monuments of Sanchi co-authored with Alfred Foucher, who famously created the category of ‘l’Art Greco-Bouddhique’ in Gandhara. However, as work by Julia

84  Shailendra Bhandare Shaw on Sanchi,25 the very site Marshall and Foucher wrote about, has shown, the identities of what constituted ‘Buddhism’ and ‘Hinduism’ were not so determinedly set up in clear-cut terms in the Early Historic period. Both religious systems borrowed freely from each other in terms of visual culture and vied for already established cults. Another interesting aspect about Marshall’s views is that he almost wants to see everything he excavates at Taxila firmly rooted and confirmed by the coin depictions, thereby creating a specific ‘time and space’ context for the local coins. He even alludes, agreeing with Tarn,26 that the elephant on some local coins must have been an ancestor of the beast named ‘Ajax’, mentioned by the classical writer Philostratus (first century CE), as resident in the temple of the sun at Taxila and worshipped or greeted by visitors to the temple! In doing so, he ignores the fact that some of these coins – particularly the ‘elephant × lion’ types – had already been found at places far away from Taxila, like Begram or Yusufzai province. In his attempt to contextualize the coins so firmly with Buddhism, Marshall also posits them chronologically into the Mauryan to post-Mauryan epoch, when according to him, “Buddhism had become the established religion at Taxila”. This interpretation is based largely on how he thinks the symbols to be ‘Buddhist’, bolstered by the occurrence of these coins in the ‘Mauryan’ layers of Bhir, but ignoring the vast numbers he found at so-called ‘Greek’ cities of Sirkap. Lastly, it might be said that Marshall’s attribution of the local coins of Taxila is itself not free from over-simplification. From the stratumand site-wise distribution charts of coins identified as such that Marshall presents and exemplified by the entries of such coins which he adds to the compendium of “Rare and Unique Coins found at Taxila between 1912 and 1934” (Chapter 39 in the report), it is quite evident that he includes a good number of types which are found across northern India at different sites into the broad label of ‘local coins’ of Taxila. Examples of such are ‘early uninscribed cast copper coins’ or EUCCCs (no. 21, 22 and 25 in the plate provided at the end of the report) of two distinct types: round coins with an elephant on the obverse and an arched hill with crescent on the reverse, and rectangular coins with four symbols on both sides. Attributional mix-ups like these limit the utility of Marshall’s data when it comes to analyzing his data in a quantified manner. The identification of ‘Mauryan’ strata at Bhir is largely on the basis of the coin hoards discovered there, namely Bhir-I, which comprised 166 coins, discovered in 1913, and Bhir-II, which contained 1,167 coins and was found in 1924. In Marshall’s 1951 report on Taxila,

Numismatics of ‘the Other’ 85 E.H.C. Walsh presents a descriptive account of the hoard as in Chapter 40. But earlier on, he published the two hoards separately as a memoir of the Archaeological Survey of India27 where much of the analysis regarding the coins can be found. Bhir-II contained chronologically earlier coins, such as Gandharan bent bars and a single Achemaenian siglos, mixed with five punch-marked silvers of the “larger and thinner” sort which Walsh contends to be “older”. The Bhir-I  hoard, in contrast, exclusively contained five unch-marked silvers of the “smaller and thicker” sort, which Walsh thinks to be “later” issues. Due to their anepigraphic nature, punch-marked coins can only be dated by other contexts. To accomplish the task of dating the deposition of these hoards, Walsh placed a considerable emphasis on the few ‘dateable’ coins contained in the hoards; in Bhir-II they are the coins of Alexander the Great and his short-lived successor Philip Arrhidaeus, whereas Bhir-I contained a single gold coin in the name of Diodotos, issued by one of the two homonymous Greek kings in Bactria who declared independence from the Seleucids. Walsh also places considerable importance on the condition of these dateable coins; in both the hoards, these Greek and Bactrian coins are said to have been in “mint condition” as compared to the punch-marked coins, so Walsh adjudges them to be the “current coinage at the time of their deposit”. Given the fact that Phillip Arrhidaeus died in 317 BCE, Walsh uses this date to suggest that “it is possible to determine within narrow limits the date of the deposit and, from it, to arrive at the date of the latest punch-marked coins in the hoard, and to estimate that of the older coins as well”. Because Bhir-I contained the coin of Diodotos, Walsh deploys a similar argument to date its deposition to 248 BCE, when such coins were in circulation. The most noteworthy aspect of the Bhir-I hoard is that 161 coins of the total of 166 bear a unique mark known as the ‘Taxila mark’ on the reverse, “though in many cases it is in such a debased form as to be scarcely recognized as such”. Noting the fact that “the two hoards were found on the same site and in distinct strata, which are respectively dated from other considerations, apart from the dates of Phillip Aridaeus (sic.) and of Diodotos”, Walsh concludes that “the later coins were first struck sometime between the close of fourth century and the reign of Diodotos”.28 Quite conveniently this is the ‘Mauryan’ epoch. Based on these dates, Walsh creates a ‘Mauryan’ identity for the site where these hoards were found. Bhir-I also contained some jewellery alongside the coins. According to Walsh, it was of a “distinctly Mauryan” character. This contention of Walsh becomes a benchmark for assessing the archaeological chronology of Taxila. The Bhir mound henceforth

86  Shailendra Bhandare emerges as the ‘Mauryan’ or ‘Indian’ city, with its chaotic organization, whereas Sirkap, with its grid plan and other civilizing architecture is recognized as the ‘Greek’ city. Much like the treatment meted out to the local Taxila coinage, this is another example of ‘Othering’, and Walsh’s analysis of the punch-marked coins underpins it by identifying Bhir as the ‘Mauryan’ city. However, Walsh’s assessment can be criticized from several angles. Most unreliable is his emphasis on the condition of the dateable coins of the hoards. Gold coins almost invariably were used more as a store of wealth due to their very high intrinsic value than a medium of circulation to effect everyday transactions. It is therefore not surprising that the gold coin of Diodotos was left in an exceptional state of preservation. The date of Diodotos, 248 BCE, therefore cannot be posited as the ‘date of current coinage’ to date the deposition of the Bhir-I; it could well have been preserved in pristine condition even at a later date. The same applies to the coins of Alexander and Phillip Arrhidaeus – their physical condition, too, can only be taken as a qualitative indicator in assessing how long they could have remained in circulation and not an absolute one. In a circulatory realm dominated by anepigraphic and largely symbolic coins, exotic foreign coins with realistic portraiture have every chance of being saved from circulation and deposited in a hoard in fresh condition. Moreover, Hellenistic and Graeco-Bactrian coins are often struck with deeply engraved dies and thus bear impressions in high relief. As such, they often show very little sign of wear even after a reasonable degree of circulation. In recent years, a large hoard of these coins was found in Bihar.29 The sample data collected for their die analysis by Zeng30 show a very high proportion of coins in a very good state of preservation that could be regarded comparable with the specimen found in Bhir-I. The date of Phillip therefore has little connection with estimating the date of deposition of the hoard. All one could say is that the coins were already struck when they were deposited alongside the punch-marked coins – but how later than their date of issue could the deposit have been made is an open-ended chronological bracket. The preponderance of coins with the so-called ‘Taxila mark’ in Bhir-I is also worth a revisit in the discussion about the date of deposition of the hoard. Studies by Elizabeth Errington31 and Robert Tye32 have indicated quite clearly that some of the so-called ‘Mauryan’ types of imperial punch-marked coins are in fact later issues and subject of a particular numismatic phenomenon involving salient punchmarked coin types (often referred to by numismatists by the GuptaHardaker classificatory system and identified by a ‘GH’ number).33

Numismatics of ‘the Other’ 87 The characteristic features of this phenomenon constitute a gradual (but often steep) debasement of precious metal contents and placement of the symbols on the flan of the coin in a ‘rigid’ manner, with their orientation vis-à-vis each other being ‘fixed’, unlike the predecessor issues where it is almost invariably random. The coins on reverse also sport a bold single mark. The phenomenon is regionally oriented – it is observed with respect to coins in Malwa, in the Ganga Plain, and in Vidarbha and Gandhara. In some regions, these act as precursors to even later punch-marked coin types, each with further distinguishing features which are also regionally oriented, such as incorporation of unique obverse and/or reverse marks. The coins in Bhir-I are typical of such a ‘regionally fossilized’ series from Gandhara, originating from a variety of imperial coins identified as GH-575 of the Gupta-Hardaker classificatory system. It is now a well-established numismatic observation that such coins, although copying prototypes which were first struck in the Mauryan or late Mauryan period, are issued considerably later than that epoch. This critique of Walsh’s assessment of the chronology of the Bhir hoards demonstrates that their ‘Mauryan’ date of deposition is far from being certain. Walsh’s statements which he adds to support this conclusion are conceivably without adducing any evidence. The “other considerations” he mentions to date the layers in which the Bhir hoards were found as ‘Mauryan’ are presumably the pre-Mauryan or Achaemenid levels discussed by Marshall,34 but the veracity of Marshall’s claim of discovering Achaemenian levels at Taxila have been questioned.35 Working within the context of the finds, Walsh remarks that the jewellery found along with Bhir-I was of a “Mauryan character” – this, too, is a vague description which lacks substance because he does not specify in which ways such a ‘character’ is assessed and attributed. Notwithstanding these facts, Walsh’s analysis of the Bhir hoards was convenient in creating the chronological narrative about Taxila because it bolstered the recognition of a ‘Mauryan’ Taxila, which in turn provided a substantial basis to Marshall’s own Essentializing arguments about how Buddhism came to be the ‘official religion’ in the region and how the local coinage (which succeeds these so-called ‘Mauryan’ coins) employed symbols which were firmly rooted in the spatio-temporal milieu of material culture known to exist at Taxila, including stupas, monasteries and pillars. This effectively facilitated their ‘Othering’ from the Greek coins which chronologically followed them in the scheme of the narrative Marshall and others established for Taxila through their writings.

88  Shailendra Bhandare R. B. Whitehead presented a ‘commentary’ on the “Rare and Unique” coins found at Taxila,36 which Marshall listed as his Chapter 39. His remarks on the local coins of Taxila reiterate the established wisdom, if only underlining the ‘Othering’ even more. He regards the local coins as predecessors to the Greek presence at Taxila. He criticises W. W. Tarn for his remarks on finding a ‘Greek type’ at Taxila (“the Greek type at Taxila ought to be discovered in Taxila’s own coinage”, vide supra) and contends against Tarn’s suggestion that the missing type of Taxila on Greek coins was the elephant, mainly because in Taxila’s local coins, the elephant is not so preponderant. But he does make an essentialist comment – that the Greeks “did derive from the Taxila was the square shape”. Here again we see the impact of ‘Othering’, originally devised by Cunningham, whereby the square/ rectangular shape of coins was adjudged as something uniquely ‘Indian’ and therefore not ‘Greek’. The ‘Othering’ is further emphasized in Whitehead’s remark on the Hirañasame coinage of Agathocles, which he regards to be a “small change of local character”, thus not being ‘Greek’ enough in its type characteristics. He also takes Hirañasame or ‘Golden Hermitage’ to be synonymous with Taxila and therefore contends Tarn’s argument that Agathocles did not rule at Taxila to be untrue. In this, he is guided by Marshall’s identification of the motif on these coins as that of a ‘lotus tree’, with a specific local Taxilan Buddhist context, which we have already seen to be just another example of how ‘Othering’ works when it comes to categorizing these coins. But apart from this one analytical remark, Whitehead does little else than link up the attributive category of ‘Taxila coins’ to their provenance (“The accepted identification of the indigenous Taxila money has been amply confirmed by actual discovery on the spot”) so far as any analytical critique goes. He refers the reader to Cunningham’s Coins of Ancient India and the “full and up-to-date treatment” by Allan in the British Museum Catalogue for further information on these coins. Curiously, he does not remark on or correct the wrong inclusion of some of the EUCCCs in the list as ‘local Taxila coins’ by Marshall even though he must have been well aware of their attribution as such. John Allan added his own “notes” to punch-marked, local, and Greek coins found in Taxila in Chapter 41 of the report.37 Marshall leaves Walsh’s and Whitehead’s contributions without adding any remarks of his own, but he annotates Allan’s section with a series of footnotes bearing his initials, offering views which are different from that of Allan’s and/or giving additional information. Allan agrees with Walsh’s estimation that the date of deposition of Bhir-I took

Numismatics of ‘the Other’ 89 place in the ‘Mauryan’ period and firmly places the local Taxila coins “between the decline of the Mauryan Empire and the Greek occupation”. According to him the contents of the Bhir-I and Bhir-II hoards are so decidedly ‘Mauryan’ that “too much stress need not be laid on the three Greek coins [from Bhir-II] which were certainly not current in Taxila but had drifted down there from the North-West, just as they still do to Rawalpindi at the present day”. He muses over the possibility of a later date of deposition, perhaps in response to the “threat from Bactrian Greeks”, and suggests that the Bhir-II hoard could even have been buried as late as 170 BCE, just before the destruction of the Bhir mound settlements. Marshall adds a cautionary footnote that “our knowledge of the history of this period is next to nil”. This is interesting because if the coins were withdrawn from circulating specie as Walsh had suggested, such a late date of burial, particularly for Bhir-I (which, according to Walsh, contained coins struck later and not represented in Bhir-II), would mean the coins had been in circulation well after the ‘Mauryan’ period. This might have meant a different approach in the ways the chronology of the hoard finds and, inter alia, the stratigraphy of sites at which they were found. However, this does not seem to have happened, probably because regarding the Bhir coins as ‘Mauryan’ neatly suited the narrative about Taxila that had been in vogue and to which Allan largely subscribed. Regarding the local coins of Taxila, Allan reiterates his agreement with Cunningham about their attribution as such, although he notes that “a number of types said by Cunningham only to be found at Taxila are conspicuous by their absence”. He also re-emphasizes his position, voiced in the British Museum Catalogue where he first classified these coins, that they form a “well define[d] series and must cover the period of autonomy between the decline of the Maurya empire and the Greek occupation”. Marshall once again adds a cautionary footnote: “The local coinage undoubtedly covers this period of autonomy, but it seems to have started well back in the third century, under Maurya rule”. In this comment, Marshall is probably guided by the Rājānka postulation of Durga Prasad and Jayaswal, who regarded the ‘arched hill with crescent’ seen on many of these coins as a ‘Mauryan’ royal emblem. Allan regards symbols such as the swastika, a hollow cross, a plant, and a chaitya as “characteristic of Taxila”. Unlike the other commentators in the report Allan also attempts to suggest an internal chronology for these coins, but in his views he is clearly guided by certain assumptions which he already had subscribed to while writing the British Museum catalogue. Thus, he is firmly of the opinion that the coins of the rectangular fabric of the ‘elephant ×

90  Shailendra Bhandare lion’ type, but where the lion is replaced by a horse, are the “probably the earliest types issued under Greek influence”, and although they are not rare, “they were found very sparingly” at Taxila.38 According to Allan, “the horse is a Greek rather than an Indian type”. Curiously enough, he comments that “the lion is also a Greek type, but not one particularly favoured by the Greeks in India” and in any case, because it is associated with symbols such as the swastika or the ‘arched hill’ on these coins, “there is no reason to think it is not the local Indian type in this case”. This is yet another instance of ‘typological Othering’, but it works very well in supporting the chronology Allan attempts to arrive at. Allan posits the type similarities between these coins and those bearing the name of Agathocles and Pantaleon to suggest that the political watershed at Taxila manifests itself at the time of Agathocles/Pantaleon. Much like Whitehead, his intention appears to be to debunk the narrative of Tarn, who did not agree with this proposition. Allan posits the hoard findings by Cunningham once again into the debate and comments, “Dr Tarn says that there is no reason to suppose that Agathocles ever ruled at Taxila. If there is one fact that the coins tell us, it is that Agathocles did rule at Taxila”.39 Allan further adduces typological clues to suggest that the local coinage at Taxila came as the immediate predecessor of Greek coins, introduced after the Greek ‘conquest’. According to him the horse which substitutes the lion on what he regards as the ‘earliest coins with Greek influence’ has direct resemblance with the horse on the copper coins of the Graeco-Bactrian king Euthydemos. He also takes the monogram ‘Ā’ occurring on some of these coins to mean either ‘Agathocles’ or ‘Ta’ for ‘Taxila’. He also draws attention to the similarity of the star symbol placed above the horse with a similar star placed above the arched hill motif on Agathocles’ Hirañasame coinage, which he views as issues of Taxila. Recapturing the numismatic analysis in Marshall’s ‘Taxila’ volumes, a few salient interpretative themes become evident. As the fundament of their analysis, all four contributors take Cunningham’s contentions as a given without really questioning it. As we have seen, Cunningham basically devised his scheme on ‘typological Othering’. The subsequent commentators work pretty much within the same frame. Marshall’s approach is more cautious as is evident from the comments he adds in footnotes to Allan’s comments. But he is still led by the logic of finding ‘Greek’ elements in Taxila’s local coins, and ‘Othering’ forms a basis of his argument too – he extends numismatic ‘Othering’ further to create a distinctly ‘Buddhist’ identity for the local coins of Taxila,

Numismatics of ‘the Other’ 91 essentially suggesting the symbols on them are ‘Buddhist’ rather than ‘Hindu’. Whitehead and Allan both try to find the ‘Greek’ type of Taxila and although they take on board Tarn’s statement that such type ought to be found in Taxila’s own local types, they attack Tarn for suggesting Agathocles did not rule at Taxila. Allan contends that the political interface between ‘independent’ and ‘Greek’ Taxila is manifested by the oblong, rectangular coins of Agathocles and Pantaleon, which act as typological successors of the ‘elephant × lion’ types, with the uninscribed coins on which the lion is replaced by a horse, also attributed to Agathocles, coming slightly earlier in the sequence. In this way, Allan firmly rejects Tarn’s contention and reinstates Agathocles at Taxila as its earliest Greek ruler.

Reassessing the ‘Greekness’ at Taxila – archaeological and numismatic realities Astonishingly, these assertions are not backed by excavated numismatic material at Taxila. The ‘elephant × lion’ type coins, supposedly the predecessors of Greek issues, are found in greater numbers at Sirkap (eighteen coins) than at Bhir (five coins). With regard to the Hirañasame coins, the number is woefully small – one coin with only Hirañasame legend was recovered from the surface, whereas four coins with the name of Agathocles alongside the Hirañasame legend were found at Sirkap. The issues of Agathocles and Pantaleon regarded to be the earliest ‘Greek’ issues of Taxila are actually absent from the archaeological record! The logic Allan applies to explain the lack of several Taxila types described by Cunningham in excavated context (“that evidences their rarity rather than casts doubt on their attribution”) may not be applied in case of the Agathocles/Pantaleon issues, because they are not rare coins. Their absence from the archaeological record is therefore a significant omission in justifying Allan’s claim that they were the earliest ‘Greek’ coins at Taxila, preceded by the local coins and copying designs from them. We have already seen that far more local Taxila coins are found at Sirkap, presumably Marshall’s ‘Greek’ city, than at the Bhir sites, deemed to be ‘pre-Greek’ or ‘Indian/ Mauryan’ in their character. A chart of stratigraphic retrieval of these coins provided by Marshall makes it even more clear; even though Marshall contends that 134 local coins were found in the supposedly ‘Mauryan’ stratum, the identity of such a stratum is based in consequence of Walsh’s analysis of the Bhir coin hoards. There appears to be a curious case of Walsh and Marshall entering into a circular agreement of each other’s stratigraphic wisdom – Marshall alludes to

92  Shailendra Bhandare certain layers as ‘Mauryan’ because Walsh discovered ‘Mauryan’ coins in them, whereas Walsh confirms the ‘Mauryan’ nature of the layers in which the punch-marked coins came from based on ‘other considerations’, probably alluding to Marshall, but does not give any idea as to what these considerations are. A type-wise break-up provided with stratigraphy on p. 757 of the report reveals some interesting facts. The most conspicuous of the local coins at Bhir mound are those which have an arched hill with crescent, or a taurine, with the Brahmi letter ‘Go’ besides it on both obverse and reverse (Figure 3.4). The weight, fabric, and manufacturing technique of coins of this type and the way in which the letter is executed leaves little doubt that it is a later type. However, a great number of these coins (67) are found in the supposedly ‘Mauryan’ layer (layer II) at Bhir. Another seemingly later type – lion to left with taurine and swastika (Figure 3.5) with a blank reverse – is also found in significant number (33), only second to the type mentioned earlier. Together, these two types constitute more than half of the total number of local coins at Taxila recovered from Bhir, and from a numismatic perspective, they are decidedly struck later than the chunky, oblong, rectangular pieces of Allan’s ‘Class I’ of the British Museum catalogue. The ‘elephant × lion’ type and varieties thereof are found in sparing numbers (only six) and out of these four come from the ‘Mauryan’ layer II. On the other hand, a greater concentration of these supposedly earlier types of the local coins is encountered not at the pre-Greek Bhir mound, but at the ‘Greek’ Sirkap. Here a total of 36 of the ‘elephant × lion’ types are found, with the number of those found in the ‘Greek’ layers (VI and V) and the ‘late Saka and Parthian’ layers (II to III) being the same: 15 each. Coins of the types preponderant at Bhir are

Figure 3.4 ‘Taxila’ coin with arched hill flanked by Brahmi letter ‘Go’

Numismatics of ‘the Other’ 93

Figure 3.5 ‘Taxila’ coin with lion and swastika; the reverse is blank

found at Sirkap in much greater numbers – those with the letter ‘Go’ and variants thereof number 201 – and here they are concentrated in maximum numbers in the ‘late Saka and Parthian’ levels, with 132 out of the total belonging to this type alone. The other preponderant type at Bhir is also encountered in a large number at Sirkap (125), and out of these 80 come from the ‘late Saka and Parthian’ layer. With such a remarkable variance in distribution, it is difficult to agree that these coins could be Essentialized as ‘Mauryan’ issues. At best, one can assume their issue must have begun in the Mauryan times, but stratigraphy shows that they certainly constituted the bulk of currency at Taxila in periods much later than the ‘Mauryan’. It is significant that their numbers appear to swell in the period after the Greek levels and, in general, these are the levels which have yielded masses of Indo-Scythian coins, so their occurrence in numbers greater than that at Bhir cannot be justified as representing an exigent ‘residual’ circulation in later times in response to a general lack of circulatory coinage. In a nutshell, therefore, the numismatic chronology of Taxila appears to be a sequence qualified by a number of essentializing contentions resulting from ‘Othering’, and the narrative emerging from it is a reflection of the segmented approach taken by numismatists who worked on the data to make the coins fit in the established framework. In view of the critique presented earlier, it is evident that there is an urgent need to set all the essentialist categorization aside and look afresh at the coins labelled as ‘local Taxila’ coins. We must investigate what they themselves reveal to us through their characteristics. With regard to their internal chronology, so far the attempts have been

94  Shailendra Bhandare based on rather feeble grounds – ‘Othering’ has facilitated their identification as ‘Indian’ coins as opposed to ‘Greek’, and this has been the basic guiding principle in deciding which of these types are earlier. Typological features such as similarity with other Greek coins, the search for ‘Greek’ types at Taxila, and deployment of aspects such as monograms have been used as further tools in the framework largely set up through this ‘Othering’. By far one of the most significant numismatic tools used to determine chronology of coins is counterstriking – wherein coins struck by one issuing authority are used as templates for manufacturing coins by another authority. This affords numismatists undeniable evidence to suggest the chronology of the coins involved in the phenomenon – if coin of ‘A’ is counterstruck by ‘B’, that would mean ‘B’ can exist only as a contemporary of ‘A’ or his successor but never as his predecessor. Some ‘local’ coins of Taxila presented hereunder throw a significant light on when they might have been struck. Two of these are from the Ashmolean Museum, University of Oxford collection (Figure 3.6 and Figure 3.7) and one has been listed on www.zeno.ru, the online database of Oriental coins, as Zeno# 23875 (Figure 3.8). The two coins from the Ashmolean collection are of the ‘square incuse’ type of local Taxila fabric belonging to Cunningham’s category of “square copper pieces of single dies, and of the standard peculiar to India”. But evident very clearly is the fact that they are struck over copper coins of the Indo-Greek ruler Apollodotos I (c. 175–165 BCE). The undertype is BN Série 6A as per Bopearachchi’s classification.40

Figure 3.6 Indo-Greek copper coins of Apollodotos I counterstruck by ‘Taxila’type devices

Numismatics of ‘the Other’ 95

Figure 3.7  Indo-Greek copper coins of Apollodotos I counterstruck by ‘Taxila’-type devices

Figure 3.8 ‘Lion × elephant’ type coin, counterstruck on Apollodotos I

The overtype, however, is of the sort Marshall et al. would attribute to the period when Taxila was ‘independent’ following the decline of the Maurya Empire. It comprises a tree in railing, a swastika, an arched hill with crescent, and a small taurine. It is listed in Allan’s British Museum catalogue41 as ‘var. c’, p. 220, Pl. XXXII, no. 4. That a coin of this sort is counterstruck on an Indo-Greek coin of the mid-second century BCE, throws open the fact the issue of coins of this type must

96  Shailendra Bhandare have continued well beyond the fall of the Mauryan empire and during the early heyday of the Indo-Greeks, just before the reign of Menander. There is no reason therefore to believe in the Essentialist premise that all such coins were pre-Greek because of their ‘Indian’ character. The coin listed on the Zeno database is even more interesting – it is of the ‘elephant × lion’ type, but like the Ashmolean coins, it, too, is counterstruck on a coin of Apollodotos I. The remnants of the base of Apollo’s tripod, the reverse device of Apollodotos I’s copper coins, is clearly evident around the top left corner on the side bearing the incused ‘lion’ punch and so are the remnants of the Kharoshthi legend that surrounds the tripod. This clearly indicates that the issue of these coins was taking place around 175–165 BCE. As such the chronological narrative that fixes these coins as ‘Indian’ prototypes for the coins of Agathocles and Pantaleon, and thereby their issue to the ‘preGreek’ epoch of independent Taxila, becomes rather tenacious. Judging by the archaeological findings at Taxila, however, these counterstruck coins come as no surprise. Marshall had already hinted at their circulation being continued at a later date, and Allan also had suggested that the dates of deposition of the Bhir mound coin hoards could have been as late as 170 BCE. The stratigraphic placement of these coins in the Taxila excavation also suggests their preponderance at sites dated later than the so-called ‘Mauryan’ settlements of Bhir. Apart from the evidence of counterstruck coins, there exist other typological aspects of the coinage which suggest that the various series which comprise it were inter-related or connected to each other irrespective of the tenets – such as single/double die striking, ‘incused’ appearance, inscribed or uninscribed – upon which Cunningham, and following him Allan, classified them. One such feature is the symbol of a standard, variously regarded by Allan and others as a ‘lotus plant’. On the thin and rectangular uninscribed series of coins (of which the inscribed Hirañasame coins are a part of), it is seen on the reverse in its most detailed form (Figure 3.9). It has a long staff in the centre, surmounted by a lotus bloom, which is flanked by two ribbons or pennants of a serpent-like nature. In its simplified form, it is evident on the round, uninscribed coin series as well (Figure 3.10). Most importantly, the same symbol appears on the oblong, rectangular coins with an incused ‘horse’ motif, which Allan regards as the ‘earliest Greek’ coin type of Taxila (Figure 3.11). None of the four writers which contribute to Marshall’s Taxila report take an account of such nuances – instead, they appear to be far more preoccupied with proving how the ‘Indian’

Numismatics of ‘the Other’ 97

Figure 3.9 ‘Taxila’ coin with elaborate lotus standard

Figure 3.10 ‘Taxila’ coin with symbolic configurations on both sides; the stylized ‘lotus standard’ symbol is seen besides the arched hill on the obverse

coin types get Hellenized into ‘Greek’ coin types, treating the bulk of the local coins summarily as a pre-Greek phenomenon. The evidence of counterstruck coins and the stratigraphy of coin finds suggest that the so-called ‘Indian’ coin types certainly continued to be issued after the Indo-Greek coins were being struck and as such formed a part of the same circulatory horizon. This important fact was sidelined in more abstract questions like finding a ‘Greek’ type of Taxila, primarily because these coins were ‘Othered’ as ‘Indian’ and the narrative of

98  Shailendra Bhandare

Figure 3.11 ‘Elephant and horse’ coin with stylized lotus standard in front of the elephant

Taxila needed to be seamlessly joined between the indigenous ‘Mauryan’ and the foreign, more civilizing ‘Greek’ period. If one overlooks the typological ‘Othering’ of these coins and accommodates the fact that the so-called ‘Indian’ and ‘Greek’ coins were being issued alongside each other, it sounds far more likely that the typological features were being drawn from the same cultural collective comprising a wide repertoire of symbols. With this in mind, Essentialist questions like whether ‘Indian’ coin types were being ‘Hellenized’ or ‘Greek’ coin types were being ‘Indianized’ really have no meaning. Lastly, the Orientalist overtones in Marshall are evident in his foreword to the ‘Taxila’ volumes. Rather romantically, he has visions of Greece in the river plains of the Punjab. He writes in the preface to the ‘Taxila’ volumes of 1951 – I still remember the thrill I got from the sight . . . I was a young man . . . filled with enthusiasm for anything Greek . . . in that far off corner of the Panjab . . . it seemed as if I had lighted on a bit of Greece itself . . . Doubtless the illusion was prompted . . . by Taxila’s historic association with Greece . . . I felt then there was something appealingly Greek in the countryside. It is no wonder that a person so vividly wishing to find a piece of Greece in the East would go on cementing the narrative created by his forerunner Alexander Cunningham. ‘Othering’ therefore predominates in his intellectual approach, but it is hardly represented in material that he excavates. One would therefore wonder whether there is really any need for Essentialism as a narrative tool in visualizing the ‘Greek’ in Gandhara.

Numismatics of ‘the Other’ 99

Epilogue So to sum up the argument made in this chapter: 1

2

3

4

5

The numismatic chronology of Taxila is a direct outcome of ‘taxonomic Othering’ of coins into Essentialist categories of ‘Indian’ and ‘Greek’ – each coming with its features which reflect an ‘Orientalist’ and antiquarian method of judging what these two labels mean. In effect, ‘Greek’ means aspects which are orderly, literate, standardized, and familiar, whereas ‘Indian’ connotes with the exotic, cryptic, and symbolic. These two categorizations are then deployed to suggest an intercultural hybrid, which is created by Greek kings who ‘Hellenize’ the ‘Indian’ coins by adopting salient features such as shape from them, while introducing ostensibly ‘Greek’ elements such as inscriptions to them. The chronology effectively feeds into an established narrative about Taxila – that it was a Mauryan province to begin with but seceded as an ‘independent state’ when the Maurya Empire fragmented. The Greeks then arrive at Taxila and build a Greek city at Sirkap. The ‘Indian’ coins of a local nature fit well into the independent phase at Taxila and give credence to the civilizing Greek phase at Taxila. The basic framework of this historic narrative is derived from textual sources. This Essentializing classification of coins therefore helped in a major way to create a narrative of ‘Greek’ Taxila. Other parallel narratives – of how the Bhir mound hoards are ‘Mauryan’, for example – cemented the story of ‘Greek’-ness in Taxila. However, archaeology as well as numismatic evidence indicate that there is nothing that justifi es this positing of the coins to create the fundament to secure the narrative of ‘Greek’-ness at Taxila. Counterstruck coins published here confirm the issue of the so-called ‘Indian’ coins well into the Greek period. In effect, they are quite clearly a part of the currency picture which is also inhabited by the Greek coins. This important aspect of their circulatory context is completely ignored when they are viewed as the ‘Other’ from a pre-Greek period. Lastly, the so-called ‘local coinage’ of Taxila is a category that requires much further research in terms of establishing the internal chronology of various constituent types. It is evident, both from the numismatic analysis and archaeological record, that localized coinages in greater Gandhara in general appear to have been struck and issued for a considerable length of time during the late Indo-Greek and Indo-Scythian periods. They obviously circulated

100  Shailendra Bhandare alongside various ‘royal’ coinages. We have very little evidence on hand to explain what precise circulatory role they played, or even the reason for their continued issue even when ‘royal’ coins were being issued simultaneously. There are instances scattered in the long history of Indian coins where coins have often been produced as commodities and/or for specific purposes, like rituals. The local coins of Taxila need to be contextualized more firmly into the realm of monetary history before they are treated as objects which reflect Essentialist ‘ethnocentrism’. Simplistic classification, like separating them into ‘square/round’, ‘inscribed/uninscribed’, or ‘uniface/double-faced’ and so on, might provide a numismatic categorization for their study, but it tells us little about how they circulated, what their monetary role was vis-à-vis other currencies in the region of greater Gandhara, and in particular, how they related to the coins of the Indo-Greeks Indo-Scythians and Indo-Parthians, alongside which they circulated. To regard them summarily as ‘Indian’ and not ‘Greek’, and therefore ‘Othering’ them from their contemporary circulatory context, does not prove helpful.

Notes 1 Plutarch, Alexander, 59 – “So, after receiving many gifts and giving many more, at last he lavished upon him a thousand talents in coined money. This conduct greatly vexed Alexander's friends, but it made many of the Barbarians look upon him more kindly.” 2 Alexander Cunningham, Taxila, or Takshasila, Archaeological Survey of India: Four Reports Made During the Years 1862–63–64–65, Government Central Press, Simla, 1871: 111–135. 3 Alexander Cunningham, Coins of Ancient India, from the Earliest Times down to the Seventh Century AD, 1891, second AES reprint, New Delhi, 2000: 52. 4 Sir John Marshall, Taxila: An Illustrated Account of Archaeological Excavations carried out at Taxila Under the Orders of the Government of India Between the Years 1913 and 1934, Cambridge University Press, Cambridge, 1951. 5 Sir John Marshall, A Guide to Taxila, 4th edition, Cambridge University Press, Cambridge, 1960: 5. 6 Sir John Marshall, A Guide to Taxila, 3rd edition, Government Press, Calcutta, 1936: 28–9. 7 Rachel Mairs, The ‘Greek Grid-Plan’ at Sirkap (Taxila) and the Question of Greek Influence in the North West, in Michael Willis edited, Migration, Trade and Peoples: European Association of South Asian Archaeologists, Proceedings of the Eighteenth Congress, London, 2005, The British Association for South Asian Studies; The British Academy, London, 2009: 135–147. 8 Cunningham, Coins of Ancient India, 1891: 60–66. 9 Cunninham, Coins of Ancient India, 1891: 65–66.

Numismatics of ‘the Other’ 101 10 Marshall, Taxila, 1951: 25–26. 11 John Allan, A Catalogue of the Indian Coins in the British Museum: Catalogue of the Coins of Ancient India, British Museum, London, 1936: CXXV‑CXXXIX, 214–238. 12 John Allan, A Catalogue of the Indian Coins, 1936: CXXXVI. 13 W.W. Tarn, The Greeks in Bactria and India, Cambridge University Press, Cambridge, 1951 (reprint 1966). 14 Tarn, The Greeks in Bactria, 1966: 135. 15 Tarn, The Greeks in Bactria, 1966: 137. 16 Tarn, The Greeks in Bactria, 1966: 143. 17 Tarn, The Greeks in Bactria, 1966: 161. 18 Tarn, The Greeks in Bactria, 1966: 163. 19 Michael Mitchiner, Indo-Greek and Indo-Scythian Coinage, Vol. 4, Hawkins Publications, London, 1975. 20 Wilfried Pieper, Ancient Indian Coins Revisited, Classical Numismatic Group, USA, 2013. 21 Osmund Bopearachchi, Monnaies gréco-bactriennes et indo-grecques, Catalogue raisonné, Bibliothèque Nationale, Paris, 1991. 22 D. Prasad, Observations on Different Types of Silver Punch-marked Coins, Their Periods and Locale, JASB-Numismatic Supplement, 1937–1938 XLVII: 61–62. 23 Marshall, Taxila, 1951. 24 Marshall, Taxila, 1951: 757–763. 25 Julia Shaw, Buddhist Landscapes in Central India: Sanchi Hill and Archaeologies of Religious and Social Change, c. 3rd century BC to 5th century AD, BASAS, The British Academy, London, 2007. 26 Tarn, The Greeks in Bactria, 1966: 164. 27 E.H.C. Walsh, Punch-marked Coins from Taxila, Memoir no. 59, Archaeological Survey of India, New Delhi, 1939. 28 Marshall, Taxila, 1951: 850–851. 29 Osmund Bopearachchi and K. Grigo, Thundering Zeus Revisited, ONS Newsletter, 169, 2001: 22–24. 30 C.D. Zeng, Some Notable Die-links among Bactrian Gold Staters, Numismatic Chronicle, 2013: 73–78. 31 Elizabeth Errington, A Survey of Late Hoards of Indian Punch-marked Coins, Numismatic Chronicle, 2003: 69–121. 32 Robert Tye, Late Indian Punch-marked Coins in the Mir Zakah II Hoard, Numismatic Chronicle, 2006: 167–171. 33 P.L. Gupta and T. Hardaker, Ancient Indian Silver Punch-marked Coins of the Magadha-Maurya Karshapana Series, IIRNS, Nasik, 1985. 34 Marshall, Taxila, 1951: 14–15 35 Pierfrancesco Callieri, India iii: Relations: Achaemenid Period, in Encyclopedia Iranica, 2004. Digital Edition, accessed at www.iranicaonline.org/ articles/india-iii-relations-achaemenid-period on 12–1–2016. 36 Marshall, Taxila, 1951: 830–842. 37 Marshall, Taxila, 1951: 853ff. 38 Alan, A Catalogue of the Indian Coins, 1936: Pl. XXXII, 17–22. 39 Marshall, Taxila, 1951: 857. 40 Bopearachchi, Monnaies gréco-bactriennes, 1991. 41 Allan, A Catalogue of the Indian Coins, 1936.

102  Shailendra Bhandare

References Allan, J. 1936. A Catalogue of the Indian Coins in the British Museum: Catalogue of the Coins of Ancient India. London: British Museum. Bopearachchi, O. 1991. Monnaies gréco-bactriennes et indo-grecques, Catalogue raisonné. Paris: Bibliothèque Nationale. Bopearachchi, O., and K. Grigo. 2001. Thundering Zeus Revisited. ONS Newsletter 169: 22–24. Callieri, P. 2004. India iii: Relations: Achaemenid Period. Encyclopedia Iranica, Digital Edition, www.iranicaonline.org/articles/india-iii-relationsachaemenid-period/ accessed on 12 January 2016. Cunningham, A. 1871. Taxila, or Takshasila, Archaeological Survey of India: Four Reports Made During the Years 1862-63-64-65. Simla: Government Central Press. Cunningham, A. 1891. Coins of Ancient India, from the Earliest Times down to the Seventh Century AD. New Delhi: Second AES reprint, 2000. Errington, E. 2003. A Survey of Late Hoards of Indian Punch-marked Coins. Numismatic Chronicle 69–121. Gupta, P.L., and T. Hardaker. 1985. Ancient Indian Silver Punch-marked Coins of the Magadha-Maurya Karshapana Series. Nasik: IIRNS. Mairs, R. 2009. The “Greek Grid-Plan” at Sirkap (Taxila) and the Question of Greek Influence in the North West, in M. Willis (ed.), Migration, Trade and Peoples: European Association of South Asian Archaeologists, Proceedings of the Eighteenth Congress, London, 2005. London: The British Association for South Asian Studies, The British Academy, pp. 135–147. Marshall, Sir J. 1936. A Guide to Taxila, 3rd edition. Archarological Survey of India. Marshall, Sir J. 1951. Taxila: An Illustrated Account of Archaeological Excavations Carried out at Taxila Under the Orders of the Government of India Between the Years 1913 and 1934. Cambridge: Cambridge University Press. Marshall, Sir J. 1960. A Guide to Taxila, 4th edition. Cambridge: Cambridge University Press. Mitchiner, M. 1975. Indo-Greek and Indo-Scythian Coinage, vol. 4. London: Hawkins Publications. Pieper, W. 2013. Ancient Indian Coins Revisited. USA: Classical Numismatic Group. Plutarch. 1919. Plutarch’s Lives with an English Translation by Bernadotte Perrin. Cambridge, MA: Harvard University Press and London: William Heinemann Ltd., Accessed through the ‘Perseus Digital Library’, Tufts University, http://data.perseus.org/texts/urn:cts:greekLit:tlg0007.tlg047. perseus-eng1/ Prasad, D. 1937–1938. Observations on Different Types of Silver Punchmarked Coins, Their Periods and Locale. JASB-Numismatic Supplement XLVII: 51–93, Pl. 6–11. Shaw, J. 2007. Buddhist Landscapes in Central India: Sanchi Hill and Archaeologies of Religious and Social Change, c. 3rd Century BC to 5th Century AD, BASAS. London: British Academy.

Numismatics of ‘the Other’ 103 Tarn, W.W. 1966. The Greeks in Bactria and India. Cambridge: Cambridge University Press, 1951 (reprint 1966). Tye, R. 2006. Late Indian Punch-marked Coins in the Mir Zakah II Hoard. Numismatic Chronicle 167–171. Walsh, E.H.C. 1939. Punch-marked Coins from Taxila. Memoir no. 59. New Delhi: Archaeological Survey of India. Zeng, C.D. 2013. Some Notable Die-links Among Bactrian Gold Staters. Numismatic Chronicle 73–78.

4 Region through text Representation of Gandhāra in the Mahābhārata Tanni Moitra

The northwestern part of the Indian subcontinent is a well-studied region. An extensive range of ancient historical sources have been employed to study the region, thereby contributing to its rich historiography. The history of scholarly interest in the region may be traced back to the early seventeenth century due to the antiquarian pursuits of the European travellers and explorers to the Indian subcontinent,1 one of the earliest being the French scholar and antiquarian named Nicholas-Claude Fabri de Peiresc (1580–1637), whose chief interest in India was to collect information about Alexander the Great.2 During the eighteenth century, other French explorers such as Colonel Gentil and General Perron collected valuable numismatic material from Punjab and also popularized them in Europe, consequently evoking a general interest among Europeans about the presence of Greek civilization in India.3 In the initial stages, the curiosity to look for Greek material remains largely motivated the European urge to study the region. Although this trend continues even in contemporary times, the nineteenthcentury marked an important shift on account of the recognition of the emphatic presence of Buddhism in the northwest. This shift was largely actuated by a couple of factors which include the decipherment of ancient scripts such as Brāhmī and Kharoṣṭhī by James Prinsep; the foundation of the Archaeological Survey of India (ASI) resulting in the intense archaeological activities and numismatic interest of its first director-general, Alexander Cunningham; and the excavation of a large number of Buddhist art objects by the ASI under James Burgess. Daniel Michon accurately summarizes the initial course of studies on the northwest as antiquarian interest in Punjab was initially dominated by the search for Alexander and evidence of the Greeks, and explorers

Region through text 105 saw their imprint on the land everywhere, often to the exclusion of other possibilities. However, in the crucial decades of the 1830s, the paradigm shifted and much of the subsequent archaeological investigation was dominated by the search for Buddhist civilization. From circa 1840 to 1900, archaeologists saw Buddhism everywhere, often to the exclusion of other possibilities, as had happened earlier regarding the Greek imprint.4 Michon’s observation is relevant to understand even the pattern of twentieth-century scholarship. The non-indigenous nature of the early historic northwest, especially Gandhāra, has been the prime focus throughout the century, and despite the major excavations focusing on early historic Gandhāra being carried out, the interest in Indian history was barely the concern. Michon argues that much like the 18th and 19th century British habit of finding traces of the Greeks under every stone, colonial archaeologists continued to posit foreign influence as the most salient cultural feature of ancient India. For example, John Marshall never referred to Taxila as a truly Indian city, but rather he discussed the phases the city had gone through as the ‘Greek-city’, the ‘Indo- Scythian city’, the ‘Indo-Parthian city’ and the ‘Kushana city’.5 This approach is also prevalent in studying Gandhāra from early Indian literature. Gandhāra is usually recognized as culturally separate from Indic traditions and is studied in the context of analyzing terms denoting the so-called ‘foreign’ communities such as yavanas and mlecchas and the perception of the early Indian texts towards them. It is often suggested by historians that in the early historic texts, the northwest is represented as ‘foreign’ or ‘other’ from the perspective of the ‘indigenous’ society of Madhyadeśa. In the late 1990s the study of Buddhism in Gandhāra received a great impetus with a series of discoveries of new Buddhist Kharoṣṭhī manuscripts in the region. The Gāndhārī manuscripts are being studied rigorously, and significant progress has been made to understand the evolving Buddhist traditions and its spread in Central and East Asia. It is thus clear that the historiography usually focuses on either the foreign cultural influences or Buddhism in Gandhāra. The indigenous or local narratives are usually not taken into account, thereby creating a limitation in our understanding of the region. This chapter seeks to understand Gandhāra from the Brahmanical epic the Mahābhārata.

106  Tanni Moitra Some of the aims are to explore what the epic authors understood as the region of Gandhāra, what constituted the region in terms of its geographical outreach, and if the authors perceived the region as integrated with the broader Madhyadeśa tradition or separate from it. The Mahābhārata is widely known as the fascinating tale of the irrevocable split of the Kuru family into two warring adversaries. It is a dramatic account of the events leading up to the decisive battle of Kurukṣetra fought between the Kauravas and the Pāṇḍavas with the help of their respective allies. The epic is also an account of the strong and determined alliances struck by the Kuru brothers for their causes. If the Pāṇḍavas had the eternal friendship of the Yādava chief Vāsudeva Kṛṣṇa, the Kauravas had Śakuni, the Gāndhāran chief, by their side. It may not be an exaggeration to state that what Kṛṣṇa was to Arjuna, Śakuni was to Duryodhana; if Kṛṣṇa was central in winning the Kurukṣetra war for the Pāṇḍavas, Śakuni was central in bringing about the war for the Kauravas. A simple narration of the core story of the Mahābhārata is enough to explain that the Kurus were not the exclusive players in the narrative. Although the Kuru kingdom was the centre of activity, Mathurā and Gandhāra were active participants in that activity. The core narrative of the epic itself makes it a valuable resource to study the region. Though the Mahābhārata claims to be history by referring to itself as itihāsam-purātanam, the historicity of the episodes, events and characters cannot be ascertained with any degree of certainty. What could, however, be stated with assurance is the fact that the region had a substantial impact on the imagination of the author(s) of the epic who brought the region within the fold of its central narrative. Gandhāra plays an integral role because two of its central epic characters, Gāndhārī and Śakuni, belong to this region. A matrimonial connection is established between Gandhāra and the Kuru kingdom by the marriage of Dhṛtarāṣṭra, the Kuru king, to Gāndhārī, the princess of Gandhāra. Gāndhārī’s son, Duryodhana, is the prime antagonist against the Pāṇḍava brothers, and her wicked and crafty brother, Śakuni, is ever present in Hastināpura instead of Gandhāra, with the prime intention of bringing about the war of Kurukṣetra.

The Mahābhārata as a historical source The uniqueness of the Mahābhārata as the longest poem in the world is well known. Although the encyclopaedic nature of the text is unique in itself, it also makes it one of the most difficult to handle in terms

Region through text 107 of its historicity, chronology, origin, growth and authorship. Scholars have been trying to answer these questions since the inception of an organized study on the epic roughly around the end of eighteenth century. In terms of date and authorship, there is unanimity among scholars that the Mahābhārata was composed over a long period and that it was not composed by a single author. Interestingly, the Mahābhārata itself offers indications about the history of its growth from a relatively shorter poem of 8,000 verses to its inflation to 100,000 verses. This process of growth was the result of the epic being told and retold by generations of poets, especially at gatherings during long sacrifices (yajn͂ as). The text talks about two major retellings, the first being the snake sacrifice organized by the Kuru king, Janamejaya, when the epic was narrated by Vaiśampāyana. The second retelling was by Ugraśravas to Śaunaka at the forest of Naimiṣa. Taking a clue from this, scholars have tried to understand the chronology of the several layers of the text. Christian Lassen was the first scholar to conduct a systematic study of the text. His aim was to reconstruct Indian geography, ethnology and pre-Buddhist history from the epics. In the process, he also examined the textual history of the Mahābhārata and suggested that the didactic passages are interpolations, whereas the original form of the epic was pre-Buddhist. He concluded that the recitation of Ugraśravas to Śaunaka was the second of the three recensions of the poem that should be dated to 460 or 400 BCE.6 Albrecht Weber looked for the origins of the epic in the Vedic hymns of praise of the heroes or patrons and argued that long epic songs were sung on the occasion of sacrificial feasts. He stressed that the Mahābhārata actually reflected a battle between the Ᾱryan people and established the Vedic elements underlying the epic view of the world by emphasizing the central position of yajn͂ a throughout the epic.7 The arguments of Joseph Dahlmann in the 1890s led to significant progress in the epic studies. He rejected the view that the Mahābhārata consisted of an original saga and that the later additions were extraneous to and inconsistent with the original material. He saw the text as the work of single author who combined the earlier myths, law books and teachings into a single whole. Therefore, the epic belongs to a single stratum. He argued that it symbolically presented conflict between good and evil and to understanding of the complex teaching of the Dharmaśāstras. He concludes that the Mahābhārata presents itself as a unified work in which there are two elements, the narrative and the epic,8 which have been welded together by one single author. His method is known as the synthetic theory.

108  Tanni Moitra E. Washburn Hopkins rejected Dahlmann’s synthetic theory and took an analytical approach to examine the process of growth of the Mahābhārata. He speaks of the four stages of the growth of the Mahābhārata extending from 400 BCE to 400 CE. Hopkins analyzed the text on the basis of disparities in language, style and meter between various parts of the epic. At its earliest stage, it was a compilation of laws in which the Pāṇḍavas were not heroes and Kṛṣṇa was not a demigod. The subsequent stages were that of the Mahābhārata tale with the Pāṇḍavas as heroes, of didactic interpolations and of later additions. Hopkins’ approach was a significant advancement which created a strong ground for proper research on the textual structure of the text. R.N. Dandekar in his The Mahābhārata: Origin and Growth,9 argues that the beginnings of the Mahābhārata can be traced to a period when the Vedic Saṃhitās had not yet come into existence. According to him, the Mahābhārata is the culmination of a long process and traces its origin from cultural and literary perspectives. He says that the beginnings of ancient Indian literature are characterized by two distinct literary traditions: the sūta tradition and the mantra tradition.10 Both were initially oral traditions that eventually got crystallized into texts. Dandekar says that the Bhārata war was an event of great magnitude, and thus numerous ballads and songs were produced by the bards around the event, leading to its crystallization into the written form of Jaya. This was the first literary form of the epic belonging to the sūta tradition. In course of time, the redactors, who were followers of Kṛṣṇa, added many legendary elements, like the Bhagavad Gītā, upon the historical elements leading to the transformation of the historical poem Jaya into the epic Bhārata. Gradually, in the final stage, a large amount of materials relating to brahmanical learning and culture got introduced into the epic and the heroes came to be represented as defenders of brahmanic faith. This is the process that he calls the brāhmanisation of the epic, which resulted in a change in the nature of the epic. The process led to the transformation of Bhārata into Mahābhārata.11 Madhav Deshpande, in his article, “Interpreting the Mahābhārata”,12 says that the received text of the Mahābhārata itself proclaims that there are three tellings of the story. The text called Jaya was composed by Vyāsa and was taught to his disciples, including Vaiśampāyana. Vaiśampāyana later narrated the story at Janamejaya’s snake-sacrificial ritual. Hearing his narration during the royal ceremony, Ugraśravas narrated the story to Śaunaka in the Naimiṣa forest. Deshpande argues that though the account of the transmission of the text is mythical,

Region through text 109 it provides useful indication about how the text of the Mahābhārata expanded into its present form. He explains that it is the current purposes of the narrators, their patrons and audiences that shape the form of a narrative. For instance, the inherited Jaya needed to serve the purpose of Janamejaya in the sense that he would be interested in Vyāsa’s narrative of victory of the Pāṇḍavas by righteous means and also to know the various feats of his ancestors. Similarly, a complete change in the narrative frame can be identified with the shift to the Naimiṣa forest, where the story is recounted by Ugraśravas at the hermitage of the sage Śaunaka and a group of sages. Because the Naimiṣa forest retelling served the needs of the brāhmaṇa renouncers, the story naturally expanded to include brāhmanical elements, thereby transforming the Bhārata to the Mahābhārata. Therefore, Deshpande establishes that it is not the characters of the story that dictates the form of the narrative act. The characters and the plots are simply the tools which are fashioned explicitly to serve the needs of the more concrete factors, namely, the narrator, the patron and the audience.13 Thus, it can be seen that studies on the epic attempt to understand the probable origin of the core story, its authors, the socio-cultural background and the process of the enlargement of the epic to its present size. Most importantly, one of the concerns of the scholars has been to separate the younger and the older layers of the epic, thereby making a clear distinction between the narrative core of the epic which is the fratricidal war and the didactic portions consisting of the post-war books of Śānti and Anuśāsana parvans. Though the period of composition of the Mahābhārata is uncertain, the scholars have agreed that the codification of the text primarily belongs to the first four centuries of the Common era.14 It is often suggested that the Brahmanical attempt at reorganization of the society is reflected in the law codes, and with the growing complexity of the social structure, it was also necessary to convey the ideas of dharma to the common people in a simple language through the composition of popular literature such as the epics.

Takṣaśilā and the Mahābhārata According to the post-Vedic literary accounts, Gandhāra is identified as a powerful political and territorial entity. Gandhāra appears in the list of the ṣoḍaśamahājanapada, or the sixteen large territorial polities, of the Buddhist canonical text, the Aṅguttaranikāya.

110  Tanni Moitra The Purāṇas and the Rāmāyaṇa identify Gandhāra as a region comprising two prosperous cities: Takṣaśilā and Puṣkarāvatī/ Puṣkalāvatī.15 The Vālmiki Rāmāyaṇa narrates that Bharata, the younger brother of Rāma, enthroned his sons Takṣa and Puṣkala in Takṣaśilā and Puṣkalāvatī, respectively (Rām, VII.114.11). There is a vibrant description of the charm of both the cities and their prosperity, and development is ascribed to Bharata, who after having established the cities in five years, returned to Ayodhyā (Rām, VII.114.12–20). Unlike the Rāmāyaṇa and the Purāṇas, the Mahābhārata offers a different account of the region. The epic does not expressly refer to the two cities – Takṣaśilā and Puṣkarāvatī – as constituting a part of Gandhāra. Whereas there is no reference to Puṣkarāvatī in the text, Takṣaśilā appears only thrice. This section explores the references of Takṣaśilā as found in the Mahābhārata, with the objective of understanding the significance of the region in the epic and its relation to Gandhāra. It is interesting to note that the Mahābhārata begins its narrative in the Ādiparvan by stating that the Kuru successor, Janamejaya, the son of Parīkṣit, conquered Takṣaśilā. His victory over Takṣaśilā is mentioned as a statement of fact without presenting any detail about the conquest. The term deśaṁ is used to identify the region. sa tathā bhrātṛn saṁdiśya takṣaśilāṁ pratyabhipratasthe taṁ ca deśaṁ vaśe sthāpayām āsa (Mbh. 1.3.18) “After instructing his brothers, he marched on Takṣaśilā and put that country in his power.”16 What immediately follows after the statement of Janamejaya’s conquest of Takṣaśilā is a long series of complex and interconnected episodes which introduces the nāga (serpent) chief Takṣaka as the prime offender of the Kurus. According to the story, Janamejaya’s father Parīkṣit was killed by Takṣaka, and this story is narrated to King Janamejaya by the brāhmaṇa called Utanka (Mbh 1.3. 185–190). Utanka himself was wronged by Takṣaka when he deceived and stole the earrings that Utanka was carrying for his guru’s wife (Mbh 1.3. 135–155). After learning about his father’s death from the brāhmaṇa, Janamejaya resolves to undertake a sarpa satra (snake sacrifice) to avenge the killing of his father by exterminating the entire race of the nāgas (Mbh 1.5. 46.35).

Region through text 111 Given that the entire episode of Takṣaka is told immediately after Janamejaya’s victory over Takṣaśilā, one may presume that Takṣaka was connected to the region, though the epic does not explicitly mention Takṣaka as an inhabitant of Takṣaśilā. According to D.D. Kosambi, however, Takṣaśilā and Takṣaka could be related on account of the fact that the terms are derived from the same root word takṣ which means ‘carpenter’. He says, “an alternative name for the Naga is Taksaka, which also means ‘carpenter’; the traditional Naga, a snake demon able to assume human form at will, is a superior craftsman”.17 He also draws our attention to the ancient nāga autochthones existent in the region around the Kashmir valley which supports the literary accounts.18 Janamejaya’s decision to undertake a snake sacrifice with the object of exterminating the nāgas instead of direct warfare is interesting to note. Considering Janamejaya’s reputation in the Brāhmaṇas and Śrauta Sūtras as a powerful king of the Kuru lineage, the poets of the epic could have represented him as having overtaken the nāgas in a battle. However, a yajña is chosen as a means of victory over them. The account of this massive snake sacrifice was probably deliberate. It helped the poets create the frame story of the epic and thereby create a grand occasion for the narration of the Mahābhārata by Vaiśampāyana to Janamejaya. The original composer of the epic, sage Vyāsa (Kṛṣṇa Dvaipāyana), is mentioned to have been present at the snake sacrifice (Mbh 1.54.1). He is requested by Janamejaya to narrate the accounts of the Kurus and Pāṇḍavas (Mbh 1. 54.17), who, in turn, order his student Vaiśampāyana to narrate the account to the king (Mbh 1.54.21). The next retelling of the epic happens during another sacrifice conducted by the brāhmaṇa Śaunaka at the forest of Naimiṣa where Ugraśravas narrates the epic. Ugraśravas was present during the snake sacrifice where the epic was formerly narrated and while concluding his narration, in the Svargārohaṇaparvan, he mentions that King Janamejaya returns from Takṣaśilā to Hastināpura after the completion of the rituals of the snake sacrifice. tato dvijātīn sarvāṁs tān dakṣiṇābhir atoṣayat pūjitāś cāpi te rājñā tato jagmur yathāgatam visarjayitvā viprāṁs tān rājāpi janamejayaḥ tatas takṣaśilāyāḥ sa punar āyād gajāhvayam (Mbh 18.50. 28–29) “King Janamejaya then gratified all the Brahmanas with copious presents. Thus worshipped by the king, they returned to their

112  Tanni Moitra respective abodes. Having dismissed those learned Brahmanas, king Janamejaya came back from Takshasila to the city named after the elephant.”19 These verses are significant because it is the only reference in the Mahābhārata which clearly ascribes a specific region to the sacrifice. This verse, by implication, also relates the narration of the Mahābhārata to Takṣaśilā. Therefore, it may be safe to assume that one of the retellings of the epic happened in Takṣaśilā, at least according to the Mahābhārata’s own admission. The narration was certainly an important event, as it turned out eventually that the narration of the epic took precedence over the snake sacrifice. Is it possible to ascribe some historic basis to the account of narration or composition of the epic in Takṣaśilā? Considering the fact that Takṣaśilā was a famous centre of learning in ancient India, it is not improbable that portions of the epic were composed here and that the region was the centre of transmission to successive generations. According to the words of Ugraśravas, his narration of the epic is based on what he had learnt from Vaiśampāyana during the snake sacrifice. Ugraśravas concludes his narration in Naimiṣāraṇya with the following verse: etat te sarvam ākhyātaṁ vaiśaṁpāyanakīrtitam vyāsājñayā samākhyātaṁ sarpasatre nr̥ pasya ha (Mbh 18.53.30) “I have now told everything that Vaishampayana narrated, at the command of Vyasa, unto the king at his snake sacrifice.”20 This suggests that the recitation of the epic at Takṣaśilā formed the basis for further retellings, which may indicate that the epic was also taught in the region. The epic is often glorified as a repository of knowledge and is recognized as a sacred scripture to be learnt. It is considered equal to that of the eminent sources of knowledge like the Vedas and Upaniṣads. Vyāsa is credited to have composed an Upaniṣad21 in the form of the epic, and its worth is considered superior to even the four Vedas put together.22 It is also claimed as the fifth Veda. Brockington argues that the recognition of the Mahābhārata as the fifth Veda “recognizes its character as a collection of ancient tales and proclaims their priestly nature and claims a measure of authority for it”.23 Some of the verses in the Ādiparvan emphasize its high educational value; it is said to be pursued by learned men because

Region through text 113 of its content and beautiful language.24 The text also asserts that the Mahābhārata must be memorized and retained by the dvijas,25 especially the brāhmaṇas, whose learning remains incomplete without the knowledge of the Mahābhārata.26 Moreover, it is also called śreṣṭhaḥ sarvāgama, or “chief among textbooks” (Mbh 1.2.33). This discussion indicates that though the references made to Takṣaśilā are limited in the text, the region is of paramount significance primarily for two associated reasons: first because the snake sacrifice of Janamejaya is said to have happened in Takṣaśilā and concomitantly the narration of the Mahābhārata. Corroborative evidence suggests that the epic was well known in the northwestern region. The ancient Indian grammarian Pāṇini, said to have lived roughly around the fifth to fourth centuries BCE and an inhabitant of Śalatura near Takṣaśilā, not only knew about the epic but also referred to the characters in the grammatical constructions in Aṣṭādhyāyī such as Yudhiṣṭhira, Arjuna and Vasudeva. According to material evidence so far, the northwest of the Indian subcontinent was among the first regions to begin writing down texts which were formerly transmitted orally. The process of writing down is estimated to have begun in the first century BCE by the Buddhist communities in the northwest. This was a momentous shift from the previous oral cultures, and according to Ingo Strauch, “within a comparatively short time, Buddhism became a true writing culture, engaged in the production, distribution and even veneration of written texts in the shape of manuscripts”.27An impressive number of manuscript findings in this region leave no doubt about the fact that the northwest, in general, had created a tradition of composing as well as writing down the erstwhile orally transmitted texts. In such a historical background, it is not improbable that the Mahābhārata was also composed in the region. Even though manuscript evidence of the epic is much later than its actual composition and writing, the oldest manuscripts written in the śāradā script in Kaśmir belong to the lost archetype of the northwestern group of manuscripts, which is also the shortest of any other known versions.28 There are, therefore, reasons to believe that activities beyond the ‘foreign’ and the ‘Buddhist’ were operating simultaneously in the northwest.

Gandhāra: the region From the references found in the Mahābhārata, Takṣaśilā appears to have a separate identity and is independent of any association with Gandhāra. There is no reference that relates Takṣaśilā to any of the

114  Tanni Moitra kings of Gandhāra mentioned in the text. Takṣaśilā was probably ascribed to the nāga chief Takṣaka over whom it is said that the Kurus gained control. Several literary evidences suggest that both Parīkṣit and his son Janamejaya may have existed historically. They are mentioned as powerful kings of the Kuru lineage and performers of various yajñas in the Brāhmaṇas, Śrauta Sūtras and Purāṇas. Whereas the Śatapatha Brāhmaṇa and the Āpastamba Śrauta Sūtra mention Janamejaya, the son of Parīkṣit, as a performer of the Aśvamedha yajña, the Pañcaviṃśa Brāhmaṇa relates him to a snake sacrifice.29 However, according to H. C. Raychaudhuri, it may be credible to believe that Janamejaya may have conquered Takṣaśilā because he is acknowledged as a “Sarva Bhūmi”, or paramount sovereign, but his association with a snake sacrifice may not be a historical fact. He states, “[A]lthough a passage of the Panchavimsa Brahmana connects a Janamejaya with the snake-sacrifice, the epic account of the Kuru king’s Sarpa-satra cannot be accepted as sober history”.30 Kuru influence over Takṣaśilā is indicated by the fact that Janamejaya is said to have conquered the region. However, the nature of control may have been in the form of a subjugation of the nāgas rather than ruling the region. This is evident from the fact that the snake sacrifice organized by Janamejaya is ultimately stopped by the intervention of Āstika, which saves the nāga chief Takṣaka from falling into the sacrificial fire. A compromise is reached between the Kurūs and the nāgas which eventually ends the long enmity between them. What then, do the authors of the Mahābhārata mean when they refer to Gandhāra? How are the people of Gandhāra, the region, its polity and society described by the authors? It emerges that Takṣaśilā is dissociated from the Gandhāra region. In fact, the epic does not refer to any city as being a part of what is described as Gandhāra. Gandhāra is conceived as a region in itself. A reference may be quoted from the Aśvamedhikāparvan which indicates that the region falling beyond modern Punjab was probably the Gandhāran region. It is mentioned that Arjuna, while following his sacrificial horse, crossed the region of ‘five waters’ (pañcanadaṁ) before reaching Gandhāra. There is no indication or description of Gandhāran prosperity. However, the country of the ‘five waters’ which Arjuna crosses before entering Gandhāra is described as prosperous and populous: tataḥ sa paścimaṁ deśaṁ samudrasya tadā hayaḥ krameṇa vyacarat sphītaṁ tataḥ pañcanadaṁ yayau tasmād api sa kauravya gāndhāraviṣayaṁ hayaḥ (Mbh 14. 84.18–19)

Region through text 115 “The sacrificial steed then proceeded along the coast of the western ocean and at last reached the country of the five waters which swelled with population and prosperity. Thence, O king, the steed proceeded to the country of Gandhāras.”31 Gandhāra is described as being governed by a ruler and having a strong army comprising the people of Gandhāra. Subala is addressed both as gandhārājaḥ and gandhāradhipatiḥ (Mbh. 1. 102.20; 1.103.17.4). Śakuni is also addressed as gandhārājaḥ, though he is never mentioned as ruling Gandhāra. Information about the descendants of Gandhāra after Subala is not clearly laid out. However, in the Aśvamedhikāparvan, Śakuni’s son is mentioned as the king who offers his submission to Arjuna after being defeated by him. An important geographical indicator of the region identified as Gandhāra is the description of mountains. Śakuni is mentioned as the native of the mountains32 (parvata), and the people of Gandhāra are referred to as the inhabitants of the mountains. Śakuni is said to have led an army during the war of Kurukṣetra, comprising the people of Gandhāra33described as invincible mountain warriors. The people of Gandhāra are described collectively as a clan of people who are accomplished fighters. They are often praised for their bravery and strength and were instrumental in protecting the Kaurava army during the battle. It is said that these mountain people are extremely difficult to overcome in combat (durjayah).34 Their competence with nails and lances find frequent mention in the epic. Bhīṣma, while explaining to Yudhiṣṭhira the art of using weapons, says that an army should use those weapons which they have become accustomed to using overtime and praises the Gandhārans for their competence with nails and lances.35

Gandhāra and the devāsura conflict The perennial conflict between devas and asuras is a recurrent theme in ancient Indian literature and is as old as the Ṛgveda. The theme also greatly motivates the Mahābhārata. The central story of the epic is fundamentally of a fratricidal war and the eventual victory of the Pāṇḍavas over the Kauravas. The Pāṇḍava brothers, individually and collectively, are represented as the perfect embodiment of Brahmanical virtues, whereas their enemies, the Kauravas, represent the force of evil. Considering Madhav Despande’s line of argument about the narrator– patron relationship in shaping the narrative material for recitation, it may be presumed that the major emphasis of Vaiśampāyana’s discourse

116  Tanni Moitra would have been on providing a legitimate ancestry to his patron, Janamejaya. In this process copious details are provided in the Ādiparvan about the origins and lineage of the Bhāratas from a distant past (Mbh. 1.90.1–95). Pāṇḍavas are, therefore, ascribed divinity by connecting them to devas (gods) by their birth, and Kṛṣṇa is elevated as an incarnation of Nārāyaṇa. Simultaneously, the enemies of the Pāṇḍavas viz., the Kauravas and Śakuni, are contemptuously classified as rākṣasas and asuras respectively: dharmasyāṁśaṁ tu rājānaṁ viddhi rājan yudhiṣṭhiram bhīmasenaṁ tu vātasya devarājasya cārjunam aśvinos tu tathaivāṁśau rūpeṇāpratimau bhuvi nakulaḥ sahadevaś ca sarvalokamanoharau (Mbh. 1. 61.85–87) “King Yudhiṣṭhira, O King, was a portion of Dharma; Bhīmsena of the Wind; Arjuna of Indra; and Nakula and Sahadeva, matchless in beauty on earth, enchanting to all the world were portions of the Aśvins.”36 yas tu nārāyaṇo nāma devadevaḥ sanātanaḥ tasyāṁśo mānuṣeṣv āsīd vāsudevaḥ pratāpavān (Mbh. 1.61. 90) “The sempiternal God of Gods Nārāyaṇa descended with a portion of himself among mankind as the majestic Vāsudeva.”37 kaler aṁśāt tu saṁjajñe bhuvi duryodhano nr̥ paḥ durbuddhir durmatiś caiva kurūṇām ayaśaskaraḥ jagato yaḥ sa sarvasya vidviṣṭaḥ kalipūruṣaḥ yaḥ sarvāṁ ghātayām āsa pr̥ thivīṁ puruṣādhamaḥ yena vairaṁ samuddīptaṁ bhūtāntakaraṇaṁ mahat paulastyā bhrātaraḥ sarve jajñire manujeṣv iha śataṁ duḥśāsanādīnāṁ sarveṣā krūrakarmaṇām durmukho duḥsahaś caiva ye cānye nānuśabditāḥ duryodhanasahāyās te paulastyā bharatarṣabha (Mbh. 1. 61. 80–84) “Prince Duryodhana, evil-spirited, evil minded, disgracer of the Kurus, was born on earth as portion of Kali; he was a creature of discord, hated by all the world; it was he, meanest of men, who caused the massacre of all the earth, he who fanned the great feud into a blaze that was to put an end to the beings. All his brothers

Region through text 117 were born among men as creatures of the Rākṣasas, one hundred in all, from Duḥsāsana onward, cruel of deeds, Durmukha, Duḥsaha, and the others who will not be named here.”38 The nature of the conflict between the Kauravas and Pāṇḍavas essentially illustrates the trend of the devāsura conflict found particularly in the Vedas and Brāhmaṇas. It has long been argued that there is an intrinsic relationship between the epics, both the Mahābhārata and the Rāmāyaṇa, to Vedic literature.39 Numerous perspectives have been put forth by scholars of Indo-European religions to explain this most repeated and central feature of Indian texts. The Indo-Iranian history has been used to explain the original conflict. Arguably, the term asura is applied as an adjective in the Ṛgveda with the meaning ‘mighty’, ‘spirited’ or ‘courageous’. The term did not have a negative or derogatory connotation and was applied to gods, persons, inanimate objects and even abstract ideas. The prominent Ṛgvedic deities like Indra, Agni and Varuṇa, as well as other deities, have been described as asura because of their might and strength (asurya, asuratva). In the later portions of the Ṛgveda, however, the term began to be used a noun and its meaning was altered to signify a ‘demonic force’; the term also ceased to be used for the deities or gods. The Iranian word ahura is understood as the Vedic asura. Scholars like P.L. Bhargava40 have drawn on the Zoroastrian scripture, Avesta, to explain the change in the meaning whereby he explains that the Iranian and Indian branch of Āryans originally had the same ancestors, which was split later; whereas the devas became the demons in Avesta, asura became demons in Vedic literature. Overtime, the usage of the term asura became crystallized to mean any force opposing devas. The Mahābhārata appears to continue this trend in which the forces hostile to the Pāṇḍavas are considered asuras. However, it is also interesting to note that whereas Duryodhana and his brothers are described as possessing the most detestable and immoral qualities, Śakuni is mentioned as a descendant of asura. The lineage of Śakuni is also found in the Ādiparvan in which his asura connection becomes evident. It is mentioned that the father of Śakuni and Gāndhārī, Subala, as well as Nagnajit, who is also identified as another king of Gandhāra, are the disciples of Prahrāda. Prahrāda is frequently mentioned in the Mahābhārata as an eminent Bahlīkā king and is identified as an asura.41 prahrādaśiṣyo nagnajit subalaś cābhavat tataḥ tasya prajā dharmahantrī jajñe devaprakopanāt gāndhārarājaputro ’bhūc

118  Tanni Moitra chakuniḥ saubalas tathā duryodhanasya mātā ca jajñāte ’rthavidāv ubhau (Mbh. 1.57.95–97) “Then were born Prahlāda’s pupil Nagnajit, and Subala, whose progeny became the scourge of the Law through the fury of the Gods- Śakuni Saubala son of the king of Gandhāra, and Gāndhārī, the mother of Duryodhana, both clever in profit.”42 The association of Subala as a disciple of Prahrāda offers clues to his identity by connecting him to the Bahlīkās. The Bahlīkās are mentioned often in the epic and have been identified with the Central Asian Bactria region. Prahrāda is mentioned as one of the three Bahlīkā brothers, the other two being Saṃhrāda and Anuhrāda. Saṃhrāda is said to have been born as King Śalya of Madra kingdom and Anuhrāda was born as King Dṛṣṭaketu. Scholars have identified that the names ending in hrāda, which literally means ‘armour’, corresponds to Iran.43 The devāsura conflict is further highlighted when Arjuna fights Śakuni’s brothers, Acala and Vṛṣaka, during the Kurukṣetra war. An interesting description of the combat is presented with Arjuna compared to Indra, and Acala and Vṛṣaka are compared to Bala and Vṛtṛa: syālau tava mahātmānau rājānau vr̥ ṣakācalau bhr̥ śaṁ nijaghnatuḥ pārtham indraṁ vr̥ trabalāv iva (Mbh. 7.29.9) “With all their might those great kings and brothers of your wife struck at Partha as Vritra and Bala once struck Indra.”44 On similar lines, Kṛṣṇa is also mentioned to have released King Sudarśaṇa from confinement by defeating the sons of the Gandhāra king, Nagnajit. Kṛṣṇa is also said to have abducted a Gāndhāran princess during a swayamvara. jitvā putrān nagnajitaḥ samagrān baddhaṁ mumoca vinadantaṁ prasahya; sudarśanīyaṁ devatānāṁ lalāmam (Mbh. 5.47.69) “Impetuously he churned up Gandhāra, And having defeated all Nagnajit’s sons, He freed from his fetters that friend of the Gods whose name is Sudarśanīya.”45

Region through text 119 tathā gāndhārarājasya sutāṁ vīraḥ svayaṁvare nirjitya pr̥ thivīpālān avahat puṣkarekṣaṇaḥ amr̥ ṣyamāṇā rājāno yasya jātyā hayā iva rathe vaivāhike yuktāḥ pratodena kr̥ tavraṇāḥ jarāsaṁdhaṁ mahābāhum upāyena janārdanaḥ pareṇa ghātayām āsa pr̥ thag akṣauhiṇīpatim cedirājaṁ ca vikrāntaṁ rājasenāpatiṁ balī (Mbh. 7.10.10–12) “With many of the earth’s regents bent to his will, the hero with lotus eyes took the hand of the Gandhara princess, and yoking to the bridal car the kings he had beaten cracked his whip across their backs as if they were so many steeds.”46 The ancient Iranian connection of Śakuni may further be established by the name Śakuni, which is also the name of a bird. It particularly relates to the importance of the Soma/Haoma in the ancient Indian and Iranian mythologies. Malati J. Shendge observes that throughout the Ṛgveda the Soma is said to have been brought by birds like the falcon or eagle, and terms indicating the bird, like Vena, Śyena, Suparṇa and Śakuna, are used as personal names in the Ṛgveda and other later literature. She cites some prominent examples of Ṛgvedic authors like Veno Bhārgava, Śyena Āgneya and Śakuni of Mahābhārata. She further suggests that “it is plausible that the Soma was brought by Varuṇa from the land of the Gandharvas probably from the land of the Gandharvas, through the good offices of the seers whose totem was Vena, Śyena or Śakuna”.47 It is probable that the bird Śakuna was a totem or a sacred symbol for the people of Gandhāra. An army of people identified as Śakuna is mentioned to have fought the battle of Kurukṣetra along with the Gāndhāras. vāmaṁ pakṣaṁ samāśritya droṇaputrāgragāḥ sthitāḥ pr̥ ṣṭhe kaliṅgāḥ sāmbaṣṭhā māgadhāḥ pauṇḍramadrakāḥ gāndhārāḥ śakuniprāgyāḥ pārvatīyā vasātayaḥ (Mbh. 7.19.10–11) “The bird’s back was made of the Kalıngas and the Ambashthas, the Magadhas and Paundras and Madrakas, the Gandharas and Shakunas and the peoples of the mountains in the East and those who dwell beyond them.”48

120  Tanni Moitra

Dicing in Gandhāra Daniel Michon, in his chapter titled “Dicing and Oracular Gambling at Sirkap”,49 has discussed the function of the 16 pieces of dice found in the archaeological excavations at Sirkap. Drawing upon Marshall’s excavation report, Michon reassigns the functions of the dice as religious or ritual objects which were hitherto understood as ‘play-things’ by Marshall in his report.50 He observes that 11 out of 16 dice were found in ‘ritually charged’ locations inside the Buddhist stūpas along with other objects used for religious rituals. Michon argues that the frequent association of the dice with ritually charged locations may indicate “oracular or ritual gambling”.51 Although Michon cites the Mahābhārata as evidence to indicate that dicing and oracular gambling were common practices in India in the beginning of the first millennium BCE, he also raises doubts regarding the similarity of the type of dice found in the excavations and those described in the text. What is crucial to note in this context, however, is that the dicing match between Yudhiṣṭhira and Śakuni forms one of the most integral events in the Mahābhārata, and Śakuni, in particular, is strongly associated with the game. Śakuni is described as being an expert player of dice, and his unique ability to control the dice to his advantage is frequently alluded to. His control over the dice is mentioned as the manifestation of his power, which he uses to defeat Yudhiṣṭhira, leading to the complete loss of the property of the Pāṇḍavas and eventually leading to their exile for 13 years. Yudhiṣṭhira, while contemplating his ill-fated decision to participate in the game, appropriately describes Śakuni’s ability to control the dice in the following words: ahaṁ hy akṣān anvapadyaṁ jihīrṣan; rājyaṁ sarāṣṭraṁ dhr̥ tarāṣṭrasya putrāt tan mā śaṭhaḥ kitavaḥ pratyadevīt; suyodhanārthaṁ subalasya putraḥ mahāmāyaḥ śakuniḥ pārvatīyaḥ; sadā sabhāyāṁ pravapann akṣapūgān amāyinaṁ māyayā pratyadevīt; tato ’paśyaṁ vr̥ jinaṁ bhīmasena akṣān hi dr̥ ṣṭvā śakuner yathāvat; kāmānulomān ayujo yujaś ca śakyaṁ niyantum abhaviṣyad ātmā; manyus tu hanti puruṣasya dhairyam (Mbh. 3. 35.5–15) “For I took the dice desirous to take Duryodhana’s kingship and kingdom away; but Subala’s son, the roguish gamester, thereupon played against me in Suyodhana’s cause. That trickster Śakuni, man of the mountains, sowed out the dice in the gaming hall and

Region through text 121 played with tricks against me who knew none. Then Bhīmasena, I saw his guile. On seeing the dice would always favour the wishes of Śakuni, even and odd, I’d still have been able to check myself. But anger destroys a person’s calm.”52 Śakuni’s expertise with the dice is not considered ordinary; his ability is considered illusionary and magical. Throughout the narrative, the most frequently used epithets for Śakuni include mahāmāya and śatamāya because of his unique skill in playing the game of dice. In the Mahābhārata, most of the references to dicing may be considered non-ritual or secular and are largely directed towards political motives. Besides the Yudhiṣṭhira–Śakuni episode, the Nala–Puṣkara episode, told in the Araṇyakaparvan, also indicates a similar pattern. What is unique about the Nala–Puṣkara story is the fact that dicing is explained as a form of secret knowledge which needs to be learnt. In the Araṇyakaparvan when an anguished Yudhiṣṭhira regrets playing the dice with Śakuni, sage Brihadaśva narrates to him the story of how Nala, the king of the Niṣadas, lost his kingdom to his brother Puṣkara in the game of dice and how he recovered his kingdom after acquiring the knowledge and secret of magically controlling the dice from King Rituparṇa (Mbh. 3. 52–79). There is also an early connection of Buddhism to dicing as indicated by the Buddhist text called the Pāśakakevali. The text discusses the details of dice throws and the oracular use of pāśakas, or dice. It is a fourth-century text found in a stūpa at Kamtura, a Buddhist cave temple in the Kucha Oasis that now forms a part of the Bower collection of manuscripts. Rudolf Hoernle, the editor of the text, is of the opinion that the text was written by the Buddhists from Kaśmir or Swāt valley, who had migrated to the Kucha region.53 Michon’s analysis of the archaeological finding of the dice as ritual objects and the presence of specialized texts such as the Pāśakakevali in the region lends credence to the fact that dicing, both secular and religious, was a prevalent practice in the northwestern region. The firm association of Śakuni, the king of Gandhāra, with dicing additionally supports this view. It is probable that the prevalence of the practice caught the imagination of the author(s) of the Mahābhārata to the extent that dicing became one of the central themes of the narrative. It is also apparent that the Brahmanical authors of the epic were in complete disapproval of such a practice of the occult and gambling. It is evident from the fact that the theme is taken up repeatedly by the authors, only to portray the devastating impact of the practice.

122  Tanni Moitra

Conclusion Although the Mahābhārata may not be used to establish the historicity of events considering the fact that the text is a literary composition, one may fruitfully employ the textual descriptions to understand the cultures and societies which were producing such texts. From the discussion so far, it is evident that Gandhāra played out in the poetic imagination in complex and nuanced ways. If it may be speculated that one of the primary levels of composition of the text may have happened in Takṣaśilā, as suggested in the chapter, one may also speculate that certain cultures, religious practices and mythological materials were picked up by the authors to weave the narratives around them. This may indicate that apart from Buddhism and foreign cultures, the Brahmanical tradition was active and participating in the region. In fact, Gandhāra appears as a major conflict zone for the Brahmanical authors of the Mahābhārata as it becomes evident from the several aspects of the description of the region and its people in the narrative. The central narrative is woven around characters strongly associated with the region; traces of an intense political struggle between Kuru and Gandhāra can be found when it is narrated that Gāndhārī tried to terminate her unusually long pregnancy of two years out of frustration when she came to know about the birth of Yudhiṣṭhira before the birth of her own son. This was probably due to the anticipation that it would take away the chance of her son to become the heir to the Kuru throne (Mbh. 1. 107.10). Moreover, after the birth of Duryodhana, Vidura is said to have foreseen a massive political disaster lurking and therefore suggests Dhṛtarāṣṭra to abandon Duryodhana immediately (Mbh. 1. 107.30–35). The contemptuous perception of the authors towards Gandhāra appears to be a consistent feature. In some instances Gandhāra is grouped within a territorial area called āraṭṭa, along with the prasthalas, madras, khashas, vasatis, sindhus and sauvıras. The group is collectively identified as vahikā (bahlikā).54 The region identified as āraṭṭa is said to lie beyond the five rivers, and the people residing in this region are collectively known as the vahikās. The region as a whole and the people comprising the region are condemned for being lawless, unrighteous and despicable.55 A sharp distinction is particularly drawn out between āraṭṭa and Kuru-Pāñcāla, Matsya, Kosala, Cedi, Naimiṣa and Magadha: kuravaḥ sahapāñcālāḥ śālvā matsyāḥ sanaimiṣāḥ kosalāḥ kāśayo ’ṅgāś ca kaliṅgā magadhās tathā cedayaś ca mahābhāgā dharmaṁ

Region through text 123 jānanti śāśvatam nānādeśeṣu santaś ca prāyo bāhyā layād r̥ te ā matsyebhyaḥ kurupāñcāladeśyā; ā naimiṣāc cedayo ye viśiṣṭāḥ dharmaṁ purāṇam upajīvanti santo; madrān r̥ te pañcanadāṁś ca jihmān (Mbh. 8.30.62–65) “The Kauravas with the Pancalas, the Salwas, the Matsyas, the Naimishas, the Koshalas, the Kasapaundras, the Kalingas, the Magadhas, and the Cedis who are all highly blessed, know what the eternal religion is. The wicked even of these various countries know what religion is. The Vahikas, however, live without righteousness. Beginning with the Matsyas, the residents of the Kuru and the Pancala countries, the Naimishas as well and the other respectable peoples, the pious among all races are conversant with the eternal truths of religion.”56 With reference to Gandhāra, it is said that the people of Gandhāra are impure, particularly because the two classes of people, the kṣatriya and the brāhmaṇa, are mixed in their duties. The king performs the role of both the sacrificer and the priest. duḥsparśaṁ śaucaṁ gāndhārakeṣu ca rājayājakayājyena naṣṭaṁ dattaṁ havir bhavet (Mbh. 8.27.80.1) “There’s no purity among the Gandhāras. Given that their king is both sacrificer and patron of the sacrifice, whatever oblation is offered shall be ruined.”57 The description of the region of āraṭṭa gives a picture of a complex composition of a population comprising several groups of people who appear to fall beyond the pale of the Āryan religion, culture and way of life. This region and its people are frequently described as difficult to control. Interestingly, however, there is a constant portrayal of religious, social and cultural engagement with the region. The narrative of King Māndhātṛ told in the Śāntiparvan may be cited in this context. In a conversation between Yudhiṣṭhira and Bhīṣma on the duties of kings, Bhīṣma relates the story of Māndhātṛ who is said to have become the king of Gandhāra when dānavas were said to have multiplied. Māndhātṛ performs a sacrifice to invoke Nārāyaṇa to understand from him the ways of commanding a deviant population who have lost their religion, law and duties. Nārāyaṇa appears in the form

124  Tanni Moitra of Indra to answer his questions, suggesting that he enforce a noble and religious way of life among the various groups of people found in the region. Māndhātṛ asks the following question: māndhātovāca yavanāḥ kirātā gāndhārāś cīnāḥ śabarabarbarāḥ śakās tuṣārāḥ kahvāś ca pahlavāś cāndhramadrakāḥ oḍrāḥ pulindā ramaṭhāḥ kācā mlecchāś ca sarvaśaḥ rahmakṣatraprasūtāś ca vaiśyāḥ śūdrāś ca mānavāḥ kathaṁ dharmaṁ careyus te sarve viṣayavāsinaḥ madvidhaiś ca kathaṁ sthāpyāḥ sarve te dasyujīvinaḥ etad icchāmy ahaṁ śrotuṁ bhagavaṁs tad bravīhi me tvaṁ bandhubhūto hy asmākaṁ kṣatriyāṇāṁ sureśvara (Mbh. 12.65.13–16) “Māndhātar said: Greeks, mountain folk, Gāndhārans, Chinese, savages, barbarians, Śakas, Tuṣāras, Kahvas, Persians, Andhras, Madrakas, Oḍras, Pulindas, Ramaṭhas, Kācas, and Mlecchas all, and men who are sons of brahmins and kṣatriyas, and also vaiśyas and śūdras: How can all of these who live within a kingdom do Meritorious and Lawful Deeds? How can all those who live as barbarians be kept with Law by men like me? This is what I want to hear. Tell me this, blessed one. For you have become friends of us kṣatriyas, O lord of the Gods.”58 Indra answers, indra uvāca mātāpitror hi kartavyā śuśrūṣā sarvadasyubhiḥ ācāryaguruśuśrūṣā tathaivāśramavāsinām bhūmipālānāṁ ca śuśrūṣā kartavyā sarvadasyubhiḥ vedadharmakriyāś caiva teṣāṁ dharmo vidhīyate pitr̥ yajñās tathā kūpāḥ prapāś ca śayanāni ca dānāni ca yathākālaṁ dvijeṣu dadyur eva te ahiṁsā satyam akrodho vr̥ ttidāyānupālanam bharaṇaṁ putradārāṇāṁ śaucam adroha eva ca dakṣiṇā sarvayajñānāṁ dātavyā bhūtim icchatā pākayajñā mahārhāś ca kartavyāḥ sarvadasyubhiḥ etāny evaṁprakārāṇi vihitāni purānagha sarvalokasya karmāṇi kartavyānīha pārthiva (Mbh. 12. 65.17–22) “Indra said: All the barbarians must obey their mother and father, the way those who live religious Patterns of Life obey their teachers and guides. And all the barbarians must obey kings.

Region through text 125 The meritorious rites prescribed in the Vedas are Lawful deeds prescribed for them. And likewise the rites for worship for the ancestors. And they should make gifts to the Brahmins at appropriate times- wells, cisterns, and shelters for sleeping. Non-injury, truthfulness, not being quick to anger, the preservation of what has been gained through inheritance and work, supporting wives and children, cleanliness, and benevolence are also prescribed for them. He who desires riches must give presents to the priests for all the sacrificial rites. All barbarians must perform the cooked offerings. Such rites as these were enjoined in the past, and they are to be performed now by all the peoples, O blameless king.”59 The fact evident in this description expresses that Gandhāra is neither ‘foreign’ nor ‘non-indigenous’. The description portrays a sense of challenge that the Brahmanical tradition was facing in the region. The discussion so far also demonstrates the plurality of the religious, political and socio-cultural space of the region where the indigenous Brahmanical tradition also played a significant part.

Notes 1 For a detailed analysis of the historiography of Early Northwest India, see Daniel Michon, Archaeology and Religion in Early Northwest India, Routledge, New Delhi, 2015. 2 Daniel Michon, Archaeology and Religion in Early Northwest India, Routledge, New Delhi, 2015: 20. 3 Ibid, p. 24. 4 Ibid, p. 43. 5 Ibid, p. 58. 6 For a detailed analysis of the history of epic studies see J.L. Brockington, The Sanskrit Epics, Brill, Leiden, 1998: 41–81. 7 Ibid, p. 44. 8 Cited in R.N. Dandekar, TheMahābhārata: Origin and Growth, University of Ceylon Review, 12.2, 1954: 68. 9 R.N. Dandekar, The Mahābhārata: Origin and Growth, University of Ceylon Review, 12.2, 1954: 65–85. 10 Ibid, p. 72. 11 Ibid, p. 73. 12 Madhav M. Deshpande, Interpreting the Mahābhārata, in T.S. Rukmani edited, The Mahābhārata: What Is Not Here Is Nowhere Else, Munshiram Manoharlal, 2005: 3–16. 13 Ibid, p. 9. 14 James L. Fitzgerald, Mahābhārata, in Sushil Mittal and Gene Thursby edited, The Hindu World, Routledge, New York, 2004: 52. 15 H.C. Raychaudhuri, Political History of Ancient India – From the Accession of Parikshit to the Extinction of the Gupta Dynasty, University of Calcutta, Calcutta, 1923: 24.

126  Tanni Moitra 16 J.A.B. van Buitenen (tr.), The Mahābhārata: The Book of the Beginning, vol. 1, University of Chicago Press, Chicago, 1973: 45. 17 D.D. Kosambi, An Introduction to the Study of Indian History, Popular Prakashan, Bombay, 1975 [1956]: 129. 18 Ibid, p. 129. 19 K.M. Ganguli (tr.), The Mahabharata of Krishna-Dwaipayana Vyasa Translated into English Prose from the Original Sanskrit Text: Swargarohanika Parva, vol. 12, MunshiramManoharlal, New Delhi, 1991 [1972]: 10. 20 Ibid, p. 10. 21 atropaniṣadaṁ puṇyāṁ kr̥ ṣṇadvaipāyano ’bravīt bhāratādhyayanāt puṇyād api pādam adhīyataḥ (Mbh 1.191–192). “In this book Kṛṣṇa Dvaipāyana has uttered a holy Upaniṣad. They who learn even a quarter couplet of the holy study of the Bhārata, and have faith in it will be purified of all their sins.” J.A.B van Buitenen (tr.), The Mahābhārata: The Book of the Beginning, vol. 1, University of Chicago Press, Chicago, 1973: 30. 22 catvāra ekato vedā bhārataṁ caikam ekataḥ samāgataiḥ surarṣibhis tulām āropitaṁ purā mahattve ca gurutve ca dhriyamāṇaṁ tato ’dhikam mahattvād bhāravattvāc ca mahābhāratam ucyate niruktam asya yo veda sarvapāpaiḥ pramucyate (Mbh. 1.1.208–209) “Once the divine seers foregathered, and on one scale they hung the four Vedas in the balance, and on the other scale The Bhārata; and both in size and in weight it was the heavier. Therefore because of its size and its weight, it is called the Mahābhārata – he who knows this etymology is freed from all sins” Ibid, p. 31. 23 J.L. Brockington, The Sanskrit Epics, Brill, Leiden, 1998: 5. 24 alaṁkr̥ taṁ śubhaiḥ śabdaiḥ samayair divyamānuṣaiḥ chandovr̥ ttaiś ca vividhair anvitaṁ viduṣāṁ priyam (Mbh. 1.1.25) “Fine words adorn it, and usages human and divine; many meters scan it; it is the delight of the learned.” J.A.B van Buitenen (tr.), The Mahābhārata: The Book of the Beginning, vol. 1, University of Chicago Press, Chicago, 1973: 21. 25 idaṁ tu triṣu lokeṣu mahaj jñānaṁ pratiṣṭhitam vistaraiś ca samāsaiś ca dhāryate yad dvijātibhiḥ. (Mbh. 1.1.24). “It is indeed a great storehouse of knowledge, rooted in the three worlds which the twice born retain in all its parts and summaries. Ibid, p. 21. 26 yo vidyāc caturo vedān sāṅgopaniṣadān dvijaḥ na cākhyānam idaṁ vidyān naiva sa syād vicakṣaṇaḥ “A Brahmin who knows the four Vedas with their branches and Upaniṣads, but does not know this epic, has no learning at all” (Mbh 1.2.235). Ibid, p. 43. 27 Ingo Strauch, Looking into Water-pots and over a Buddhist Scribe’s Shoulder – On the Deposition and the Use of Manuscripts in Early Buddhism, ASIA,68.3, 2014: 797–798. 28 Vishnu S. Sukthankar, The Ādi Parva, Bhandarkar Oriental Research Institute, Poona, 1933: XXX. 29 H.C. Raychaudhuri, Political History of Ancient India – From the Accession of Parikshit to the Extinction of the Gupta Dynasty, University of Calcutta, Calcutta, 1923: 9.

Region through text 127 30 H.C. Raychaudhuri, Political History of Ancient India – From the Accession of Parikshit to the Extinction of the Gupta Dynasty, University of Calcutta, Calcutta, 1923: 10. 31 K.M. Ganguli (tr.), The Mahabharata of Krishna-Dwaipayana Vyasa Translated into English Prose from the Original Sanskrit Text: Aswamedha Parva, vol. 12, MunshiramManoharlal, New Delhi, 1991 [1972]: 144 32 gāndhārarājaḥ śakuniḥ pārvatīyo nikartane yo 'dvitīyo 'kṣadevī (Mbh, 5. 30. 27.1) “The mountain king of Gandhāra, Śakuni, Unmatched at cutting and playing dice, Who raises the pride of Dhṛtarāṣṭra, Do ask good friend, that hypocrite’s health.” J.A.B van Buitenen (tr.), The Mahābhārata: The Book of the Effort, vol. 3, University of Chicago Press, Chicago, 1978: 248 33 taṁ sarvataḥ śakuniḥ pārvatīyaiḥ; sārdhaṁ gāndhāraiḥ pāti gāndhārarājaḥ (Mbh. 6.20.8) “Sakuni, the ruler of the Gandharas, followed with mountaineers of Gandhara placed all around”. K.M. Ganguli (tr.), The Mahabharata of Krishna-Dwaipayana Vyasa Translated into English Prose from the Original Sanskrit Text: Bhisma Parva, vol. 5, MunshiramManoharlal, New Delhi, 1990 [1970]: 44. 34 gāndhāribhir asaṁbhrāntaiḥ pārvatīyaiś ca durjayaiḥ (Mbh. 8.31.12) “Gandhara horsemen armed with bright lances, and many mountaineers difficult to defeat”. K.M. Ganguli (tr.), The Mahabharata of Krishna-Dwaipayana Vyasa Translated into English Prose from the Original Sanskrit Text: Karna Parva, vol. 7, MunshiramManoharlal, New Delhi, 1990 [1970]: 113 35 gāndhārāḥ sindhusauvīrā nakharaprāsayodhinaḥ "ābhīravaḥ subalinas tadbalaṁ sarvapāragam (Mbh. 12.102.4) “Gāndhāras, Sindhus, Sauvīras, and Ābhīras fight with claws and darts and are very potent. Their armies are fully conversant with all weapons and are courageous.” James. L. Fitzgerald (tr.), The Mahābhārata: The Book of Peace, University of Chicago Press, Chicago, 2004: 423. 36 J.A.B van Buitenen (tr.), The Mahābhārata: The Book of the Beginning, vol. 1, University of Chicago Press, Chicago, 1973: 154. 37 Ibid, p. 154. 38 Ibid, p. 153. 39 For an analysis of the relationship of the epics with the Vedic literature, see J.L. Brockington, The Sanskrit Epics, Brill, Leiden, 1998. 40 P.L. Bhargava, The Word Asura in the Ṛgveda, Annals of the Bhandarkar Oriental Research Institute, 64.1.4, 1983: 119–128. 41 saṁhrāda iti vikhyātaḥ prahrādasyānujas tu yaḥ sa śalya iti vikhyāto jajñe bāhlīkapuṁgavaḥ (Mbh. 1. 61.30) “The second of the Asuras, Śalabha, becomes the Bāhlikā king, Prahrāda.” J.A.B. van Buitenen (tr.), The Mahābhārata: The Book of the Beginning, vol. 1, University of Chicago Press, Chicago, 1973: 152. 42 Ibid, p. 135. 43 Prods Oktor Sklervo, Eastern Iranian Epic Traditions II: Rostam and Bhīṣma, Acta Orientalia Academiae Scientiarum Hung, 51.1.2, 1998: 159–170. 44 Pilikian Vaughan (tr.), Mahābhārata. Book Seven, Droṇa, vol. 1, Clay Sanskrit Library, New York University Press and the JJC Foundation, New York, 2006, p. 257.

128  Tanni Moitra 45 J.A.B. van Buitenen (tr.), The Mahābhārata: The Book of the Effort, vol. 3, University of Chicago Press, Chicago, 1978: 304. 46 Pilikian Vaughan (tr.), Mahābhārata. Book Seven, Droṇa, vol. 1, Clay Sanskrit Library, New York University Press and the JJC Foundation, New York, 2006, p. 103 47 Malati J. Shendge, The Civilized Demons: The Harappans in Rigveda, Abhinava Publications, New Delhi, 1977, p. 109 48 Pilikian Vaughan (tr.), Mahābhārata. Book Seven, Droṇa, vol. 1, Clay Sanskrit Library, New York University Press and the JJC Foundation, New York, 2006: 171. 49 Daniel Michon, Archaeology and Religion in Early Northwest India, Routledge, New Delhi, 2015: 152–200. 50 John Hubert Marshall, Taxila: An Illustrated Account of Archaeological Excavations Carried Out at Taxila under the Orders of the Government of India Between the Years 1913 and 1934, Cambridge University Press, 1951. 51 Daniel Michon, Archaeology and Religion in Early Northwest India, Routledge, New Delhi, 2015: 183. 52 J.A.B. van Buitenen (tr.), The Mahābhārata: The Book Forest, vol. 2, University of Chicago Press, Chicago, 1975: 290. 53 Cited in Daniel Michon, Archaeology and Religion in Early Northwest India, Routledge, New Delhi, 2015: 183. 54 Prasthalā, Madra Gandhāra, Āraṭṭa nāmatah., Khasah. vasātisindhusauvīrā iti prāyo vikutsitāḥ (Mbh. 8.44.45) “The Prasthalas, Madras, Gandharas, Khashas, Vasatis and Sindhus and Sauvıras are known as Āraṭṭas”. Adam Bowles (tr.), Mahābhārata, Book 8, Karṇa, vol 1. Clay Sanskrit Library, New York University Press and the JJC Foundation, New York, 2006: 441. 55 yad anyo ’py uktavān asmān brāhmaṇaḥ kurusaṁsadi pañca nadyo vahanty etā yatra pīluvanāny api śatadruś ca vipāśā ca tr̥ tīyerāvatī tathā candrabhāgā vitastā ca sindhuṣaṣṭhā bahir gatāḥ āraṭṭā nāma te deśā naṣṭadharmān na tān vrajet vrātyānāṁ dāsamīyānāṁ videhānām ayajvanām (Mbh. 8.30.34–37). “This is what another brahmin said to us at the Kuru’s court: ‘Five rivers flow where the forests of pilu trees are: the Shatadru, the Vipasha, the Airavati is the third, the Chandrabhaga and the Vitasta. A sixth, the Sindhu, flows from the other side of the mountain. These lands, called the Arattas, are lawless and one should avoid them. The gods, ancestors and brahmins do not accept anything from vagrants, from the Dasamıyas, from Vahikas who don’t perform sacrifices and from Vahıkas who have no law. This is what’s taught.” Adam Bowles, (tr.), Mahābhārata, Book 8, Karṇa, vol 1. Clay Sanskrit Library, New York University Press and the JJC Foundation, New York, 2006, p. 437. 56 K.M. Ganguli (tr.), The Mahabharata of Krishna-Dwaipayana Vyasa Translated into English Prose from the Original Sanskrit Text: Karna Parva, vol. 7, MunshiramManoharlal, New Delhi, 1990 [1970], p. 110 57 Adam Bowles (tr.), Mahābhārata, Book 8, Karṇa, vol. 1. Clay Sanskrit Library, New York University Press and the JJC Foundation, New York, 2006: 397. 58 James L. Fitzgerald (tr.), The Mahābhārata: The Book of Peace, University of Chicago Press, Chicago, 2004: 328. 59 Ibid, pp. 328–329.

Region through text 129

References Primary Sources Bowles, A., tr. 2006. Mahābhārata, Book 8, Karṇa, vol. 1. New York: Clay Sanskrit Library, New York University Press and the JJC Foundation. Buitenen, van J.A.B., tr. 1973–1978.The Mahābhārata [Books 1–5], 3 vols. Chicago: University of Chicago Press. Fitzgerald, J., tr. 2004. The Mahābhārata [Books 11 and 12], vol. 7. Chicago: Chicago University Press. Ganguli, K.M., tr. 1973–1975. The Mahabharata of Krishna-Dwaipayana Vyasa Translated into English Prose from the Original Sanskrit Text, 19 vols. New Delhi: Munshiram Manoharlal. Vaughan, P., tr. 2006. Mahābhārata. Book Seven, Droṇa, vol. 1. New York: Clay Sanskrit Library, New York University Press and the JJC Foundation.

Secondary Literature Bhargava, P.L. 1983.The Word Asura in the “Ṛgveda”. Annals of the Bhandarkar Oriental Research Institute, 64.1‑4: 119–128. Brockington, J. 1998.The History of Epic Studies, in The Sanskrit Epics. Leiden: Brill. Dandekar, R.N. 1954.The Mahābhārata: Origin and Growth. University of Ceylon Review 12.2. Deshpande, M. 2005.Interpreting the Mahābhārata, in T.S. Rukmani (ed.), The Mahābhārata: What Is Not Here Is Nowhere Else. New Delhi: Munshiram Manoharlal. Fitzgerald, J. 2004. Mahābhārata, in S. Mittal, and G. Thursby (eds.), The Hindu World. New York: Routledge. Kosambi, D.D. 1975 [1956]. An Introduction to the Study of Indian History. Bombay: Popular Prakashan. Michon, D. 2015. Archaeology and Religion in Early Northwest India. New Delhi: Routledge. Raychaudhuri, H.C. 1923. Political History of Ancient India – From the Accession of Parikshit to the Extinction of the Gupta Dynasty. Calcutta: University of Calcutta. Shendge, M. 1977. The Civilized Demons: The Harappans in Rigveda. New Delhi: Abhinava Publications, p. 109. Sklervo, P.O. 1998. Eastern Iranian Epic Traditions II: Rostam and Bhīṣma. Acta Orientalia Academiae Scientiarum Hung 51.1–2: 159–170. Strauch, I. 2014. Looking into Water-pots and over a Buddhist Scribe’s Shoulder – On the Deposition and the Use of Manuscripts in Early Buddhism. ASIA 68.3: 797–798. Sukthankar, V.S. 1933. The Ādi Parva. Poona: Bhandarkar Oriental Research Institute.

5 Charles Masson A footloose antiquarian in Afghanistan and the building up of numismatic collections in the museums in India and England Sanjay Garg Museums, the world over by and large, hold portable antiquities and/ or works of art and/or other ‘collectibles’. Also, as compared to some of the major museums of the West, Indian museums, more specifically the public museums in India, generally draw their collections from state sources – mainly from the finds of state-sponsored excavations, or treasure trove finds, or state-financed acquisitions. A few Indian museums have also received generous bequests of the collectors and connoisseurs of art objects, but their number as compared to the museums in the West is much smaller. For example, the majority of the collection of the Fitzwilliam Museum in Cambridge (UK) comes from the collections of the fellows of different colleges, which, in turn, havebeen loaned to the Fitzwilliam Museum. Similarly, the nucleus of the British Museum that was opened in 1759 was formed by the extensive collection of scientific and literary objects collected by Sir Hans Sloane (1660–1753), comprising 117,000 items, including 45,000 books and 5,000 manuscripts. In fact after Sloane’s death in 1753, his entire collection was purchased under an Act of Parliament (26 Geo II cap 22) for a sum of 20,000 GBP. Again, if we see the whole gamut of the portable antiquities, we find that due to the sheer prolificacy of their production, coins of the bygone era constitute the most commonly encountered portable antiquity, and thus constitute a major proportion of the total holdings of many museums, both in India and worldwide. In the National Museum, New Delhi, for example, out of 200,000 objects in its collection, 120,000 are coins and other numismatic objects. As such, the importance of the coin collections and the people who collect these ‘metallic mirrors’ (a term borrowed from Bhandare 2006)

Charles Masson 131 of the past is well recognized in the making of museums’ collections. And this leads us to the theme of this chapter, which is devoted to one of the greatest coin collectors of all times, who, even one and a half centuries after his death, is better known by his pseudonym: Charles Masson. Between 1833 and 1838, Masson explored or excavated nearly 50 monuments in the vicinity of Peshawar and Kabul and amassed a staggering amount of antiquities. These included a large number of coins, of which the British Museum received a substantial share. Other recipients of Masson’s coin collection were the Asiatic Society, Calcutta [Kolkata], Ashmolean Museum, Oxford, and the Fitzwilliam Museum, Cambridge. The story of Masson’s antique-collecting escapades in Afghanistan during the years preceding the First Anglo-Afghan War (1838–1842) and the subsequent dispersal of his fabulous collection of antiquities, which included seals, beads, ingots, weights, ornaments, plaques, sculptures, reliquaries (caskets, boxes, and bowls), arrowheads, discs, amulets, buttons, etc., together with a large number coins, in various museums of India and England lie scattered in the official archives of the British East India Company as well in his private correspondence, which constitute the main source of this study.

Charles Masson: A footloose antiquarian No comprehensive account of Masson’s life and works was available until 1986, when Sir Gordon Whitteridge, who had served as the British Ambassador to Afghanistan between 1965 and 1968, published the first biography of Charles Masson (Whitteridge 1986). As Whitteridge’s account did not incorporate the archival sources available in India, I surveyed the sources available in the National Archives of India, and an abstract of the findings was submitted to the International Numismatic Congress, Brussels (Garg 1991). Later, in 1993, the British Museum, which had received a substantial share of Masson’s collection, launched a project called ‘the Masson Project’. This project was aimed at identifying and documenting his collections in the British Museum and elsewhere and reconstructing the archaeological record of the sites, many of which no longer exist. The project has resulted in a series of valuable contributions from the pen of the curator of the project and an eminent scholar of Gandhara studies, Dr. Elizabeth Errington (q.v.). As such, a short career graph of our protagonist may suffice here to situate the subsequent discussion (Table 5.1). This segment

132  Sanjay Garg Table 5.1  Charles Masson (1800–1853): a career graph 1800–1821 1822–1827 1827–1832 1832–1834 1834–1838 1838–1839 1840–1842 1842–1853

Early Life in England Army service in India Travels Antiquarian researches in Afghanistan British spy/antiquarian researches in Afghanistan Quitting of government service; antiquarian researches resumed Retreat from Afghanistan Back in London

is substantially drawn from my earlier works (Garg 1991, 1998, and 2014). Practically nothing is known of our hero’s early life except for the fact that he was born on 16 February 1800 in London as James Lewis. Like many young and aspiring youths of his age, on 5 October 1821 he enlisted in the armed forces of the East India Company. He gave his age as 19 though he was in fact almost 22. The recruiting register describes him as 5' 3" (170 cm) tall, fair-complexioned, and with hazel eyes and brown hair. The following year, on 17 January 1822, along with 290 fellow recruits aboard the East Indiaman Duchess of Athall, Lewis sailed for Bengal. From 1822 to 1827, he served in the 3rd Troop of the 1st Brigade of the Bengal Horse Artillery. In 1827 when his regiment was posted at Agra, he deserted from the army. Under the self-assumed citizenship of Kentucky (USA) and a fake name, Charles Masson, he wandered over a large tract of land – ‘penniless at first, usually alone and on foot’. He traversed through Rajputana to Bahawalpur (1827); to Peshawar (1828) and thence through the Khyber Pass to Jalalabad and Kabul (1828); then to Baluchistan and Sind (1829) and finally to Persia (1830). In Persia, for the first time since his desertion from the army, Masson came in contact with a British official. To Major David Wilson, British resident at Bushire, Masson introduced himself as an American traveller from Kentucky and acted so well under his assumed persona and a pseudonym that poor Wilson could never doubt the ‘genuineness’ of his guest’s claim. Wilson forwarded to the government of Bombay a large number of Masson’s papers relating to his description of the prevailing conditions in Afghanistan, including an account of the city of Herat. These papers aroused great interest of the English authorities in India. Diligent enquiries then followed which eventually ‘unmasked’ Masson.

Charles Masson 133 From Persia Masson returned to Kabul in May 1832, which was to become his home for the next six years. It was in the Kabul bazaar that Mir Karamat Ali, a hawk-eyed British ‘news writer’ (intelligence agent) spotted Masson – shabbily dressed and barefoot, a green cap on his shaven head, red-bearded, ‘and a faqir or dervish drinking cup slung over his shoulder’ (Grey 1929: 188). From the end of 1832 until 1838, Masson made the city of Kabul his headquarters and devoted himself to antiquarian researches in Kohistan and in the neighbourhood of Jalalabad. Meanwhile realizing the importance of Masson’s first-hand knowledge of that politically sensitive land of Afghanistan and his acquaintance with its people, the East India Company decided to use Masson for its political gain. A Royal Warrant of Free Pardon to the deserter James Lewis was obtained and, in 1834, he was appointed a newswriter at Kabul on a monthly salary of Rs. 250. From 1834 until 1838, Masson very reluctantly performed his dual role as a (re)searcher of antiquities and a government agent. In 1838, in the wake of declaration of the First Anglo-Afghan War (1838–1842), Masson resigned from the services of the East India Company and once again engaged himself in his favourite pursuit of exploration of antiquities, and we find him making excursions into Shahbazgarhi and the Yusufzai districts. At the former, he discovered an Ashokan rock edict. In 1839, suffering from ill health, Masson spent most of his time at Tatta (Sind) writing his narratives and his memoir for H.H. Wilson’s Ariana Antiqua. In the early part of 1840, having dispatched his manuscripts for publication in England, Masson decided to resume his travels and antiquarian researches. According to his own estimate, he intended to complete his work in Afghanistan within two years. However, while on his journey to the Afghan land, he was estranged in the strife-torn principality of Kalat. He was taken prisoner by one of the local insurgent leaders and in the loot that ensued he lost ‘a large accumulated stock of manuscripts and papers, the fruit of above fifteen years’ labour and inquiry’. Later, when the insurgents chose him to negotiate their case with Captain Bean, political agent at Quetta, the latter arrested him on suspicion of being a Russian spy. He was finally released in January 1841 after which he returned to Karachi and thence sailed to Bombay, arriving there in October 1841. Masson had by now decided to return to England. He sailed from Bombay on 1 November 1841 by a steamer to Suez, crossed Egypt overland, and after making a small stopover at Paris finally reached London in March 1842.

134  Sanjay Garg From the moment he returned to England he demanded financial compensation, not only for his scandalously unjust treatment at Quetta but also for what he described as ‘the trifling sums doled out to me on account of the researches and antiquities collected in Afghanistan’ (quoted in Alder 1975: xiv). Dejected with the bureaucratic apathy in England, Masson contemplated returning to Afghanistan and resuming his antiquarian researches with the assistance of the Austrian government. This, too, did not work. He died in despair on 5 November 1853.

Significance of Masson’s numismatic discoveries Masson’s contribution in the field of numismatic studies is most remarkable. The value as well as the quantum of coins collected by him during his thirteen years’ labour in Afghanistan far exceeds the general expectations. ‘When Masson began his investigations he had most of this untilled field, wide both in space and time, virtually to himself. No wonder he succeeded beyond his wildest dreams’ (Alder 1975: VII). It was Masson’s numismatic discoveries in Afghanistan that helped in deciphering the ancient Indian scripts – the Brahmi and Kharoshthi – which ‘made a vast wealth of inscriptional and textual material available for the first time, for use in attempting to reconstruct a historical picture of the past’ (Errington and Curtis 2014: 25). Brahmi was used extensively on the bilingual coins of the GrecoBactrian kings that ruled over a vast tract of land south of Hindu Kush c. second century BCE. Though prior to discoveries of these coins attempts were made by Charles Wilkins (1749–1836) and Henry Colebrooke (1765–1837) to decipher this script on the basis of various epigraphs, the presence of Greek alphabets on these bilingual coins provided a definite clue to assign the correct phonetic value to a majority of characters of this enigmatic script. Thus in 1836, Brahmi legend on the coin of Agathocles (c. 190–180 BCE) was correctly read by the Norwegian scholar Christian Lassen (1800–1876) (Errington and Curtis 2014: 21–22). By the next year, the task of deciphering the rest of the Brahmi characters was completed by James Prinsep (1799– 1840) with the help of Alexander Cunningham (1814–1893). Another ancient script widely used in the Gandhara region was the Kharoshthi. It was used on the bilingual (also some rare monolingual) coins of the Indo-Greeks, Indo-Scythians, and the Kushans from c. second century BCE to third century CE. Masson, who calls this

Charles Masson 135 unknown script Pehlavi or Bactrian, was the first to make a successful attempt towards its decipherment in 1835. As in the case of Brahmi, ‘the realisation that the reverse legends on the bilingual issues of the Bactrian and Indo-Greeks were direct translations of the Greek on the obverse of the coins was the starting point for deciphering Kharoshthi’ (Errington and Curtis 2014: 23). Once again, Prinsep carried forward the spade work done by Masson and completed the task. According to Errington and Curtis, ‘greater emphasis should be placed on Masson’s contribution, not in decipherment of Kharoshthi per se, but in supplying all the raw material – in the form of legends diligently transcribed from his huge collection of coins’ (Errington and Curtis 2014: 25). A tribute to Masson’s contribution, not only to the field of numismatics, but also to the general history of this ‘broad sweep of lands on and beyond India’s north-west frontier’ about which nothing was known ‘from the time of Alexander’s death down to the main Muslim invasions of the 12th century’ (Alder 1975: vii), has been recorded in the Centenary Volume of the Asiatic Society of Bengal by Hoernle (1885: 30–31), who paraphrases the words of H.H. Wilson (Wilson 1841: 12): Among the coins discovered by him are not only new ones of Greek princes already known, but also those of several whose names are not mentioned in history as Antialkidas, Lysias, Agathocles, Archebias, Panthaleon, and Hermaeus. He also found coins of the king whose titles are only specified as the Great King of Kings, the Preserver, and others whose names, although assuming a Greek form, indisputably denote barbaric or Indo-Scythic princes – Undopherres, Azes, Azilises, Kadphises and Kanerkes. The first step in the series of the Bactrian numismatic discovery was thus accomplished, and the great object of later investigations became only to complete and extend the structure, of which such broad foundations had been laid. Or, as Gary Alder writes: His collection of coins, ranging over 15 centuries, was unsurpassed at a time when they were proving in the hands of scholars to be an unprecedently effective tool for illuminating the darkness of post- Alexandrine chronology. (Alder 1975: xvii)

136  Sanjay Garg The estimates for the number of coins collected by Masson from Afghanistan vary from 60,000 to 120,000 pieces. Wilson observed: The far greatest proportion, judging from Mr. Masson’s collection, must have been too much injured by time and corrosion to have had any other than metallic value; but from the same accumulations we may infer that great numbers of coins, of high numismatic interest, must have perished in the indiscriminate destruction to which the whole have for so long a time been condemned. (Wilson 1841: 12)

Dispersal of Masson’s coin collections According to Masson, ‘The whole of the coins, and other antiquities from Begram, with several thousands of other coins brought to light in various parts of Afghanistan, have been forwarded to the Honourable the East India Company’ (Masson 1842, vol. III: 149). Under an arrangement, Masson sent his collection of coins and other antiquities at intervals to Bombay, from where it was transmitted to London and deposited there in the East India Company Museum in Leadenhall Street. Asiatic Society/Indian Museum, Kolkata (55 coins) However, when reports of the dispatch of Masson’s collection of antiquities to England reached Bengal, James Prinsep, secretary of the Asiatic Society, requested the government to make that collection available for his examination before it was sent to England. The government of India acceded to Prinsep’s request and on 26 June 1837 instructed the government of Bombay that the coins and relics received from Masson should first be sent to Calcutta for examination. However, by the time these instructions were received at Bombay, the government of Bombay had already sent two lots of Masson’s collection – the first lot comprising coins and other articles collected by Masson during 1833 and 1834, and the second lot comprising coins, etc., collected during 1835 – to England on 24 January and 26 April 1837, respectively. The next lot of Masson’s collection of coins and relics, which was received at Bombay through Colonel Pottinger, was forwarded to Calcutta by the ship John Adams on 15 October 1838. Its receipt

Charles Masson 137 was acknowledged by the secretary to the government of India in a letter dated 19 December 1838, stating that the collection ‘will be forwarded to the Court of Directors after examination of the Asiatic Society of Bengal and the learned and curious on such subject at this Presidency’. Meanwhile, things had changed in Bengal. James Prinsep had left Calcutta for England in October 1838, where he eventually died on 22 April 1840. ‘The Asiatic Society declined to undertake the examination and report upon these collections in the absence of its late Secretary [James Prinsep] and other Members who had given attention to the subject.’ In a meeting of the Asiatic Society held in January 1839, it was found that no one was knowledgeable enough to be entrusted with the task of examining Masson’s collection. The society, therefore, resolved that the collection be immediately shipped to England. Lord Auckland, the governor-general of India, however, was anxious to retain such coins for the Museum of the Asiatic Society as may be found duplicates. The Governor-General is unwilling to deprive the Asiatic Society of the advantage of enriching its Museum with such duplicate specimens of coins as may prove of value in completing defective series, or in forwarding the researches of local antiquities. (National Archives of India, Foreign Department, Political Branch, Consultations dated 27 February 1839, No. 13) Accordingly on 23 February 1839 the collection was made over to one William Cracroft, who was requested by the government to make a selection of the duplicate coins for the Asiatic Society of Bengal. Banking more on his common sense than any expert knowledge, Cracroft earnestly took upon the task of examining Masson’s collection and finally selected a total of 55 coins for the Asiatic Society’s Museum. The details of these coins are generally missing in the published accounts of Masson’s coins, which are provided primarily by Elizabeth Errington. These details are, however, available in the records preserved in the National Archives of India:   4   2 10   6

Gold coins with Arabic Inscriptions -do- Nagree Inscriptions Silver Mahomadens -do- Bull and Horse

138  Sanjay Garg 16 - Sassanian   1 -do- Device of Bird   1 -do- Menander Eramaious   4 -do- Boodhists   7 -do- Probably Boodhists  1 Menander  1 Eukratides Copper  1 Applodotus’  1 Unknown Sd/- W. Cracroft. (Source: National Archives of India, Foreign Department, Political Branch, Consultations dated 27 February 1839, No. 18) The remainder of the collection was sent to England with Cracroft on board the ship Repulses. In 1904 the coin collection of the Asiatic Society of Bengal was transferred to the Indian Museum and, together with the coins from Masson’s collection, it has remained there ever since. British Museum, London After its arrival in London, Masson’s collection was initially deposited in the Museum of the East India Company. The British Museum initially received two lots from this collection: 116 coins in 1838 and 59 coins in 1845. Later, after the closure of the company’s India Museum in 1879 all its antiquities became the property of the India Office. In 1882 the India Office presented a selection of 2,420 coins from ‘miscellaneous sources’ to the British Museum which included an estimated 2,000 coins from the Masson collection. Out of this, a selection of Greek and Scythian coins was included in Percy Gardner’s Catalogue (1886). A large portion of the coin collection of the company’s India Museum was auctioned by Sotheby, Wilkinson and Hodge for the government of India in August 1887. These included 3,697 coins from Masson’s collection, many of which eventually found their way into the British Museum through other coin collectors, notably Alexander Cunningham. In 1995 6,641 coins from Masson’s collection were ‘discovered’ amongst the residue of the India Office collection comprising a total of 10,445 coins that were lying in the British Library; in 2007 these were transferred to the British Museum on permanent loan. Out of

Charles Masson 139 these, until January 2012, only 332 coins could be identified using archival records. Fitzwilliam Museum, Cambridge Between 1906 and 1912, c. 600 coins from this collection were transferred from the India Office to the Fitzwilliam Museum, Cambridge. Ashmolean Museum, Oxford The Ashmolean Museum, Oxford, is also reported to have received some coins from Masson’s collection. The details of this transfer are not readily available.

Epilogue Gary Alder notes that Masson was the victim of his own success as a finder of ancient coins and other antiquities . . . Unfortunately for Masson, Afghanistan in the 1830s was beginning to interest his countrymen for reasons very much larger and more compelling than those of mere scholarship. The happy hunting-ground of his carefree travels and investigations was rapidly becoming a highly sensitive political and strategic middle-ground between two expanding European empires. (Alder 1975: vii) Although Masson was able to leave Afghanistan just before the situation turned worse, [h]e failed to extract from the Company the full compensation he demanded or even all the copper coins that were rightfully his; he failed to sell what coins he had to foreign museums or institutions; [and] he failed to persuade the Austrian government to finance any further researches. (Alder 1975: XV) H.H. Wilson, the company’s librarian in London, ‘was always the determining voice in assessing the sums to be paid to Masson’

140  Sanjay Garg (Alder 1975: XIV), and he was most conservative in his valuation. But it was not the monetary deprivation that prompted Masson to accuse Wilson of ‘treachery’. Masson believed that Wilson had deceived him over his editing of Ariana Antiqua, published in 1841. This sumptuous official publication was intended to reveal to the world Masson’s collections, researches, and drawings and put him, in Pottinger’s words, ‘at the top of all the antiquarians of Europe’. Instead, the book appeared as Wilson’s, with Masson, notwithstanding the generous things said about him in the text, merely as on-the-spot researcher and the contributor of one chapter (Alder 1975). Masson unleashed his wrath in the form of marginalia recorded in his hand in his personal copy of the Ariana Antiqua that is now preserved in the Bodlean Library in Oxford. To Wilson’s note in the preface, that ‘from the year 1834 until 1837 Mr. Masson was sedulously employed [in the pursuit, in which he had engaged with equal intelligence and zeal,] on behalf and at the expense of the East India Company’ (p. v) – Masson writes ‘falsehood’ and goes on to explain that: Mr. Masson proposed to the Bombay Govt., it furnish with funds to expend such funds on research on account of the Govt, of course to hand them over the results. When eventually Mr. Masson learned that the Govt consented to advance money, Mr. Masson routinely transferred to Govt the collections made at his own expence and then in his possession. They comprised relics from the Topes at Daraunta, the first year’s collection of coins from Beghram, besides other and various antiquities. The Bombay Govt was willing to have defrayed the cost of the operations and authorized Mr. Masson to appropriate for such purpose one half of the sums remitted to him – but the interference of Capt. Wade prevented this arrangement being available to Mr. Masson. Consequently, Mr. Masson conducted his researches throughout at his own cost. (Wilson 1841 – Masson’s copy: v) There are many passages of this marginalia that give us a new insight into the circumstances of the coin collecting in this terra incognita – as Afghanistan was then known to the Westerners. To many coins that have been referred to as rare by Wilson, Masson says these are a common sight, and viceversa. An interesting note by Masson is about the number of coins he discovered at Begram. Wilson wrote: ‘Mr. Masson continued his researches at this place [Begram] during four succeeding years [from 1833], and collected in

Charles Masson 141 this interval above thirty thousand coins’ (p. 11; emphasis added). To this Masson noted: Mr. Masson’s researches at Beghram extended from 183(6) to 1838. His collections comprised nearly or quite 120,000 silver and copper coins, a few gold ones and a very few engraved gems. Also many thousands of rings, seals, brass and copper cylinders and other antiquities. (Wilson 1841 – Masson’s copy: 11) A thorough assessment of Wilson’s Ariana Antiqua in light of Masson’s personal notes and remarks in his own copy would definitely give a new insight to our understanding of the rich and prolific antiquarian researches that laid the foundation of the Gandharan studies during the first half of the nineteenth century. It is hoped that someone would find this a worthwhile undertaking.

Appendix Charles Masson’s records in the NAI

I. Foreign (Political) Sl. Year No. 1

Date

1834 6 Feb

Number Subject 11

Dost Muhammad Khan’s employment of Masson

Remarks

2

4 May

6–8

Lt. Col. Pottinger authorized to correspond with Masson regarding his antiquarian researches in Afghanistan

3

10 Dec

23

Researches and discoveries in the antiquities of Afghanistan by Masson

4

24 Dec

28–31

Researches and discoveries in the antiquities of Afghanistan by Masson; pecuniary assistance granted to Masson

57–60

Appointment of Masson as newswriter at Kabul; allowances assigned to him

5

1835 7 Jan

6

23 Jan

45–47

Afghanistan’s political state and resources, by Masson; MohunLall to furnish books for perusal to Masson

7

30 Mar

46

Ludhiana political agent’s intimation to Masson of his appointment as newswriter at Kabul

8

6 Apr

42–43

Indus map supplied to Masson (Continued)

Charles Masson 143 I. (Continued) Sl. Year No.

Date

Number Subject

Remarks

 9

20 Apr

11–12

Appropriation of a moiety of advance given to Masson, sanctioned; Lt. Col. Pottinger authorized to correspond with Masson regarding his antiquarian researches in Afghanistan; Lt. Col. Pottinger prohibited from making pecuniary advances to Masson

10

4 May

6–8

Orders regarding dispatch of antiquarian relics in Kabul and Jellalabad

11

4 May

117– 125

Claims by Shikarpuri banker and Keramat Ali against Masson

12

18 May

57–58

Bombay government authorized to make advances to Masson on account of antiquarian researches in Afghanistan; Lt. Col. Pottinger authorized to correspond with Masson regarding his antiquarian researches in Afghanistan

13

25 May

30–31

Report on the strength of the army and resources of Dost Muhammad Khan in his hostilities against Maharajah Ranjeet Singh

14

6 July

16

Masson on Dost Muhammad Khan’s proceedings affecting the political aspects of affairs in Kabul

15

20 July

14–15

Masson’s Report on Dost Muhammad Khan’s position

16

10 Aug

10–12;

Masson’s conduct towards Col. Henry Pottinger; Bombay government’s complaint of his conduct; Lt. Col. Pottinger’s complaint of Masson’s conduct (Continued)

144  Sanjay Garg I. (Continued) Sl. Year No.

Date

Number Subject

17

10 Aug

29–31

Masson on political state of affairs in Afghanistan

18

28 Sept

38

Capt. Wade’s dispute with Pottinger regarding Masson’s employment

19

5 Oct

52–53

Indus map supplied to Masson

20

12 Oct

69–71

Acknowledgements of Masson’s appointment at Kabul; introductory letter to Dost Muhammad Khan requested by Masson on his appointment as the British newswriter at Kabul; Ludhiana political agent warned regarding the appointment of Masson as newswriter at Kabul

21

2 Nov

56–57

Kabul trade report by Masson

22

3 Nov

26–28

Designation of Masson in Kabul

23

30 Nov

26–28

Dost Muhammad Khan’s flight from Peshawar reported by Masson; travelling allowance refused to Masson

71

Publication of the report on the trade of Kabul by Masson

24.

1836 8 Feb

Remarks

25

15 Feb

34–36

Ludhiana Political Agent’s instructions to Masson regarding his duties at Kabul

26

22 Feb

29

Extracts of letters from Mr. Masson regarding affairs of Kabul

27

9 May

26–29

Explanation of Masson’s conduct in connection with researches into Afghan antiquities

28

18 July

12–16

Pecuniary advances to Masson on account of antiquarian researches

29

22 Aug

46–51

Condition of Persia by Masson (Continued)

Charles Masson 145 I. (Continued) Sl. Year No.

Date

Number Subject

30

21 Nov

31–32

Political intelligence from Kabul furnished by Masson; his resignation as newswriter at Kabul

31

12 Dec

6–8

Bombay Government authorized to make advances to Masson on account of antiquarian researches in Afghanistan

32

20 Feb

22–23

Bombay government authorized to make advances to Masson on account of antiquarian researches in Afghanistan; his resignation as newswriter at Kabul

33

26 June

14–15

Advances to Masson on account of antiquarian researches in Afghanistan; orders regarding dispatch of antiquarian relics in Kabul and Jellalabad

34

4 Sept

11

Antiquarian researches of Masson dispatched to England; Bombay government’s proceedings regarding amount overdrawn by Masson

35

28 Sept

80–83

Masson’s services placed at the disposal of Capt. Burnes at Kabul

36

1 Nov

9

Kutch resident’s advances to Masson adjusted

37

29 Nov

10

Lt. Col. Pottinger’s disbursements to Masson in excess of authorized amount passed

34–40

Capt. Burnes’ acknowledgement of the aid received from Masson while on mission to Kabul

38.

1838 31 Jan

Remarks

(Continued)

146  Sanjay Garg I. (Continued) Sl. Year No.

Date

Number Subject

Remarks

39

25 Apr

52

Bombay government’s request for orders regarding coins collected by Masson

40

9 May

42–43

Kabul intelligence furnished by Masson

41

30 May

11

Governor-general’s appreciation of the services of Masson

42

6 June

18–19, Masson’s services placed at the 21–22 disposal of Capt. Burnes at Kabul

43

20 June

36–37

Capt. Burnes to employ Masson in transcribing dispatches for record in Ludhiana political agent’s office

44

27 June

23

Disposal of coins collected by Masson

45

4 July

53–54

Qassid’s charges incurred by Masson

46

15 Aug

25–26

Court’s orders regarding Masson’s antiquarian researches in Afghanistan; donation to Masson; publication of his Afghanistan memoirs

47

19 Dec

21–22

Researches and discoveries in the antiquities of Afghanistan by Masson

48

1839 20 Feb

137– 138

Resignation by Masson of his appointment as newswriter in Kabul

49

27 Feb

121

Coins and relics collected in Afghanistan by Masson sent to the Asiatic Society

50

27 Mar

13–20

Expediency of forwarding to England coins and relics collected in Afghanistan by Masson (Continued)

Charles Masson 147 I. (Continued) Sl. Year No.

Date

Number Subject

Remarks

51

17 Apr

8–9

Dispatch to England of Afghan relics and coin collected by Masson

52

29 May

46–47

Donation to Masson; manuscripts of Masson delivered to him at Tatta

53

31 July

34

Payment of donation authorized by Court of Directors to Masson

54

16 Oct

19

Disposal of coins collected by Masson

II. Foreign (Secret) Sl. Year No.

Date

Number Subject

Remarks

55.

1833 19 Mar

30–31

Masson reported to be travelling in Afghanistan

56.

1834 19 June

1–4

Particulars of Masson; pardon granted to him; advances given to him on account of antiquarian researches in Afghanistan; Central Asia intelligence to be procured by him

57.

1835 27 July

1

Pardon granted to Masson

58.

1837 23 Jan

30

Afghanistan intelligence furnished by Masson; Central Asia intelligence to be procured by him

59.

6 Mar

14

Kabul intelligence furnished by Masson

60.

1838 3 Oct

74–75

Qassid’s charges incurred by Masson; recall of Masson

61.

3 Oct

80–81

Dost Muhammad Khan’s intrigues with Russian agent reported by Masson (Continued)

148  Sanjay Garg II. (Continued) Sl. Year No.

Date

Number Subject

62.

17 Oct

71–80

Masson’s memorandum for counteracting Dost Muhammad Khan’s policy

63.

14 Nov

52

Kabul intelligence furnished by Masson

64.

14 Nov

81–83

Masson ordered to assist Dr. Lord in his mission to Peshawar

65.

1839 9 Jan

40–42

Instructions to Masson to send his dispatch through Col. Wade

66.

1840 17 Dec

8–11

Antecedents of Masson

67.

21 Dec

51–53

Explanation of Masson regarding visit to Khelat

68.

28 Dec

40

Antecedents of Masson

69.

1841 1 Feb

103–05

Detention of Masson at Khelat

70.

8 Feb

107– 108

Detention of Masson at Quetta

71.

1 Mar

92

Compensation to Masson for his detention at Quetta

72.

18 Oct

81–83

Compensation refused to Masson for losses sustained at Khelat

Remarks

III.  Foreign (Foreign) Sl. Year No. 73.

Date

1840 27 July

74.

9 Nov

75.

1844 11 Jan

76.

11 May

Number Subject 140–42

Bill for Qassids employed by Masson

97–101

Vindication of the alleged connection of Masson in the revolution in Khelat

1–3

Antecedents of Masson

1–3

Observations on claims preferred by Masson

Remarks

Charles Masson 149 IV.  India Political Despatch to Court Sl. Year No.

Date

Number Subject

77.

1836 25 July

20

Orders regarding dispatch of antiquarian relics in Kabul and Jellalabad

78.

1837 10 April

18

Afghanistan intelligence furnished by Masson

79.

1839 6 March

14

Coins and other relics collected by Masson in course of his antiquarian researches in Afghanistan

Remarks

V.  India Political Despatch from Court Sl. Year No. 80.

Date

1842 28 Sept.

Number Subject 21

Remarks

Allowances claimed by Masson for services rendered to the late Sir A. Burnes in Afghanistan

VI.  India Despatch to Secret Committee Sl. Year No.

Date

Number Subject

81.

1834 19 June

3

Pardon granted to Masson

82.

1835 5 March

2

Masson on political condition of Afghanistan; pardon granted to Masson

83.

1840 16 Nov.

125

Vindication of the alleged connection of Masson in the revolution in Khelat

Remarks

VII.  India Despatch from Secret Committee Sl. Year No.

Date

Number Subject

84.

1835 9 Feb

Nil

85.

1841 29 Jan.

No. 699 Explanation of Masson

Pardon granted to Masson

Remarks

150  Sanjay Garg VIII.  Bombay Political Despatch to Court Sl. Year No.

Date

Number Subject

86.

1834 5 Aug.

87.

1835 11 March 8

Memorandum on antiquarian researches in Kabul and Afghanistan; pecuniary assistance granted to Masson

88.

1835 31 Dec.

62

Orders regarding dispatch of antiquarian relics in Kabul and Jellalabad; pecuniary assistance granted to Masson

89.

1837 24 Jan

2

Researches and discoveries in the antiquities of Afghanistan by Masson

90.

1837 26 Apr

12

Researches and discoveries in the antiquities of Afghanistan by Masson

91.

1838 21 Nov.

94

Orders regarding dispatch of antiquarian relics in Kabul and Jellalabad

92.

1839 12 Jan.

2

Orders regarding dispatch of antiquarian relics in Kabul and Jellalabad

13 Apr.

20

Orders regarding dispatch of antiquarian relics in Kabul and Jellalabad

11 Sept.

31

Orders regarding dispatch of antiquarian relics in Kabul and Jellalabad

93.

14

Services tendered by Masson in regard to antiquarian researches in Afghanistan

1840 25 March 6

Antiquarian researches of Masson in Afghanistan

20 May

15

Antiquarian researches of Masson in Afghanistan

94.

1841 28 Jan.

2

Application of Masson for copper coins rejected

95.

1841 28 Dec.

75

Donation to Masson

Remarks

Charles Masson 151 IX.  Bombay Political Despatch from Court Sl. Year Date No.

Number Subject

96. 1835 18 Nov. 15 97. 1838 2 May

7

98. 1840 25 May 7 99. 1842

9

Remarks

Orders regarding dispatch of antiquarian relics in Kabul and Jellalabad Orders regarding dispatch of Reproduced antiquarian relics in Kabul and in Ariana, Jellalabad Preface Coins and relics collected by Masson in Afghanistan Donation to Masson for proceeding to England

Source: National Archives of India

References Alder, G.J. 1975. See Masson 1842, repr. Graz, Austria. Bhandare, S.2006.A Metallic Mirror: Changing Representations of Sovereignty During the Raj, in H.P. Ray (ed.), Coins in India: Power and Communication. Mumbai: Marg Publications, pp. 84–97. British Museum, Masson Project. www.britishmuseum.org/research/research_ projects/all_current_projects/masson_project.aspx/ accessed on 14 April 2015. British Museum, Masson Project. Afghanistan 1809–1838: Sources in the India Office Records. www.bl.uk/reshelp/findhelpregion/asia/afghanistan/ afghanistancollection/1809to1838/sources1809to1838.html/ accessed on 14 April 2015. Desmond, R.1982. The India Museum, 1801–1879. London: Her Majesty’s Stationery Office (HMSO). Errington, E.1997.The British East India Company Collection, in P. Callieri (ed.), Seals and Sealings from the North-West of the Indian Subcontinent and Afghanistan (4th century BC‑11th century AD). Naples: Istituto Universitario Orientale and Istituto Italiano per l'Africa e l'Oriente, pp. 18–23. Errington, E. 1999. Rediscovering the Collections of Charles Masson, in M. Alram, and D.E. Klimburg-Salter (eds.), Coins, Art and Chronology: Essays on the pre-Islamic History of the Indo-Iranian Borderlands. Vienna: Istituto Universitario Orientale and Istituto Italiano per l'Africa e l'Oriente, pp. 207–237. Errington, E. 1999‑2000. Numismatic Evidence for Dating the Buddhist Remains of Gandhāra. Silk Road Art and Archaeology6, Papers in Honour of Francine Tissot 191–216. Errington, E. 2001 [2003]. Charles Masson and Begram. Topoi, Oriental Occident 11.1: 1–53.

152  Sanjay Garg Errington, E. 2002a. Ancient Afghanistan Through the Eyes of Charles Masson: The Masson Project at the British Museum. International Institute for Asian Studies Newsletter March: 8–9. Errington, E. 2002b. Discovering ancient Afghanistan: The Masson Collection. Minerva 13.6, November‑December: 53–55. Errington, E., and V.S. Curtis, eds. 2014.From Persepolis to the Punjab: Exploring Ancient Iran, Afghanistan and Pakistan. London: British Museum Press, 2007; repr. as From Persia to Punjab: Exploring Ancient Iran, Afghanistan and Pakistan. Mumbai: Chhatrapati Shivaji Maharaj Vastu Sangrahalaya. Gardner, P. 1886. The Coins of the Greek and Scythic Kings of Bactria and India in the British Museum. London: Trustees of the British Museum. Garg, S. 1991. Charles Masson’s Numismatic Explorations in Afghanistan: A Survey of Archival Sources in the National Archives of India. Abstracts of Papers, XI International Numismatic Congress, Brussels, September, p. 112. Garg, S.1998. Coins Collected by Charles Masson in Afghanistan, in A.K. Jha, and S. Garg (eds.), Ex-Moneta: Essays on Numismatics, Archaeology and History in Honour of Dr. D.W. MacDowall, vol. II. New Delhi: Harman Publishing House, pp. 349–358. Garg, S. 2016. New Light on Charles Masson, in V. Widorn, U. Franke, and P. Latschenberger (eds.), Proceedings of the 20th Conference of the ‘European Association for South Asian Archaeology and Art’. Vol. 2: Contextualizing Material Culture in South and Central Asia in Pre-Modern Times. Turnhout, Belgium: Brepols Publishers. Hambley, G. 1974. See Masson 1842, repr. Karachi. Hoernle, A.F.R. 1885. Centenary Volume of the Asiatic Society of Bengal, 1784–1883 [Part II: Archaeology, History, Literature & C]. Calcutta: Asiatic Society of Bengal. Kalra, M.A. 2013. Charles Masson and the Relic Deposit of Tope Kelan. British Museum Blog. http://blog.britishmuseum.org/2013/12/03/charles-massonand-the-relic-deposit-of-tope-kelan/ accessed on 15 April 2015. Lambrick, H.T. 1977. See Masson 1843, repr. Karachi. Masson, C. 1834. Memoir on the Ancient Coins Found at Beghrām, in the Kohistān of Kābul. Journal of the Asiatic Society of Bengal 3: 153–175. Masson, C. 1836a. Second Memoir on the Ancient Coins Found at Beghrām, in the Kohistān of Kābul. Journal of the Asiatic Society of Bengal 5: 1–28. Masson, C. 1836b. Note on an Inscription at Bamyan. Journal of the Asiatic Society of Bengal 5: 188. Masson, C. 1836c. Third Memoir on the Ancient Coins Discovered at a Site Called Beghrām in the Kohistān of Kābul. Journal of the Asiatic Society of Bengal 5: 537–548. Masson, C. 1836d. Notes on the Antiquities of Bamian. Journal of the Asiatic Society of Bengal 5: 707–720. Masson, C. 1837. Suggestions on the Sites of Sangala and the Altars of Alexander, Being an Extract from Notes of a Journey from Lahore to Karychee, Made in 1830. Journal of the Asiatic Society of Bengal 6: 57–61.

Charles Masson 153 Masson, C. 1842. Narrative of Various Journeys in Balochistan, Afghanistan and the Panjab, 3 vols. London: Richard Bentley (repr.Karachi: Oxford University Press, 1974, with an introduction by Gavin Hambley; repr. Graz (Austria): Akademische Druck-u. Verlagsanstalt, 1975, with an introduction by Gary J. Alder). Masson, C. 1843. Narrative of a Journey to Kalat: Including an Account of the Insurrection at the Place in 1840, and a Memoir on Eastern Balochistan, vol. 4 of Narrative of Various Journeys in Balochistan, Afghanistan and the Panjab. London: Richard Bentley (repr.Karachi: Oxford University Press, 1977 with an introduction by H.T. Lambrick). Masson, C. 1846. Narrative of an Excursion from Peshawar to Shah-Baz Garhi. Journal of the Royal Asiatic Society 8: 292–302. Masson, C. 1848. Legends of the Afghan Countries: In Verse, with Various Pieces, Original and Translated. London: J. Madden. Masson, C. 1850. Route from Seleucia to Apobatana, According to Isidorus of Charax. Journal of the Royal Asiatic Society XII: 97–124. Masson Manuscripts, British Library India Office Collections MSS Eur. (listed in George R. Kaye and E.H. Johnston, Catalogue of Manuscripts in European Languages, vol. II, part II, section II; Masson Papers, British Library India Office Collections [uncatalogued Masson manuscripts] His Majesty's Stationery Office (HMSO) for the India Office). Omrani, B. 2008. Charles Masson of Afghanistan: Deserter, Scholar, Spy. Asian Affairs 39.2, July: 199–216. Pottinger, H. 1833. Letter no. 456, dated 27 November 1833, to Charles Norris, Chief Secretary to Government, Bombay: British Library India Office Collections, MSS Eur. E 161/I, f. 3(9). Prinsep, J. 1835. Further Notes and Drawings of Bactrian and Indo-Scythic Coins. Journal of the Asiatic Society of Bengal 4: 327–348. Richardson, E. 2013. Mr Masson and the Lost Cities: A Victorian Journey to the Edges of Remembrance. Classical Reception Journal 5.1: 84–105. Ross, F.E. 1933. New Light on Charles Masson. Indian Antiquary 62: 221–222. Whitteridge, G. 1986. Charles Masson of Afghanistan: Explorer, Archaeologist, Numismatist and Intelligence Agent. Warminster: Aris & Phillips. Wilson, H.H. 1841. Ariana Antiqua: A Descriptive Account of the Antiquities and Coins of Afghanistan: With a Memoir on the Buildings Called Topes, by C. Masson, Esq. London: East India Company (repr. Delhi: Oriental Publishers, 1971; Masson’s copy in Bodlean Library, Oxford, Acc. no. Afghan 1d.1). Wilson, H.H. 1850. On the Rock Inscriptions of Kapur di Giri, Dhauli and Girnar. Journal of the Royal Asiatic Society 12: 153–251.

6 The collection of Gandharan art in the residence of the Malakand Political Agent, Khyber Pakhtunkhwa Province, Pakistan Pia Brancaccio The collection of Gandharan art from the residence of the Malakand Political Agent in Malakand District, Khyber Pakhtunkhwa Province, Pakistan, consists of fifty-three sculptures collected between 1895 and the early 1930s still on display in the house garden (Figure 6.1). These Buddhist sculptures were virtually unknown to the public until a recent documentation project was carried out by Dr. Luca Maria Olivieri, director of the Italian Archaeological Mission in Pakistan.1 An analysis of the newly recovered colonial correspondence preserved in the agency illuminates the history of the Malakand collection as well as the processes by which early collections of Gandharan art, now in Indian and Pakistani museums, were formed. The sculptures from the Malakand Political House were re-discovered in the summer of 2008 by Luca Maria Olivieri who was invited by Arshad Khan, the then District Coordinator Officer – Commandant Malakand Levies, to assess the historical relevance of all archival and archaeological material kept in that office since its establishment. A total of about 600 archival folders dating between 1895 and 1947 were reviewed, containing a variety of documents shedding light on the modern history of the region with references to colonial political matters. Three folders providing information on surveys and archaeological explorations conducted in Swat prior to the year of partition of British India, as well as letters referring to visits by personalities and diaries of political agents with immediate bearings on the history of collecting, were analyzed by Olivieri in a volume entitled Sir Aurel Stein and the Lord of the Marches2; the book also includes a catalogue of the Gandharan sculptural fragments from the Malakand House completed by the present author.3 These pieces have been damaged by prolonged weather exposure and by defacement, yet they are

Figure 6.1 View of the garden with sculptures set in the wall, Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan Source: All photographs are by Luca Maria Olivieri. Used with permission

156  Pia Brancaccio extremely important because of their direct link with archaeological materials that ended up in major museum collections in the colonial and post-colonial period.

The collection of Gandharan Buddhist sculptures The Political House, the official residence of the Malakand Political Agent, is a compound located immediately below the fort overlooking the Malakand Pass (Figure 6.2). Alfred Foucher, who visited the place in 1896, described it in 1901 as a small house tastefully decorated with ‘trinkets’ coming from different parts of the region.4 Today fifty-three Buddhist sculptures still remain within the compound, set in the walls adjacent to the Guesthouse and the Residence (Figure 6.1). Sculptures nos. 1 to 40 are cemented into the north wall in proximity of the Guesthouse; a beautiful, large Bodhisattva (no.41) is placed along the steep staircase leading to the offices (Figure 6.3); and the remaining pieces (nos. 43 to 51) are affixed into the east wall next to the Residence.5 A special placement has been reserved for three harmikas creatively assembled to form a sundial in the centre of the garden that represents an interesting example of colonial reuse of Gandharan art (Figure 6.4). The Political House sculptures are for the most part without provenance. They consist of stupa architectural fragments such as harmikas and broken cornices collected from Buddhist sacred areas in Swat and adjacent valleys (Figure 6.5), a few damaged narrative panels of the Buddha’s life (Figure 6.6), and a number of fragmentary images of Buddhas and Bodhisattvas, both seated and standing (Figure 6.7). Most impressive is the Bodhisattva sculpture measuring 1.2 metres in height, attached to a corner of the garden (Figure 6.3). The imposing figure depicted in abhayamudra is adorned with exquisite jewellery and has curly hair flowing down onto the shoulders; it has a broad and muscular chest, and is wearing the traditional garments paridhana and uttariya. When Luca Olivieri saw the piece, the lower part of the sculpture was buried under a few centimetres of debris that, once removed, revealed the base of the sculpture with the exquisite treatment of the feet.6 The Malakand Agency sculptures today constitute the leftovers of a significant assemblage of sculpture dispatched over time to major museums such as the Indian Museum in Calcutta, the Lahore Museum, the Peshawar Museum, and, most recently, the Swat Archaeological Museum in Saidu Sharif, Khyber Paktunwa province of Pakistan.

Figure 6.2  View of the Malakand Pass, Khyber Pakhtunkhwa Province, Pakistan

Figure 6.3 Sculpture of a Bodhisattva, Gandharan art (1st–3rd c. CE), Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan

Figure 6.4 Sundial built with reused Gandharan sculptures, Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan

Figure 6.5  Architectural element from a Buddhist stupa, Gandharan art (1st–3rd c. CE), Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan

Figure 6.6 Fragment of a relief depicting the Buddha’s Parinirvana, Gandharan art (1st–3rd c. CE), Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan

Figure 6.7 Headless sculpture of a seated Buddha, Gandharan art (1st–3rd c. CE), Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan

162  Pia Brancaccio

Archival information related to the Malakand Political House collection and its relationships with early museum collections of Gandharan art Archival information uncovered in the Political House sheds light on the long history of the fragments that still remain in the garden, beginning with Major Arthur Harold Deane, who was appointed as the first Political Agent in Malakand in 1895. Deane was a prominent personality in the colonial history of the Northwest Frontier, ending his career as the first Chief Commissioner of the Northwest Frontier reporting directly to the Viceroy of India, a post he held until the time of his death in 1908. Major Deane was seriously interested in Gandharan art and archaeology, as shown in his Note on Udyana and Gandhara published in 1896 by the Journal of the Royal Asiatic Society of England and Ireland.7 A manuscript of this work was also found in the Malakand Political House,8 and its opening lines read: As the Swat valley, and neighbourhood, which constitute the principal portion of the old province of Udyana, have hitherto been inaccessible for archaeological research, the following rough notes (made during the little time at my disposal as Chief Political Officer with the Chitral Relief Force, and lately as Political Officer for Dir and Swat) may induce others better qualified to devote some attention to this interesting neighbourhood. They are principally connected with the travels of the Chinese pilgrim Huan Tsiang, as given in Beal’s ‘Buddhist Records of the Western World’. Much like Homer’s epics ignited early archaeological research in the Aegean Sea, the travel accounts of the Buddhist Chinese monks Xuanzang and Song Yun, published in English by Samuel Beal, respectively, in 1883 and 1884, elicited interest in the Buddhist archaeology of the Swat valley. Arthur Harold Deane, prior to becoming Political Agent, had already shown a keen interest in documenting the artistic and historical heritage of the Northwest. While holding the post of Assistant Commissioner in Mardan, Deane had become fully aware of the relevance of archaeological material from Pakistan and Afghanistan. The task of recording new Gandharan archaeological finds fell within the competence of his office between 1884 and 1895, and it was Deane who in 1888 first reported the discovery of the Buddhist site of Sikri; after the Archaeological Survey inspected it, Deane excavated the site in 1889, bringing to light the well-known Sikri stupa

Collection of Malakand Political Agent 163 now in the Lahore Museum.9 Deane also documented many objects of epigraphic relevance, often seeking the expert opinion of illustrious Indologists like Emile Senart.10 Major Deane’s appointment as first Political Agent in Malakand marked the establishment of British control over the turbulent Frontier region, even if tribal opposition continued strong until 1917, when the creation of the Yusufzai State of Swat by Miangul Badshah pacified de facto the region. As a Political Agent, Major Deane held administrative and judiciary functions that included the supervision of a police force known as the Swat Levies, as well as the engagement with tribal groups residing in Swat and adjacent valleys. It is to Major Deane that we owe the first initiatives concerning the protection of the archaeological heritage of the Swat valley. The British campaigns in the independent tribal territories had opened up new opportunities for archaeological explorations. The ruling law at Deane’s time in Malakand, the Treasure Trove Act of 1878, was weak and not applicable to the Frontier territories; therefore, many sites had already been looted and destroyed. In his note on Uddiyana published in 1896, Deane is already lamenting “the considerable damage done in places by irresponsible digging” especially in the Kafirkot area of the Swat valley, from where a few sculptures had already been sent to the Imperial Museum in Calcutta.11 When he was assigned to Malakand in 1895, the protection of the artistic and archaeological heritage of Gandhara had already become an issue of significant proportions. In his correspondence dated a few months after he became Political Agent, Deane notes: In addition to the looting of archaeological objects to feed to the antiquities market there is the locals’ lack of consideration for antiquities; in fact they smash it up wherever they find it. Another threat is the attitude of the very men in the service of the army or the British administration: There are 2 classes of men who damage ruins by hunting for sculptures. Sepoys who like to take a piece or 2 for their officers [. . .]. As regards officers and others, if they have the opportunity they have no hesitation in pulling places to pieces or pieces of sculptures for themselves or their Messes. The harm done is in the destruction of a ruin without having a proper plan or record made of it.12 Deane’s role as Political Agent from 1895 to 1901 represents a key moment in the history of archaeology in Swat: for the first time an attempt was made to control excavations and to regulate the

164  Pia Brancaccio unrestrained harvest of Gandharan sculpture. Deane started confiscating looted sculptures, gathering many of them in the Political House; he then became the official conduit through which many of such pieces ended up in colonial museums. Correspondence from the Malakand archive shows that just a few months after Deane took office, the government of Bengal acquired many sculptures collected by Deane and Major Maisey of the 30th Punjab Infantry based in Dargai. This acquisition consisted of about 200–300 architectural pieces and Gandharan sculptures from Swat, as well as 47 fragments found in the southern slopes of Malakand.13 With the goal of making the Calcutta Museum worthy of the capital of British India, the secretary general of the government of Bengal dispatched Surgeon-Major L.A. Waddell to carry out archaeological research in Swat. In his report, Waddell expresses great respect for Major Deane whom he describes as “a well-known archaeologist who for many years has been zealously and most successfully exploring the Buddhist remains of Peshawar and its frontier countries”.14 Waddell confirms that Deane had generously offered to the Indian Museum several sculptures from the Swat valley already in his possession – likely kept in the Malakand Political House or in the nearby fort. The delegate of the government of Bengal who delivered in 1896 the Deane-Maisey collection to the Indian Museum in Calcutta was Alexander Caddy. While in the region, Caddy completed a survey of the Swat valley, took several photographs, and got involved in the excavation of an ‘Ionic’ temple in a military encampment at Malakand.15 Photos taken in 1896 by Alexander Caddy, now in the British Library, constitute precious documentation of many Gandharan artworks that left the Malakand Political Agency destined for major colonial museum collections.16 This gives us a sense of the incredible amount of Gandharan sculpture from Swat that was originally assembled by Deane in the Malakand Political House. The correspondence kept in the agency also reveals that such large shipments of Gandharan art from Malakand to Calcutta were justified by the fact that the relatively new Imperial Museum had little Buddhist art in its collection if compared to the Lahore Museum established in 1865.17 However, additional documents show that the Lahore collection continued to grow with Deane’s support, as ‘4 crates with 11 stones’ were received by the Lahore Museum in November 1895.18 An entry in Deane’s diary dated April 1896 explains that a provision was established by which the museums of Lahore and Calcutta were given preference of choice over any archaeological objects found in the Malakand Agency.19 A certain

Collection of Malakand Political Agent 165 rivalry must have existed between these two colonial museums over appropriation of Gandharan sculptures. The second important phase in the development of the Malakand collection took place in the early 1900s. Having been promoted to Lieutenant Colonel in 1900, Deane left the Malakand Agency in 1901 to become the first Chief Commissioner of the North West Frontier Province reporting directly to Viceroy Lord Curzon. The departure of Deane from Malakand did not mean that the Political Agency ceased to be involved in the supply of Gandharan art to colonial museums. Later archival documents from the agency show that in 1907 Dr. D. B. Spooner, Superintendent, Archaeological Survey, Frontier, requested the then Political Agent F. C. Minchen to send ethnographic material and other material kept in the House to the newly formed Peshawar Museum.20 Several sculptures were sent from Malakand to the Peshawar Museum sometime before 1922 and then again in 192421; this shipment includes reliefs that are still on display in the Peshawar Museum.22 It should be noted that most of them consist of narrative panels of the Buddha’s life, a genre conspicuously scarce from the Malakand collection as it is today. One can surmise that perhaps the intact narrative panels of the Buddha’s life left the Political House for Peshawar, whereas fragments of broken sculptures and architectural elements remained there. The last dispatch of sculptures to Peshawar in 1924 is linked to a third important phase in the history of the collection in the Malakand Political House associated with the appointment of Lt. Colonel H. Stewart as Political Agent between July 1922 and October 1924. As indicated by a letter dated November 1923, it appears that Stewart carried out several archaeological excavations in Swat without properly consulting with the Archaeological Survey of India or following the instructions of the central government. He was badly scolded in a semi-official letter that instructed him to report to the local government all discoveries of objects of archaeological interest.23 The letter also ordered him not to remove objects which were still “in situ” and that “statues, etc. worthy of preservation of which the provenance is unknown, should be removed to a Museum”. The same communication also made clear that Stewart had to stop digging and had to compile two inventories of objects – one of the archaeological material excavated by him, and another list of all the Gandharan pieces (statues, etc.) that were already in the House by the time he became Political Agent. This clearly shows that when Stewart moved to Malakand in 1922, a Gandharan art collection existed within the residential compound.

166  Pia Brancaccio An interesting reference to the state of preservation of the Gandharan pieces from the Malakand Political House can be found in a letter dated January 1924 in which Stewart denied any involvement in ‘illegal’ digging.24 Stewart claimed that when the General Commander-in-Chief, India, Lord H. S. Rawlinson (G.C.S.I.) visited Malakand he saw nothing of incredible value in the Political House – only “old remnants and debris which have been lying in the Political Agent’s verandah for years, and which, owing to their mutilated state cannot be catalogued”.25 This would suggest that sculptures collected by Deane and deemed too fragmentary to display in imperial museums were still sitting in the garden in 1924. It appears that Stewart was not being honest about his archaeological activities and antiquarian interest – he was indeed a collector of Gandharan sculptures. Apparently he tried to blow smoke on the whole story by intentionally confusing his collection with these “old remnants and debris” lying at Malakand. His collection was quite large and partly deposited at Malakand. Another archival note from the Malakand Agency signed by Hargreaves, Superintendant of the Archaeological Survey of India, challenged Stewart’s sincerity: apparently a British officer had seen Stewart’s private collection of Gandharan art at Amandarra, not Malakand, and a few locals had testified to their involvement in Stewart’s excavations at Gunyar and Malakand, as well as in an expedition to retrieve sculptures from Upper Swat, which refers to the state of Swat.26 A large part of Stewart’s private collection was eventually donated to Peshawar; however, he was allowed to retain forty minor fragments for his personal collection.27 The standing Bodhisattva from the collection of the Political House in Malakand is the only piece that has secure provenance and can be associated with the figure of E. H. Cobb, Political Agent from mid-1932 to mid-1933. Correspondence between Cobb and Sir John Marshall dated November 1932 informs us that this Bodhisattva was found right by the Shahkot Pass.28 In particular, Document 324 refers to Cobb’s ‘fresh discovery’: “a life-size image of a man hewn out of solid stone with only the head and the arm missing, depicted as draped with cloth and wearing necklaces and armlets, otherwise in perfect condition”.29 Cobb also points to the spot where the statue was found: the area of Palai located to the south of the Shahkot Pass, described as particularly rich in archaeological remains. He reports seeing there “a circle of enormous stone Buddhas still in situ”30 and that the sculpture in question was excavated from a mound of broken images. In the same letter to Marshall, Cobb expresses his personal interest for Gandharan art and archaeology, requesting a bibliography on the subject;

Collection of Malakand Political Agent 167 in particular, having already read the works by Foucher and Stein who both passed through the Political House, respectively, in 1896 and 1903, he asks Marshall the whereabouts of Sir Harold Deane’s notes that in fact were already kept in the agency’s archive. Cobb concludes his letter by inviting Marshall to spend a week or so in Malakand to explore the Shahkot Pass and nearby areas that were so archaeologically rich. In the summer of 2008, he surveyed the area in question and documented a large urban settlement of more than four hectares dating to the third and fourth centuries CE, as well as a Buddhist sacred area with a monumental stupa (Figure 6.8). The final chapter of the Malakand Political House collection takes places in the 1990s when one last group of 134 sculptures was transferred from the agency to the Swat Museum in Saidu Sharif. These sculptures have been registered under the label MK, and some of them are now on display in Gallery no. 6.1 of the museum, which was recently rebuilt thanks to the efforts of the ACT project led by the Italian Mission (Figure 6.9).31 To sum up, the sculptural fragments remaining today in the garden of the Political House tell a long story

Figure 6.8  View of Palai, Shahkot Pass, Khyber Pakhtunkhwa Province, Pakistan

168  Pia Brancaccio

Figure 6.9 Seated Buddhas from Malakand (MK 8, MK 9, MK no number), Gandharan Art (1st-3rd c. CE), Swat Museum, Saidu Sharif, Khyber Pakhtunkhwa Province, Pakistan

made of discoveries, acquisitions, and displacements that begins with Major Harold Deane in 1895 and ends in the 1990s. The life of the Malakand collection is interwoven with some of the earliest and richest museum collections of Gandharan art – those in the Lahore, Calcutta, and Peshawar Museums and, most recently, the Swat Museum in Saidu Sharif.

The British reuse of Gandharan sculptures in the Malakand Political Agency garden One last issue concerning the history of Gandharan sculpture from the Malakand Political House remains to be explored; it relates to the actual placement of the sculptural fragments into the garden walls – a practice that also raises interesting questions about the reception and reuse of Gandharan art among British officers residing in the colony. Unfortunately, no documents from the Malakand Agency explicitly mention the plastering of sculptures in the garden. The only information we have is recorded on the sundial of the Political House garden – it states that the clock was erected on May 12, 1937, to commemorate the coronation of George VI (Figure 6.10). The sundial pedestal made from three superimposed Gandharan harmikas was probably the work of E. H. Cobb, Political Agent in Malakand between 1936 and 1937, who also added the Bodhisattva to the collection when he was

Figure 6.10 Close-up of a Sundial built with reused Gandharan sculptures viewed from above, Garden of Malakand Political House, Malakand District, Khyber Pakhtunkhwa Province, Pakistan

170  Pia Brancaccio previously posted there in 1932. It is likely that the Malakand sculptures were cemented in the current position prior to Cobb’s arrival in 1932, yet this remains pure speculation. Certainly Major Deane was not responsible for it; in a letter dated in 1895, he condemns the custom of reusing of Gandharan fragments, making special reference to British officers who like to put sculptures in their messes – an allusion to the Mardan Guides Mess where Gandharan art had been plastered in the officers’ dining hall.32 A sundial was also installed in the mess garden, as the practice of reusing Gandharan sculpture in gardens within military grounds must have been an established custom: in the Northwest Frontier Province Gandharan sculptures were installed in the Peshawar Residency Garden before 1888, in the Guides Mess at Mardan, and in the Political House at Malakand. Documentation from the British Library shows another remarkable example of sculptural reuse outside the northwestern regions in the Artillery Mess Garden in Sagar, Madhya Pradesh, where officers erected pavilions made of repurposed Gupta pieces.33 To conclude, I hope that this brief discussion of sculptures from the Political House at Malakand has contributed another small piece of evidence to the early history of museum collections of Gandharan art. In addition to tracing links between the Malakand collection and pieces from Swat now in the holdings of Lahore, Calcutta, Peshawar, and Swat museums, the wealth of archival material from the Malakand Political Agency sheds further light on the history of archaeology and museum building in British India.

Notes 1 I would like to thank Dr. Luca Maria Olivieri for his tireless work conducted in the Swat Valley of the Khyber Pakhtunkhwa Province, Pakistan, and for allowing me to study the sculptures from the Political House in Malakand. 2 L.M. Olivieri, Sir Aurel Stein and the Lord of the Marches: New Archival Material, Sang-e-Meel Publications, Lahore, 2015. 3 P. Brancaccio, Catalogue of the Malakand Gandhara Collection, in L.M. Olivieri edited, Sir Aurel Stein and the Lord of the Marches: New Archival Material, Lahore, Sang-e-Meel Publications, 2015: 285–305. 4 ‘décorée avec goût de divers bibelot du pays [. . .]‘des étendards prix aux Svàtis, en passant par les longs pierre à fusils dont se servaient jadis les Afghans [. . .] et les longs poignards triangulaires, en forme de couteaux de boucherie [. . .].’ A. Foucher, Sur la Frontière Indo-Afghane, Hachette, Paris, 1901: 126. 5 The sculptures are numbered as in P. Brancaccio, Catalogue of the Malakand Gandhara Collection, in L.M. Olivieri edited, Sir Aurel Stein and the

Collection of Malakand Political Agent 171 Lord of the Marches: New Archival Material, Sang-e-Meel Publications, Lahore, 2015: 285–305. 6 P. Brancaccio, Catalogue of the Malakand Gandhara Collection, in L.M. Olivieri edited, Sir Aurel Stein and the Lord of the Marches: New Archival Material, Sang-e-Meel Publications, Lahore: 2015: 298. 7 A.H. Deane, Note on Udyana and Gandhara, Journal of the Royal Asiatic Society of England and Ireland, 1896: 665–675. 8 Document no. 10 in L.M. Olivieri, Sir Aurel Stein and the Lord of the Marches: New Archival Material, Sang-e-Meel Publications, Lahore, 2015: 38. 9 For a thorough discussion of Deans’ involvement in Gandharan archaeology and collectiosn see E. Errington, The Western Discovery of the Art of Gandhara and the Finds of Jamalgarhi, Ph.D. Thesis, London, 1987, vol. 2: 363–366. and V. Sarkosh Curtis, Persepolis to Punjab, British Museum Press London, 2007: 226. 10 In an 1889 note on Indian epigraphy Senart thanks Captain Deane for sending him six seals (two originals and four impressions) for study purposes; apparently the seals in question had been uncovered and acquired by Deane himself in the Kabul Valley. E. Senart, Sur Quelques Pierres Gravees Provenant de Caboul, Journal Asiatique, 58, April‑June 1889. 11 A.H. Deane, Note on Udyana and Gandhara, Journal of the Royal Asiatic Society of England and Ireland, 1896: 664. 12 Document no. 17 in L.M. Olivieri, Sir Aurel Stein and the Lord of the Marches: New Archival Material, Sang-e-Meel Publications, Lahore, 2015: 41–42. 13 Document no. 2, Ibid, pp. 34–35. 14 British Library, Buddhist Sculptures from the Swat Valley 10031163. 15 K. Behrendt, Alexander Caddy’s 1896 report: the Chakdara Ionic temple and other sites, in L.M. Olivieri edited, Sir Aurel Stein and the Lord of the Marches: New Archival Material, Sang-e-Meel Publications, Lahore: 2015: 255–262. 16 See for example the British Library photos nos. 10031166–70 taken by Alexander Caddy’s in 1896, where one can easily identify Gandharan reliefs currently in the holding of the Indian Museum, Calcutta. 17 Document no. 2 in L.M. Olivieri, Sir Aurel Stein and the Lord of the Marches: New Archival Material, Sang-e-Meel Publications, Lahore, 2015: 35. 18 Document no. 7, Ibid, p. 36. 19 Document 15, Ibid, pp. 40–41. 20 See for example Document no.75, Ibid, p. 62. 21 Document no.130, Ibid, pp. 80–81. 22 Twenty-nine sculptures from Malakand are in the holdings of the Peshawar Museum. The following Malakand pieces are published in the museum catalogue: PM 01926, PM 02732, PM 02741, PM 00418, PM 02746, PM 01259, PM 01093, PM 00427. I. Ali and M.N. Qazi, Gandharan Sculptures in the Peshawar Museum, Hazara University Mansehra NWFP, 2008. 23 Document no.121 in L.M. Olivieri, Sir Aurel Stein and the Lord of the Marches: New Archival Material, Sang-e-Meel Publications, Lahore, 2015: 75–76.

172  Pia Brancaccio 4 Document no.124, Ibid, p. 77. 2 25 Document 124, Ibid, p. 77. 26 Document 125, Ibid, pp. 77–78. 27 Documents 128–131, Ibid, pp. 81–82. 28 Documents 324–325, Ibid, pp. 160–163. 29 Document 325, Ibid, p. 163. 30 Ibid. 31 The following Malakand sculptures are currently on display in the Swat Museum, Saidu Sharif: MK 20 in Gallery no. 5.1; MK 8, MK 9, MK NO NUMBER in Gallery no. 6.1; MK 25 in Gallery no. 6.7. 32 For a thorough discussion of Gandharan art in the Guides Mess see E. Errington, The Western Discovery of the art of Gandhara and the finds of Jamalgarhi, Ph.D. Thesis, London, 1987, vol.: 210–214. This is how the Guides Mess was described by Officer Colonel Jenkins in 1882: “There are at present four stone figures let into the wall of the mess room of the Corps of Guides. There is one head of a figure on a wooden stand and a stone slab which is used as a table. There are also several small pieces of stone carving, some of which have been built into the pedestal of a sun dial and some let into a fire-place.” E. Errington, The Western Discovery of the Art of Gandhara and the Finds of Jamalgarhi, Ph.D. Thesis, London, 1987: 212. 33 British Library, Collection of Sculptures in the Artillery Mess Gardens Sagar 10031302 and 10031303.

References Ali, I., and M.N. Qazi. 2008. Gandharan Sculptures in the Peshawar Museum. Mansehra: Hazara University NWFP. Deane, A.H. 1896. Note on Udyana and Gandhara. Journal of the Royal Asiatic Society 665–675. Errington, E. 1987. The Western Discovery of the Art of Gandhara and the Finds of Jamalgarhi. Ph.D. Thesis. London. Errington, E. 2007. Exploring Gandhara, in E. Errington and V. Sarkosh Curtis (eds.), From Persepolis to Punjab. Cambridge, London: British Museum. Foucher, A. 1901. Sur la frontière indo-afghane. Paris: Hachette. Foucher, A. 1917. The Beginnings of Buddhist Art, and Other Essays in Indian and Central-Asian Archaeology. London: Humphrey Milford. Olivieri, L.M. 2015. Sir Aurel Stein and the Lord of the Marches: New Archival Material. Lahore: Sang-e-Meel Publications. Senart, E. 1889. Sur Quelques Pierres Gravees Provenant de Caboul. Journal Asiatique 58.

7 Vajīrasthāna/Bazira and beyond Foundation and current status of the archaeological work in Swat Luca M. Olivieri The beginning The historical and political backgrounds that allowed the Italian Archaeological Mission of the then IsMEO1 to organize the first reconnaisances in the valley and to establish a permanent institution in Saidu Sharif have been discussed in recent works2 and they do not need to be repeated here. What may be really important to underline, instead, is the value and the weight of the consequences of that event for the archaeological study in Swat. In the aftermath of the British conquest of the Malakand Pass (1895), and after the establishment of an administrative and military outpost at the bottom of the Swat valley (Malakand Agency), the immense archaeological heritage of the valley started being heavily harvested. “Graeco-Buddhist” sculptures were amateurishly dug, zealously confiscated, and sometimes bargained for and acquired by British officers and native traders. In the absence of a clear legal context, a deluge of carved stones was earmarked for different destinations: military messes and private collections, museums in British India and Great Britain, auction houses and the antiquary market in Europe, and, from there to non-British museums, etc. (see a letter from Aurel Stein of 1913).3 After 1904, despite the introduction of the Ancient Monument Protection Act (VIII, 1904), the legal situation in Swat remained unclear, being a territory lying beyond the border of British India. For example, when Stein was planning an excavation campaign at Amluk-dara or Barikot (two sites he had surveyed in 19264), he did not receive any reply to the question “what exactly is the archaeological legal position of Indian States”.5 After the 1930s amendments to the Act VIII, 1904, made clear that entities other than the Archaeological Survey of India (ASI) were allowed to carry out excavations

174  Luca M. Olivieri under a license issued by the ASI. Evert Barger (University of Bristol) and Philip Wright (Victoria and Albert Museum) in this framework came to Swat in 1938 on their way to Afghanistan.6 Their trip was put under an ASI license, and in fact, as per the license’s terms, they published their first (and only) report in the Memoirs of the ASI (MASI).7 Still, the Act VIII, 1904, was not applicable to the Swat state. Therefore, they should have arranged a sort of private agreement with the Swat state that eventually let the booty be divided up with the Wali (Ruler) of Swat (thus forming the first nucleus of the so-called “Wali Swat Collection”, in the Swat Museum after 1963). The British share is now in the possession of the British Museum and of the Victoria and Albert Museum. However, their pillage was so successful and their booty so rich and heavy that the two gentlemen were left with no option but to throw the discarded pieces into the refilling of their pits, as was discovered in recent fieldwork at the site of Gumbat.8 After the Partition, Giuseppe Tucci (then president of IsMEO) secured for his institute a Pakistani excavation license in 1956 (his first visit to Swat had occurred in 1955).9 It is almost certain that he asked the Department of Archaeology and museums in Karachi (which at that time was advised by Raul Curiel,10 but de facto directed by F. A. Khan), the same question Stein had posed, that is, “what was the legal status of archaeology in the Swat State”. Nothing had changed between 1931 and 1955. After independence, both in India and Pakistan, the Act VIII, 1904 (with the 1933 amendments) was still the ruling law. Differently than Stein, this time Tucci must have received advice in Karachi that he found reasonably practicable. In Italy, where Giuseppe Tucci came from, archaeology, both with the meaning of patrimony and field activity, was perceived as a state matter, regulated by the state law, executed by means of strict procedures, and managed by a complex system of control directly under the authority of the central government. The cultural patrimony was conceived of as property of the state, and therefore excavations were carried out under its authority, and more importantly, the finds – being public property – were placed in the custody of the state itself in governmental public repositories (i.e. museums). Therefore, from Tucci’s perspective, archaeological fieldwork would have been possible only if the following prerequisites were satisfied: the existence of a clear legal framework and the existence of a public repository for the finds. With regard to the first point, it is now clear that Tucci helped the Wali to introduce Act VIII, 1904, in Swat (as an amendment to the instrument of accession which regulated the relationship between Pakistan and Swat) and proceed consequently to the

Vajīrasthāna/Bazira and beyond 175 establishment of the Swat Museum, which was directly controlled, as per law, by a an officer from the Civil Service of the Government of Pakistan11 (Figure. 7.1). The Swat Museum was initially perceived by Domenico Faccenna (the first and unforgettable Direttore of the Italian Mission for 40 years) as a necessity to give a proper space to the collection of the Wali.12 The plan radically changed after the deluge of sculptures and finds that had appeared since the first year of excavation at Butkara I and Udegram (1956).13 Initially the new findings were kept in a storeroom in the library of the local Jahanzeb College. Finally, in 1958 (after the Act VIII, 1904, was extended to Swat) the new museum was established in the town of Saidu Sharif, the capital city of the Swat state. In this same year, Rome inaugurated the Museo Nazionale di Arte Orientale (now “Giuseppe Tucci National Museum of Oriental Art”), initially conceived of as a house for previous IsMEO collections, but then especially meant to house the finds exported to Italy from Swat/Pakistan (on a sort of sine die loan, because the objects were, legally speaking, the property of Pakistan).14 What is now more important in the context of the present chapter is to define the innovative program and model of archaeological

Figure 7.1 The new Swat Museum: front view Source: Aurangzeib Khan; © ISMEO Italian Archaeological Mission

176  Luca M. Olivieri intervention that was outlined by Domenico Faccenna from the outset of the IsMEO Mission in Swat. The program included the following points: (1) a comprehensive archaeological mapping, (2) the urgency of a methodologically advanced system of excavation, (3) the definition of a standardized cataloguing system, and (4) regular publication of the archaeological reports.15 All these points were achieved, one by one, albeit not always systematically. Excavations started in 1956. The introduction of stratigraphic methodology since the early excavations of Buddhist sites in Swat was transformative.16 In the subcontinent a reliable excavation method was introduced by M. Wheeler after 1942,17 but never utilized for the excavation of a Buddhist site in Gandhara. The larger excavations in the Taxila and Peshawar valleys carried out during the tenure of J. Marshall as Director General of the ASI were basically carried out with a methodology that Wheeler described as pseudo-scientific.18 Therefore, the excavations opened by Faccenna in Swat were the first carried out following a modern scientific method, and the collection of sculptures recovered from those sites, now in the Swat Museum and partly in Rome, contained the first Gandharan sculptures, and were possibly amongst the very few ones that included a reliable archaeological record and metadata.19

Barikot: latest research and results The following pages will provide the reader with an overview of the main results obtained by the Mission in the last five years of fieldwork at the site of Barikot in the framework of the ACT (Archaeology, Culture, Tourism) field-school project.20 The sculptural material illustrated in the following figures is now in the new Swat Museum, where it was put on display only 12–18 months after their recovery. The new display of the museum was built upon preliminary studies of the new material. Therefore, the Swat Museum collection is entirely based on local materials coming from scientific excavations, and the entire educational project is based on the recent results presented in a clear and comprehensible way through a real visual path. Between 2007 and 2009, the Swat region was the theatre of serious insurgent activity, which was annihilated by an extensive military intervention. In the newly pacified Swat, from 2011 to 2016, the Italian Mission/ACT managed to carry out twelve archaeological campaigns in Barikot, two at Gumbat and three at Amluk-dara21 (Figure 7.2).

Figure 7.2  Map of Swat Source: Courtesy by Karel Fritz and Daniel Neil, University of Vienna, Department of Geography and Regional Research – modified by Luca M. Olivieri; © ISMEO Italian Archaeological Mission

178  Luca M. Olivieri The three sites are situated about 5 km from each other. The settlement of Barikot lies on the Swat River in a strategic position and naturally occurring fortifications. The Buddhist establishments of Gumbat and Amluk-dara lie in two side valleys, which were part of the ager of the ancient city. Understanding the link between the city and Buddhist monasteries is actually one of the most urgent issues for the archaeology of the northwest are of the subcontinent.22 This has not been possible so far because of the chronic absence of data, which is due to the reluctance to excavate historic urban sites23 and to the difficulties involved in obtaining reliable data from Buddhist monastic sites, in most cases excavated with outdated methods and little competence. Bazira/Beira//Vajirasthāna//Bīr-koṭ/Barikot The site of Barikot (Bīr-koṭ-ghwaṇḍai; 34°40'51"N, 72°12'46"E; approx. 799 m asl), after a preliminary investigation of some protohistoric tombs discovered by Tucci himself,24 became a major excavation project for the Italian Archaeological Mission from 1978, when G. Stacul began to investigate the flat area located south of the hill that gives its name to the site. The work of Stacul also brought to light important architectural remains of a large settlement of the historical period that was the subject of a successful series of excavations begun in 1984.25 The studies carried out so far have confirmed the great archaeological and historical importance of this urban site, located in a strategic position dominating the roads leading to the upper and lower valley of Swat and to Buner, and have highlighted the astonishing occupational continuity and the impressive stratigraphic sequence that make Barikot a crucial site for understanding of the evolutionary processes of the region: Stacul’s studies, also confirmed by the excavation data collected on top of the hill, have provided evidence of a human presence from the second millennium BCE.26 The archaeological data and the study of the historical sources have provided sound ground to the identification, first proposed by A. Stein and G. Tucci, of this site with the city of Bazira reported by classical historians as being conquered by Alexander in 327 BCE.27 Early mention of the town as Beira or Bazira is made in the accounts of Alexander’s expedition (Curtius Rufus, VIII 10, 22 and Arrian, Anabasis, IV 27, 5); as Vajirasthāna, it is mentioned in a ninth-century CE inscription from Barikot. The inscription, which was found on the top of the Barikot hill and is now in the Lahore Museum (inv. no. 119), mentions

Vajīrasthāna/Bazira and beyond 179 the foundation of a (sacred) building in Vajirasthāna by three individuals, whose names are not preserved, in the time of the Hindu Shah Jaypaladeva (ninth to tenth centuries CE).28 The toponym vajrasthāna has been interpreted differently: certainly the hypothesis that it may refer to Waziristan (sic) has to be rejected, whereas what was wisely suggested by Tucci is certainly more probable, that is, that the name indicates the sthāna of Vajira (i.e. Bazira or Barikot [in Pashto: Bīrkoṭ]). From 1998 to 2000 the Italian Mission discovered and excavated a religious Brahmanic building on the top of the hill, a temple whose early phase is positively dated to the Turki Shahi phase, seventh to eighth centuries CE.29 Recently Professor Oskar von Hinüber has studied the surviving text of the Barikot inscription at the Lahore Museum. The toponym vajirasthāna is clearly present in the inscription, but a great part of the text is unfortunately lost (personal communication by O.v.H.). The linguistic analysis of the toponym vajirasthāna led Matteo De Chiara to the interpretation of vajira-sthāna as ‘strong place’, ‘fortified place’, or better ‘The Strong’. The form vayira, well attested in the Gāndhārī language in inscriptions from the region dating about the first century CE (CKI 249, CKI 52, CKI 367), can explain the local form Beira (pronounced [ve(j)irə], see Tribulato and Olivieri, forthcoming) reported by Curtius Rufus (“Beira incolae vocant”: VIII 10, 22) (S. Baums, personal communication). Outline of the excavation’s history The ancient city, located to the west of the modern village, occupies an area of 10 ha and is surrounded on the north by a crescent-shaped hill (ghwaṇḍai), bathed by the Swat River and on the south by the Kandak and Karakar streams. Despite some interruptions due to international and internal situations, the excavations have continued until the present day; seven trenches were opened on the alluvial plain at the foot of the hill and four trenches on the two artificial terraces on the top of the hill (see above). The fieldwork on the upper hill was preceded by a complete survey of the site and all the emerging structures.30 So far, 12 trenches have been excavated: BKG 1 (1984) at the eastern limits of the ancient city, BKG 2 (1984 and 1987) at the southern foot of the hill; BKG 3 (1987) across the south stretch of the ancient city wall; BKG 4–5, 11, and 12 at the southwest corner of the city (1990–1992, 2011–2016); BKG 6 through 9 (1998–2000) on the upper hill; and BKG 10 (2006) at the southeast corner of the city.31

180  Luca M. Olivieri In a first step, from 1984 to 1990, five trenches were opened (BKG 1, BKG 2, BKG 3, and BKG 4–5) which, for the periods following the protohistoric phases, allowed a stratigraphic sequence to be reconstructed in the plain ranging from the second century BCE until the fourth century CE and continuing in the hillside area until the Islamic period (thirteenth to fourteenth centuries CE). The remains of an imposing Indo-Greek fortification wall (hereafter “Wall”) enclosing the southern part of the city were documented in trenches BKG L and M,32 then again in trenches BKG 3 and BKG 4–5: until now a stretch of the south side of the Wall, which runs east to west, and another longer one, that from the southwest corner proceeds in a north-northwest direction. Both have rectangular bastions at regular intervals of about 27–28 m. The bastion at the southwest corner, on the other hand, is pentagonal in shape (BKG 4–5). Even though some hypotheses have been advanced regarding the whole perimeter of the Wall, this aspect remains uncertain.33 Since 1998, we have extended excavations up to the hill that, due to its crescent shape, protects the area of the ancient city (ancient akropolis or citadel), and we have opened four more trenches (BKG 6, BKG 7, BKG 8, and BKG 9). The earliest materials are dated to the second millennium BCE, but the major structures brought to light are dated to the second century BCE (a portion of the upper Wall), to the first to second centuries CE (BKG 7 and 9), and to the Shahi and Ghaznavid periods (seventh to twelfth centuries CE) (BKG 6).34 The most important find in this area is, without doubt, the Brahmanical temple partially excavated during the campaigns of 1999 and 2000: the east terrace on the top of the hill, supported by a high retaining wall, is almost entirely occupied by the remains of this imposing building, rectangular in shape, and decorated with stucco architectural elements and reliefs. During the excavation, remains of the stucco decoration and some fragments of two marble worship images, stylistically assignable to the art of the Shahi period (seventh to eleventh centuries CE), were brought to light. In 2006 we opened a new trench (BKG 10) with the aim of finding useful data for the alignment of the east section of the city’s defenses. Although no sections of the Wall were found, the data collected during the campaign, including the position of the southeastern corner bastion, provided enough elements for a hypothetical reconstruction of the entire urban circuit. The largest excavated area, BKG 11 lies at the north end of the trench BKG 4–5 dug in 1990–1992 (Figure 7.3).35 The new area was

Vajīrasthāna/Bazira and beyond 181

Figure 7.3 Trench BKG 11 seen from the northwest Source: Luca M. Olivieri; © ISMEO Italian Archaeological Mission

selected with the aim of exposing the largest possible area of the ancient city within the original limits provided by the south and west segments of the city wall.36 After having marked out the area, we divided it in two portions west and east of a strip 2 m wide, oriented north to south, that was kept unexcavated and used as a service track for wheel-barrows. The two portions were then internally divided into eight sectors each, orthogonal to the service path, and numbered from north to south, from 1 to 8. It was then considered convenient to combine each adjacent pair of sectors into larger sectors that were consequently labeled from the north as BKG 11 W and BKG 11 E, 1–2, 3–4, 5–6, and 7–8. The sectors are separated by baulks left unexcavated and used as service corridors for the passage of the wheel-barrows used for removing waste soil. The total excavated area covers c. 6000 sqm; it has the shape of a trapezoid with its major base on the northwest. The west side is c. 140.00 m long, the east side, 105.00 m long, and its width ranges from

182  Luca M. Olivieri 70.00 m (north) to 60.00 m (south). The sectors are approximately 30.00 m wide and 25.00 m long. The surface clearly slopes towards the north and is marked on the west side by an artificial step created by the subterranean presence of the ancient city wall. The excavation/ conservation work lasted for more than 450 working days, distributed in 12 seasons from April 2011 to October 2016. In 2014 to 2016 we worked also on another two trenches outside the southwest stretch of the defensive Wall, immediately south of trench BKG 4–5 (BKG 12 and 12E).37 Compared chronology of BKG trenches The dating of the fortified city with its defensive Wall and the akropolis (upper hill) with their associated layers and materials seemed since the beginning clearly defined by the coin and inscription assemblage (eight Indo-Greek coins, three punch-marked and local coins) and three inscribed potsherds with Greek letters.38 However, a definitive chronology for the Defensive Wall layers has been positively obtained: both radiocarbon (C14) and thermoluminescence (TL) analyses have given a consistent c. mid-second–century chronology. The 2016 excavation campaigns and the data collected (and the associated new C14 dates) added new important details to the picture. In simple words, the ancient Barikot revealed the totally unexpected presence of a “Bhir Mound or Charsadda II A-D Horizon”, that is, a rich town (urbs opulenta in the words of Curtius Rufus) which had been already established when it was fortified in the mature IndoGreek period. Regarding the finds of these early historical phases, NBPW is well documented in the Indo-Greek phase in association with vessels with thick red slip, identical to the so-called “plates à poisson” found at Ai Khanoum and in many other sites (Termez, Kurganzol, etc.) in Graeco-Bactrian contexts. Below the Indo-Greek phase, the 2016 excavation revealed the existence of a continuing sequence that goes back to Graeco-Bactrian (Macrophase 3a1), Mauryan times (Macrophase 2b) (amongst the other materials, the widespread presence of Baroque Lady terracotta figurines, along with local Indian coins, is noteworthy), and to the Achaemenian cultural horizon (featured by the presence of typical tulip-shaped bowls and pear-shaped jars) (Macrophase 2a). In addition to these earlier cultural phases, the later cultural phase during which the city was fortified and maintained is defined

Vajīrasthāna/Bazira and beyond 183 as Macrophase 3a2–4 and lasted from the mid-second century BCE to the mid-first century CE, from the Indo-Greek times to the SakaParthian period (Macrophase 3b). In the early Kushan phase the city wall was abandoned and simply served as substruction for the new phase of life of the city. We have defined two cultural phases, early and late Kushan, Macrophases 4 and 5, culminating with the abandonment of the built-up area towards the end of the third century CE. The temptative correspondence between the various BKG trenches and between the latter and the other urban and Buddhist sites excavated with a reliable stratigraphic methodology is provided in Tables 7.1 and 7.2. It is worth mentioning that the two tables, which are based on the current available bibliography and data, are not definitive tools, but rather the graphic representation of a “work in progress”, an item which is shared with reader in order just to foster the discussion. Pre-wall phases In the outer area, phases prior to the construction of the Wall show signs of pits and masonry foundations with a superstructure of mud brick. The materials indicate an Early Bronze Age chronology dated around the mid-second millennium and featured by a well-known Late Harappan black-on-red painted ware. Macrophase 0 includes the remains of pit dwells or granaries, with or without pebble-masonry elevations, and small stone structures, including bases for pilasters. These structures either stand on or are cut out of a clay layer that partially covers the alluvial slope. The floors are made of clay with the addition of a few slabs; the superstructures, probably mud brick, have been poorly preserved. Macrophase 0 ends with a period of abandonment characterized by clay layers in which the foundation trenches of some new structures of Macrophase 1 were dug.39 Other unexpected data were revealed by the 2016 excavations outside the defensive Wall. The materials associated with Phase 1 of BKG 4–5 are typical of the Late Bronze Age horizon, which has been positively dated to end of the second millennium/early beginning of the first millennium BCE.40 The layers of Macrophase 1 are covered by a deposit of compact soil with materials from destroyed clay structures of Macrophase 2, whose surface corresponds to the level associated with the construction in Macrophase 3a of the defensive Wall and the excavation of the enscarpment.

Amluk-dara (AKD 1)

earthquake*

Gumbat - Balo Kale (GBK 1)

Saidu Sharif I (SS I) Panr I (P I)

Buddhist sacred areas

*

Butkara I (B I)

Late Buddhist rockreliefs

Ghurids and Khwarezmian Shahs

11th CE

9th-10th CE

MACROPHASE 4

MACROPHASE 5

Only BKG 1

Buddhist Sacred Area

1st-2nd CE

2nd CE

3rd CE (first half)

3RD CE (SECOND HALF)

EARLY KUSHANS

KUSHANS

LATE KUSHANS

KUSHANO-SASANIANS

Kidarites

Hephtalites

TURKI-SHAHI

HINDU-SHAHI

GHAZNAVIDS

Local

-

10th-11th CE

Local Timurids

14th CE

Moghuls

YUSUFZAI

DYNASTIES

-

16th CE

Relative Chronology ABSOLUTE CHRONOLOGY (14CFROM TRENCHES BKG1,3,11,12)

4TH CE (BEGINNING)

Barama (BA I)

MACROPHASE 6

earthquake

Hindu Temple? Castle

*

Mosque Castle Muslim Graveyard

earthquake

Muslim Graveyard

Top-hill "Castle" Upper area

5TH-6TH CE

BKG 4-5

Gogdara 3

Lower area "Bazar"

MACROPHASE 7

earthquake

*

Hindu Temple Castle?

BKG 12

Outside the city wall

Buddhist complexes in agris

7TH-8TH CE

Only BKG 3: Phase 4

Only BKG 3: Phase 5

Only BKG 2

BKG 1 BKG 2

BKG 3/BKG 4-5 BKG 11

Mosque? Castle

BKG 6 BKG 7 BKG 8 BKG 9

Muslim Graveyard Only in BKG 3

Top-hill

Upper area

Lower area

Oḍi/Ora (Udegram)

MACROPHASE 8

MACROPHASE 9

MACROPHASE 10

MACROPHASE 11

MACROPHASE 12

BKG MACROPHASES

Vajīrasthāna/Bazira (BKG)

Urban settlements

Tables 7.1–7.2  Concordance table between BKG and other excavated sites in Swat

KUSHANA

SASANIAN

Huna

SHAHI

EARLY ISLAMIC

CULTURAL CONTACT PHASES

MACROPHASE 2A2

MACROPHASE 2B

MACROPHASE 3A1

MACROPHASE 3A2

MACROPHASE 3A3

MACROPHASE 3A4

MACROPHASE 3B

MACROPHASE 4

BKG MACROPHASES

BKG 3/BKG 4-5 BKG 11

Lower area

BKG 6 BKG 7 BKG 8 BKG 9

Top-hill

earthquake

BKG 1

Upper area

BKG 12

BKG 4-5

Outside the city wall

Urbansettlements

Vajīrasthāna/Bazira (BKG)

Lower area "Bazar" Gogdara 3 Top-hill "Castle" Upper area

Oḍi/Ora (Udegram)

Barama (BA I) Gumbat-Balo Kale (GBK 1) Amluk-dara (AKD 1)

Saidu Sharif I (SS I)

Buddhist sacred areas

Panr I (P I)

Buddhist complexes in agris

Butkara I (B I)

Relative Chronology

6TH-4TH BCE

3RDBCE

2NDBCE(SECOND HALF)

1ST CE

BKG 1,3,11,12)

(14CFROM TRENCHES

ABSOLUTE CHRONOLOGY

ASSAKENOI-ASVAKA

APRACA

ODI

LOCAL DYNASTIES

ACHAEMENIDS

Macedonians

MAURYA

GRAECO-BACTRIAN

INDO-GREEK

SAKA

PARTHIANS

EARLY KUSHANS

DYNASTIES

ASSAKENOI

MAURYA

GREEK

SAKA-PARTHIAN

KUSHANA

CULTURAL CONTACT PHASES

MACROPHASE 2A1

MACROPHASE 0

MACROPHASE 1C

MACROPHASE 1C MACROPHASE 1B

Gogdara 4 UDG

Reabilitation phases Abandonment or re-utilization phases Data from other trenches of the same site Destruction phases Obliterated layers

Macrophase Historical data Living phases.1 (Urban/Monumental) Living phases.2 (Village/non-Monumental) Graveyard

Gogdara 1, 2

DARADAS-KAMBOJAS (MAHAJANAPADAS)

Unclear cultural limit

Archaeological deposit continues to the present

Archaeological deposited excavated below

Lower excavated limit

Upper limit of the archaeological deposit

Missing data No data

Period IV of the Ghalegai sequence (1700-1400 BCE)

Periods V to VII of the Ghalegai sequence END 2ND MILLENNIUM/ BEGINNING 1 MILLENNIUM BCE

Vajīrasthāna/Bazira and beyond 187 Early urban phases Before the construction of the Wall (Macrophase 3a3), the entire area was characterized by a steeply inclined artificial mound entirely occupied in Graeco-Bactrian, Mauryan and Achemenian horizons by a settlement (Macrophases 3a1, 2b, and 2a) revealed both inside the city and outside the (later) defensive Wall. The construction of the defensive Wall was preceded by the cut of a long stepped foundation trench dug into an intensively artificially modified slope that had almost totally obliterated the preexisting stratigraphy of Macrophase 2, revealing the far earlier Iron Age structures. In this way the latter appeared to us as being directly cut by the Indo-Greek Wall, which was actually not the case. At a distance of about 4.50 m from the Wall, an escarpment was cut in front of the latter (evidence of a real ditch has not been found).41 The cut partially exposed the earlier stratigraphy down to a depth of about 5 m below the foot of the Wall. The escarpment cuts the profile of the original slope (at the southwest corner of the urban circuit, in trench BKG 12, it reached a depth of about 6 m). The escarpment was later provided with a walled embankment, or a berm, a sort of proteichisma, that finds a clear comparison in the external defence of the city of Sirkap. The model of massive ramparts with rectangular towers is the typical Hellenistic model employed to hold out against attacks of siege artillery and machines. This model is dominant in the Hellenized East up to Greek Bactria and independently developed in the subcontinent (the model is also discussed in theoretical treaties).42 The same model is replicated in a later phase at Sirkap, where its association to a SakaParthian chronology clearly marks the cultural continuity until the beginning of the Common Era. It also seems clear from the Barikot data that the new rulers – quoting Francfort – “les Parthes encore philllènes”,43 wanted to present themselves in Gandhara as the true successors of the Hellenized élites. It is evident that – at least in Swat – the Indo-Greek military foundations at Udegram and Barikot are constructed on the very same preexisting towns where the Macedonians had established garrisons, namely Ora and Bazira. The two sites look like colonial foundations organized in the lower city and akropolis where the local element was certainly absent from the élites. Certainly Barikot, the true gateway of inner Swat due to its exceptional topographical position, had the function of military stronghold, or katoikia, keeping watch over the Swat valley and the roads leading to west and south Udegram, and was a larger centre, similar to a subordinate or dependant pòlis.44 Both Udegram and Barikot – the two sites are located at a distance of less than 10 km from each other – show that the Graeco-Bactrian and

188  Luca M. Olivieri Indo-Greek acculturation phases radically changed the cultural material. The interesting fact is that the new material from Macrophase 2a-b and 3a1, unmistakenly position Swat in wider cultural horizons: eastern Iranian (Achemenian = 2a), northern Indian (Mauryan = 2b), and Graeco-Bactrian (3a1).45 As far as the chronology of the Wall is concerned, it is evident that the annexation of Swat to the Indo-Greek controlled territories happened in a mature phase of their power in north India (middle or end of the second century BCE). It is possible that Swat was fortified by the Indo-Greek in a specific moment, to create a defensive salient at the time of Menander, during the conflicts with the Graeco-Bactrian King Eucratides (c. 150 BCE). The southwestern quarters of the city from the Sakas to the Kushanshahr THE STRUCTURAL SEQUENCE INSIDE THE CITY (BKG 4–5 AND 11)

Period IV (=Macrophase 3b) is closely related to Period III (the IndoGreek Wall); the new structures actually re-use material from the walls of the previous period, and their foundation trenches are often cut into the last floor of that period. Findings and coins ascribed to the Saka dynasty, allow this period to be dated to between the first century BCE and the middle of the first century CE. The numismatic finds and the material context (typically Kushan) also suggest a dating of the structures of the following Period V (Phase 4) from the middle of the first century CE to the middle of the second century CE. It is probably in Period VI (=Macrophase 4) that the city Wall loses its defensive function: the wall appears partially collapsed and on its razed upper surface a masonry drain was built. The most important structure of this period is a small sacred area. This small Buddhist shrine is characterized by a court with a stupa in the centre; on the north side of the court we also found three shrines, built at a later stage (Period VII and VIII). Once again, the numismatic findings allow us to contextualize the period between the mid-second century and the beginning of the third century CE. Period VII (=Macrophase 5) represents the peak of the building expansion of the inner city, and it is characterized by large residential units separated by lanes and squares, or public open spaces, sometimes provided with pit wheels. Among the finds from the layers of this period, in addition to Buddhist votive stelae, Kushan, late Kushan, and Kushano-Sasanian coins are particularly significant and suggest a date around the third century CE (Figure 7.4).46

Figure 7.4a BKG 11/4–5: orthophoto Source: Francesco Genchi; © ISMEO Italian Archaeological Mission

Figure 7.4b BKG 11/4–5: plan with residential units Source: Ivano Marati; © ISMEO Italian Archaeological Mission

Vajīrasthāna/Bazira and beyond 191 In Period VIII (=Macrophase 5) the structures adjacent to the city Wall started to be abandoned. However, in cultural terms, there seems to be a substantial cultural continuity between Period VII and Period VIII that, from a chronological point of view, seems to fall at the end of third century CE. In Period IX (=Macrophase 6), the city structure was extensively abandoned, and the houses, largely deserted, were left in ruins, transformed into a sort of makeshift slum. THE LIVING CONTEXT OF THE LATE KUSHAN/KUSHANO-SASANIAN CITY (MACROPHASE 5)47

The combined analysis of the masonry uncovered in trenches BKG 4 to 5, 11, and 12 indicated that, during its final urban phase, or Periods VII to VIII (the latter C14-dated to the second half of the third century CE),48 the southwest quarter of the city was divided into 11 singlefloor dwelling units of different sizes (from 300 to 700 square metres), always arranged around a central courtyard, sometimes provided with domestic worship areas. The blocks were served by a network of communicating streets, and the main street ran intra muros along the western section of the walls. Dwelling unit B has an open courtyard flanked by a continuous low bench characterized on one side (the north) by the presence of three votive niches. In one of them Buddhist steles were found in situ, together with a statuette in the round depicting a kneeling man carrying a lamp holder (Figure 7.5). An open raised shrine runs along the entire western side in communication with the courtyard through two short staircases. In front of the main niche an altar with ex voto was uncovered (including a laminated armour bracelet) and a large stone basin of the ‘alms bowl’ type. On the lateral benches numerous fragments of shell and luxury pottery bracelets were found, with both materials deliberately broken (Olivieri 2017). The building’s typology (defined as “Sacred Building B”), closely resembles that of the Complex E of Kara-tepe (Figure 7.6). A few metres northward, excavation revealed a second building (defined as “Temple B”) similar to the previous one and datable to the same chronological horizon (see Figure 7.6). The two buildings are linked and interconnected through a raised corridor. Temple B features a raised rectangular paved space closed on three sides and open to the north where at least three quadrangular bases were originally meant to support huge wooden pillars. Although only three bases have been discovered (we are close to the northern limit of the trench), it is

Figure 7.5 BKG 11, Sacred Building B: detail of the niche of the cultural court Source: Luca M. Olivieri; © ISMEO Italian Archaeological Mission

Figure 7.6 BKG 11, Units B and D: axonometric restitution Source: Francesco Martore and C. Moscatelli; © ISMEO Italian Archaeological Mission

Vajīrasthāna/Bazira and beyond 193 highly possible that the building featured a four-pillar or four-column façade in view of the discovery of a rough altar right in the lower space between the second and the third bases. A flight of steps at the southern end of the platform gave access to a lower open area that featured a rectangular tank. Close to the tank, a distiller was documented in situ, and two fireplaces were documented a few steps from the central altar (Moscatelli, Olivieri and S. Niaz Ali Shah 2016). In the debris of the collapsed south wall of courtyard in front of Temple B was found a stele representing Hariti (Figure 7.7). Dwelling unit D (Figure 7.6) has a central courtyard used as a cooking area, but housing a small Buddhist shrine (Figure 7.8), in the collapsed debris of which decorative sculpture materials were found. The corridor leading to the courtyard from the main street yielded, among other material (jars and millstones along the sides), a small stele depicting an unknown bearded male deity sitting in European style holding a chalice and a goat’s head as attributes (Figure 7.9). The stele was found inside a small stone cyst beside a fireplace (Olivieri 2014b). Dwelling unit K has an alignment that is different from the others and may be linked to the original alignment of the Period V and VI structures (Macrophase 4). The layout is particularly interesting as it consists of a rectangular enclosure around a rectangular building in the same axis made up of a courtyard and with a distyle building open to the north (Figure 7.10). In front of the latter, in the courtyard, stands a small shrine, the interior of which still houses a miniature stupa. Also here numerous ex voto were found, in particular, horse figurines originally with (now missing) riders and lions (Figure 7.11). The dystile building (defined as “Temple K”) has an open antecella with traces of continued combustion on the clay floor, an adjacent cell, and a side corridor leading to a rear chamber or thesauron (that also can be accessed via the cell), in which a deposit of valuable objects, no doubt donations, was discovered. The latter include intact pottery (belonging to the three classes documented in Building B), a glass ampulla imported from the West, and a precious elephant’s tusk. Temple K, of which earlier examples may be found in the Central Asian area, displays clear similarity with the temple of Mohra Maliaran in Taxila (Olivieri 2017). Ample traces of two successive earthquakes in the space of less than 50–70 years have been clearly documented as third-century CE stratigraphy. This fact, alongside the political upheaval represented by the collapse of the Kushan Empire, eventually led to the city being abandoned. The urban elites comprising the Kushan system of local alliances, or the new families ruling the city during the early

Figure 7.7  Temple B: the small stele of Hariti BKG 3636 (w. 16.2 cm, h. 24.4 cm, t. 5.1 cm) Source: C. Moscatelli; © ISMEO Italian Archaeological Mission

Figure 7.8 BKG 11, Unit D: the Buddhist shrine Source: Luca M. Olivieri; © ISMEO Italian Archaeological Mission

Figure 7.9 BKG 11, Unit D: the non-Buddhist miniature stele BKG 2304 (h. 17.2 cm, w. 8.5 cm, t. 2.8 cm)

Figure 7.10 BKG 11, Temple K Source: Luca M. Olivieri; © ISMEO Italian Archaeological Mission

Figure 7.11 BKG 11, Temple K: axonometric restitution Source: Francesco Martore; © ISMEO Italian Archaeological Mission

198  Luca M. Olivieri Kushano-Sasanian period, would probably have had less financial power, or interest, regarding the maintenance of the complex metropolitan system. A complex series of events eventually led the urban elites to abandon the city when it was already partly in ruins at the very beginning of the fourth century CE. When the city was just a field of ruins, it was re-occupied and transformed into a sort of slum by non-urban or low-class settlers (C14-dated to the early years of the fourth century CE).49 The city in its environment: urbs et ager in ancient Swat The two sacred areas of Gumbat and Amluk-dara were certainly founded between the first and second centuries CE.50 The site of Gumbat comprises three monumental terraces, the first characterized by a large stepped substruction wall and marked by three flanking principal monuments of the same size, two stupas, and a central vihara (Figures 7.12–7.13). The latter may well have been constructed as early as the late first century CE in view of the C14 dating of the wooden lintel of one of the upper loopholes of the cell. The dating obtained for the other three beams found during the restoration work to prop up the southeastern lean of the dome have a later dating (second century CE). This fact seems to suggest that the double dome was constructed, or rather reconstructed, at a later stage. The terrace is enhanced with lesser monuments in the subsequent two structural phases. The great stupa of Amluk-dara, one of the most majestic and best conserved in Gandhara, evidences a complex construction, renovation, and reuse lasting from the second to the tenth centuries CE (Figures 7.14–7.15). Originally decorated with schist, with a monumental staircase, as early as the mid-third century, the main monument has been completely reshaped, most probably to compensate for the destruction caused by seismic events (as documented in the nearby Barikot area). The stupa was redecorated in coloured stucco and organogenic limestone known locally as kanjur. The access staircase has been lengthened and the stupa equipped with a front niche, certainly of stuccoed limestone, which likens the monument to the models dominant in Afghan Kapisa, such as Shewaki and Top Dara, towards the mid first millennium. The construction of a small Kashmiri-style temple with a truncated pyramid roof supported by a false dome and with a square base on top of the fourth phase of the staircase has been dated to the sixth century. The sacred area, although completely buried and

Figure 7.12 Gumbat-Balo Kale: a view from the west Source: Edoardo Loliva; © ISMEO Italian Archaeological Mission

Figure 7.13 Gumbat-Balo Kale: restitution of the terrace seen from the east Source: Francesco Martore; © ISMEO Italian Archaeological Mission

Figure 7.14 Amluk-dara: the stupa terrace Source: Edoardo Loliva; © ISMEO Italian Archaeological Mission

Figure 7.15 Amluk-dara: axonometric restitution of the stupa terrace Source: Francesco Martore; © ISMEO Italian Archaeological Mission

Vajīrasthāna/Bazira and beyond 201 partially collapsed, continued to be used, as is shown by the very late stucco covered pavements on which worship continued to be practised. This is attested, for example, by the find of an excellent Gandharan panel depicting the birth of Siddhartha having been re-used found up against the outer wall of the first body of a smaller stupa when the latter had been almost completely buried (Figure 7.16). The C14 dating of this phase sets it in the sixth to seventh centuries, that is, contemporary with the similar re-utilization phases at Sahri Bahlol.51 The C14 datings of the final phase lie between the seventh and tenth centuries. Clear traces of a late earthquake that certainly occurred during this time span are marked by the collapse of the chattravali and reveal the definitive abandonment of the sacred area. After comparing the data from Barikot with those obtained at Gumbat and Amluk-dara, the overall hypothetical picture emerging is as follows. In the first century CE, Parthians of northwest India are the heirs of the Indo-Greeks. This is demonstrated by the fact that the Indo-Greek military foundations display a structural continuity in the Parthian age, as attested in Barikot and Udegram, but also by the

Figure 7.16a Amluk-dara: panel AKD 95 in its stratigraphic context Source: Luca M. Olivieri; © ISMEO Italian Archaeological Mission

202  Luca M. Olivieri

Figure 7.16b Amluk-dara: panel AKD 95: the Birth of Siddhartha (w. 36.5 cm, h. 19.0 cm; t. 4.5 cm) Source: Edoardo Loliva; © ISMEO Italian Archaeological Mission

fact that, in this period, the new foundations reproduce Hellenistic fortification models, such as at Sirkap (Taxila). The Saka-Parthian period sees the beginning of the expansion of the Buddhist foundations. The development of the monastic foundations in Swat continues uninterrupted during the next two centuries, reducing the living space available to the indigenous communities of Swat but also involving them.52 The monastic foundations played an increasingly important economic and productive role.53 During the Kushan phase, Barikot lost its aspect as a military outpost after Swat ceased to be a frontier zone and effectively became a metropolitan territory of the Kushan-controlled territories.54 The now-settled urban area of Barikot had a complex cultural fabric: in its sacred architecture typically Gandharan shrines mingled with Central Asian templar models; the material culture is dominated by

Vajīrasthāna/Bazira and beyond 203 paddle-and-anvil–made Indian pottery forms and Indian shell ornaments, as well as with copper ornaments typical of the traditional east Iranian koiné. The divinities worshipped are those belonging to the Buddhist world, even though in the third century, small portable steles often show local divinities. This fact could actually mean that the process of merging with the local culture, albeit during a phase of Kushan ‘globalization’, was proceeding successfully, thanks no doubt to the dynamic role placed by the monastic communities. Natural disasters, as well as the collapse of the Kushan power vis-à-vis the emerging role of the Sasanians, led the city to be abandoned around the middle of the third century CE. Nevertheless, in the country areas the Buddhist communities managed to cope with the political crisis, although not everywhere or in equal degrees. And now a few words about the sculptural production. The recent archaeological results clearly demonstrate that in the mid-third century CE the Buddhist domestic and public sanctuaries at Barikot re-employed composite Gandharan sculptural material which had already been used in different contexts. Only medium and small steles remained in production in this period. The steles, only occasionally featuring Buddha, generally represent Bodhisattva, and local and Indian, often female, divinities, and were intended for private worship. At the same time, in the functioning Buddhist sacred areas, schist sculpture, now found only in reuse contexts, was no longer produced and was largely replaced by stucco decoration. The introduction of both stucco and limestone (the former, among other things, being the product of the latter), brought not only a different artistic sensitivity, but, above all, a new masonry technique. It would not be too bold to interpret these new techniques and aesthetic sensitivity as a spillover (via Afghanistan) of the contemporary art of Iranian stucco.

Notes 1 Istituto Italiano per il Medio e l'Estremo Oriente (1933–1996; from 1996 to 2011 IsIAO: Istituto Italiano per l'Africa e l'Oriente); today ISMEO (International Association of Studies on the Mediterranean and the Orient). 2 L.M. Olivieri, Outline History of the IsIAO Italian Archaeological Mission in Pakistan (1956–2006), in L.M. Oliveri edited, East and West, 2006, 56 (1–3) (Special Issue for the 50th Anniversary of the IsIAO Italian Archaeological Mission in Pakistan) 2006: 23–43. T. Tanweer, The Italian Archaeological Mission. A Preliminary Archival Study, in Ghani-urRahman and L.M. Olivieri edited, Italian Archaeology and Anthropology in Northern Pakistan (1955–2011), Journal of Asian Civilizations, Special Issue, 34.1, 2011: 41–49.

204  Luca M. Olivieri 3 Document 113, letter to R.L. Kennion, Political Agent Malakand, 12 July 1913; L.M. Olivieri, M.A. Stein and the ‘Lords of the Marches’: New Archival Materials. ACT Reports and Memoirs, Archival Studies, 1. Lahore: Sang-e-Meel, 2015a: figs. on pp. 353–355). 4 Aurel Stein, An Archaeological Tour in Upper Swāt and Adjacent Hill Tracts, Memoirs of the Archaeological Survey of India, 42, Government of India Central Publication Branch, Calcutta, 1930. 5 Document 312, letter to B.J. Gould, Political Agent Malakand, 31 July 1931; see figs. on pp. 151–152; see also L.M. Olivieri, “Frontier Archaeology”. Sir Aurel Stein, Swat and the Indian Aornos, South Asian Studies, 31.1, 2015b: 58–70. The campaign, planned in 1933 was eventually cancelled (see the reasons in Olivieri M.A. Stein and the ‘Lords of the Marches’ and Olivieri, “Frontier Archaeology”). 6 These issues are extensively debated in Olivieri, M.A. Stein and the ‘Lords of the Marches’. 7 E. Barger and Ph. Wright, Excavations in Swat and explorations in the Oxus territories of Afghanistan, Memoirs of the Archaeological Survey of India, 64, Government of India Press, Delhi and Calcutta, 1941. 8 L.M. Olivieri, with others, The Last Phases of the Urban site of Bir-kotghwandai (Barikot). The Buddhist sites of Gumbat and Amluk-dara (Barikot), ACT Reports and Memoirs, 2, Sang-e-Meel, Lahore, 2014: 291, 303. 9 See details in L.M. Olivieri, Notes on the Problematic Sequence of Alexander’s Itinerary in Swat: A Geoarchaeological Study, East and West, 46.1.2, 1996: 45–78. 10 See L.M. Olivieri, Outline History of the IsIAO Italian Archaeological Mission in Pakistan (1956–2006), in L.M. Oliveri edited, East and West, 56.1.3, 2006 (Special Issue for the 50th Anniversary of the IsIAO Italian Archaeological Mission in Pakistan): 26. 11 T. Tanweer, The Italian Archaeological Mission: A Preliminary Archival Study, in Ghani-ur-Rahman and L.M. Olivieri edited, Italian Archaeology and Anthropology in Northern Pakistan (1955–2011), Journal of Asian Civilizations, Special Issue, 4.1, 2011: 43, 54. 12 See above; see also Olivieri, Outline History of the IsIAO Italian Archaeological Mission: 28–29. The establishment of a Museum in Swat was first discussed with the Wali by Barger and Wright (Eid. 1941: 13). 13 Faccenna, and Taddei 1962, Faccenna and Gullini 1962. 14 It was not just a coincidence that, in 2011, as soon as Swat was reopened to foreign researchers after a dramatic standoff caused by the Taliban insurgency, the Italian Mission was busily supporting the reconstruction of the Swat Museum, whose structure had been irreversibly weakened by an earthquake and a bomb blast. For all information concerning the rebuilding of this museum, completed in 2013 within the framework of the ACT project, as well as the history of the previous structure with its many expansions (1958, 1963, 1969, 1992), the reader may refer to a recent monograph (Marati and Vassallo 2013) now available online as a free ebook (Eid. 2014). 15 See Olivieri Olivieri, Outline History of the IsIAO Italian Archaeological Mission: 28.

Vajīrasthāna/Bazira and beyond 205 16 There was a strong difference between the excavation methodology utilized by G. Gullini at Udegram and the one introduced by Faccenna at Butkara I. The difference is clearly understandable from their respective reports (Faccenna and Gullini 1962). Gullini’s basic archaeological unit is a “stratum”, which more precisely indicates what since L. Wooley’s time was better defined as a period’, that is, the summa of all the layers that were surfaces/structures at the same time (Wooley 1961: 24). In Gullini’s terminology a ‘stratum, rathen than a ‘layer’, is what in modern archaeology is called ‘structural’ period, that is, a composite set of multiple events (multiple stratigraphic units) that went under various and multiple processes of depositional/post-depositional events in a sequence that lasted for a certain amount of time. Faccenna’s terminology follows Wheeler’s concepts (Wheeler 1954), as he documented more correctly each ‘layers’, that is, singular events that occurred one time in one place (see also Olivieri 2014a). 17 The method was elaborated by Wheeler in his outstanding Roman excavations in Britain and disseminated to the general public and academics only about 20 years later (M. Wheeler, Archaeology from the Earth, Oxford University Press, Oxford, 1954). 18 M. Wheeler, Review to J. Marshall, Taxila, 3 vols., The Antiquaries Journal, 32.3.4, 1952: 204–206. 19 All the other major Gandharan collections in the world (with the exception of the Peshawar University Museum) comprise either antiquary assortments or sets of excavated materials gathered through unscientific procedures. The situation did not change. It should be also admitted that only a few current Gandharan specialists have an experience in field archaeology, and practically none have directed stratigraphic excavations. As a consequence, the field is currently shared by field archaeologists who do not know Buddhist art and Buddhist art specialists who are not familiar with the archaeological data. Faccenna’s baton was not picked up by his colleagues, as the very few professional field archaeologists working in Gandhara were either involved in different projects or left Pakistan for other fields. Curiously, the overall methodological legacy of Faccenna was therefore taken up only by younger collaborators who became more and more involved in settlement archaeology, a field that was touched upon by Faccenna only en passant. Some of them, though, extended their experience to the excavation of Buddhist sites. Just a few months ago, the ACT project published the first archaeological manual for university students ever published in Pakistan, where the methodology of Faccenna is updated with the more advanced trends and described in a very practical way. The manual, published in August 2014 in printed form (Olivieri 2014a), will be available online as a free ebook by March 2017. 20 The ACT project is financed by the Pakistani-Italian Debt Swap Agreement. The project started in 2011 and continued throughout 2016. In this contribution we are dealing only with the historic phases of archaeological fieldwork. Large-scale excavations were carried out at two Late Bronze age graveyards, the report on which has been recently published (Vidale, Micheli, Olivieri, 2016; Vidale and Micheli, 2017). The activity of the ACT project has been published in eight volumes of the series ACT Field

206  Luca M. Olivieri School Reports and Memoirs (other five volumes are in press, or in preparation) (see Bibliography). 21 The excavation report was published in 2014 (Olivieri et al., 2014). Other useful updates can be found in Olivieri 2014b, Meister, Olivieri and Vidale, 2015. 22 See Allchin, 1995. 23 After Marshall’s Sirkap excavation (Taxila), which was carried out between 1913 and 1934 (Marshall 1951), a sector of Sirkap was excavated in 1944–1945 under Wheeler’s guidance (Gosh 1948). After the Partition, Bala Hisar (Charsadda) was sondaged by Weeeler in 1958 (Wheeler 1962), Udegram by Gullini in 1956–1962 (Faccena and Gullini 1962), and Barama by Faccenna in 1963–1964 (Faccenna 1964). The Shaikhan-dheri excavation (Charsadda) in 1963–1964 (Dani 1965–66) was supposed to be a project comparable to Sirkap in terms of extension, but unfortunately was terminated after the second campaign, and one of the most promising ancient urban sites of the subcontinent started sinking under the expansion of the modern village. Nowadays, no trace is left of the ancient city. The 1967 small-scale digging at Damkot (Abdur Rahman 1968–1969) yielded controversial results, party due to the difficulties of applying a proper methodology to sloping and heavily eroded layers (a common crux for field archaeologists). After the excavations at the historic settlement at Barikot started in 1984, only three excavation projects were attempted: in three campaigns between 1993 and 2000 a British-Pakistani team reexcavated some of Wheeler’s trenches and made some new limited trial pits at Bala Hisar (Charsadda) (Coningham and Ihsan Ali eds. 2007). The second project is composed of a scattered series of diggings carried out at Aziz-dheri between 1993 and 2012 by different Pakistani teams (see Nasim Khan 2010). The third and most successful project was the one carried out by a second British-Pakistani team at the Akra mound near Bannu (Magee et al. 2005). 24 S. Tusa, Notes on Some Protohistoric Finds in the Swāt Valley (Pakistan), East and West, 31.1.2, 1981: 99–120. 25 See an overview in L.M. Olivieri, The Survey of the Bir-kot Hill: Architectural Comparisons and Photographic Documentation. Bīr-koṭ-ghwaṇḍai Interim Reports I, IsIAO Reports and Memoirs, Series Minor, IV. Rome: Istituto Italiano per l'Africa e l'Oriente, 2003: 23–24. 26 Stacul 1978, Id. 1980; Olivieri in Callieri et al., Bīr-koṭ-ghwaṇḍai, Swāt, Pakistan. 1998–1999 Excavation Report, East and West, 50.1.4, 2000: 194–204. 27 See Stein 1930: 28; Tucci 1958: 296 and 327, fn. 28. 28 On that, see again Tucci 1958: 296, fn. 28. 29 Callieri et al. 2000, Eid. 2000–2001. 30 L.M. Olivieri, The Survey of the Bir-kot Hill: Architectural Comparisons and Photographic Documentation: Bīr-koṭ-ghwaṇḍai Interim Reports I, IsIAO Reports and Memoirs, Series Minor, IV. Rome: Istituto Italiano per l'Africa e l'Oriente, 2003. This and the following paragraph was partly elaborated in collaboration with my collegaue Luca Colliva (University of Bologna). I owe him my gratitude for his help.

Vajīrasthāna/Bazira and beyond 207 31 The following list includes the excavation reports so far published (BKG 10 is still unpublished; the preliminary survey of the upper hill area also should be included; Olivieri 2003): BKG 1: Callieri, Faccenna, and Filigenzi 1984. BKG 2: Callieri, Faccenna, and Filigenzi 1984; Callieri, Filigenzi, and Stacul 1990; Callieri et al. 1992. BKG 3: Callieri, Filigenzi, and Stacul 1990. BKG 4–5: Callieri et al. 1992; Olivieri 1993. BKG 6–9: Callieri et al. 2000; Callieri et al. 2000–2001. BKG 11: Olivieri et al. 2014; Olivieri 2013; Olivieri forth. BKG 12: Olivieri 2015d; Afridi, Iori and Olivieri 2016. 32 A. Filigenzi and G. Stacul, Pakistan 1: The Italian Archaeological Mission in Pakistan, East and West, 35.4, 1985: 436–439. 33 See again Olivieri, The Survey of the Bir-kot Hill. 34 Callieri et al. 2000; Callieri et al. 2000–2001. 35 See above: P. Callieri et al. Bīr-koṭ-ghwaṇḍai 1990–1992. A Preliminary Report on the Excavations of the Italian Archaeological Mission, IsMEO, Annali dell'Istituto [Universitario] Orientale di Napoli, 1992, 52 (4) 1992, Suppl. 73. and L.M. Olivieri, Excavations at Bir-kot-ghwandai (Swat) 1992: Preliminary Report, Pakistan Archaeology, 28, 1993: 103–116. 36 See the final report: Olivieri et al., The Last Phases of the Urban Site of Bir-kot-ghwandai, 2014. 37 Olivieri 2015; E. Iori, L.M. Olivieri, and A. Afridi, Urban Defenses at Bīr-koṭ-ghwaṇḍai, Swat (Pakistan): Data from the 2015 Excavation Campaign, Pakistan Heritage, 7, 2016: 73–94. L.M. Olivieri, and E. Iori, Early-historic Data from the 2016 Excavation Campaigns at the Urban Site of Barikot, Swat (Pakistan): A Shifiting Perspective, in A. Hardy and L. Greaves edited, 23rd South Asian Archaeology and Art Conference, forthcoming. 38 The third inscription was found in 2016 (Olivieri and Iori, forth.). It is being published in Tribulato and Olivieri, forth. For the other inscriptions, see Rougemont 2012: nos. 85 and 86 (see also no. 87 from Udegram, “Bazar”, Stratum V = Phase 3 in Table 1.2). 39 Stacul, 1987. 40 Olivieri and Iori, forth. These dates match those obtained from the newly excavated Late Bronze/Early Iron Age graveyards (Vidale, Micheli, Olivieri 2016; Vidale and Micheli, 2017). The new chronology is radically changing both the chronology and cultural perspectives of the transition phase from protohistory to the early historic period in Swat and Gandhara. See more in Olivieri and Iori, forth. 41 In another paper we proposed that the ditch was already dug in Macrophase 2 for the earlier urban defense (evidence of possible clay rampart was found) (Olivieri and Iori, forth). However, it is now clear that intensive levelling work has caused the total obliteration of the post–Iron Age stratigraphy all along the defensive Wall (both inside and outside). A similar situation was recently documented during the reassessment of the preBuddhist stratigraphy at the site of the monastery of Saidu Sharif I. There,

208  Luca M. Olivieri some large-scale levelling work performed at the time of the construction of the Buddhist complex (first quarter of the first century CE) radically removed all the previous stratigraphy, also partly cutting the upper layers of a late-protohistoric graveyard (fourth century BCE), which appeared to the archaeologists as though it was directly cut by the foundation walls of the monastery (Olivieri 2016a). Instead, the recent study of a long section outside the monastery area revealed that the graves were covered by burial mounds, as well as by a subsequent stratigraphy, which was artificially removed inside the monastery area. The same situation appeared to have happened at Barikot as well. 42 See references in Callieri et al., 1992: 34. See the seminal Callieri, 2007. 43 The quotation is from Franfort 1979. Interesting evidence about the organization and functions of the extramural area during the SakaParthian period comes from trench BKG 12E located on the area outside the bastion close to the southwest corner of the defensive wall (Period IV, Macrophase 3b). In the phase related to the first century BCE, besides the reconstruction of the bastion, partially collapsed after a natural event, some additional structures were built in order both to reinforce the stability of the bastion (i.e. an escarpment) and to facilitate the runoff of the rainwater from the area close to the Wall down to the ditch. Once this operation of reinforcement and prevention was completed, further efforts were directed towards the construction of a pit well and related structures. The presence of an external pit well clearly connected with the urban centre raises some issues. A possible explanation for the location of the pit well in the outer area might be the presence of a city gate in the immediate surroundings. This was certainly not a main gateway, but possibly a secondary gate nearby, which may to some extent be compared to the function of the so-called ‘water-gates’. The latter were meant to draw water in security and have been hypothesized at Sirkap (see Iori, Olivieri, Afridi, 2016). 44 Ora was certainly a regional capital. 45 On the Mauryan phase in Swat, see Iori 2016; on the Achaemenian phase in Swat and at Barikot, see Olivieri and Iori, forth. 46 D.W. MacDowall and P. Callieri, A Catalogue of Coins from the Excavations at Bīr-koṭ-ghwaṇḍai 1984–1992: Bīr-koṭ-ghwaṇḍai Interim Reports II, IsIAO Reports and Memoirs, Series Minor, IV. Rome: Istituto Italiano per l'Africa e l'Oriente, 2003: 27–90. 47 See Olivieri 2013 and Olivieri 2016b. 48 Olivieri et al. 2014. 49 See Olivieri 2012, Cupitò and Olivieri 2013, Olivieri et al. 2014. 50 See besides Olivieri et al., 2014, also the hypothesis presented earlier in Olivieri, Vidale et al. 2006. On the previous research on these two sites, as well as for the previous graphic documentation, see Faccenna and Spagnesi 2014. 51 See K. Beherendt, The Ancient Reuse and Recontextualization of Gandharan Images: Second to Seventh Centuries CE, South Asian Studies, 25.1, 2009: 11–27. 52 See L.M. Olivieri, Behind the Buddhist Communities. Subalternity and Dominancy in Ancient Swat, in Ghani-ur-Rahman and L.M. Olivieri edited, Italian Archaeology and Anthropology in Northern Pakistan

Vajīrasthāna/Bazira and beyond 209 (1955–2011) Journal of Asian Civilizations, Special Issue, 34.1, 2011: 123–151; L.M. Olivieri, Talking Stones: Painted Rock Shelters of the Swat Valley, ACT Report and Memoirs, Series Minor, 2. Lahore: Sang-e-Meel, 2015c. 53 See Olivieri, Vidale et al., 2006. 54 See L.M. Olivieri, Notes on the Problematic Sequence of Alexander’s Itinerary in Swat: A Geoarchaeological Study, East and West, 46.1.2, 1996: 45–78.

References Allchin, F.R. 1995. The Archaeology of Early Historic South Asia, in F.R. Allchin (ed.), The Archaeology of Early Historic South Asia: The Emergence of Cities and States. Cambridge: Cambridge University Press, pp. 3‑9. Barger, E., and Ph. Wright. 1941. Excavations in Swat and Explorations in the Oxus Territories of Afghanistan. Memoirs of the Archaeological Survey of India, 64. Delhi and Calcutta: Government of India Press. Behrendt, K. 2009. The Ancient Reuse and Recontextualization of Gandharan Images: Second to Seventh Centuries CE. South Asian Studies 25.1: 11‑27. Callieri, P. 2007. Barikot, an Indo-Greek Urban Center in Gandhāra, in D.M. Srinivasan (ed.), On the Cusp of an Era: Art in the Pre-Kuśāṇa World. Leiden and Boston: Brill, pp. 133–164. Callieri, P., D. Faccenna, and A. Filigenzi. 1984. Pakistan 1: Excavations and Researches in the Swat Valley – Bīr-koṭ-ghwaṇḍai. East and West 34.4: 493–500. Callieri, P., A. Filigenzi, and G. Stacul. 1990. Bir-Kot-Ghwandai, Swat: 1987 Excavation Campaign. Pakistan Archaeology 25: 183–192. Callieri, P., P. Brocato, A. Filigenzi, L.M. Olivieri, and M. Nascari. 1992. Bīr-koṭ-ghwaṇḍai 1990–1992: A Preliminary Report on the Excavations of the Italian Archaeological Mission, IsMEO. Annali dell’Istituto [Universitario] Orientale di Napoli 52.4, Supplement: 73. Callieri, P., L. Colliva, R. Micheli, N. Abdul, and L.M. Olivieri. 2000. Bīrkoṭ-ghwaṇḍai, Swāt, Pakistan: 1998–1999 Excavation Report. East and West 50.1–4: 191–226. Callieri, P., L. Colliva, and N. Abdul. 2000–2001. Bir-kot-ghwandai, Swat, Pakistan. Preliminary Report on the Autumn 2000 Campaign of the IsIAO Italian Archaeological Mission in Pakistan. Annali dell’Istituto [Universitario] Orientale di Napoli 60–61: 215–232. CKI = B. Stefan and G. Andrew. Catalog of Gāndhārī Texts. https://gandhari. org/catalog/ Coningham, R., and A. Ihsan, eds. 2007. Charsadda: The British-Pakistani Excavations at the Bala Hisar. Society for South Asian Studies Monograph No 5. Oxford: British Archaeological Reports International Series 1709. Cupitò, M., and L.M. Olivieri. 2013. Architectural and Infra-structural Evidence of Re-use of Residential Units in Macro-phase D, Sector W of Bīrkoṭ-ghwaṇḍai/Barikot. Journal of Asian Civilizations 36.1: 41–81.

210  Luca M. Olivieri Dani, A.H. 1965–1966. Shaikhan Dheri Excavations, 1963 and 1964. Ancient Pakistan II: 1–407. Faccenna, D. 1964–1965. Preliminary Report on the 1963 Excavation Campaign of Barama I (Swat-Pakistan). East and West 15.1–2: 7–23. Faccenna, D., and G. Gullini. 1962. Mingora: Site of Butkara I: Udegram. IsMEO Reports and Memoirs, I. Rome: Istituto Italiano per il Medio ed Estremo Oriente. Faccenna, D., and P. Spagnesi. 2014. Buddhist Architecture in the Swat Valley, Pakistan: Stupas, Viharas, a Dwelling Unit. ACT Reports and Memoirs, Special Volume. Lahore: Sang-e-Meel Publications. Faccenna, D., and M. Taddei. 1962. Sculptures from the Sacred Area of Butkara I (Swāt, W. Pakistan). IsMEO Reports and Memoirs, II, 1–2. Rome: Istituto Italiano per il Medio ed Estremo Oriente. Filigenzi, A., and G. Stacul. 1985. Pakistan 1: The Italian Archaeological Mission in Pakistan. East and West 35.4: 436–439. Francfort, H.-P. 1979. Les fortifications en Asie Centrale de l’age du bronze à l’epoque kouchane. Paris: Commission des fouilles et missions archéologiques. Gosh, A. 1944–1945. Taxila (Sirkap). Ancient India 4: 66–79. Iori, E. 2016. The Early-Historic Urban Area at Mingora in the Light of Domenico Faccenna’ Excavations at Barama – I (Swat). Frontier Archaeology 7: 99–112. Iori, E., L.M. Olivieri, and A. Afridi. 2016. Urban Defenses at Bīr-koṭghwaṇḍai, Swat (Pakistan): Data from the 2015 Excavation Campaign. Pakistan Heritage 7: 73–94. MacDowall, D.W., and P. Callieri. 2003. A Catalogue of Coins from the Excavations at Bīr-koṭ-ghwaṇḍai 1984–1992: Bīr-koṭ-ghwaṇḍai Interim Reports II. IsIAO Reports and Memoirs, Series Minor, IV. Rome: Istituto Italiano per l’Africa e l’Oriente, pp. 27–90. Magee, P., C.A. Petrie, R. Knox, F. Khan, and K. Thomas. 2005. The Achaemenid Empire in South Asia and Recent Excavations in Akra in Northwest Pakistan. American Journal of Archaeology 109: 711‑741. Marati, I., and C. Vassallo. 2013. The New Swat Archaeological Museum: Architectural Study, Master Plan and Execution. ACT Reports and Memoirs, 1. Lahore: Sang-e-Meel Publications, pp. 1–15. Marati, I., and C. Vassallo. 2014. The New Swat Archaeological Museum: Architectural Study, Master Plan and Execution. ACT Reports and Memoirs, 1. Bologna: BraDypus, pp. 1–15. http://books.bradypus.net/sites/ default/images/free_downloads/new_swat_archaeological_museum.epub/ Marshall, J. 1951. Taxila: An Illustrated Account of Archaeological Excavations Carried out at Taxila Under the Orders of the Government of India Between the Years 1913 and 1934, I‑III. Cambridge: Cambridge University Press. Meister, M., L.M. Olivieri, and M. Vidale. 2015. Gumbat-Balo Kale (Swat) Architectural Analysis, Conservation, and Excavation (2011), in V. Lefèvre (ed.), South Asian Archaeology and Art 2012, 2: South Asian Religions and

Vajīrasthāna/Bazira and beyond 211 Visual Forms in Their Archaeological Context (Indicopleustoi: Archaeologies of the Indian Ocean 12). Paris: Brepols, pp. 553–566. Moscatelli, C., L.M. Olivieri, and A.S. Syed Niaz. (2016). A Late Kushan Urban Temple from Bazira/Vajīrasthāna: Data from the 2016 Excavation Campaign at Barikot, Swat. Pakistan Heritage 8: 49–61. Nasim Khan, M. 2010. The Sacred and the Secular: Investigating the Unique Stūpa and Settlement Site of Aziz Dheri, Peshawar Valley, Khyber Pakhtunkhwa, Pakistan, 1‑3. Peshawar: M. Nasim Khan. Olivieri, L.M. 1993. Excavations at Bir-kot-ghwandai (Swat) 1992: Preliminary Report. Pakistan Archaeology 28: 103–116. Olivieri, L.M. 1996. Notes on the Problematic Sequence of Alexander’s Itinerary in Swat: A Geoarchaeological Study. East and West 46.1‑2: 45‑78. Olivieri, L.M. 2003. The Survey of the Bir-kot Hill: Architectural Comparisons and Photographic Documentation: Bīr-koṭ-ghwaṇḍai Interim Reports I. IsIAO Reports and Memoirs, Series Minor, IV. Rome: Istituto Italiano per l’Africa e l’Oriente. Olivieri, L.M. 2006. Outline History of the IsIAO Italian Archaeological Mission in Pakistan (1956-2006), in L.M. Oliveri (ed.), East and West 56.1–3 (Special Issue for the 50th Anniversary of the IsIAO Italian Archaeological Mission in Pakistan): 23–43. Olivieri, L.M. 2011. Behind the Buddhist Communities: Subalternity and Dominancy in Ancient Swat, in Ghani-ur-Rahman and L.M. Olivieri (eds.), Italian Archaeology and Anthropology in Northern Pakistan (1955‑2011), Journal of Asian Civilizations, Special Issue, 34.1: 123–151. Olivieri, L.M. 2012. When and Why the Ancient Town of Barikot Was Abandoned? A Preliminary Note Based on the Last Archaeological Data. Pakistan Heritage 4: 109–120. Olivieri, L.M. 2014a. Digging up: Fieldwork Guidelines for Archaeology Students. ACT Report and Memoirs, Series Minor, 1. Lahore: Sang-e-Meel. Olivieri, L.M. 2014b. The Last Phases at Barikot: Urban Cults and Preliminary Chronology: Data from the 2012 Excavation Campaign in Swat. Journal of Inner Asian Art and Archaeology 6: 7–40. Olivieri, L.M. 2015a. M.A. Stein and the ‘Lords of the Marches’: New Archival Materials. ACT Reports and Memoirs, Archival Studies, 1. Lahore: Sang-e-Meel. Olivieri, L.M. 2015b. Frontier Archaeology: Sir Aurel Stein, Swat and the Indian Aornos. South Asian Studies 31.1: 58–70. Olivieri, L.M. 2015c. Talking Stones: Painted Rock Shelters of the Swat Valley. ACT Report and Memoirs, Series Minor, 2. Lahore: Sang-e-Meel. Olivieri, L.M. 2015d. Urban Defenses at Bīr-koṭ-ghwaṇḍai, Swat (Pakistan): New Data from the 2014 Excavation Campaign. Ancient Civilizations from Scythia to Siberia 21.1: 183–199. Olivieri, L.M. 2017. The Last Phases at Barikot: Urban Cults and Sacred Architecture. Data from the Spring 2013 Excavation Campaign in Swat. Journal of Inner Asia Art and Archaeology 7: 7–30.

212  Luca M. Olivieri Olivieri, L.M., and E. Iori. (forthcoming). Early-historic Data from the 2016 Excavation Campaigns at the Urban Site of Barikot, Swat (Pakistan): A Shifiting Perspective, in A. Hardy, and L. Greaves (eds.), 23rd South Asian Archaeology and Art Conference. Olivieri, L.M., and M. Vidale, with contributions by A. Nasir, T. Saeed, L. Colliva, R. Garbini, L. Langella, R. Micheli, and E. Morigi. 2006. Archaeology and Settlement History in a Test-Area of the Swat Valley: Preliminary Report on the AMSV Project (1st Phase), in Id (ed.), East and West 56.1–3, Special Issue for the 50th Anniversary of the IsIAO Italian Archaeological Mission in Pakistan: 73–150. Olivieri, L.M., with others. 2014. The Last Phases of the Urban Site of Birkot-ghwandai (Barikot): The Buddhist sites of Gumbat and Amluk-dara (Barikot). ACT Reports and Memoirs, 2. Lahore: Sang-e-Meel. Rahman, A. 1968‑1969. Excavation at Damkot. Ancient Pakistan IV: 103–250. Rougemont, G. 2012. Corpus Inscriptionum Iranicarum: II.I. Inscriptions greques d’Iran et d’Asie centrale. London: Corpus Inscriptionum Iranicarum and School of Oriental and Africal Studies. Stein, A. 1930. An Archaeological Tour in Upper Swāt and Adjacent Hill Tracts. Memoirs of the Archaeological Survey of India, 42. Calcutta: Government of India Central Publication Branch. Tanweer, T. 2011. The Italian Archaeological Mission: A Preliminary Archival Study, in Ghani-ur-Rahman, and L.M. Olivieri (eds.), Italian Archaeology and Anthropology in Northern Pakistan (1955-2011), Journal of Asian Civilizations, Special Issue, 34.1: 41‑49. Tribulato, O., and L.M. Olivieri. (forthcoming). Writing Greek in the Swat Region: A New Graffito from Barikot (Swat). Zeitschrift für Papyrologie und Epigraphik. Tusa, S. 1981. Notes on Some Protohistoric Finds in the Swāt Valley (Pakistan). East and West 31.1‑2: 99‑120. Vidale, M., and R. Micheli. 2017. New Evidence from Protohistoric Graveyards of the Swat Valley (Khyber Pakthunkhwa, Pakistan): Observations on Funerary Practices and Absolute Chronology. Antiquity 356: 389–405. Vidale, M., R. Micheli, and L.M. Olivieri, eds. 2016. Excavations at the Protohistoric Graveyards of Gogdara and Udegram. ACT Reports and Memoirs, III. Lahore: Sang-e-Meel Publisher. Wheeler, M. 1952. Review to J. Marshall, Taxila, 3 vols. The Antiquaries Journal 32.3‑4: 204‑206. Wheeler, M. 1954. Archaeology from the Earth. Oxford: Oxford University Press. Wheeler, M. 1962. Chārsada: A Metropolis of the North-West Frontier. Oxford: Oxford University Press. Wooley, L. 1961. The Young Archaeologist. Edimbourgh: Edimbourgh University Press.

8 The beginning and development of Gandhāran collections in German public museums Britta Schneider1

This chapter focuses on the beginnings of the collection of Gandhāran sculptures in the Museum für Völkerkunde (Ethnological Museum) in Berlin. Its collection is the oldest and largest in Germany. The collections in the Ethnological Museums of Munich and Stuttgart are much smaller and younger.

Berlin – from Kunstkammer to museum At present the Gandhāran Collection in Berlin is housed in the Museum für Asiatische Kunst (Asian Art Museum). This museum emerged from the unification of the Museum für Indische Kunst (Museum of Indian Art) and the Museum für Ostasiatische Kunst (Museum of East-Asian Art) in 2006. Both of them can be traced back to the Ethnologische Sammlung (Ethnological Collection), a department of the Königliche Museen zu Berlin (Royal Museums of Berlin),2 and beyond that to the private collections of the Prussian kings. After the end of the Napoleonic occupation of Berlin in 1806, King Friedrich Wilhelm III (1797–1840) opened the way for several reforms. The renovation of the educational system became the key reform. Wilhelm von Humboldt (1767–1835) was the central figure in this process. He wanted education for every citizen. He was convinced that a nation had to be educated to make progress. As a result of the reform, the new educated bourgeoisie demanded access to the art treasures of the nation. In 1815 the Prussian king took the initiative for transforming his private collections of paintings and sculptures into a museum. His ancestors had started collecting works of art, weapons and curiosities from about the seventeenth century onwards. This type of private royal collection was commonly called ‘Cabinet of Curiosities’ or ‘Kunstkammer’. The king entrusted Friedrich Schinkel (1781–1841) with the plan for a museum. The architect chose a

214  Britta Schneider location directly opposite the northern side of the king’s city palace. The museum, the first in Prussia, opened in 1830. Admission was free and entry was allowed to all. Alexander von Humboldt, Wilhelm’s older brother, and Carl Ritter made the suggestion to establish an ‘Ethnological Collection’ as well. The king took up this proposal, and a second museum building became necessary. The Neues Museum (New Museum) built by Schinkel’s student Friedrich August Stüler (1800–1865) was opened in 1859. In 1856 the ethnological, pre-historic and Egyptian collections were moved from the palace to the New Museum. In 1869 the Ethnological Collection got its own assistant-director, Adolf Bastian (1826– 1905). In the same year, Bastian together with Rudolf von Vichow and other scholars launched the ‘Berliner Gesellschaft für Anthropologie, Ethnologie und Urgeschichte’ (Berlin Society for Anthropology, Ethnology and Prehistorical Research). This society meticulously pursued the foundation of an independent Ethnological Museum in Berlin. Finally in December 1873, the German Emperor Wilhelm I granted the ‘Ethnological Collection’ the status of an independent museum with a building of its own. For the Königliches Museum für Völkerkunde (Royal Ethnological Museum) a location to the south of Tiergarten was chosen.3 In 1876 Bastian was appointed director. In 1886 the emperor opened the museum. But the installation of some exhibits remained well behind schedule. The first official guide to the museum published in 1888 mentioned that from the Indian Department which consisted of three parts (South Asia, South-East Asia and the Indian Archipelago), so far only the collection of South Asian artefacts was on display. In cabinet 27 several sculptures and replica from Gandhāra were exhibited.4

The German discovery of Gandhāra How and when did the German scholarly community become aware of Gandhāra, its history and art? Carl Ritter (1779–1859), the German pioneer in the field of geography, was fascinated by the new finds from the northwest of British India published by Mountstuart Elphinstone5 and Alexander Burnes6 through articles in the Asiatick Researches and the Journal Asiatique. He was aware that these publications were hardly known in Europe and especially in Germany. So in 1838 he published a German summary of the finds by Elphinstone (Manikyala) and Burnes (Bamiyan) in his book Die Stupas (Topes) oder die architectonischen Denkmale an der Indo-Baktrischen Königsstraße und die

Gandhāran collections in German museums 215 Colosse von Bamiyan (The Stupas and Architectural Monuments on the Indo-Bactrian Royal Trunk Road and the Colossi of Bamiyan). With his book Ritter wanted to inspire his readers so that they would go and conduct field investigations.7 In the same year Christian Lassen (1800–1876), a Norwegian Sanskrit scholar who had studied in Germany and France and became professor in Bonn, Germany (1840), published his decipherment of the Kharosthi script.8 He worked on it in parallel to James Prinsep who also published his results in 1838. Between 1847 and 1861 Lassen published a first overview of everything that was known to this date about older Indian philology and history. His magnum opus Indische Altertumskunde (Indian Antiquities) consisted of four volumes.9 So far the exploration of Gandhāra was concentrated not only in Germany, but elsewhere, as well on geography, coins and inscriptions. Only very few sculptures from the northwest, mainly chance finds, had been published, most of them in English-language journals. In 1857 Carl Friedrich Koeppen’s (1808–1863) very popular10 first volume on the life of the Buddha and the history of Buddhism was published.11 Koeppen, a follower of the Young Hegelian movement, was a historian and journalist, not a Sanskrit scholar. In his book he showed a thorough understanding of the contemporary research on Buddhism, and especially the translations of the records of the Chinese pilgrims.12 After 1870 the British colonial administration in India began excavating Gandhāran sites in what they called a systematic way and published the results.13 In Germany the first article that dealt exclusively with Greek influences on Indian art and especially on Gandhāran sculptures was written by Ernst Curtius (1814–1896).14 He was a German archaeologist who, besides his professorship at the faculty of Archaeology at the University of Berlin, held the position of a director of the Antiquarium (a collection of small artefacts from antiquity – for example vases, terracottas, engraved gems) in the Altes Museum, Berlin (1872–1896). Curtius was convinced that classical archaeology should not refuse to consider the new finds from the northwest Frontier Area, although the closer examination would properly become the task of the Orientalist.15 Curtius mentioned in his article that recently selected samples of Gandhāran sculptures had arrived in Berlin.16 One of the drawings published by Curtius shows the head of a Bodhisattva that was found by Gottfried Wilhelm Leitner and belonged to his private collection (Figure 8.1). The four other pieces were probably from the Lahore Museum. All sculptures were only known to Curtius through photographs.

216  Britta Schneider

Figure 8.1 Examples of Greek art in India Source: Ernst Curtius, Griechische Kunst in Indien. Archäologische Zeitung 8 (33), 1876, p. 90‒95, Taf. 11, https://archive.org/details/archaeologischez33deut

The Gandhāran collection in Berlin until World War I Right from the beginning all Royal Museums of Berlin were supported by patrons and enthusiasts who were travelling around the world on behalf of the museums and donated their collections or granted financial support to enlarge the museums’ collections.

The nucleus of the collection The selected samples of Gandhāran sculptures and photos which were mentioned by Curtius in his article17 were given to the Ethnologische Sammlung by Fedor Jagor (1814–1900), a private traveller and selftaught Ethnologist with a keen interest in developing the museums in Berlin. On behalf of the Royal Museums of Berlin he made three

Gandhāran collections in German museums 217 long-term voyages to South and South East Asia, the Indian Archipelago and the South Sea.18 On his second long-term journey from 1873 to 1876 he also travelled to Peshawar where he arrived in April 1874. In a letter to the director of the Ethnological Department Adolf Bastian dated April 23, 1874, Jagor wrote that he was allowed by the Commissionaire of Peshawar, Captain Edward Lacon Ommanney, to select six Gandhāran sculptures from the Peshawar Museum.19 His choice was as follows: two complete figures standing ♀20(MIK I 32, MIK I 215), one complete figure sitting ♀ (MIK I 74), one broken sitting (MIK I 566), one head (MIK I 120, all five made of argillaceous shale; one head made of lime mortar. (MIK I 224) In addition, he was given two reliefs (MIK I 91 and I C 5995 – this relief was lost during World War II) which had been excavated from a heap of rubble outside the Hasht Nagari Gate21 in the south of Peshawar when Jagor visited the place. Jagor gives the conditions as more or less mutilated with fusion on all surfaces. Beside the sculptures Jagor also sent an album of 51 photos to Berlin. He wrote that a rich collection of the famous sculptures from Takht-i Bahi was in the Lahore Museum and that by order of General Cunningham the most interesting of them were photographed.22 This is only partly correct. Because this album seems to agree with the “Descriptive List of Selected Buddhist Sculptures in the Lahore Central Museum” written by Cunningham and published 1873,23 Takhti-i Bahi is only one of several find-spots Cunningham mentioned. Mixing up of provenances was not rare in the early days of exploring Gandhāra, as Elizabeth Errington has shown in her doctoral thesis.24 Therefore we cannot be sure that the six sculptures Jagor selected for the Berlin Museum were really found in Takht-i Bahi. The report of F. H. Wilcher25 about the campaign of the Sappers and Miners at Takht-i Bahi from January 23, 1871, until April 12, 1871, which he led, is silent about the place they brought their finds from. In his letter to Adolf Bastian, Jagor mentioned that he made an excursion to Jamalgarhi, where he saw a large number of highly mutilated sculptures. Although he searched for several hours, he did not find anything he would have liked to take with him. Therefore, he had the hope that it might be possible to get a Jamalgarhi piece from General Cunningham. Jagor expected to make his acquaintance in Simla.26

218  Britta Schneider In 1890 Jagor donated one more sculpture to the Königliches Museum für Völkerkunde: a Bodhisattva (MIK I 441, old I C 23860), which he had kept in his private apartment in Berlin for years.27 The sculpture was from Jamalgarhi, he wrote.28 So far I have not been able to find any hint in the archives that Jagor really managed to get a sculpture directly from Cunningham himself, but of course it cannot be ruled out. Jagor’s whole collection of ethnological material was so large that it was impossible to put it on display in the Neues Museum itself. Hence, Jagor himself arranged an exhibition of his donations in an annex, the so-called Bergakademie.29 The Gandhāran pieces seemed to have been there too, because they are not mentioned in the early guides of the royal museums.30 Three years before Jagor’s journey to India, in 1871, the director general of the Royal Museums received a letter from the Reichskanzleramt (Imperial Chancellor’s Office). Therein the Imperial German consul Gumpert based in Bombay informed the ministry that Professor Dr. G. W. Leitner offered a number of Graeco-Indian and Buddhist antiquities at the exchange value of 600 Rupies (that were the calculated costs of his trip) as a whole or shared with the museum in Vienna to the Ethnological Collection, Berlin.31 The Hungarian-born Gottlieb Wilhelm Leitner (1840–1899) had a talent for languages. When only 15 years old (1855), he became interpreter to the British Commissariat during the Crimean War (1853–1856). In 1861 he was appointed professor of Arabic with Muhammadan Law at King’s College, London. In 1862 he received an MA and PhD from the University of Freiburg, Germany. Shortly afterwards in the same year he was naturalized as a British subject. Two years later Leitner was appointed to the post of principal of the Government College at Lahore and spent the next 15 years in India. In the years 1866–1869 he made several excursions to the north of the British Territory (Hunza, Gilgit, Kashmir), learned many of the local languages (partly by inviting locals to live in his house in Lahore) and published books on their vocabulary and grammar.32 During his Christmas holidays in 1870 he paid a visit to the ruins of Takht-i Bahi. He wrote: We began to dig after removing the slates obstructing the way, at the third house in the second row on the extreme east of the city. Half a foot below the surface we came to a circular slab, under which a female statue was found. Another slab, with broad lines, concealed the figure of a warrior, whilst a third with numerous and narrow lines covered a carved group of boys.33

Gandhāran collections in German museums 219 After three days in Takht-i Bahi Leitner had to return to Lahore, but he made arrangements to continue the search. Some months after Leitner’s excursion, the letter from the Imperial Consul arrived in Berlin together with several photographs and a newspaper article written in Urdu containing several drawings of Leitner’s finds.34 Yet the director of the Ethnologische Sammlung Adolf Bastian hesitated. On the one hand he wanted to be sure that Leitner was the legal owner of the artefacts – he recalled the Elgin’s marbles scandal. On the other hand he was not sure whether the material and the condition of the pieces would survive transportation to Germany.35 Because of this Bastian wanted to make inquiries on both points to Georg Bühler in Bombay. Unfortunately Bühler’s reply is not preserved in the records of the Museum für Völkerkunde. In the end Berlin did not purchase anything. Nevertheless Leitner kept in touch with the museum, and around 187536 he donated several casts and copies of sculptures from his private collection and from the Lahore Museum Collection to the Berlin Museum. The guide to the Royal Museums from 1880 mentions these Graeco-Buddhistic artefacts as free standing in room F and G of the Neues Museum.37 The term Graeco-Buddhistic for sculptures from Gandhāra was first introduced by George Pearse in an article on his work at Taxila,38 but Gottlieb Wilhelm Leitner made the term popular by his constant publishing39 and commitment to show his collection to the public in Vienna, London, Florence and Berlin.40 Elizabeth Errington wrote in her PhD thesis41: “The so-called Leitner collection was thus the first to arouse general interest in the art of Gandhāra in Europe itself.”

The first expansion In 1890 Adolf Bastian travelled to India. In Peshawar he succeeded in getting three sculptures as a gift for the museum from Colonel Robert Warburton.42 And in Lahore he ordered two casts to be made from sculptures in the Lahore Museum.43 One year later, in 1891, the collection of Gandhāran sculptures in Berlin saw its first major growth. Through the procuration of the German consul in Calcutta, Baron von Heyking, the Royal Ethnological Museum received 52 Gandhāran sculptures. They came as a kind of compensation. Von Heyking described the deal in a letter to Reichskanzler Leo von Caprivi that in autumn 1890 on a journey through the Punjab he got notice that his semi-official request to get permission to excavate Graeco-Buddhist sculptures in the neighbourhood of Peshawar was turned down by the director

220  Britta Schneider of the Lahore Museum, John Lockwood Kipling. Kipling claimed ownership of these artefacts for his museum alone. He declared the assignment of anything taken out of British soil to a foreign museum as illegal.44 During a dinner the opportunity arose for von Heyking to speak with the lieutenant governor of Punjab, James Lyall. Sir James Lyall kindly replied that he could not interfere with the affairs of the Lahore Museum, but it would be his pleasure to give order to his Deputy Commissionaire Mr. Merk to compile a certain number of sculptures from the northern Peshawar District between the Kabul River and the mountains.45 Based on all material that was now in the Königliches Museum für Völkerkunde (64 original sculptures, 15 casts and an uncertain number of photos) Grünwedel published his Buddhistische Kunst in Indien in 1893.46 In Appendix II he listed all original sculptures and casts the museum owned at that time.47

The second expansion Four years after Gottlieb Wilhelm Leitner’s death in 1899, the connection of the Leitner family with the Königliches Museum für Völkerkunde was renewed. His widow Olympia and his son Henry offered Leitner’s whole collection kept in Leitner’s museum at the Oriental Institute, Woking (Surrey, England) to the Berlin Museum.48 Albert Grünwedel, who had been Direktorialassistent since 1896, wrote on July 27, 1903: In my opinion we should not miss the last chance to purchase original material from Gandhāra. I am at any time willing to travel to London to see the things in situ.49 About a week later Grünwedel was on his way to Woking. After his return he once more argued for the purchase: To export such pieces is nowadays prohibited and it is almost hopeless to get originals. If it would not have been Leitner’s wish that his collection one day should go to a German Museum, the American competitors would again have snatched everything away from us.50 Although the museum had to cope with financial problems and needed the help of sponsors, in the end it succeeded so that in 1904

Gandhāran collections in German museums 221 and 190751 over 550 original Gandhāran sculptures together with about 100 replicas – mostly of artefacts from the Lahore Museum – arrived in Berlin. Thanks to the financial support of three of the most important sponsors of the Royal Museum, Gustav Jacoby, Magarete Krupp and James Simon, the purchase of yet another significant Gandhāran collection became possible in 1907. About 80 stone sculptures from the London-based Mansel Longworth Dames (1850–1922) found their new home in Berlin.52 Right from the opening day all branches of the Königliches Museum für Völkerkunde faced a severe problem: the lack of space to exhibit all its treasures. The cabinets got crammed with artefacts and the staircases were turned into galleries. In 1911 the majority of the now huge Gandhāran Collection had to be located far away from the rest of the Indische Sammlung on the second and third floors of the museum.53 This was not a satisfactory solution. A new concept for all departments had to be formulated: the division of the material in a public and a study collection. In 1912 the director decided to give the order to build a new museum complex in Dahlem, in the southwest of Berlin. Work started in 1914 and was stopped due to World War I two years later.

After World War I until World War II After the war the construction work at Dahlem was continued on a much smaller scale. Instead of a four-wing building – one wing each for America, Africa, the Pacific and Asia – only the Asian building was completed. Because of this, only a study collection was transferred, and the public exhibition in the Königgrätzer Straße was rearranged. After its re-opening in 1926, the masterpieces of the Gandhāran Collection were shown at the basement in rooms II (Figures 8.2 and 8.3) and III, in four glass cabinets in room IV54 and as part of an arrangement that illustrated the development of Buddhist art in India in room XXVIII. Replicas of Gandhāran sculptures and photos were not integrated in the new displays. About ten years after the end of World War I the museum continued to enlarge its Gandhāran Collection. It now started buying from the international art market. In 1930 and 1931 the museum bought around 80 artworks mostly made of stucco from Ram Dass, an art dealer based in Rawalpindi. In 1934 and 1937 Ram Dass sold several stone sculptures to the Staatliches Museum für Völkerkunde.

222  Britta Schneider

Figure 8.2 Gandhāra exhibition (1926), room III towards room II, Museum für Völkerkunde, Berlin Source: © Staatliche Museen zu Berlin, Museum für Asiatische Kunst; photographer: Max Krajewsky, 1926

Although the museum started as early as 1934 to make preparations in the event of a war and evacuated the first precious objects in 1938, it was in the end not possible to pack each and every piece and carry them away to the salt mines in Kaiseroda and Grasleben. Due to World War II more than 2,100 objects from the Indian collection were lost,55 amongst them 188 Gandhāran artefacts and 51 casts of Gandhāran sculptures. “In 2002, permission was granted for the first time to view around 20 per cent of the collection’s missing holdings, in the storerooms of the Hermitage in St. Petersburg, where they have been held since the Second World War.”56 The following exceptional Gandhāran sculptures from the Staatliches Museum für Völkerkunde are part of these 20 per cent: • I C 36836 stele of the Bodhisattva Avalokitesvara, purchased in 1910 from E. Elsmie; • I C 37058 stele of the Bodhisattva Maitreya, purchased in 1910 from Colonel Carr-Calthrop;

Gandhāran collections in German museums 223

Figure 8.3 Gandhāra exhibition (1926), room II, wall 2, Museum für Völkerkunde, Berlin Source: © Staatliche Museen zu Berlin, Museum für Asiatische Kunst; photographer: Max Krajewsky, 1926

•

I C 37060 standing Buddhas in niches with Vajrapani in between, purchased in 1910 from Colonel Carr-Calthrop; • I C 43399 a Buddha stele, purchased in 1934/37 from Ram Dass; • I C 43400 a Bodhisattva stele, purchased in 1934/37 from Ram Dass.57

After World War II The old building of the Staatliches Museum für Völkerkunde was destroyed and Dahlem became the new permanent home of the ethnological collections. In 1955 plans were made for an exhibition on Turfan and Gandhāra. For this, the upper entrance hall of the building had to be modified. In the same year the director of the Indische Abteilung, Herbert Härtel (1921–2005), started negotiations with the Americans to return the property of the department from the art collecting point

224  Britta Schneider at the German town of Celle.58 Two years later the exhibition ‘Turfan and Gandhāra’ was opened and a small catalogue59 was published. During this time Herbert Härtel revived the old discussion of separating Indian fine art from South Asian applied art and ethnological material. In autumn 1962 the president of the Stiftung Preussischer Kulturbesitz (founded in August 1957) declared on January 1, 1963, the foundation of the new Museum für Indische Kunst. It is still the only institution of this kind in Germany. The collection of stone and stucco sculptures from Gandhāra that had its beginning in 1874 had selectively expanded its holdings since 1964 and now has one of the most interesting collections, for example, on representations of the Buddha’s life.

Munich In Munich, the development from a royal Kunstkammer to museum is similar to that in Berlin in the nineteenth century. The Bavarian royal family, the Wittelsbacher, was fascinated by collecting ethnological objects, and several family members brought artefacts back from their voyages.60 King Maximilian I, Joseph (1756–1825) and his son, King Ludwig I (1786–1868) were especially interested in ethnology, financed expeditions and bought collections from all over the world. The Museum für Völkerkunde (Museum of Ethnology) was founded in 1862 and is the oldest in Germany. But since 1911, it has had Gandhāran art. The first piece, a small Bodhisattva (MFK 11–432),61 was purchased together with a Chinese vase and a Persian tile. The lot was offered by M. Bing, an art dealer from Paris.62 In the same year the museum’s director, Lucian Scherman, travelled to India and Burma to enlarge the museum’s collection. On November 10, 1911, he and his wife arrived in Peshawar, where they visited the museum. Director Hargreave showed them around, and in his travel diary Scherman describes in great detail how the Gandhāran sculptures were arranged and displayed.63 Two days later he was invited to dinner at the home of Sir Roos-Keppel (chief commissioner of the northwest Frontier Province), who made to Scherman the gift of a Gandhāran sculpture showing the birth of Siddhartha Gautama (Inv.-Nr. Pe 85)64 from his private collection.65 On another occasion Hargreave took Scherman on an excursion to the excavation at Shah-ji ki Dheri. Lucian Scherman did not receive or buy any other Gandhāran sculptures during his journey. Collecting Gandhāran objects was never pursued systematically by the Staatliches Museum für Völkerkunde, Munich. Here and there

Gandhāran collections in German museums 225 smaller items which were offered directly to the director were purchased. That was also the case in 1931 when some stucco heads, which in 1930 had been purchased by the Staatliches Museum für Völkerkunde, Berlin from Ram Dass, were offered by the head of the Indische Abteilung Otto Kümmel to director Scherman66 who selected five pieces (MFK nos. 31-16-1-5). Exchanging or reselling art from one museum to another was normal in those days. This practise established links between museums and scholars and sometimes it financed a museum’s daily routine.67 Fourteen out of the eighteen pieces altogether were exhibited in 1998.68 Munich was the first museum to showcase the narrative reliefs and sculptures around a stūpa (unfortunately Munich does not own scenes from all major events of the life of the Buddha). About ten years ago the complete Indian exhibition went into the reserve collection. This situation has not changed; nevertheless, the Indian department tries to enlarge its collection. In 2013 the Freundeskreis of the Museum made the gift of a standing Buddha to the collection. The so far unpublished sculpture is 90 cm high. Its provenance is unknown, as is often the case with artefacts from the international art market (the sculpture was sold by Michael Woerner Oriental Art Ltd).

Stuttgart The third museum that must be mentioned in the context of Gandhāran collections in Germany is the Linden Museum in Stuttgart. This museum’s nucleus is no former royal collection. In 1882 leading tradesmen from the region established the Württembergischer Verein für Handelsgeographie und Förderung der Deutschen Interessen im Ausland. The society’s aim was the promotion of geography, trade and culture. In 1889 the society’s president Duke Karl von Linden (1838–1910) transformed the trade-geographic museum into a scholarly based ethnological museum. Since 1973 the Linden Museum has been operated by the federal state Baden-Württemberg and the city of Stuttgart. It thus became an official state museum. Until the 1970s the museum did not own a single piece of Gandhāran art. At that time a change in the old ethnological concept of gathering material goods from different cultures was made. From now on the idea was to document all aspects of nonEuropean cultures, including the so-called advanced civilizations. In addition to a remarkable collection of Cola bronzes and sculptures from Thailand, the museum has built up a small but fine collection of Gandhāra art.69

226  Britta Schneider

Perspective of the Gandhāran collection in Berlin In December 2015 the Indian exhibition of the Museum für Asiatische Kunst closed its doors in Dahlem forever. Together with all other nonEuropean collections the museum moved to the Humboldt Forum in the reconstructed city palace of the Prussian kings and German emperors and will hopefully re-open in 2019. According to the new exhibition concept, it will be possible to show more Gandhāran material – stone and stucco – to the public in the future.

Notes 1 First of all I would like to thank Himanshu Prabha Ray, who suggested the topic to me. Beyond that I am much obliged to Martina Stoye, curator of South Asian art, Museum für Asiatische Kunst, Berlin, for her generous supply of information on the various donations and purchases and her keen interest in continuing the exploration of the records; Silvia Davoli, for allowing me to read her unpublished lecture on G.W. Leitner; Anke Zenner (library and archives, Museum für Völkerkunde, Berlin) and Karin Guggeis (archives, Museum fünf Kontinente, München), for access to the records; Caren Dreyer (Museum für Asiatische Kunst, Berlin), for the supply of old photos; Juhyung Rhi, for his comments regarding my lecture in Delhi; and last but not least special thanks to Andrea Schlosser and Johannes Schneider for proofreading. 2 After World War I the name had to be changed to Staatliche Museen zu Berlin. 3 In 1880 the directors of the Königliche Museen zu Berlin (Royal Museums of Berlin) decided that in the future only the hohe Kunst, which in those days meant the art of Europe and the Near East, should be exhibited on Museum Island. 4 Königliche Museen zu Berlin, Führer durch die Sammlung des Museums für Völkerkunde – Erster Nachtrag – Indische Sammlung, Herausgegeben von der Generalverwaltung. Berlin, Spemann, 1988: 22. 5 Mountstuart Elphinstone, An account of the kingdom of Caubul, and its dependencies in Persia, Tartary, and India comprising a view of the Afghan nation, and a history of the Doorannee monarchy, 2 vols. London, Bentley, 1815. 6 Alexander Burnes, Travels into Bokhara being the account of a journey from India to Cabool, Tartary and Persia – also, narrative of a voyage on the Indus from the sea to Lahore, with presents of the king of Great Britain; performed by order of the supreme government of India, in the years 1831, 32, and 33, London, Murrey, 1834. 7 Carl Ritter, Die Stupa's (Topes) oder die architectonischen Denkmale an der Indo-Baktrischen Königsstraße und die Colosse von Bamiyan, Nicolaische Buchhandlung, Berlin, 1838, p. v. 8 Christian Lassen, Zur Geschichte der griechischen und indoskythischen Könige in Baktrien, Kabul und Indien durch Entzifferung der Alt-Kabulischen Legenden auf ihren Münzen, Bonn, H. B. König, 1838.

Gandhāran collections in German museums 227 9 Christian Lassen, Indische Altertumskunde, 4 vols., Bonn, H. B. König, 1847–1861. 10 Markus Mode, Ein vergessener Anfang: Carl Ritter und die “Kolosse von Bamiyan”: Zum 220. Geburtstag des großen deutschen Geographen. www.orientarch.uni-halle.de/ca/bam/bamiyanx.htm (accessed on 10th August 2015). 11 Carl Friedrich Koeppen, Die Religion des Buddha: Erster Band: Die Religion des Buddha und ihre Entstehung, Schneider, Berlin, 1857. 12 Koeppen used for his book the works of Jean-Pierre Abel- Remusat and Stanislas Julien. 13 For a complete overview see Elisabeth Errington, The Western Discovery of Gandhāra and the Finds of Jamālgarhī, Unpublished Ph.D. thesis, School of African and Oriental Studies, London, 1987. 14 Ernst Curtius, Die Griechische Kunst in Indien, Archäologische Zeitung, 8.39, 1875: 90‒95, Taf. 11. 15 “Unter den Gebieten, welche neuerdings der kunstgeschichtlichen Forschung eröffnet sind, ist ein weit entlegenes und von der deutschen Wissenschaft noch kaum beachtetes, ein Gebiet, dessen eigentliche Bearbeitung die Aufgabe der Orientalisten bleiben wird, dessen ernstliche Berücksichtigung aber auch die klassische Archäologie nicht ablehnen darf” (Curtius, Die Griechische Kunst in Indien: 90). 16 Curtius, Die Griechische Kunst in Indien: 91. 17 Curtius, Die Griechische Kunst in Indien: 91, 94. 18 1859–61, 1873–76 and 1890–93. 19 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta Pars. I. B. 13 letter from F. Jagor to A. Bastian dated 23rd April 1874: ‘Captain E. L. Ommanney H.M. Beng. Staff Corps. Dep Commissoner v Peshawar hat mir erlaubt sechs Stück auszuwählen’. 20 Here Jagor clearly made a mistake by using the symbol for females; he did not seem to be aware of the real gender and identity of the figures. 21 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta Pars. I. B. 13 letter from F. Jagor to A. Bastian dated 23rd April 1874: ‘Ich erhielt ferner 2 Reliefs mit kl. Figuren die während meiner Anwesenh[ei]t in einem Schutthüg[e]l (bei) Peshawar vor dem Hushnagar Thor ausgegraben wurden.’ 22 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta Pars. I. B. 13 letter from F. Jagor to A. Bastian dated 23rd April 1874: ‘Von den berühmten Skulpturen aus Takht i Bhai (buddhistisch mit griechischem Einfluss) befindet sich eine reiche Samml[un]g im Lahore Museum; die interes[s]anteren davon hat Gen. Cunningham photographieren lassen, 51 Stück, die sie mit der nächsten Calcutta Send[un]g erhalten’. 23 Punjab Government Gazette, Supplement, 24th July 1873: 631–636. 24 One example of swapping provenances is the case of the so-called Takht-i Bahi inscription. See Errington, Western Discovery: 115. 25 Punjab Government Gazette, Supplement 6th August 1874: 528–532. 26 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta Pars. I. B. 13 letter from F. Jagor to A. Bastian dated 23rd April 1874: “Wie ich später erfuhr, sind aber mehrere oder gar viele Karrenladungen voll mehr oder weniger gut erhaltener Skulpturen

228  Britta Schneider an Gen. Cunningham gesandt, wahrscheinl[ich] nach Simla, wo ich seine Bekanntschaft zu machen hoffe. Vielleicht gelingt es dort von ihm etwas zu erhalten.” 27 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta 20/ E 1212/90 short notice on a piece of paper from Jagor: “Bitte eine baktrische Skulpture aus Jamagarhai bei mir abholen lassen. Mit freundlichem Gruss, Ihr Jagor Enckeplatz 4.” 28 Albert Grünwedel, who published the sculpture in his book Buddhistische Kunst in Indien. 2. Auflage. Berlin, Spemann, 1900, p. 161 Nr. 89, gave as its provenance Takht-i Bahi; also the old index card of the museum correctly says Jamalgarhi. This seems to be another case of mixing up provenances. 29 Amtliche Berichte aus den Königlichen Kunstsammlungen, 1, January 1, 1880: XI. 30 Königliche Museen zu Berlin, Führer durch die Königlichen Museen, Weidmannsche Buchhandlung, Berlin, 1880. In those days it was common to put everything on display. No reserve collections existed. 31 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta E 581/71, first folio. 32 For further information on G.W. Leitner see W.D. Rubinstein, Leit ner [formerly Sapier], Gottlieb Wilhelm (1840–1899). Oxford Dicionary of National Biography. 2004 online version, www.oxforddnb.com. oxforddnb.emedia1.bsb-muenchen.de/view/article/51109. 33 Gottlieb Wilhelm Leitner, The Languages and Races of Dardistan, Goverment Central Book Depot, Lahore, 1877 Appendix, I f. This is a reprint of an article by Leitner in Indian Public Opinion, Lahore 11th February 1871. 34 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta E 581/71. 35 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta E 581/71 letter by Adolf Bastian dated July 8, 1871. 36 So far I have not been able to see the correspondence between Leitner and the EC/REM dealing with the donation of ca. 10 casts. 37 Königliche Museen zu Berlin, Führer, 1880: 148. 38 Georg Godfrey Pearse, Literary Intelligence. Journal of the Asiatic Society of Bengal, 34 (Nos. I to IV), 1865: 113. 39 For example: Gottfried Wilhelm Leitner, Graeco-Buddhistic Sculptures, The Art Journal, November 1874: 324; and a reprint of several articles in Leitner, Dardistan, 1877. 40 Exhibitions of Gandhāran Art organized by Leitner: 1871 Exhibition of Gandhāran sulptures at the “Unteres Belvedere”, G. W. Leitner, Altund Neu-Indische Kunstgegenstände aus Professor Leitner's Jüngster Sammlung. Wien, K.K. Österreichische Museen, 1883: 1; 1873 Exhibition of Gandhāran sculptures from Leitner’s private collection together with photographs and casts from the Lahore Museum on the occasion of the Great Exhibition in Vienna, John Forbes Watson, A Classified and Descriptive Catalogue of the Indian Department Vienna Universal Exhibition, Allan and Co., London, 1873; 1874 Exhibition of Gandhāran sculptures from Leitners private collection at the Royal Albert Hall, Elisabeth Errington, London, The 1878 Florence Exhibition of Gandhāran

Gandhāran collections in German museums 229 Sculptures. Maurizio Taddei (ed.), Angelo de Gubernatis: Europa e Oriente nell'Italia umbertina. Vol. II. Napoli, 1997, p. 140; 1878 Exhibition of Leitners “new material from Swat” in Florence on the occasion of the fourth international Congress of Orientalists organised by A. de Gubertinatis, Errington, Florence; 1883 Exhibition of Gandhāran Sculptures at the “K.K. Österreichisches Museum für Kunst und Industrie” Vienna, Leitner, Kunstgegenstände, 1883; 1884 foundation of the Leitner Museum, Woking. 41 Errington, Western Discovery, 1987, p. 160. 42 Sitzung von 20. December 1890. Reise des Herrn Bastian. Zeitschrift für Ethnologie, 22, 1890, p. 613. 43 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta E 749/90 and E 739/91. 44 Acta E 868/91: “meine einige Monate vorher in halbamtlicher Form gestellte Bitte, ein[ig]e graeco=buddhistische Skulpturen aus der Umgegend Peshawars für das Berliner Museum zu erhalten und dieselben eventuell auf dessen Kosten ausgraben lassen zu dürfen, auf Widerstand bei der Direktion des Museums zu Lahore gestos[s]en war. Der Direktor Mr. Kipling hatte in höchst engherziger Weise die Monopolosierung des Besitzes dieser Altertümer für seine Anstalt beansprucht und die Ueberlassung solcher aus britischem Boden gehobenen Kunstschätze an ein fremdländisches Museum als unzulässig bekämpft.” 45 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta E 868/91: Sir James Lyall erwiederte sehr liebenswürdig, er könne zwar angesichts der vorausgegangenen Verhandlungen mit der Museumsdirektion zu Lahore nicht die Genehmigung geben, daß für das Berliner Museum Ausgrabungen gemacht würden, er werde sich aber freuen wenn der Deputy Commissioner Mr. Merk mir persönlich einige Skulpturen, welche bei Peshawar zu finden wären, zur Verfügung Stellen würde. Mr. Merk, der bei dem Gespräch zugegen war, kam meinem Wunsche mit der größten Bereitwilligkeit entgegen und sandte mir einige Monate darauf eine Anzahl Skulpturen, welche im Norden des PeshawarDistriktes zwischen dem Kabul-Flusse und den nach Norden die Grenze bildenden Gebirgen aufgefunden worden sind. 46 Albert Grünwedel, Buddhistische Kunst in Indien, Berlin, Spemann, 1893; 2nd enlarged edition, Berlin, Spemann, 1900; English translation Agnes C. Gibson, Buddhist art in India. Revised and enlarged by James Burgess, Quaritsch, London, 1901, the list of original sculptures and casts in the REM, Berlin is not included. 47 Contrary to Jagors note E1212/90 that MIK I 441 (old IC23860) was from Jamalgarhi, Grünwedel, Buddhistische Kunst, 1900: 161, Nr. 89 attributes it to Takht-i Bahi. 48 The records dealing with these transactions will be published in an article in the “Indo-Asiatische Zeitschrift”. 49 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta E 958/03 notice by Grünwedel dated July 27, 1903: “Nach meiner Meinung dürfen wir uns diese einzige Gelegenheit noch Originale aus Gandhāra zu erwerben nicht entgehen lassen. Ich bin

230  Britta Schneider jederzeit bereit nach London zu reisen, um die Dinge an Ort und Stelle zu besichtigen.” 50 Staatliche Museen zu Berlin – Preußischer Kulturbesitz, Ethnologisches Museum, Acta E 958/03 report by Grünwedel dated August 31, 1903: “Solche Stücke zu exportieren ist heute verboten und es ist geradezu hoffnungslos, Originale zu erhalten. Wenn Leitners Wunsch nicht gewesen wäre, dass seine Sammlung mal in ein Deutsches Museum käme, würde uns amerikanische Concurrenz auch hier alles vor der Nase weggenommen haben; man sollte sich also die Gelegenheit nicht entgehen lassen, diese ausserordentliche, nie wiederkehrende Sammlung zu erwerben.” 51 Leitner not only collected Gandhāra, but also held a certain amount of pieces from Bihar, chiefly Bodh Gaya. They arrived in Berlin with the second shippment. 52 Other parts of the collection of M. L. Dames went to the Britsh Museum and the Victoria and Albert Museum, London. 53 Königliche Museen zu Berlin, Führer durch das Museum für Völkerkunde, 15, Auflage, Berlin, Reimer, 1911: 257. 54 Staatliche Museen zu Berlin, Führer durch das Museum für Völkerkunde I – Schausammlung, Verlag Walter de Gruyter & Co., Berlin und Leipzig, 1929, pp. 7–10. 55 Dokumentation der Verluste: Band III Museum für Indische Kunst, SMB, Berlin, 2002: 7–8. 56 www.smb.museum/museen-und-einrichtungen/museum-fuer-asiatischekunst/ueber-die-sammlung/kunstsammlung-sued-suedost-und-zentralasien.html (accessed on 8th August 2015) 57 Dokumetation, 2002: 39f., 44f. For modern photos of the five sculptures refere to: https://commons.wikimedia.org/wiki/Category:Art_of_India_ in_the_Hermitage (accessed on 8th August 2015). 58 Gerd Höpfner, Hundert Jahre Museum für Völkerkunde – Abteilung Südasien. Baessler-Archiv, Neue Folge, 21, 1973: 522. 59 Herbert Härtel, Turfan und Gandhāra – Frühmittelalterliche Kunst Zentralasiens. Stiftung Preussischer Kulturbesitz Staatliche Mussen– Indische Abteilung, Berlin, 1957 (2. Auflage. Berlin, 1964). 60 Claudius C. Müller (ed.), Exotische Welten – aus den völkerkundlichen Sammlungen der Wittelsbacher 1806–1848, München, J.H. Röll, 2007. 61 See Walter Raunig, Dorothee Schäfer and Wolfgang Stein (eds.), Indien – Reigen der Götter. (Mare Erythraeum 2), München, Staatliches Museum für Völkerkunde, München: 22 62 Museum fünf Kontinente, München, Archive, correspondence with M. Bing from 1911. 63 Museum fünf Kontinente, München, Archive, Lucian Scherman Notizbuch 10.11.1911. 64 See Walter Raunig, Dorothee Schäfer, Wolfgang Stein (eds.), Indien – Reigen der Götter (Mare Erythraeum 2). München, Staatliches Museum für Völkerkunde, München, p. 26. 65 Museum fünf Kontinente, München, Archive, Lucina Scherman Notizbuch 12.11.1911: “Er hat in seiner Wohnung ausgezeichnette Teppiche . . . gute alte Waffen, schöne Bronzen u[nd] eine Anzahl guter Gandhāraskulpturen. Eine derselben schenkt er uns – die besten, sagt er, hat er dem Kronprinzen s[einer] Z[eit] geschenkt”.

Gandhāran collections in German museums 231 66 Museum fünf Kontinente, München, Archive, letter from Otto Kümmel to Schermann, dated 22 nd December 1930: “Sehr verehrter Herr Kollege, . . . Ferner noch etwas, das ich Ihnen diesmal verkaufen möchte. Wir haben eine Anzahl Köpfe und Statuetten, Stucco, in Indien selbst erworben, die entweder aus den Hadda-Funden sind, was ich für das wahrscheinlichste halte, oder aus ganz ähnlichen (wohl Raub-)Grabungen stammen. Für uns ist es etwas zu viel, und ich möchte einiges abgeben.” 67 For a overview of this practise please refer to Beatrix Hoffmann, Das Museumsobjekt als Tausch- und Handelsgegenstand: Zum Bedeutungswandel musealer Objekte im Kontext der Veräußerungen aus dem Sammlungsbestand des Museums für Völkerkunde Berlin. (Kunstwissenschaft, Bd. 33), LIT Verlag, Berlin, 2012. 68 Walter Raunig, Dorothee Schäfer and Wolfgang Stein (eds.), Indien – Reigen der Götter (Mare Erythraeum, II), Staatliches Museum für Völkerkunde, München, 1998: 19–26. 69 Friedrich Kussmaul (ed.), Ferne Völker, Frühe Zeiten: Kunstwerke aus dem Linden-Museum Stuttgart, Staatliches Museum für Völkerkunde, Band 2: Orient, Südasien, Ostasien. Recklinghausen, Aurel Bongers, 1982.

9 Decoding Gandharan art The making of museum collections in India Himanshu Prabha Ray

Museums are institutions that express and support the activity of collecting, yet collecting is by no means restricted to them. However, in their public function as repositories of objects and loci of scholarly and didactic activity, museums score over private collections. An important issue relates to the role of museums: were these conceived as centres of public history or as locales of scholarly research or as centres that projected an ‘authoritative’ version of the past? This final chapter in the volume follows up on issues raised in the introduction about the production of knowledge about Gandhara in the colonial period and the means through which Gandhara school of art became a separate field of study, often seen as distinct from other regional developments of art and architecture associated with Buddhism in the subcontinent. More recently issues of national and community identity have gained primacy as museums increasingly need to raise resources for many of their scheduled activities, especially in the countries of North America. Thus the Gandhara gallery at the Asian Art Museum in San Francisco acknowledges the contribution of Isha and Asim Abdullah, who dedicate it ‘to their beloved Pakistan, one of the oldest cradles of human civilization’ (Figure 9.1), notwithstanding the fact that Pakistan came into being with the partition of the subcontinent in 1947, and the sculptures in the gallery date to the early centuries of the Common Era.1 Besides, the museum labels link the sculptures to the life of the Buddha, rather than to the region where they were found, namely Gandhara. Can the life of the Buddha be understood in terms of the present political boundaries of Pakistan without reference to other Buddhist sites in the subcontinent? Can the heritage of Pakistan be restricted to Gandharan art and not take cognizance of early Hindu temples dated from sixth to eleventh centuries found along the Indus?2 The Asian Art Museum has sixty objects from Gandhara, which is perhaps the largest group of sculptures from South Asia. These

Making of museum collections in India 233

Figure 9.1 Sign in South Asia Gallery Asian Art Museum, San Francisco Source: All photographs are from the author’s personal collection

were collected by Avery Brundage (1887–1975), the fifth president of the International Olympic Committee, and gifted to San Francisco in 1960, as he felt that the city provided a gateway to Asia and could help “San Francisco and the Bay Area become one of the world’s greatest centers of Oriental culture”. No doubt museum collections and their study cannot be divorced from the larger cultural contexts of philanthropy or national identity formation.3 This raises a more fundamental issue of identity of a cultural object displayed in a museum.4 Are aesthetic and cultural values inherent in objects? Can one speak of multiple identities: one, drawing on the context of its use in a living culture; and two, the museum context? Richard Davis has referred to the power and authority of the museum curator and the academic to prioritize knowledge over belief and living practices.5 In the post-colonial period, material representations of culture have become potent means to legitimate an ethnic group’s claims to unique identity and political power and to their attempt

234  Himanshu Prabha Ray to create a sense of unity among themselves, as evident from the example of the Asian Art Museum. Another example may be quoted from a symposium organized under the sponsorship of the Aesthetics Project supported by Christie’s titled Treasure and Tragedy: From Oxus to Indus: Gandhara, Bactria and Bamiyan Symposium and curated by a professor at the School of Arts and Aesthetics, Jawaharlal Nehru University on 27 and 28 January 2016 in New Delhi.6 The website of the symposium refers to the cosmopolitan culture of ancient Gandhara, due to its location at the cross-roads of the ancient world and continues: It is sadly ironic that this place of cultural confluences has become the site of culture wars. If some sites and objects such as the Bamiyan Buddhas or the collections of the Kabul Museum perished through deliberate acts of destruction, much else has been lost as the collateral damage of unending disturbance in the region, or may yet be lost in reckless ‘development’. The website also bemoans the fact that Gandharan art has been marginalized in the art history of post-Independence India and has become the art of nowhere.7 It is significant that a recent publication of the curator of the symposium drawing on a Getty Foundation Project on museums makes almost no reference to Gandhara, further corroborating the influence of the patron and the sponsor in not only museum displays, but also academic priorities.8 The catalogue of the Gandharan collection in the Metropolitan Museum, New York, raises an important issue, that is, that much of what is known of the art and architecture of the region originates from religious architecture and adequate attention has not been paid to settlement sites and contexts of finds not religious in nature. For example, Kurt Behrendt suggests that the earliest period of artistic activity in the region is evident from ritual dishes, luxury objects and trade goods found in urban contexts.9 As outlined in the introduction the trade networks that linked the region of Gandhara to the land and maritime routes provided a larger context to the art and need to be factored into any discussion of its artistic output. By highlighting the distinctive features of the five major museum collections of Gandhara sculptures in India, viz. Indian Museum, Kolkata; Chhatrapati Shivaji Maharaja Vastu Sangrahalaya, Mumbai; National Museum, Delhi; Government Museum, Chandigarh; and Heras Institute Museum, Mumbai, this chapter draws attention to the impact of the process of ‘collecting’ in an understanding of the cultural

Making of museum collections in India 235 heritage of Gandhara. It is important to factor in the history of these collections in order to appreciate their cohesiveness as a group, for example, in the case of the Indian Museum, Heras Institute and the Government Museum, Chandigarh. In other cases, collections were acquired or donated by private collectors, and though they often contain superb specimens aesthetically, they are somewhat disjointed as a group. Collecting thus raises geographical questions of source, of circulation and of consumption both on the part of collectors and/ or scholars. The housing of the collections in unique architectural settings, often in close proximity to other institutions of power and authority in nineteenth- and early twentieth-century India, was as important as the making of the collection and provides clues to the continued appeal of the collections. In the nineteenth century, the word Gandharan was used to describe finds coming from a series of culturally related areas beyond the Peshawar plains, such as the Swat valley, the Buner and Taxila regions, eastern Afghanistan and even parts of Kashmir. Gradually the architectural and cultural contexts of these finds were lost as the objects found their way to museums in the subcontinent, Europe and the United States and Gandhara came to be studied as one of the schools of art. The twentieth century saw the maturing of the two disciplines of art history and archaeology in the Indian subcontinent and the emergence of Gandhara and Mathura as the two major schools of art in the Kushan period (CE 40–360). In this consolidation of the discipline of art history, museum collections contributed in a big way to framing the history of ‘Indian’ art. Thus in early twentieth-century India and elsewhere, the purpose of the museum shifted from being an institution to house Gandharan artefacts to one that provided a foundation to the study of the art history of Gandhara in its various locations both chronologically and spatially. This chapter moves away from the art historical understanding of sculptures in museums to focus on the process through which these objects were acquired, which made the collections of the five museums discussed here distinctive. These atypical features, it is suggested, are ironed out when sculptures and other artefacts are considered out of context with no discussion on the mode of acquisition. The first issue that needs to be addressed is the extent to which museums were “initiated out of necessity by archaeological surveyors as early as the eighteenth century, as vessels to support and store the fruits of scholarly research”.10 As discussed in the next section, the antiquarian enterprise in the eighteenth and nineteenth centuries was more about acquiring marketable objects rather than about scholarly pursuits.11

236  Himanshu Prabha Ray This leads to a critical issue of the twenty-first century, that is, the ethics of collecting and the continued trafficking of antiquities leading to loss of heritage and a destruction of the patrimony of the world.12 It is important for scholars and researchers to appreciate and accept their responsibility by refraining from legitimizing de-contextualized works of art that continue to find their way into museums. The patron/ collector is by no means a symbol of the past, but continues to exert enormous power over the field of the visual arts and the career paths of several upcoming artists and researchers,13 as is evident from the history of the collections discussed in this chapter. We start with the nineteenth century and the beginnings of the history of museums in the subcontinent.

The Asiatic Society and the Indian Museum The Asiatic Society Museum provided the nucleus of the Indian Museum, Calcutta, which was founded as a public museum under the Act XVII of 1866. The location of the first public museum in the country at Calcutta was no surprise, because the city had served as the capital since 1773 when the East India Company had consolidated its rule in the region. An important location was chosen for the museum at the edge of the main square, the Maidan, and along the main thoroughfare (i.e. Chowringhee). The museum moved into its present premises on Chowringhee in 1875, and archaeology was one of the five sections, the others being zoological and ethnological, geological, art and industrial. Clearly its mandate was much larger than a study of the past. It was housed in an impressive building, which was designed by the architect Walter L. B. Granville (1819–1874) to emulate the British Museum in London established in 1753. Built at a cost of £140,000, the Indian Museum was located in an area that had several impressive buildings, such as that of the Geological Survey of India and the Government College of Art & Craft. Granville also designed the General Post Office and the High Court buildings. By 1914, the Indian Museum was the largest institution of its kind in the subcontinent and received the lion’s share of imperial funding. The Indian Museum’s collection reflected the vision of the members of the Asiatic Society and the results of their antiquarian searches. In this it was different from other contemporary museums, which focussed on industrial art. Interest in Gandhara started as early as 1833 when a portion of the stupa at Manikyala was deposited in the Asiatic Society Museum. Between 1848 and 1876, Alexander Cunningham, the director-general of the newly established Archaeological

Making of museum collections in India 237 Survey of India, explored sites such as Jamalgarhi, Sahri Bahlol, Takht-i-Bahi and Kharkai and sent the sculptures collected to the Indian Museum. In 1870, some stucco heads collected by J. C. Demrick from the stupa site at Shah-ji-ki-Dheri near Peshawar were sent to the Indian Museum. Dr. John Anderson and James Wood-Mason were in charge of the organization of the archaeological and zoological galleries, which opened to the public in 1878. The same year, Cunningham provided the blueprint for display of the archaeological collections in the Indian Museum. In 1886 the government of Punjab sent a few pieces to the museum, which had been excavated by Major H. H. Cole at Sanghao in 1883 from the sites of Koi Tangi and Nathu. A major part of the Gandharan sculptures (i.e. 414) come mainly from Loriyan Tangai in the Swat valley, collected by A. E. Caddy in 1895– 1896. ASI gave on permanent loan sculptures from the excavations at Taxila, Sahri Bahlol, Charsadda and Takht-i-Bahi, and a few objects were added through purchase and exchange.14 The Indian Museum Act of 18 March 1910 made the director-general of archaeology one of the ex-officio trustees of the museum vide section 2 (1)a and resulted in closer interaction between the Archaeological Survey of India and the Indian Museum.15 As a result of the then Director-General John Marshall’s (1876–1958) initiatives, a large number of sculptures from Gandhara belonging to the personal collection of Lord William Brailsford, who was once the secretary to the viceroy, were transferred to the museum from Shimla, where they were lying in the godown of the Viceregal Lodge. In addition ASI loaned for exhibition 165 specimens excavated from various sites of Taxila, such as Bhir mound, Sirkap, Chirtope, Mora Moradu and Jaulian and 71 Gandhara sculptures in stone and stucco excavated by the Superintending Archaeologist of the Frontier Circle from Sahri Bahlol and Takht-i-Bahi.16 Museum records show that the collection comes from Kabul, Jamalgarhi (Figure 9.2), Sahri Bahlol, Takht-i-Bahi, Kharkai, Shah-ji-ki-dheri, Koi Tangi, Nathu, Loriyan Tangai, Charsadda, Mardan, Sanghao, Rhode (near Sanghao), Mer Jan, Mian Khan, sites around Peshawar, Zinthara (Khyber Pass), Rawalpindi, Dheri Sohan, Yusufzai (Figure 9.3) and sites around Taxila such as Sirkap, Chir Tope and Jaulian.17 In 1921 Ramaprasad Chanda (1873–1942) was appointed superintendent in ASI and placed in charge of the Archaeological Section of the Indian Museum. Chanda had earlier been associated with the collections of the Varendra Research Society at Rajshahi in present Bangladesh. In March 1921 sixty-six antiquities from the godown of the Peshawar Museum were shifted to Calcutta, and in April 1922,

Figure 9.2 Indian Museum Kolkata no. G16 A23286. Relief showing consolation to Ananda from Jamalgarhi.

Figure 9.3 Indian Museum Kolkata no. 2377 A 23244 Rishyasringa Jataka from Yusufzai

Making of museum collections in India 239 Chanda travelled to Peshawar and acquired a number of Gandhara stucco heads from the sites of Sahri Bahlol and Takht-i-Bahi for the Indian Museum.18 Thus the 1,602 Gandharan objects in the Indian Museum acquired prior to 1923 were the result of archaeological excavations in the region and the close association with the ASI which made this possible. The largest number of Gandharan antiquities come from Loriyan Tangai (414) followed by those from Taxila (178), Jamalgarhi (153) and Yusufzai area (103). A majority of the sculptures (713) remain without provenance.19 At present only 75 of these sculptures are on display, though Anderson’s 1883 Catalogue refers to 171 Gandharan sculptures on display in the two galleries labelled ‘Asokan’ and ‘Indo-Scythian’. Clearly during the nineteenth century, Gandharan sculptures enjoyed a pre-eminent position, which has been somewhat reduced as the museum acquired other objects subsequently.20 Thus the Indian Museum is one of the few museums that contain sculptures acquired directly from archaeological sites, rather than through the mediation of the art market. In this it bears similarities to the Heras Institute Museum in Mumbai, though there are also several differences between the two, the most conspicuous being that the latter is a private museum and is much smaller in scale.

Heras Institute Museum in Mumbai The Museum of the Heras Institute of Indian History and Culture is a private collection acquired by Father Henry Heras S. J. (1888– 1955) who founded the Indian Historical Research Institute in 1926, which was renamed Heras Institute after his death. Born in Barcelona, Spain, Father Heras came to India in 1922 to teach Indian history at St. Xavier’s College in Bombay, which had been founded in 1869 by German Jesuits. Like the library, the museum was set up to provide reference material for teaching and research. The largest single category of objects in the museum comprises Buddhist sculpture from sites in Quetta collected by Father Heras. The museum boasts of a total of thirty-six pieces, which includes eighteen heads of stucco and stone and eighteen stone sculptures (Table 9.1). The sculptures have been dated from first to fourth century CE, and the collection comprises hand-picked panels depicting scenes from the life of the Buddha, as well as some beautiful stucco heads and Buddha and Bodhisattva images. A prized relief in the collection represents Ambapali (Amrapali), the well-known courtesan of Vaishali, donating a mango grove to the Buddha (Figure 9.4). A striking feature of the collection is the number of panels showing monks as

240  Himanshu Prabha Ray Table 9.1  Table of eighteen stone sculptures in the Heras Institute Museum 30.22

Standing moustached Bodhisattva wearing sandals and a jewelled turban; arm broken below the elbows. The pedestal depicts a seated Buddha in dhyāna mudra in the centre flanked by two devotees with folded hands.

30.23

Standing Bodhisattva with bejewelled turban, necklace and partly broken halo. The arms and feet of the image are damaged.

30.24

Standing Bodhisattva under arch. The lower part of the figure is broken.

30.25

Standing Bodhisattva under arch with hand on hip; lower part of the panel is damaged.

30.27

Broken panel showing arch with standing Buddha and two female devotees.

30.29

Standing Buddha with halo behind the head and feet and arms damaged.

30.30

Damaged image of seated Buddha with two devotees on his left, one of them a monk.

30.32

Seated Buddha image with left hand holding the edge of the robe, and the right arm is broken.

30.35

Architectural fragment in two registers with seated Buddha in dhyāna mudra with devotees; he is flanked by two bhikkhus in the lower register.

30.36

Panel showing two scenes. The one on the left shows standing Buddha and Vajrapani with five of Devadatta’s assassins, one of them holding a stick. The partly broken right panel shows seated Buddha in preaching mudra with two devotees, including one monk.

30.37

Panel showing Amrapali’s gift to the Buddha. The left side panel shows seated Buddha in abhaya mudra surrounded by eight devotees, including Vajrapani, and the right panel depicts the seated Buddha in abhaya mudra under a tree, with two male figures in the background, perhaps with flowers in their hands. On his left stands Vajrapani with two male devotees and on his right Amprapalli is shown with her head covered and holding a pot accompanied by a female attendant with folded hands.

30.38

Fragmentary panel showing Buddha under an arch on the left hand panel with two bhikkhus with folded hands facing him on his left.

30.39

Three seated Buddha images in dhyāna mudra in a three-arched pavilion; each on a lotus.

30.42

Seated Buddha in abhaya mudrā under a tree on an asana marked with dharmacakra and two deer. Three devotees stand around him, of whom two are bhikkhus. (Continued)

Making of museum collections in India 241 Table 9.1 (Continued) 30.43

Fragment showing veneration of the Buddha in abhaya mudra seated under a tree flanked by two standing devotees. The scene to the left, separated by a pillar is damaged.

30.50

Buddha seated on lotus under a bejewelled tree in dhyāna-mudrā accompanied by two standing Bodhisattvas, one with a water bottle. At the base are two seated devotees with folded hands.

30.51

The Miracle of Sravasti depicted in two registers. The top register shows seven standing Buddha images with the right hand in abhaya mudra. The lower register has the seated Buddha in the centre in abhaya mudra surrounded by six devotees, four of them being monks.

30.71

Circular base of a stupa sculpted with four scenes, each separated from the other by pillars. Three of the panels show seated Buddha flanked by two devotees, while the fourth has two monks showing surprise.

Source: Prepared by the author

Figure 9.4 Heras Institute collection no. 30.37. Amrapali’s gift to the Buddha.

devotees (Figure 9.5). The focus of the display relates to the life of the Buddha and highlights scenes associated with the Master. The architectural context of the images or the cultural landscapes within which the monasteries were located find little mention.

242  Himanshu Prabha Ray

Figure 9.5 Heras Institute collection no. 30.43. Buddha in congregation.

Thus the Heras Institute Museum provides a contrast to the Indian Museum, both in terms of its small numbers and the care with which the reliefs have been selected. This brings me to another museum in Mumbai where the focus shifts to collectors who have played an important role in building up the museum. Chhatrapati Shivaji Maharaj Vastu Sangrahalaya (CSMVS) or the former Prince of Wales Museum of western India The Museum was founded in the early twentieth century and, unlike the Indian Museum, was the result of the impetus provided by prominent citizens of Bombay to erect a memorial to the visit of the Prince of Wales (later King George V) to the city. The Foundation Stone of the Museum was laid by the prince on 11 November 1905, and the museum was named Prince of Wales Museum of Western India. It was established at the southern tip of the island and completed in 1914, but was not opened to the public until 1922. The Scottish architect George Wittet (1878–1926) was chosen after an open competition in 1909. Wittet is known for the Indo-Saracenic style of architecture of which this museum is one of the best examples (Figure 9.6). He also

Figure 9.6 The imposing building that houses the collections of Chhatrapti Shivaji Maharaj Vastu Sangrahalaya, Mumbai

244  Himanshu Prabha Ray designed the Gateway of India, an imposing structure at the water’s edge of the harbour located close to the museum. The dome of the museum building was designed after the Gol Gumbaz of Bijapur, and the finial was copied from the Taj at Agra. The nucleus of the sculpture collection of the museum was formed either by gifts or by permanent loans from various institutions such as the Asiatic Society of Bombay, the defunct Archaeological Museum, Poona, the Archaeological Survey of India and the Victoria and Albert Museum, Bombay. At present, CSMVS has a total of 198 Gandharan sculptures of which 186 are of stone and 8 of stucco plus 12 sculptures from the Karl J. Khandalavala collection. Several other well-known collectors also contributed to CSMVS, such as Seth Purushottam Mavji, Sir Akbar Hydari collection and the Karl and Meherbai Khandalavala Trust. In his catalogue, Moti Chandra refers to the lopsided development of the collections as a result of reliance on gifts, resulting in an overemphasis on one school such as Gandhara and the neglect of others.21 Two exquisite pieces of Gandharan art were gifted by Karl J. Khandalavala and include a narrative panel showing the departure of Prince Siddhartha from the palace (Figure 9.7), a standing Bodhisattva Maitreya and a kneeling adolescent devotee. The focus of the sculptures displayed in the Sculpture Gallery at CSMVS is largely regional, including the ninth Asokan edict from Sopara, a yaksha from the cave temple at Pitalkhora and so on. A standing image of Shiva from the Baijanath Mahadeva temple at Parel, about 12 kilometres north of the museum, occupies pride of place. The image was in active worship until recently. Stylistically, it is similar to the sculptures at the caves at Elephanta and belongs to the same period (i.e. fifth to sixth centuries CE). A much more imposing 3.48-metre-high monolithic bas relief of Shiva continues in worship at Parel; one of the original seven islands of Bombay and a replica has been displayed in the museum. The in situ image was declared a nationally protected monument in 1985 and a small temple constructed to house it. Clearly the shift of images from living sites to museums is a complex process and transformations of identity/ies defy categorization. Perhaps one of the only archaeological sites from which the antiquities reached the museum was that of Mirpur Khas about 68 kilometres east of Hyderabad, Sindh (now in Pakistan). The site was known since 1859, and some of the antiquities found were deposited in the Karachi museum. The Buddhist stupa at Kahujodaro was part of a ruined Buddhist establishment situated in one of the large mounds extending over

Making of museum collections in India 245

Figure 9.7 CSMVS departure of Prince Siddhartha

12 hectares of land to the north of the present town of Mirpur Khas, the district headquarters of Thar and Parkar in the Sindh province of Pakistan. The stupa structure, made of sun-dried bricks, predates the decorative terracottas and bricks, assigned to the fifth century CE, and is perhaps coeval with the Gandharan sites of the Peshawar and Swat valleys. Henry Cousens (1854–1933) excavated the site in 1909–1910 and dug a 3-metre well down the middle of the mound through the hard sun-dried brickwork.22 At a depth of 7.6 metres, the excavator found a chamber with a stone coffer. Within this cavity in the lower stone was a small crystal bottle (which contained ash), while around it were sprinkled a number of offerings consisting of coral beads (63), crystal beads, drilled and undrilled (7), two small crystals cut to simulate diamonds, very small seed

246  Himanshu Prabha Ray pearls (30), four gold beads, ten copper coins, some lumps of charcoal, a few grains of wheat and some other small odd beads and chips.23 In 1919 the terracotta images and bricks excavated from the Buddhist stupa of Mirpur Khas were given to the museum by its excavator, Henry Cousens (Figure 9.8). The images had all originally been painted, those of the Buddha having red robes and a golden-coloured

Figure 9.8 CSMVS Buddha image from Mirpur Khas

Making of museum collections in India 247 complexion with black eyes and hair. This is also perhaps the only museum in India which contains objects from the archaeological site of Mirpur Khas. Another unique collection is that of fourth- to sixthcentury terracotta heads from the Buddhist monastic site of Akhnoor, about 30 km from Jammu, which is said to be influenced by the art of Gandhara. Thus unlike the display at the Indian Museum, Kolkata, which is framed by the Buddhist art of Bharhut, Mathura and Nalanda, CSMVS presents regional linkages between Gandhara, Kashmir and Sindh. A third strategy evident in museum collections in India is acquisition through redistribution, a good example being that of Chandigarh Museum established in the post-Independence period, which received its collections from the Lahore Museum after the division of antiquities at the time of the Partition of the subcontinent in August 1947. The Government Museum and Art Gallery, Chandigarh Lahore Central Museum, popularly known as Anarkali Museum, had existed since 1856, and the collection, as with other local museums such as at Peshawar, seems to have evolved from chance finds and donations. The earliest Gandhara accession listed at Lahore in 1867 is a carved stone figure from Shah-ji-ki-dheri donated by Major C. H. Hall.24 In 1869 the museum received thirty-four pieces from the site of Rokhri in the Mianwali district largely due to the efforts of Alexander Cunningham. Six of these sculptures were transferred in 1947 to Chandigarh Museum.25 Under John Lockwood Kipling’s curatorial direction from 1875 to 1893, the size and scope of the Lahore Museum expanded dramatically. Kipling, who was also principal of the Mayo School of Art, established connections with numerous archaeologists and government officials whom he convinced to donate or lend significant finds to the museum’s collection.26 Henry Hardy Cole collaborated with Kipling, who sent many of his art school graduates to work with Cole. In return, Cole supplied artefacts and plaster casts from his archaeological surveys in the Punjab to Lahore Museum. In 1909 the Lahore Museum acquired a number of Gandhara sculptures from the private collection of Major Bamandas Basu, a medical practitioner and literary figure. Basu had gathered these sculptures when he was a civil surgeon at Peshawar from the site of Hoti Mardan in Yusufzai Province.27 Of the 962 pieces in the Lahore Museum, 627 pieces were transferred to East Punjab before Partition and are now in the Chandigarh Museum.28 Even though the collection of Lahore Museum had been

248  Himanshu Prabha Ray divided between the new nation-states of India and Pakistan, the question of the capital of the now-truncated state of Punjab had still to be resolved. In March 1948 the government of Punjab, in consultation with the government of India, approved the area at the foothills of the Shivaliks as the site for the new capital, which the French architect Le Corbusier was entrusted to design. One of the buildings that the French architect planned for the new city was that of the Government Museum and Art Gallery. Situated close to the city centre, the museum has a sprawling and extensive campus at one side of which is located the Government College of Art. The museum was inaugurated on 6 May 1968 under the initiative and active support of Late Dr. M. S. Randhawa, renowned connoisseur and patron of art and the then chief commissioner of Chandigarh. Nearly three decades after the opening of the museum, a colloquium on Gandhara Art was organized in March 1998 in which Dr. Saifur Rahman Dar, former director of Lahore Museum, also participated. Issues of chronology, identification of sculptures and a system of classification were discussed. These deliberations resulted in the publication of a catalogue of sculptures of the museum, along with some of the papers that were presented, though the history of the collections does not find detailed discussion. The provenance of 406 of the total of 627 sculptures in the museum at Chandigarh is not available. The remaining sculptures come from the following sites, with a large number of images from Sikrai or Sikri29: Attock/ Manikyala (nos. 276–277, 344, 348, 352) Charsadda (nos. 343, 345, 353–358, 613) Yusufzai (nos. 278–280, 283, 288, 292–293, 295–301, 304–306, 308, 314–5, 319, 325–327, 329–331) • Jamalgarhi (nos. 44, 175, 182, 183, 185, 187, 189, 190–194, 200, 209, 225–227, 230–233, 239, 241–243, 248, 262–264, 266) • Karamar Hill (nos. 346–347, 349) • Mian Khan (nos. 322) • Muhammad Nari (nos. 340–342, 505; Fig. 10.6) • Naogram (no. 54) • Nathu, Lower Monastery (nos. 282, 285, 289, -290, 294, 303, 307, 311, 313, 318, 323–324, 337, 416) • Rokhri (nos. 48–50) • Sahri Bahlol (no. 38) • Sanghao (281, 284, 286, 291, 310, 316–7, 320–1, 332–336, 338–339) • Sikri/ Sikrai (nos. 514–608) • • •

Making of museum collections in India 249 • •

Skarah Dheri (nos. 432) Takht-i-Bahi (no. 221)

Thus with several unusual pieces, the Government Museum and Art Gallery in Chandigarh boasts one of the richest collections in India, though in terms of numbers it comes closer to the collection of the National Museum in New Delhi, which I will discuss in the next section. Here it may be appropriate to refer to the other important collections in the museum that come from the Buddhist site of Sanghol, 40 km from Chandigarh on the highway to Ludhiana. Until 1948 Sanghol formed a part of the former Princely State of Patiala and was transferred to Ludhiana District on 25 January 1950. In 1964 the Punjab Legislature passed the Punjab Ancient and Historical Monuments and Archaeological Sites and Remains Act, and around the same time the state established an archaeological cell under the Director of Archives and Curator of Museums. It was this newly established cell that started excavations at Sanghol under the charge of R. S. Bisht on 20 December 1968. G. B. Sharma, a local resident who

Figure 9.9 Stucco Buddha head from Sanghol in the site museum

250  Himanshu Prabha Ray

Figure 9.10 Railing pillar from Sanghol in the site museum

later joined the State Department, continued the work, though with gaps, until 1985.30 Given the limited resources of the State Department, both in terms of finances and trained personnel, the excavations were restricted in nature. It is, however, to the credit of the excavators working under difficult conditions that they discovered the stone railings of the stupa buried neatly in its vicinity. This discovery brought Sanghol back into the limelight and drew the attention of institutions

Making of museum collections in India 251 of the government of India, such as the National Museum and the Archaeological Survey of India. An exhibition of selected pieces of railing pillars from Sanghol was arranged in the National Museum, New Delhi, which was inaugurated by the President of India, and a catalogue of Kushan sculptures was published jointly by the Department of Cultural Affairs, Punjab, with the National Museum.31 Large-scale excavations were conducted at Sanghol jointly between the Archaeological Survey of India and the Department of Cultural Affairs, Archaeology and Museums, Punjab, over at least four seasons until 1990. In 1990 the Department of Cultural Affairs, Archaeology and Museums, Punjab, established a site museum to display the rich archaeological heritage of Sanghol, including several from the almost 15,000 antiquities from the site (Figures 9.9 and 9.10). In addition some of the railing pillars continued to be displayed at the Government Museum and Art Gallery, Chandigarh. How do these displays of Gandhara sculptures compare with those at the National Museum, New Delhi?

National Museum, Delhi The proposal for a museum in Delhi was first mooted when the capital was shifted from Calcutta to Delhi in 1925.32 At that time the suggestion was to set up an Ethnological Museum in Delhi, though this had to be shelved due to lack of funds. In 1936 a five-year plan for the development of Delhi was contemplated, and the requirement of a central museum worthy of the capital found mention in it. There was, however, no unanimity on the site to be selected for the purpose.33 Mortimer Wheeler, director-general of the Archaeological Survey of India, strongly supported the setting up of a national museum and in 1945 drew up a detailed proposal for the purpose.34 In his frame of reference the proposed museum, termed Central Indian Museum of Art, Archaeology and Anthropology, was to be an institution “where the public and the student can obtain anything approaching a general conspectus of the development of Indian civilisation during the many thousands of years in which man has inhabited the subcontinent”. The memorandum relating to the proposal for a central museum was discussed at the interdepartmental meeting of the department of Planning and Development on 10 September 1945. Though the meeting approved of the scheme in principle, the additional secretary of the Finance Department was doubtful about the priority that could be accorded to the scheme.35 Thus the government decided to appoint a small committee to provide the details for the establishment of such

252  Himanshu Prabha Ray a museum with special regard to its functions and powers, its general administration, its internal organization, its site and its building (including the possibility of progressive construction). The committee was composed of Sir Maurice Gwyer, Sir John Sargent, H.A.N. Medd, V. Narahari Rao, H.V.R. Iyengar, G. S. Bozman, R.E.M. Wheeler, S. N. Sen, B. S. Guha, D. M. Sen and Dr. N. P. Chakravarti, OBE, Secretary.36 Thus the National Museum was initiated under the aegis of the Archaeological Survey of India, but there was still a long way to go before these recommendations could be implemented and the museum became a reality. An exhibition was organized in 1947 at Burlington House, London, to display masterpieces collected from state and archaeological museums all over the subcontinent. On their return a year later, these objects were housed in Rashtrapati Bhavan and subsequently constituted the nucleus of the National Museum. It is significant that Gandharan sculptures formed a part of the London exhibition, which was later displayed with great fanfare in New Delhi and “orchestrated as a great state event” and is said to have marked the transition from colonial to national status.37 The museum continued to be in the charge of the ASI until 1958 when it became an independent institution under the control of the government of India. A search committee was set up, which led to the appointment of Dr. Grace Morley as the first director of the National Museum, New Delhi, she having earlier worked as head of the Museums Division of UNESCO. She joined the museum on 8 August 1960 and continued to hold its charge for six more years. When she joined the National Museum, most of the objects were in the Durbar Hall of the Rashtrapati Bhavan, and there were hardly any galleries installed in its new building at Janpath. But she gave the Indian government an assurance that the museum would be opened to the public during the month of December (i.e., within four months after her joining the museum) and kept that promise. The core of the National Museum’s early collections resulted from purchases, as well as gifts by wealthy Indian families who had amassed or inherited Indian objects at a time when European items dominated the definitions of aesthetics and Indian objects were barely admired for their artistic value. As the National Museum’s collecting committee was composed of seasoned connoisseurs and scholars, including Moti Chandra, Karl Khandalavala and Rai Krishnadasa, it seems likely that the random nature of their collecting was in part due to this practice of purchasing near-complete private collections. The Ministry of Education proposed that letters of request specifically be sent to

Making of museum collections in India 253 the maharajas of Bikaner, Udaipur, Jaipur, Mysore, Rampur, Jodhpur, Gwalior and Hyderabad in order to secure their family treasures for the National Museum.38 Imre Schwaiger (1868–1940) was, in his age, a well-known collector and dealer of art. He was based in London and Delhi between the two world wars. He was among the first to collect Nepalese art and organized the first exhibition of Nepalese bronzes in Calcutta in 1912. Many important pieces of the Victoria and Albert Museum, particularly of its Mughal collection, but also four pieces of Gandhara art, came from him. The first few objects to be acquired for the National Museum were purchased from Schwaiger, and subsequently the museum received gifts, permanent loans from the Archaeological Survey of India and objects in exchange until 1987. No further additions were made to the Museum Collection after this date. The National Museum has a total of 688 objects from Gandhara, including 101 objects in stone, 160 in metal and ivory, 162 in stucco and 91 in terracotta. The single largest category of 174 objects comprises jewellery, coins, shell and glass; a large number of these were recovered from Marshall’s excavations at Taxila. There are twentyfive narrative reliefs in stone from the life of Shakyamuni Buddha in the National Museum collection, and other stone sculpture includes images/heads of the Buddha and Bodhisattvas. Notable objects in the collection include a relic casket from Taxila, a silver scroll inscription and broken halo of an image with an inscription. The display in the sculpture galleries in the National Museum is chronologically arranged, and the guiding principle is the dynastic history of ancient India. The historical period starts with the Maurya and Sunga rules presented through stone and terracotta pieces and then moving to the beginning of the Common Era. The next gallery presents sculpture from three different art schools: Gandhara, Mathura and Amaravati from three different regions of the subcontinent. The regions of Gandhara and Mathura represent the two centres of the Kushan Empire where two different schools of art flourished, though both were inspired by Buddhism. Gandhara produced images of the Buddha and narrative sculptures in black schist, whereas the material used in Mathura was red sandstone. In her book, Grace Morley39 highlights exchanges of sculptural traditions between the two schools of art, and this is also a theme dealt with by other art historians.40 In the introduction mention was made of stone images from Mathura being found in the region of Gandhara or present-day Pakistan. Clearly stone images travelled long distances in the past and linked diverse centres in networks of ideas and knowledge exchange.

254  Himanshu Prabha Ray In addition to the major collections discussed earlier, there are others which could be brought into the discussion, such as the Bharat Kala Bhavan of Banaras Hindu University. The Gandhara collection of Bharat Kala Bhavan includes 31 stone sculptures of schist and 23 stucco heads. The provenance is not known, as most of the sculptures were acquired through transfer after Independence.41 In addition, the State Museum at Lucknow, the Archaeological Museum at Mathura and the Allahabad Museum at Allahabad have ninety-four, forty, and twenty-five objects of Gandharan art, respectively.42 As is evident from the brief survey presented here, museums in India hold substantial collections of Gandharan art, including both stone and stucco specimens. This raises the issue of the framework adopted by respective museums for framing and presenting the sculptures over time. A question often asked is: How did museum objects that were collected and categorically assigned to the imperial canons of Indian art history shift in meaning to assume a national significance associated with cultural pride, heritage and modernity?43 Perhaps the earliest catalogue that provides insights into the issue is John Anderson’s 1883 catalogue of the Indian Museum, Calcutta, which provides the first attempts at classifying sculptures chronologically. John Anderson (1833–1900) was a Scottish anatomist and zoologist who worked as the superintendent of the Indian Museum from 1865 to 1887. The catalogue of the archaeological collections was published by the trustees, and Anderson expresses his debt to the work of Alexander Cunningham in preparing the catalogue. The division and display of the collection into four galleries were based on Cunningham’s recommendation and included the following: Asoka, Indo-Scythian, Gupta and Mahomedan. The sculptures in the Indo-Scythian gallery dated to the first three centuries of the Common Era and included those from Mathura, Amaravati and Gandhara. The Gandhara images were collected from the Yusufzai area of the Punjab from the sites of Sahri Bahlol, Takht-i- Bahi, Kharkai and Jamalgarhi. Anderson ascribes them to the Indo-Scythian period based on Cunningham’s account assigning them to the period of the rule of Kanishka and his successors from 40 BCE to 200 CE.44 Anderson points out the influence of Greek art on them and records that many of them would have been covered with gold leaf. How is this affiliation to the patronage of the Kushans and the impact of Greek influence reflected in post-Independence writings? To what extent did art historians or ‘nationalist ideologues’, as termed by Kavita Singh, curator of the 2016 symposium under the Aesthetics Project, reject Gandhara art as “being the product of ‘Western influence’ and therefore extraneous to Indian tradition”?45 The answer

Making of museum collections in India 255 is evident in the writings of Calambur Sivaramamurti (1909–1983), museologist and art historian, who started his career as superintendent, Archaeological Section, Indian Museum and then shifted to the National Museum, New Delhi, under Grace Morley. Subsequently he took charge as director of the National Museum. Sivaramamurti started his account of the art of Gandhara as follows: “North Western India has been the meeting place of many cultures, and the influence of the Greeks in the wake of Alexander’s invasion has left a permanent impress on the sculpture in this area.”46 This clearly does not support Kavita Singh’s contention of Gandhara being relegated to ‘nowhere art’. Similarly unsubstantiated is her statement that the National Museum, New Delhi’s, narrative on Gandharan sculpture is aligned “firmly with the nationalists.”47 On the contrary, one could also argue that the arrangement of Gandharan sculptures in the Kushan gallery is a continuation of the chronological framework first adopted by Anderson in the colonial period at the Indian Museum, Calcutta.

Final notes The organizing principle in most museum collections discussed in this chapter is either dynastic history of the subcontinent or a chronological framework for the different schools of Indian art. The usual emphasis is on identification of the sculptures and presentation of scenes from the life of the Buddha. There is no master template that all museums follow; instead, Gandharan sculptures are framed within early historic images from Buddhist sites as available in the museum’s collection. Thus at the Indian Museum it is Bharhut and Mathura, whereas at CSMVS sculptures from Mirpur Khas and Akhnoor are prominent. Almost absent is a contextualization of the sculptures in their architectural setting or any discussion on how the collections were made, the only exception being CSMVS where Gandhara sculptures are displayed in a gallery dedicated to Karl J. Khandalavala. Except perhaps for the National Museum, coins are rarely mentioned, though it is evident that these were placed in stupa deposits.48 The emphasis continues to be on style and form rather than on the cultural and social context of the sculptures. Peter Skilling has convincingly shown that images of the Buddha are not merely ‘art objects’; they are products of complex ideologies. “Across Asia, images multiplied and played multiple roles – bringing rain, warding off disease, offering protection and victory in war, and acting as tribute in diplomatic missions”.49 How does archaeology help unravel these multiple functions? Archaeology as it has developed over

256  Himanshu Prabha Ray the last three decades helps define a context for both relic and image worship. It maps Buddhist sacred spaces both horizontally across the physical and natural landscape, but also vertically in time, highlighting antecedents of religious sites and subsequent transformations. A good example is the archaeological work being done in Swat and published elsewhere in this book.

Notes 1 This is a part of a trend in art history of the region as evident from an earlier exhibition curated by Christian Luczanits on “Gandhara – the Buddhist Heritage of Pakistan. Legends, Monasteries and Paradise” at the Kunst- und Ausstellungshalle der Bundesrepublik Deutschland in Bonn together with Michael Jansen and was responsible for its catalogue (2008). More recently, the Asia Society, New York curated The Buddhist Heritage of Pakistan: Art of Gandhara from 9 August 2011 to 30 October 2011. 2 Michael Meister, Temples of the Indus: Studies in the Hindu Architecture of Ancient Pakistan, Brill, Leiden, 2010. 3 Arjun Appadorai and Carol Breckenridge, Museums are Good to Think: Heritage on View in India, Saloni Mathur and Kavita Singh edited, No Touching, No Spitting, No Praying: The Museum in South Asia, Routledge, New Delhi, 2015: 174. 4 Jyotindra Jain, Museum and Museum-like Structures: The Politics of Exhibition and Nationalism in India, Exhibitionist, Spring 2011: 50–55. 5 Richard Davis, Lives of Indian images, Princeton University Press, New Jersey, 1997: 6. 6 www.theaestheticsproject.com/events-january-2016.php (accessed on 15th June 2016). 7 “Although Gandhara and Bactria today lie outside India’s borders in Pakistan, Afghanistan and Tajikistan, the region lies at the heart of periods central to Indian history, for it was the birthplace of the Mauryan dynasty and the capital of the Kushanas. Colonial interest in ‘Greco-Buddhist’ art prompted the founding of the Archaeological Survey of India and several museums in India have rich collections of art from the region. Yet these have fallen into oblivion: if some groups in Pakistan and Afghanistan have recently disavowed Gandharan and Bactrian art for being non-Islamic, in India these were rejected decades before by nationalist ideologues, for being the product of ‘Western influence’ and therefore extraneous to Indian tradition. Intolerance can take many forms, and this art seems to have suffered all of them in recent times. An intensely cosmopolitan expression that once fused elements taken from everywhere, Gandharan and Bactrian art, it seems, has now become the art of nowhere. It is hoped that this symposium will encourage us to look at the again, and with new eyes.” www.theaestheticsproject.com/events-january-2016.php (accessed on 15th June 2016). 8 Saloni Mathur and Kavita Singh edited, No Touching, No Spitting, No Praying: The Museum in South Asia, Routledge, London, New York and New Delhi, 2015.

Making of museum collections in India 257 9 Kurt A. Behrendt, The Art of Gandhara in the Metropolitan Museum of Art, Yale University Press, New Haven and London, 2007: 19. 10 Kristy K. Phillips, The National Museum of India: A Museum to and of the Nation, www.ep.liu.se/ecp/022/009/ecp072209.pdf (accessed on 30th January 2016). 11 Daniel Michon, Archaeology and Religion in Early Northwest India: History, Theory, Practice, Routledge, New Delhi, 2015: 13–46. 12 Neil Brodie, Jennifer Doole and Colin Renfrew (eds.), Trade in Illicit Antiquities: The Destruction of the World’s Archaeological Heritage, McDonald Institute Monographs, Cambridge, 2001: 4: “It will be difficult to stem looting until people understand the importance of context, and that their heritage is worth more in every sense than the usually paltry sums that it can generate in the short-term when sold to middlemen.” 13 Tom Phillips Review of Philip Blom, To Have and To Hold: An Intimate History of Collectors and Collecting, Allen Lane, London, 2002, www. theguardian.com/books/2002/aug/03/highereducation.news1 (accessed on 1st February 2016). 14 R.C. Sharma, Introduction, Anasua Sengupta and Dibakar Das, Gandhara Holding in the Indian Museum (A Handlist), Indian Museum, Calcutta, 1991. 15 Archaeological Survey of India – Annual Report 1909–1910, Archaeological Survey of India, New Delhi, 2002 (first published 1914): 9. 16 Archaeological Survey of India – Annual Report 1922–1923, Archaeological Survey of India, New Delhi, 2003 (first published 1924): 140–141. 17 R.C. Sharma, Introduction, Anasua Sengupta and Dibakar Das, Gandhara Holding in the Indian Museum (A Handlist), Indian Museum, Calcutta, 1991. 18 Archaeological Survey of India – Annual Report 1921–1922, Archaeological Survey of India, New Delhi, 2002 (first published 1923): 110. 19 R.C. Sharma, Introduction, Anasua Sengupta and Dibakar Das, Gandhara Holding in the Indian Museum (A Handlist), Indian Museum, Calcutta, 1991: Appendix IV. 20 John Anderson, Catalogue and Handbook of the Archaeological Collections in the Indian Museum, part 1, Calcutta, 1883: Asoka and IndoScythian Galleries. 21 Moti Chandra, Stone Sculpture in the Prince of Wales Museum, The Board of Trustees of the Prince of Wales Museum of Western India, Bombay, 1974: IX. 22 Henry Cousens, Buddhist Stupa at Mirpur Khas, Sind, Archaeological Survey of India Annual Report 1909–1910, Archaeological Survey of India, New Delhi, 2002 (first published 1914): 80–92. 23 Cousens, Buddhist Stupa at Mirpur Khas: 84. 24 Errington The Western Discovery of the Art of Gandhara: 184. 25 Carrie Ann LaPorte, Displaying Empire? The Architecture and Development of Museums in Nineteenth Century India, Ph.D. Thesis, University of Pennsylvania, 2003: 248. 26 LaPorte, Displaying Empire?, 2003: 248. 27 Madhuparna Roychowdhury, Displaying India’s Heritage: Archaeology and the Museum Movement in Colonial India, Orient Blackswan, New Delhi, 2015: 126.

258  Himanshu Prabha Ray 28 Elizabeth Errington, The Western Discovery of the Art of Gandhara and the Finds of Jamalgarhi, Ph.D. Thesis, SOAS, London, 1987: 206. The Government Museum and Art Gallery, Chandigarh, India: A Handy Guide, Chandigarh, 2003: 4. 29 D.C. Bhattacharyya (ed.), Gandhara Sculpture in the Government Museum and Art Gallery, Government Museum and Art Gallery, Chandigarh, 2002. 30 G.B. Sharma, Coins, Seals and Sealings from Sanghol, Chandigarh: Department of Cultural Affairs, Archaeology and Museums, Punjab 1986. 31 S.P. Gupta edited, Kushana Sculptures from Sanghol (1st-2nd century AD): A Recent Discovery, New Delhi: National Museum 1985. 32 Himanshu Prabha Ray, Colonial Archaeology in South Asia (1944–1948): The Legacy of Sir Mortimer Wheeler in India, Oxford University Press, New Delhi, 2007: 122–30. 33 Archaeological Survey of India File no. 25/25/44–1944. 34 Archaeological Survey of India File no. 25/4/45: Archaeological Survey of India, Memorandum no. I on the proposal for a Central Indian Museum of Art, Archaeology and Anthropology dated 15 January 1945 issued by REM Wheeler. 35 Archaeological Survey of India file no. File no. 25/4/45. 36 The British Library MSS Eur D1224/45: Central National Museum of Art, Archaeology and Anthropology (Report of the Gwyer Committee). 37 Kristy K. Phillips, The National Museum of India: A Museum to and of the Nation, www.ep.liu.se/ecp/022/009/ecp072209.pdf (accessed on 30th January 2016), 101. 38 Kristy K. Phillips, The National Museum of India: A Museum to and of the Nation, www.ep.liu.se/ecp/022/009/ecp072209.pdf (accessed on 30th January 2016): 105–107. 39 Grace Morley, Indian Sculpture; Foreword Kapila Vatsyayan, Roli Books, New Delhi, 2005: 25. 40 J. E. Van Lohuizen-de Leeuw, Gandhara and Mathura: Their Cultural Relationship, Pratapaditya Pal edited, Aspects of Indian Art, E. J. Brill, Leiden, 1972: 27–43. 41 Shashi Asthana and Bhogendra Jha, Gandhara Sculptures: Bharat Kala Bhavan, Banaras Hindu University, Varanasi, 1999. 42 N. P. Joshi and R. C. Sharma, Catalogue of Gandhara Sculptures in the State Museum, Lucknow, The State Museum, Lucknow, 1969: 48. 43 Kristy K. Phillips, The National Museum of India: A Museum to and of the Nation, www.ep.liu.se/ecp/022/009/ecp072209.pdf (accessed on 30th January 2016): 95. 44 John Anderson, Catalogue and Handbook of the Archaeological Collections in the Indian Museum, Trustees, Calcutta, 1883, part 1: 199. 45 Quoted at the beginning of this paper. 46 C. Sivaramamurti, A Guide to the Archaeological Galleries of the Indian Museum, Indian Museum, Calcutta, 1954: 6. 47 Kavita Singh, The Museum is National, Saloni Mathur and Kavita Singh edited, No Touching, No Spitting, No Praying: The Museum in South Asia, Routledge, New Delhi, 2015: 114. [107–131]. 48 Elizabeth Errington, Gandhara Stupa Deposits, Arts of Asia, volume 28, no. 2, March-April, 1998:

Making of museum collections in India 259 49 Peter Skilling, Buddhism and the Circulation of Ritual in Early Peninsular Southeast Asia, Pierre-Yves Manguin, A. Mani and Geoff Wade edited, Early Interactions between South and Southeast Asia: Reflections on Cross-Cultural Exchange, Institute of Southeast Asian Studies, Singapore, 2011: 370.

References Anderson, J. 1883. Catalogue and Handbook of the Archaeological Collections in the Indian Museum, part 1. Calcutta: Board of Trustees. Appadorai, A., and C. Breckenridge. 2015. Museums are Good to Think: Heritage on View in India, in M. Saloni and S. Kavita (ed.), No Touching, No Spitting, No Praying: The Museum in South Asia. New Delhi: Routledge, pp. 173–183. Archaeological Survey of India – Annual Report 1909–1910. 2002. New Delhi: Archaeological Survey of India (first published 1914). Archaeological Survey of India – Annual Report 1921–1922. 2002. New Delhi: Archaeological Survey of India (first published 1923). Archaeological Survey of India – Annual Report 1922–1923. 2003. New Delhi: Archaeological Survey of India (first published 1924). Asthana, S., and J. Bhogendra. 1999. Gandhara Sculptures: Bharat Kala Bhavan. Varanasi: Banaras Hindu University. Behrendt, Kurt A. 2007. The Art of Gandhara in the Metropolitan Museum of Art. New Haven and London: Yale University Press. Bhattacharyya, D.C., ed. 2002. Gandhara Sculpture in the Government Museum and Art Gallery. Chandigarh: Government Museum and Art Gallery. Blom, P. 2002. Phillips Tom Review of Philip Blom, To Have and To Hold: An Intimate History of Collectors and Collecting. London: Allen Lane. www. theguardian.com/books/2002/aug/03/highereducation.news1/ accessed on 1 February 2016. Brodie, N., D. Jennifer, and R. Colin, eds. 2001. Trade in Illicit Antiquities: The Destruction of the World’s Archaeological Heritage. Cambridge: McDonald Institute Monographs. Cousens, H. 2002. Buddhist Stupa at Mirpur Khas, Sind. Archaeological Survey of India Annual Report 1909–1910. New Delhi: Archaeological Survey of India (first published 1914), pp. 80–92. Davis, R. 1997. Lives of Indian Images. New Jersey: Princeton University Press. Errington, E. 1987. The Western Discovery of the Art of Gandhara and the Finds of Jamalgarhi. Ph.D. thesis. London: School of Oriental and African Studies. Errington, E. 1998. Gandhara Stupa Deposits. Arts of Asia 28.2, March‑April: 80–87. Gupta, S.P., ed. 1985. Kushana Sculptures from Sanghol (1st‑2nd century AD): A Recent Discovery. New Delhi: National Museum. Jain, J. 2011. Museum and Museum-like Structures: The Politics of Exhibition and Nationalism in India. Exhibitionist Spring: 50–55.

260  Himanshu Prabha Ray Joshi, N.P., and R.C. Sharma. 1969. Catalogue of Gandhara Sculptures in the State Museum, Lucknow. Lucknow: State Museum. LaPorte, C.A. 2003. Displaying Empire? The Architecture and Development of Museums in Nineteenth Century India. Ph.D. thesis. University of Pennsylvania. Lohuizen-de Leeuw, J.E. Van. 1972. Gandhara and Mathura: Their Cultural Relationship, in P. Pal (ed.), Aspects of Indian Art. Leiden: E. J. Brill, pp. 27–43. Mathur, S., and K. Singh, eds. 2015. No Touching, No Spitting, No Praying: The Museum in South Asia. London, New York and New Delhi: Routledge. Meister, M., 2010. Temples of the Indus: Studies in the Hindu Architecture of Ancient Pakistan. Leiden: Brill. Michon, D. 2015. Archaeology and Religion in Early Northwest India: History, Theory, Practice. New Delhi: Routledge. Morley, G. 2005. Indian Sculpture. Foreword by Kapila Vatsyayan. New Delhi: Roli Books. Moti, Chandra 1974. Stone Sculpture in the Prince of Wales Museum. Bombay: Board of Trustees of the Prince of Wales Museum of Western India. Phillips, Kristy K. The National Museum of India: A Museum to and of the Nation. www.ep.liu.se/ecp/022/009/ecp072209.pdf/ accessed on 30 January 2016. Ray, H.P. 2007. Colonial Archaeology in South Asia (1944–1948): The Legacy of Sir Mortimer Wheeler in India. New Delhi: Oxford University Press. Roychowdhury, M. 2015. Displaying India’s Heritage: Archaeology and the Museum Movement in Colonial India. New Delhi: Orient Blackswan. Sharma, G.B. 1986. Coins, Seals and Sealings from Sanghol. Chandigarh: Department of Cultural Affairs, Archaeology and Museums, Punjab. Sharma, R.C. 1991. Introduction, in A. Sengupta, and D. Das (eds.), Gandhara Holding in the Indian Museum (A Handlist). Calcutta: Indian Museum. Singh, K. 2015. The Museum Is National, in S. Mathur and K. Singh (ed.), No Touching, No Spitting, No Praying: The Museum in South Asia. New Delhi: Routledge, pp. 107–131. Sivaramamurti, C. 1954. A Guide to the Archaeological Galleries of the Indian Museum. Calcutta: Indian Museum. Skilling, P. 2011. Buddhism and the Circulation of Ritual in Early Peninsular Southeast Asia, in P.-Y. Manguin, A. Mani, and G. Wade (eds.), Early Interactions Between South and Southeast Asia: Reflections on Cross-Cultural Exchange. Singapore: Institute of Southeast Asian Studies, pp. 367–380.

Index

Achaemenian/Achaemenid 6, 8, 87, 182, 185 Agathocles 22, 74 – 81, 88, 90 – 1, 96, 134 – 5 Ai Khanum 35 – 6, 45 Ajax 84 Akhnoor 247, 255 Alexander (Sikandar) (the Great) 1, 2, 9 – 10, 33 – 4, 52, 54, 70, 75, 82, 85 – 6, 104, 135, 178, 255 Allen, John 76 – 8, 81 – 3, 88 – 91,  96 Amluk-dara 173, 176, 178, 198, 200 – 2,  204 Ancient Monument Protection Act (VIII, 1904) 173 Anderson, John 2, 237, 239, 253 – 5 Aṅguttaranikāya 109 Apollodotos 77 – 9, 94 – 6 Arrian (Arrianus) 10 – 11, 70, 178 Ashmolean Museum 23, 72, 94, 131, 139 Asian Art Museum (Museum für Asiatische Kunst, Berlin) 24, 213 Asian Art Museum (San Francisco) 233 Asiatic Society (Kolkata) 2, 23, 131, 135 – 8, 146, 162, 236; (Bombay) 244 Aśoka(n) 2, 56, 75, 79, 83, 239 Aśokan rock and pillar edicts 15, 20, 35, 37, 38, 83, 244, 254 Bahlīkā 117 – 18,  122 Bamiyan 20 – 1, 214 – 15,  234 Barikot (Bazira/Beira/Vajirasthāna/ Bīr-koṭ-ghwaṇḍai) 23, 57, 173, 176 – 82, 187 – 98, 201 – 3

Beal, Samuel 47, 50, 53, 162 Begram 12 – 13, 84, 136, 140 Beni Hissar 2 Bhabha, Homi 71 Bhargava, P.L. 117 Bhir mound 27, 71, 82, 84, 86 – 9, 91 – 2, 182, 237; (coin) hoard 84 – 7, 88 – 9, 91, 96,  99 Brahmi 18, 20 – 1, 74, 92, 104, 134 – 5 British Museum 23, 72, 76, 81, 88 – 90, 92, 95, 130 – 1, 138, 174, 236 Burgess, James 104 Burnes, Alexander 145 – 6, 149, 214 Butkara 14, 60, 175, 184 – 5, 205 Caddy, Alexander 164, 237 Chanda, Ramaprasad 237, 239 Charsadda 182, 237, 248 Chhatrapati Shivaji Maharaj Vastu Sangrahalaya (Mumbai) 242 Cobb, E.H. 23, 166 – 8, 170 coins: Buddha on coins 1, 4 – 5; counterstruck/striking (coins) 94 – 7, 99; Indo-Scythian 93, 100, 134; Kuninda coins 18; Mauryan 9, 80 – 2, 84, 86 – 7, 89, 91 – 3, 95 – 6, 99; punch-marked 7, 9, 18, 72, 76, 80, 85 – 8, 92, 182; in stupa deposits 4; Taxila coins 72 – 82, 86, 88 – 9, 91 – 5, 97, 99; uninscribed 72, 76, 82 – 3 Cole, Henry Hardy 237, 247 Coningham, Robin 50 counter-struck see coins Cousens, Henry 245 – 6

262 Index Cunningham, Alexander 6, 14, 21, 50, 52 – 3, 70 – 1, 72 – 5, 76, 78, 88, 89 – 91, 94, 96, 98, 138, 217 – 18, 236 – 7, 247,  254 Curtis, Sarkosh 135 Dahlmann, Joseph 107 Dandekar, R.N. 108 Darius 8 Deane, Major Arthur Harold 162 – 7,  170 Demetrios 41, 77 – 9 Deshpande, Madhav 108 – 9, 115 Devāsura conflict 115 – 19 Dharmarajika stupa 16 – 17, 83 Dharmaśāstra 107 Dice/dicing 120 – 21 Diodotos 85 – 6 Droysen, Johann Gustav 71 Durga Prasad, Babu 80, 89 East India Company 2, 23, 131 – 3, 136, 138, 140, 236 Ed-Dur 13 – 14,  136 Elphinstone, Mountstuart 1, 214 Eucratides 22, 188 Faccena, Dominico 176 Faxian (Fo-Hian) 48, 50 – 1, 53, 57, 60 Fitzwilliam Museum (Cambridge) 23, 130 – 1,  139 Foucault, Michel 48 Foucher, Alfred 3 – 5, 15, 83 – 4, 156, 167 Gāndhārī (wife of Dhṛtarāṣṭra) 105 – 6, 118,  122 Gāndhārī language 179 Gāndhārī Prakrit 20 – 2, 33, 37 – 40, 55 – 6,  105 Government Museum (Chandigarh) 24, 234 – 5, 247 – 9,  251 Grünwedel, Albert 220 Gumbat 174 Harappan 14, 183 Hargreave, H. 166, 224 Härtel, Herbert 223 – 4 Hastinapur 111 Hathial 6

Hēliódōros 36, 39 – 41 Heras, Henry S.J. 239 Heras Institute Museum (Mumbai) 234, 239 – 42 Hirañasame (coinage) 88, 90, 96 Hoernle, Rudolph 121 Hopkins, E. Washburn 108 Hoti Mardan 3, 247 Humboldt, Alexander von 214 Humboldt, Wilhelm 213 Humboldt Forum 226 Huntington, John 5 Huntington, Susan 5 Indian Museum (formerly Imperial Museum) (Kolkata) 2, 136, 138, 156, 164, 238 Indo-Scythian 2, 105, 239, 254; see also coins Jamalgarhi 37, 217 – 18, 237 – 9, 248, 254 Janamejaya 107, 109 – 11, 113 – 14 Jātakas 54, 57 – 60; Kṣāntivādin (Palli: Khantivādī) 58; Maitrībala 58; Mṛga 59; Rishyasringa 238; Sibi 57 – 9; Śyāma 59; Viśvantara 57, 59; Vyāghrī 57, 59 Jayaswal, K.P. 80 Kandahar 35 Kashmir Smast 19 Khandalavala, Karl 244, 252, 255 Kharkai 237, 254 Kharoṣṭhī 8, 18, 20 – 1, 39, 55 – 6, 96, 104 – 5, 134 – 5,  215 Kipling, John Lockwood 47, 220, 247 Kipling, Rudyard 47 – 8 Koeppen, Carl Friedrich 215 Kosambi, D.D. 111 Krishnadasa, Rai 252 Kuninda see coins Kushan(a) 5, 14 – 15, 19, 21, 105, 183, 188, 193, 202 – 3, 235, 251, 253, 255 Lahore (Central) Museum 2 – 3, 47 – 8, 156, 163 – 4, 178 – 9, 215, 217, 219 – 21, 247 – 8 Lassen, Christian 107, 134, 215

Index  263 Leitner, G.W. 3, 52, 215, 218 – 20 Linden Museum of Ethnology (Stuttgart) 24, 213, 225 Loriyan Tangai 237, 239 Madhyadeśa 105 – 6 Mahājanapada/ ṣoḍaśamahājanapada 6, 109, 186 (table) Manikyala 1, 7, 214, 236, 248 Marshall, John 6 – 7, 22, 71, 74 – 6, 81 – 4, 87 – 92, 95 – 6, 98, 105, 120, 166 – 7, 176, 205, 237, 253 Mathura 2, 7 – 9, 16 – 18, 33, 106, 235, 247, 253 – 5 Mathura Style/School 7, 17 Maurya/n 15, 75, 78, 80, 83 – 6, 89, 96, 98 – 9, 182, 185, 187 – 8 Menander (Ménandros) 33 – 4, 37 – 8, 41, 188 Metropolitan Museum (New York) 234 Michiner, Michael 78 – 9 Milindapañha (Questions of Menander) 15, 33 – 5, 41 – 2 Mirpur Khas 244 – 7, 255 Mleccha 105, 124 Moti Chandra 244, 252 Mūlasarvāstivādavinaya 56, 59 Musée Guimet 3 – 4 Museo Nazionale di Arte Orientale (now Giuseppe Tucci National Museum of Oriental Art (Rome) 175 Museum for Ethnology (Museum für Völkerkunde, Berlin) 213 – 14, 219, 222 – 3 Museum for Indian Art (Museum für Indische Kunst, Berlin) 224 Nāgasena 15, 33 – 4, 42 National Museum (New Delhi) 24, 130, 175, 234, 249, 251 – 5 Nearchus 10 Onesicritus 10 Osterhammer, Jűrgen 48 Panini 8 – 9,  113 Pantaleon 22, 74, 75 – 81, 90 parinirvāṇa 54, 160

Pāśakakevali 121 Periplus (Maris Erythraei) 9, 11 – 13 Peshawar Museum 2, 18, 156, 165 – 6, 168, 170, 217, 219, 224, 237, 247 Peshawar valley/plains 5 – 6, 52, 176, 220, 235, 245 Philostratus 84 Pieper, Wilfried 79 – 80 Plutarch 34, 70 Porus 70, 75 Prinsep, James 23, 104, 134 – 7, 215 punch-marked see coins Puṣkalāvatī 37, 110 Rabatak 15, 19 Ram Dass 221, 223, 225 Ranjit Singh (Maharaja) 1 Rawlinson, H.S. 166 Raychaudhuri, H.C. 114 Relic/reliquary 1, 5, 8, 22, 34, 37 – 8, 40, 54 – 7, 60, 131, 136, 140, 143, 145 – 7, 149 – 51, 253,  256 Ritter, Carl 214 – 15 Royal Asiatic Society (London) 52 Sahri Bahlol 3, 201, 237, 239, 248, 254 Saïd, Edward 71 Śakuni 115 – 21 Sanchi 83 – 4 Sanghao 237, 248 Sanghol 16 – 17, 249 – 51 Sarai Khola 6 Scherman, Lucien 224 – 5 Schwaiger, Imre 253 Senart, Emile 163 Shahbazgarhi 4, 15, 133 Shah-ji-ki-dheri 16, 224, 237, 247 Shahnameh 2 Sikri (Sikrai) 162, 248 Sirkap 7, 13, 18, 71, 77, 82, 84, 86, 91 – 3, 120, 187, 202, 237 Slone, Hans 130 Song Yun 162 Sonkh 18 Spooner, D.B. 165 State Museum for Ethnology (Munich) 224 – 5

264 Index Stein, Aurel 155, 167 Stewart, H. 165 – 6 sundial 156, 159, 168 – 70 Surkh Kotal 19, 35 Swat 51 – 2, 60, 173; see also Udyāna Swat Museum 156, 167 – 70, 174 – 6 Takht-i-Bahi 3, 52, 217, 237, 239, 249, 254 Tarn, William 34 – 7, 77 – 9, 84, 88, 90 – 1 Taxila (Takṣaśilā, Takkasilā) 6 – 9, 10 – 13, 15, 18, 22, 36 – 7, 40, 57 – 8, 70 – 2, 75, 78 – 82, 84, 87 – 8, 90, 98 – 100, 105, 109 – 14, 122, 176, 193, 202, 219, 235, 237, 239, 253 Taxila Museum 7 Treasure Trove Act 1878 163 Tucci, Guiseppe 54, 56 – 7, 174 – 5, 178 – 9 Tylos 14

Udegram 175, 184 – 5, 187, 201, 205 Udyāna (Uḍḍiyāna, Swat) 7, 51 – 9, 162 – 3 Ugraśravas 107 – 8 Vaiśampāyana 107 – 8, 111 – 12,  115 Ventura, Jean-Baptiste 1 Victoria and Albert Museum 174 Vidiśā 36 – 7,  40 Waddell, L.A. 164 Walsh, E.H.C. 76, 81 – 2, 84 – 9, 91 – 2 Weber, Albrecht 107 Wheeler, Mortimer 6, 50, 76, 251 – 2 Whitehead, R.B. 76, 81, 88, 90 – 1 Wilson, H.H. 133, 139 – 41 Xuanzang (Hsüan-tsang, Huan Tsiang, Hwen-Thiang) 21, 50 – 1, 53, 60, 75, 162 yajn͂a 107, 111 yavana 9, 105, 124 Yusufzai 3, 52, 133, 239