The Cambridge World History of Genocide: Volume 1, Genocide in the Ancient, Medieval and Premodern Worlds [New ed.] 110849353X, 9781108493536, 9781108655989

Citation preview

the cambridge world history of

GENOCIDE

Volume I offers an introductory survey of the phenomenon of genocide. The first five chapters examine its major recurring themes, while the further nineteen are specific case studies. The combination of thematic and empirical approaches illuminates the origins and long history of genocide, its causes, consistent characteristics and the connections linking various cases from earliest times to the early modern era. The themes examined include the roles of racism, the state, religion, gender prejudice, famine and climate crises, as well as the role of human decisionmaking in the causation of genocide. The case studies cover events on five continents, ranging from prehistoric Europe and the Andes to ancient Israel, Mesopotamia, the early Greek world, Rome, Carthage and the Mediterranean. It continues with the Norman Conquest of England’s North, the Crusades, the Mongol Conquests, medieval India and Viet Nam, and a panoramic study of premodern China, as well as the Spanish conquests of the Canary Islands, the Caribbean and Mexico. B E N K I E R N A N is the Griswold Professor of History Emeritus at Yale University and was founding Director of Yale’s Genocide Studies Program. His book Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (2007) won numerous awards, including a gold medal for the best work of history, awarded by the Independent Publishers Association. T . M . L E M O S is Professor of Hebrew Bible at Huron University College and a member of the graduate school faculty and the Centre for Transitional Justice and Post-Conflict Reconstruction at the University of Western Ontario. T R I S T A N S . T A Y L O R is Senior Lecturer in Classics and Ancient History at the University of New England. He was a Visiting Fellow in Genocide Studies in the Yale University Genocide Studies Program in 2013–14 and was awarded a UNE Partnerships Grant for his work on comparative genocide studies in the Roman world.

t h e ca m b r i d g e w o r ld h i s t o r y o f

GENOCIDE General Editor Ben Kiernan, Yale University

Split into three volumes, The Cambridge World History of Genocide offers an analytical survey of genocide across six continents from prehistory to the twenty-first century. Combined, they compare and contrast cases in multiple different cultures and contexts, demonstrating common themes and sharp variations that have developed over time. By examining the long-term and immediate causes of genocide, these essays emphasize that genocidal intent has historically been shaped by structural factors and human decision-making. Featuring over 80 essays from experts across the field, together they cover ancient Carthage, the Holocaust, medieval Crusader massacres, Mongol conquests, the extermination of Indigenous peoples in European settler colonies in the Americas, Africa and Australia, as well as prehistoric mass graves from the Alps to the Andes, and the Rohingya genocide in Myanmar. A much-needed addition to genocide studies, these volumes reveal how genocide is a world historical phenomenon that has operated under different names and capacities, but possesses similar key characteristics. V OLUME I Genocide in the Ancient, Medieval and Premodern Worlds EDITED BY BEN KIERNAN , T . M . LEMOS AND TRISTAN S . TAYLOR V OLUME I I Genocide in the Indigenous, Early Modern and Imperial Worlds, from c. 1535 to World War One EDITED BY NED BLACKHAWK , BEN KIERNAN , BENJAMIN MADLEY AND REBE TAYLOR

EDITED BY BEN

V OLUME I I I Genocide in the Contemporary Era, 1914–2020 KIERNAN , WENDY LOWER , NORMAN NAIMARK AND

SCOTT STRAUS

THE CAMBRIDGE WORLD HISTORY OF

GENOCIDE *

VOLUME I

Genocide in the Ancient, Medieval and Premodern Worlds *

Edited by

BEN KIERNAN Yale University T. M. LEMOS Huron University College, University of Western Ontario TRISTAN S. TAYLOR University of New England, Australia

University Printing House, Cambridge C B 2 8B S, United Kingdom One Liberty Plaza, 20th Floor, New York, N Y 10006, USA 477 Williamstown Road, Port Melbourne, V I C 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781108493536 D O I: 10.1017/9781108655989 © Cambridge University Press 2023 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2023 Printed in the United Kingdom by TJ Books Limited, Padstow, Cornwall A catalogue record for this publication is available from the British Library. A Cataloging-in-Publication data record for this book is available from the Library of Congress Three-volume set I S B N 978-1-108-75973-1 Hardback Volume I I S B N 978-1-108-49353-6 Hardback Volume II I S B N 978-1-108-48643-9 Hardback Volume III I S B N 978-1-108-48707-8 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Contents

List of Figures page x List of Maps xii List of Tables xiii List of Contributors to Volume I xiv General Editor’s Acknowledgements xx General Editor’s Introduction to the Series: Genocide: Its Causes, Components, Connections and Continuing Challenges 1 ben kiernan

Introduction to Volume I 31 t. m. lemos, tristan s. taylor and ben kiernan

part ithemes of genocide through history 1 . Genocide before the State? helle vandkilde

59

2 . The Religion–Genocide Nexus 86 steven leonard jacobs 3 . Genocide and Gender: Dynamics and Consequences adam jones and wendy lower

103

4 . Genocide, Starvation and Famine 127 bridget conley and alex de waal 5 . Climate, Violence and Ethnic Conflict in the Ancient World 150 francis ludlow, chris morris and conor kostick

vii

Contents

part iithe ancient world 6 . Genocide in Ancient Israelite and Early Jewish Sources t. m. lemos

185

7 . Genocide in Ancient Mesopotamia during the Bronze and Iron Ages t. m. lemos and seth richardson 8 . Urbicide in the Ancient Greek World, 480–330 B C E paul cartledge

209

235

9 . Violence, Emotions and Justice in the Hellenistic Period 257 michael champion 10 . A Tale of Three Cities: The Roman Destruction of Carthage, Corinth and Numantia 278 tristan s. taylor 11 . Caesar’s Gallic Genocide: A Case Study in Ancient Mass Violence tristan s. taylor

309

12 . Genocidal Perspectives in the Roman Empire’s Approach towards the Jews 330 gil gambash 13 . Religious Violence in the Later Roman Empire: The Tetrarchic Persecutions, 302–313 C E 353 carl r. rice 14 . Genocide, Extermination and Mass Killing in Chinese History 376 victoria tin-bor hui

part iiithe medieval world and early imperial expansions 15 . William the Conqueror’s Harrying of the North, 1069–1070: What, If Not Genocide? 403 c. p. lewis

viii

Contents

16 . Genocidal Massacres of Jews in Medieval Western Europe, 1096–1392 425 maya soifer irish 17 . Crusaders and Mass Killing at Jerusalem in 1099 448 thomas a. fudge 18 . The Albigensian Crusade and the Early Inquisitions into Heretical Depravity, 1208–1246 470 mark gregory pegg 19 . Mongol Genocides of the Thirteenth Century 498 timothy may 20 . Viêṭ Nam and the Genocide of Champa, 1470–1509 523 george dutton 21 . Genocidal Massacres in Medieval India raziuddin aquil

547

22 . Mass Extermination in Prehistoric Andean South America 572 danielle kurin 23 . The Spanish Destruction of the Canary Islands: A Template for the Caribbean Genocide 594 igor pe´ rez tostado 24 . Genocidal Massacres in the Spanish Conquest of the Americas: Xaragua, Cholula and Toxcatl, 1503–1519 622 harald e. braun Index 648

ix

Figures

1.1 The ordered abnormal grave from Wassenaar (Haag, Netherlands), c.1700 B C E. (© Leendert Louwe Kooijmans 1993. Reproduced by permission of Leendert Louwe Kooijmans) page 79 5.1 Top panels of a carved gypsum relief from the North Palace in Nineveh, Iraq, 645–640 B C E, depicting irrigation of lands with an aqueduct. (Photo: Francis Ludlow) 161 5.2 Bottom two panels of a carved gypsum relief from the North Palace in Nineveh, Iraq, 645–640 B C E, showing a Neo-Assyrian attack on an Arab encampment. (Photo: Francis Ludlow) 163 5.3 Gypsum relief from the North Palace in Nineveh, Iraq, 645–640 B C E, showing the sacking of an Egyptian stronghold by Neo-Assyrians under King Ashurbanipal. (Photo: Francis Ludlow) 165 5.4 Levels of conflict and sociopolitical stress recorded in Neo-Assyrian sources in the years before and after the dates of four volcanic eruptions, 750–650 B C E. 177 6.1 The Moabite Stone, discovered in 1868, bearing an inscription left by the ninth-century B C E Moabite king Mesha. (Print Collector / Contributor / Getty Images) 191 7.1 Assyrian relief from the palace of Sennacherib. (Photograph from Werner Forman / Getty Images) 225 10.1 Ruins of Numantia. (Photo: Ben Kiernan) 290 10.2 Reconstruction of the ruins of Numantia. Museo Numantino, Ministerio de Cultura. (Photo: Ben Kiernan) 290 10.3 Inscription erected by Mummius dedicating the temple to Hercules Victor. Inscriptiones Latinae Selectae 20. (Center for Epigraphical and Palaeographical Studies, The Ohio State University) 299 11.1 Bust of Julius Caesar. Posthumous, but thought to be first century B C E . Museo Pio-Clementino, Musei Vaticani. (Photograph from Getty Images) 310 13.1 Porphyry sculpture portrait of the Four Tetrarchs, c.300 C E. Originally in Constantinople, now at St Mark’s Basilica in Venice. (Photograph from Education Images / Getty Images) 359 16.1 Representation of a massacre of the Jews in 1349. Antiquitates Flandriae, Royal Library of Belgium, MS 1376/7. (Courtesy Jewish Women’s Archive) 442 18.1 Béziers. (Photograph from Getty Images) 471

x

List of Figures 19.1 Chinggis Khan statue at the Chinggis Khan Statue Complex in Tsonjin Boldog, Mongolia. (Photo: Timothy May) 505 19.2 The siege of Alamu¯t, from Rashı¯d al-Dı¯n’s Jami’ al-Tawarikh. (Courtesy Bibliothèque nationale de France) 516 20.1 Cham Tower still standing at Vijaya (1). (Photo: George Dutton) 538 20.2 Cham Tower still standing at Vijaya (2). (Photo: George Dutton) 539 20.3 Cham Hindu temples at Tháp Chàm, near Phan Rang, Vietnam. (Photo: Ben Kiernan) 542 21.1 Qutb Minar, Delhi. (Photograph from frentusha / Getty Images) 5 53 21.2 Tughlaqabad fort. (Photograph from Getty Images) 564 22.1 Diverse manifestations of deadly head wounds. 576 23.1 Torre del conde, Gomera, Canary Islands c.1450. Photograph 1890–5. (Courtesy the Colección José A. Pérez Cruz, Fundación para la Etnografía y el Desarrollo de la Artesanía Canaria (FEDAC) / Cabildo de Gran Canaria/Biblioteca Virtual del Patrimonio Bibliográfico - España) 607 23.2 Luis Arencibia Betancort, Hautacuperche, 2007, bronze, Valle Gran Rey, Gomera. (Photo by Ángel Guzmán Ramírez) 615 24.1 La Matanza de Cholula, Lienzo de Tlaxcala. 639 24.2 La Matanza de Templo Mayor – Toxcatl, Codex Dúran. 641

xi

Maps

1.1 Selected prehistoric abnormal graves and mass fatality sites. (© Helle Vandkilde) page 70 6.1 Ancient Israel and surrounding regions in the Iron Age. (Cartography by Shane Kelley) 186 7.1 Mesopotamian and other Near Eastern sites discussed. (Cartography by Steven Townshend) 211 8.1 Ancient Greece and Sicily. (Cartography by the Ancient World Mapping Center, University of North Carolina) 238 9.1 Sites of city destruction in Hellenistic Greece, 223–217 B C E. (Cartography by the Ancient World Mapping Center, University of North Carolina) 258 10.1 The Mediterranean of the Roman Republic, showing relative positions of Carthage, Corinth and Rome. (Cartography by the Ancient World Mapping Center, University of North Carolina) 279 10.2 The Iberian peninsula in the second century B C E. (Cartography by the Ancient World Mapping Center, University of North Carolina) 292 11.1 Peoples and main settlements of Gaul at the time of Julius Caesar. (© Citypeek/Wikimedia commons) 314 12.1 The Roman Near East in the early second century C E (Cartography by Gil Gambash) 332 13.1 Selected sites in the late Roman Empire. (Cartography by Tom Elliott, with modifications by the Author) 354 14.1 Expansion of the Qin Empire, c.300 – c.210 B C E. (Cartography by Bill Nelson) 392 14.2 Expansion of the Han Empire, c.140 B C E – 220 C E. (Cartography by Bill Nelson) 396 15.1 Northern England in the time of William I. 404 16.1 Anti-Jewish violence in western Europe, 1096–1392. (Cartography by Amy Ferguson, GIS/Data Center, Rice University) 426 17.1 The siege and capture of Jerusalem, July 1099. (Cartography by Bill Nelson) 456 18.1 Southern France c.1200 (From M. G. Pegg, A Most Holy War (Oxford University Press, 2008), p. 9). 473 19.1 The Mongol Empire, 1250–60. (Cartography by Mapping Specialists, Fitchburg, Wisconsin) 507 20.1 The Vietnamese attack on Champa, 1471. (Cartography by George Dutton) 525 22.1 Andean South America showing sites of prehistoric mass extermination. (Cartography by Danielle Kurin) 573 23.1 Spain, North Africa and the Canary Islands in 1544 (Portolan Map). 596

xii

Tables

2.1 7.1 14.1 14.2 14.3

‘Religion’ as a participating factor in genocide Casualties of military campaigns of Rimuš and Naram-Sin Common words for ‘extermination’ in dynastic histories Common words for ‘killing’ in dynastic histories Qin mass killing, 356–236 B C E

xiii

page 95 214 378 379 391

Contributors to Volume I

R A Z I U D D I N A Q U I L is Professor of History at the University of Delhi. He has published widely on political culture and religious traditions in medieval and early modern India. His books include The Muslim Question: Understanding Islam and Indian History (Penguin, 2017) and Days in the Life of a Sufi: 101 Enchanting Stories of Wisdom (Pan Macmillan, 2020). He has also co-edited, with Partha Chatterjee, History in the Vernacular (Permanent Black, 2008), with David Curley, Literary and Religious Practices in Medieval and Early Modern India (Routledge, 2016), and with Tilottama Mukherjee, An Earthly Paradise: Trade, Politics and Culture in Early Modern Bengal (Routledge, 2020). H A R A L D E. B R A U N is Reader/Associate Professor in European History (1300–1700) at the University of Liverpool. His main field of research is early modern Iberian political culture, and his work connects the histories of political thought, religion, law and violence. His publications include Juan de Mariana and Early Modern Spanish Political Thought (2007), Contexts of Conscience in Early Modern Europe (2004; with E. Vallance), The Renaissance Conscience (2011; with E. Vallance), Theorising the Ibero-American Atlantic (2013; with L. Vollendorf), The Transatlantic Hispanic Baroque (2014; with J. Pérez-Magallón) and Jesuits as Counsellors in the Early Modern World (2017). P A U L C A R T L E D G E is A. G. Leventis Senior Research Fellow of Clare College, Cambridge, and author, co-author, editor or co-editor of some thirty books, the most recent Thebes: The Forgotten City of Ancient Greece (Picador, Abrams, 2020). He is A. G. Leventis Professor of Greek Culture emeritus, Faculty of Classics, Cambridge University, and currently president of the Hellenic Society, UK, an honorary citizen of Sparta, Greece, and Commander of the Order of Honour, Hellenic Republic. M I C H A E L C H A M P I O N is Associate Professor in Late Antique and Early Christian Studies at the Institute for Religion and Critical Inquiry, Australian Catholic University (ACU). He directs ACU’s Node of the Australian Research Council Centre of Excellence for the History of Emotions. He is the author of Explaining the Cosmos: Creation and Cultural Interaction in Late-Antique Gaza (2014) and co-editor, with Lara O’Sullivan, of Cultural Perceptions of Violence in the Hellenistic World (2017). He has also co-edited Understanding Emotions in Early Europe (2015; with Andrew Lynch) and Before Emotion: The Language of Feeling 400–1800 (2019; with Juanita Feros Ruys and Kirk Essary).

xiv

List of Contributors to Volume I B R I D G E T C O N L E Y is the research director of the World Peace Foundation and an associate research professor at the Fletcher School of Law and Diplomacy, Tufts University. She is the author of Memory from the Margins: Ethiopia’s Red Terror Martyrs Memorial Museum (Palgrave, 2019) and editor of How Mass Atrocities End: Studies from Guatemala, Burundi, Indonesia, the Sudans, Bosnia-Herzegovina, and Iraq (Cambridge University Press, 2016). She has published on issues relating to the 1992–5 war in Bosnia, mass atrocities and genocide, and how museums engage on human rights issues. A L E X D E W A A L is the executive director of the World Peace Foundation at the Fletcher School of Law and Diplomacy, Tufts University. Considered one of the foremost experts on Sudan and the Horn of Africa, his scholarly work and practice has also probed humanitarian crisis and response, human rights, HIV/AIDS and governance in Africa, and conflict and peace-building. His latest books are New Pandemics, Old Politics: Two Hundred Years of War on Disease and its Alternatives (Polity Press, 2021) and Mass Starvation: The History and Future of Famine (Polity Press, 2017). G E O R G E D U T T O N is a professor in the Department of Asian Languages and Cultures at the University of California, Los Angeles. A specialist in early modern Vietnamese social and intellectual history, he is the author of A Vietnamese Moses: Philiphe Binh and the Geographies of Early Modern Catholicism (University of California Press, 2017), winner of the inaugural Fondação Oriente Book Prize, and The Tay Son Uprising: Society and Rebellion in Eighteenth Century Vietnam (University of Hawai’i Press, 2006). He is also editor of Voices of Southeast Asia (M. E. Sharpe, 2014) and co-editor of Sources of Vietnamese Tradition (Columbia University Press, 2012), named a Choice outstanding title. T H O M A S A . F U D G E is a historian of Christianity and Professor of Medieval History at the University of New England, Australia. Previously, he has held academic appointments in the United States and New Zealand. He holds a PhD in medieval history from Cambridge University and a PhD in theology from Otago University, New Zealand. The author of seventeen books, over the past thirty years he has focused his research on heresy, Jan Hus and Hussite history, patterns of religious reform and revolt in fifteenth- and sixteenthcentury Europe, and the later crusades, especially those directed at heretics. G I L G A M B A S H is a classical historian studying the ancient Mediterranean. He is the former chair of the Department of Maritime Civilizations, former director of the Recanati Institute for Maritime Studies at the University of Haifa, and the co-founder and director of the Haifa Center for Mediterranean History. Next to a range of Mediterranean topics, he specialises in Roman imperialism and its relationship with the provinces. He wrote his chapter for this volume while holding a Leverhulme Visiting Professorship at the Institute of Classical Studies, University of London. V I C T O R I A T I N - B O R H U I is Associate Professor in Political Science at the University of Notre Dame. She received her PhD in Political Science from Columbia University and her BSSc in Journalism and Communications from the Chinese University of Hong Kong. Hui examines the centrality of war in Chinese history, and has authored War and State Formation in Ancient China and Early Modern Europe (Cambridge University Press, 2005).

xv

List of Contributors to Volume I She also studies Hong Kong’s struggle for freedom, and has testified to a United States congressional hearing and written for the Journal of Democracy, Foreign Affairs, Washington Post’s Monkey Cage and The Diplomat. M A Y A S O I F E R I R I S H is an associate professor of history at Rice University whose research focuses on religious violence and toleration, and explores the legal, social and economic situation of religious minorities in Iberian Christian societies. She is the author of many articles, including ‘Beyond convivencia: critical reflections on the historiography of interfaith relations in Christian Spain’. Her first book, Jews and Christians in Medieval Castile: Tradition, Coexistence, and Change, which explores the changes in Jewish– Christian relations in the kingdom of Castile between the eleventh and fourteenth centuries, was published in 2016. She is currently working on a new book, The Politics of Persecution in Medieval Spain: Toward the Anti-Jewish Riots of 1391. S T E V E N L E O N A R D J A C O B S is Professor of Religious Studies and Emeritus Aronov Endowed Chair of Judaic Studies at the University of Alabama. His research foci include issues of translation with regards to the Hebrew Bible and Dead Sea Scrolls, Jewish– Christian relations, the Holocaust and historical genocides, and the nexus between religion and violence. He is the author, editor and/or translator of more than fifteen books and more than fifty refereed articles and book reviews in both scholarly and popular journals. A D A M J O N E S is Professor of Political Science at the University of British Columbia Okanagan. He is the author of Genocide: A Comprehensive Introduction (3rd ed., Routledge, 2017) and author or editor of some fifteen other books, mostly on genocide and crimes against humanity. He serves as executive director of Gendercide Watch, a Web-based educational initiative, and has worked as an expert consultant with the United Nations Office on Genocide Prevention and the Responsibility to Protect. B E N K I E R N A N is A. Whitney Griswold Professor of History emeritus and former Professor of International and Area Studies at Yale University. Founding director of both the Cambodian Genocide Program and the Genocide Studies Program from 1994 to 2015, he has also served as chair of Yale’s Council on Southeast Asia Studies. His books include How Pol Pot Came to Power (1985), The Pol Pot Regime: Race, Power and Genocide in Cambodia under the Khmer Rouge, 1975–1979 (1996), Genocide and Resistance in Southeast Asia (2007), Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (2007) and Viê ̣t Nam: A History from Earliest Times to the Present (2017). C O N O R K O S T I C K is a historian and writer living in Dublin. He is currently Postdoctoral Research Fellow with the Climates of Conflict in Ancient Babylonia project, funded by the Irish Research Council and based in the Trinity Centre for Environmental Humanities, Trinity College Dublin. D A N I E L L E K U R I N, Associate Professor of Anthropology at the University of California, Santa Barbara, specialises in the analysis of human bones. She investigates how natural and social disasters impact human bodies in ancient, historic and forensic contexts. While her

xvi

List of Contributors to Volume I primary area of interest is Latin America, Kurin has done similar work in other parts of the world. She also serves as an expert consultant in cases involving trafficked skeletons, mummies and other cultural heritage, and has aided law enforcement, disaster survivors and victims’ families in the identification of unknown human remains stemming from mass casualty events. T. M. L E M O S is Professor of Hebrew Bible at Huron University College, Ontario, and a member of the graduate school faculty and affiliate of the Centre for Transitional Justice and Post-Conflict Reconstruction at the University of Western Ontario. In 2022 she was Gerstein Distinguished Visiting Professor of Jewish Studies and Religion at the University of Toronto. Lemos has co-edited three volumes and is the author of two books, including Violence and Personhood in Ancient Israel and Comparative Contexts (Oxford University Press, 2017), as well as various articles and essays on topics related to the history of violence. Her current project is on the comparative history of dehumanisation. C. P . L E W I S has worked on English and Welsh history in the long eleventh century since his Oxford DPhil of 1985. His current interests in those fields include the prosopography of landowners named in Domesday Book (www.pase.ac.uk) and the making of Domesday Book, on which a monograph co-authored with Stephen Baxter and Julia Crick is forthcoming. Between 1982 and 2009 he worked for the Victoria County History on the local history of places in Cambridgeshire and Sussex, and the city of Chester, and has also published on onomastics and the historiography of English local history. W E N D Y L O W E R is John K. Roth Professor of History and George R. Roberts Fellow at Claremont McKenna College. She is the author of Nazi Empire-Building and the Holocaust in Ukraine (2005), The Diary of Samuel Golfard and the Holocaust in Galicia (2011) and The Ravine: A Family, a Photograph, a Holocaust Massacre Revealed (2021), and co-editor of Shoah in Ukraine: History, Testimony, Memorialization (2008; with Ray Brandon). Her book Hitler’s Furies: German Women in the Nazi Killing Fields (2013) was a finalist for the National Book Award and has been translated into twenty-three languages. She is a co-editor of Volume III of The Cambridge World History of Genocide. F R A N C I S L U D L O W is Associate Professor of Medieval Environmental History at Trinity College Dublin. He has held postdoctoral fellowships with the Harvard University Centre for the Environment and Department of History (2011–13), the Rachel Carson Centre for Environment and Society, LMU Munich (2013–14), the Yale Climate and Energy Institute and Department of History (2013–16), and a Marie Skłodowska-Curie Individual Fellowship (2016–18) at Trinity College Dublin. Presently he is principal investigator of the Irish Research Council-funded Climates of Conflict in Ancient Babylonia project. T I M O T H Y M A Y (PhD, University of Wisconsin-Madison) is Professor of Central Eurasian History at the University of North Georgia. He is the author of The Mongol Art of War (2007), The Mongol Conquests in World History (2012), The Mongol Empire (2018), The Mongols (2019) and Simply Chinggis (2021), and the editor of The Mongol Empire, A Historical Encyclopedia (2016). He was awarded his university’s Distinguished Teaching Award in

xvii

List of Contributors to Volume I 2021. When not teaching courses on the Mongol Empire, the Crusades or Islamic history, he practises the dark arts of administration as Associate Dean of Arts and Letters. C H R I S M O R R I S studied geography and soil science, obtaining a PhD at Queens University Belfast, before having a thirty-year career as a government statistician, working in fields as varied as GDP, demography and urban development. Since retirement he obtained postgraduate degrees in reconciliation studies and human geography before starting research towards a second PhD at Trinity College Dublin. This is based on quantitative analysis of a 120-case sample of classical and medieval interactions of nomad groups or groups engaged in holy war with other groups that they have encountered, taking account of their ecological, economic and cultural circumstances. M A R K G R E G O R Y P E G G is Professor of History at Washington University in St Louis. He is the author of Beatrice’s Last Smile: A History of the Medieval West, 200–1500 (Oxford University Press, 2022). I G O R P E´ R E Z T O S T A D O is Associate Professor of Early Modern History at the University Pablo de Olavide, Spain. He studies early modern migration, exile and violence. He is the author of Irish Influence at the Court of Spain in the Seventeenth Century (2008) and AngloSpanish Relations during the English Civil Wars: Assassination, War and Diplomacy in Early Modern Europe (forthcoming). He co-edited Ireland and the Iberian Atlantic (2 vols., 2010 and 2020) and Los exiliados del rey de España (2015) and most recently edited A Cultural History of Genocide in the Early Modern World (2021). C A R L R . R I C E is a doctoral candidate in ancient history at Yale University. A firstgeneration university student, he completed his undergraduate education at West Virginia University and his master’s degree at North Carolina State University. His work examines the relationship between religion, law, society and culture in the Roman and late Roman Empire. S E T H R I C H A R D S O N is managing editor of the Journal of Near Eastern Studies and a historian of the ancient Near East at the Oriental Institute of the University of Chicago. His work centres on issues relating to ancient state sovereignty and political subjectivity, especially the problem of the collapse of the First Dynasty of Babylon, the history of violence, and he is a generalist-comparativist. His recent output includes several studies of slavery, royal narratives, social bandits, and the social and legal symbolism of animals. All his work is freely available online. T R I S T A N S. T A Y L O R is a senior lecturer in Classics and Ancient History at the University of New England, Australia, and University Associate of the University of Tasmania, Australia. He completed a PhD in classics at Yale University (2010) after completing an MA in Roman law and undergraduate studies in law and classical civilisation at the University of Tasmania. He has been a visiting fellow in Genocide Studies at Yale University (2013–14) and a visiting scholar in Classics at the University of Texas, Austin (2015). He is editor of the Cultural History of Genocide in the Ancient World (2021), as well as works on modern and Roman law.

xviii

List of Contributors to Volume I H E L L E V A N D K I L D E holds the chair in prehistoric archaeology at Aarhus University in Denmark. She has written widely about the European final Stone Age and the Bronze Age, and led research projects on major issues such as mobility, identity and conflict. A current focus includes patterns of violence at the threshold of the Bronze Age in Europe, in particular among highly mobile third millennium B C E Corded Ware populations, whose genetic ancestry traces back to pastoral herders of the Eurasian Steppe Zone. The coeditor of Warfare and Society: Archaeological and Social Anthropological Perspectives (2006), her monographs include From Stone to Bronze (1996), Culture and Change (2007) and Pile in Scania (2017).

xix

General Editor’s Acknowledgements

I wish to thank my fellow editors of the three volumes in this series, without whom it could not have been produced: Ned Blackhawk, Tracy Lemos, Wendy Lower, Benjamin Madley, Norman Naimark, Scott Straus, Rebe Taylor and Tristan Taylor. Despite the subject matter, it has been a pleasure working on a multi-year project with this group of superb scholars. The other seventy-four authors also deserve all our thanks for their generous participation and expertise. I am also grateful to David Simon, director of the Genocide Studies Program at Yale University, and to the MacMillan Center for International and Area Studies at Yale, the Yale Council on Southeast Asia Studies, and the Orville H. Schell Jr Center for International Human Rights at Yale Law School, for their support for this project from its inception. Bill Nelson’s cartographic skills are evident in the seventeen maps he prepared for the three volumes. Tyrell Haberkorn and Bernt Hagtvet kindly sent along important long-forgotten and unpublished texts, respectively. As ever, Thavro Phim, Heng Samnang and Eve Monique Zucker were helpful colleagues on matters Cambodian and Southeast Asian, as was Puangthong R. Pawakapan, who also provided valuable assistance with locating some maps that were difficult to find. My thanks go also to Michael Adas, Rolena Adorno, Abbas Amanat, Omer Bartov, Yehuda Bauer, Doron Ben-Atar, Christopher Browning, Frank Chalk, Israel Charny, Daniel Chirot, Susan E. Cook, Kai Erikson, Robert Gellately, Greg Grandin, Richard Hovannisian, Paul Kennedy, Harold Koh, Howard R. Lamar, Noel Lenski, Victor Lieberman, Ramsay MacMullen, Robert Melson, John Merriman, Raya Morag, James C. Scott, Mark Selden, Ian Shapiro, James Silk, Gaddis Smith, Roger W. Smith, Timothy Snyder, Jonathan Spence, Joanna Waley-Cohen, Robin W. Winks and Jay Winter, for their encouragement and support.

xx

General Editor’s Acknowledgements

A number of other colleagues, now sadly departed and much missed, long inspired and assisted me in the fields of Genocide Studies and Southeast Asian Studies. I wish to express my gratitude to them here: Ivo Banac, Johanna Bodenstab, Vahakn N. Dadrian, Drew S. Days III, Geoffrey Hartmann, Irving Louis Horowitz, Paula Hyman, Kurt Jonassohn, Leo Kuper, Dori Laub, Erik Markusen, Colin Tatz, Henry A. Turner and Patrick Wolfe, as well as the Southeast Asia scholars Melanie Beresford, Herbert Feith, John D. Legge, J. A. C. Mackie and Merle C. Ricklefs. I would also like to record my appreciation for the work of my literary agent, Lisa Adams. At Cambridge University Press, I wish to thank Elizabeth Hanlon, Victoria Inci Phillips, Emily Sharp and Stephanie Taylor, for their very professional assistance on this large project. My wife, Glenda Gilmore, is a fine editor in her own right; her mark is also visible in many of these pages. My heartfelt thanks go to her, as well as to Mia-lia Kiernan, Derry Kiernan, Polina Adamovich and Miles Johnson.

xxi

General Editor’s Introduction to the Series Genocide: Its Causes, Components, Connections and Continuing Challenges ben kiernan Seven weeks after the fall of the Pol Pot regime in 1979, the first noncommunist journalists visited the Cambodian capital. One wrote: ‘Phnom Penh . . . is in the image of the rest of the country. There is no drinking water, no telephone, no mail service, no transport, no registry office, no money, no markets, hardly any electricity, hardly any schools, hardly any medical dispensaries. The city is so quiet that bird-song has a sinister ring to it.’1 Sixteen years later, another journalist reported a similar silence in another country that had just suffered a genocide: ‘The nights were eerily quiet in Rwanda. After the birds fell silent, there were hardly even any animal sounds. I couldn’t understand it. Then I noticed the absence of dogs.’2 This was not a new theme. In Chapter 7 of this volume, T. M. Lemos and Seth Richardson describe an ancient Mesopotamian vision of genocide that also included eerie silence, and darkness devoid even of wild animals. Genocide has a way of imposing silence. Part of its purpose is to erase history, and human voices, if not sound itself. This series of three volumes aims to contribute to breaking the silence that so often follows genocidal outbreaks. The volumes, in combination, attempt to document and understand the global phenomenon of genocide. They include more than eighty substantive chapters, five in this volume on overarching themes, and more than seventy-five in all three volumes on individual case studies from prehistory to the present. 1 Jean-Pierre Gallois, Agence-France Presse, report from Phnom Penh, 25 March 1979, quoted in Ben Kiernan, ‘Kampuchea 1979–1981: national rehabilitation in the eye of an international storm’ in Southeast Asian Affairs 1982 (Singapore: Institute of Southeast Asian Studies, Heinemann, 1982), pp. 167–95, at p. 167. 2 Philip Goureveitch, We Wish to Inform You that Tomorrow We Will Be Killed with Our Families: Stories from Rwanda (New York: Picador, 1998), p. 147. I am grateful to David Simon for bringing this passage to my attention.

1

ben kiernan

Yet this series cannot claim to be comprehensive. Many of our contributing authors have researched cases that had previously been identified but have hitherto received relatively little scholarly attention, such as the premodern Vietnamese genocide of Champa, examined by George Dutton in Chapter 20 of this volume.3 In other identified cases, our authors have uncovered a paucity of evidence that genocide actually occurred; see for instance Cornelia Soldat’s chapter in Volume I I on Ivan the Terrible’s ‘Sack of Novgorod’. But we have unfortunately been unable to include some genocides, such as – to cite another imperial Russian case – the mass murder, according to one estimate, of 600,000 Circassians in the mid-nineteenth century.4

Definitions The term ‘genocide’, as a way of describing the ‘practice of extermination of nations and ethnic groups’, was coined in 1943, when Raphael Lemkin (1900–59) penned the preface to his 1944 book Axis Rule in Occupied Europe.5 Just five years later the United Nations General Assembly unanimously adopted the Genocide Convention, which came into force as an international treaty in 1951.6 But it took until 1998 for the first genocide perpetrator to be convicted in an international trial – for a crime committed in 1994 in Rwanda, a genocide examined by Scott Straus in Volume I I I of The Cambridge World History of Genocide. The first genocide conviction of a former head of state in an international court came only in 2018, for crimes committed in Cambodia in 1977–8, which are also examined in Volume I I I.7 3 Bernard Fall wrote that ‘the case of the Chams . . . amounted to veritable genocide’ (The Two Viet-Nams (New York: Praeger, 1963), pp. 13–14). 4 Kurt Jonassohn and Karin Solveig Björnson include ‘The Sack of Novgorod in 1570 by Ivan the Terrible’ as a case of ‘genocidal massacre’ in their generally excellent book, Genocide and Gross Human Rights Violations in Comparative Perspective (New Brunswick: Transaction, 1998), pp. 202–5. On the Circassian case, see e.g. Yehuda Bauer, ‘Reflections on cruelty and sadism’, Yad Vashem Studies (online) 49:2 (2021), 37–64, citing John F. Baddeley, The Russian Conquest of the Caucasus (Mansfield Center, CT: Martino, 2006); and Walter Richmond, The Circassian Genocide (Brunswick: Rutgers University Press, 2013). 5 Raphael Lemkin, preface dated 15 November 1943, Axis Rule in Occupied Europe: Laws of Occupation, Analysis of Government, Proposal for Redress (Washington, DC: Carnegie Endowment for International Peace, 1944), p. xi. 6 Convention on the Prevention and Punishment of the Crime of Genocide, Paris, 9 December 1948, United Nations, Treaty Series, vol. 78, no. 1021, www.hrweb.org/leg al/genocide.html. 7 See also Ben Kiernan, ‘The Pol Pot regime’s simultaneous war against Vietnam and genocide of Cambodia’s ethnic Vietnamese minority’, Critical Asian Studies 53:3 (2021), 342–58.

2

General Editor’s Introduction to the Series

Yet it is widely acknowledged that the Nazi regime committed genocide against Jews, Romani and others during World War Two, even though the crime had no legal status at that time. The Nazi defendants at the Nuremberg trials (1945–6) were initially charged with genocide, though they were ultimately convicted of crimes against humanity and aggression.8 Indeed ‘genocide’ is often considered a twentieth-century crime rather than a retroactive post-war legalism or an analytical framework for understanding historical events. Yet Lemkin, a Polish Jewish jurist, conceived of genocide as a phenomenon that predated the Holocaust. He considered the Armenian genocide during World War One to have been a similar crime, and already during the 1930s he had actively worked for its international recognition as an event whose repetition should be criminalised and hopefully prevented by international legal prohibition. In 1933 he termed it a case of ‘barbarity’, and proposed that ‘an international treaty should be negotiated declaring that attacks upon national, religious and ethnic groups should be made international crimes’.9 In fact, Lemkin considered genocide to have much older roots. He had set about writing – but did not complete before his death – a three-volume history of genocide from ancient times, in which he argued that the phenomenon had ‘followed humanity throughout history’. For his first and second volumes, covering antiquity and the Middle Ages, Lemkin drafted chapters on ancient Assyria, the Albigensian war, the Mongol conquests and Spanish persecutions of Moors and Moriscos; and he planned eighteen other chapters, for example on ‘Biblical Genocide’, ‘Carthage’, ‘Genocide in Gaul’ and ‘Crusades’. Although he considered that ‘the last centuries have been particularly abundant in genocide cases’, and he planned forty-one chapters for his third volume on ‘Modern Times’, Lemkin pointed out that in his view the causes began far earlier: ‘One of the basic reasons for genocide is a conflict of cultures as it appeared for example in the encounter between migrating nomadic societies and sedentary ones.’10 This first volume of The Cambridge World History of Genocide tests that view, among many others.

8 Nuremberg Trial Proceedings, vol. I, Indictment, Count Three: War Crimes, VIII, Statement of the Offence, (A) Murder and Ill-Treatment of Civilian Populations . . . para 2, https://avalon.law.yale.edu/imt/count3.asp. 9 Lemkin, Axis Rule in Occupied Europe, p. xiii. 10 Raphael Lemkin, Lemkin on Genocide, ed. Steven L. Jacobs (Lanham, MD: Lexington Books, 2012), pp. 5, 17–19, 55–185. See also Michael A. McDonnell and A. Dirk Moses, ‘Raphael Lemkin as historian of genocide in the Americas’, Journal of Genocide Research 7:4 (December 2005), 501–29.

3

ben kiernan

The inaugural Cambridge World History of Genocide helps fill a need that arises from a popular misunderstanding that genocide, like its name, began only in the twentieth century. These volumes do so by showcasing much of the superb research and scholarship on genocide through the ages. In the last few decades, following the path that Raphael Lemkin and his peer Hersch Lauterpacht pioneered, genocide scholars have developed the robust new field of Genocide Studies. One of its hallmarks is the acknowledgement that although the term ‘genocide’ did not exist before World War Two, similar terms did, and more importantly, similar concepts and practices existed.11 One example is illustrated by premodern usage of the term ‘extermination’. The Latin term exterminare originally meant ‘drive beyond the boundaries’, and it retained this meaning well beyond the eighth century C E, when the Venerable Bede employed it probably in that sense of ‘expel’. But in another context, Bede (and others to follow) did also use exterminare to mean ‘kill them all’, in the modern sense of the word ‘exterminate’. He wrote of England in the late seventh century: ‘After Caedwalla had gained possession of the kingdom . . . he also captured the Isle of Wight . . . and he endeavoured to exterminate (exterminare) all the natives by merciless slaughter (ac stragica caede), and to replace them by men from his own kingdom’.12 Many other examples show that well before 1900, let alone 1939, what we call genocide was in no way unthinkable. Rather, people described it using different terminology, for instance as ‘annihilation of a whole people’, ‘extermination’ or ‘generall massacre’, even ‘unpeopling’.13 The German term Völkermord, a synonym for ‘genocide’ which Lemkin also used but ultimately avoided, appeared as early as 1831.14 And of course in earlier times humans committed genocide in different ways, depending on variable historical, 11 See e.g. Frank Chalk and Kurt Jonassohn, The History and Sociology of Genocide: Analyses and Case Studies (New Haven: Yale University Press, 1990); Ben Kiernan, ‘Is “genocide” an anachronistic concept for the study of early modern mass killing?’, History 99:336 (July 2014), 442–60, https://doi/10.1111/1468-229X.12062. 12 Bede quoted in James E. Fraser, ‘Early medieval Europe’ in The Oxford Handbook of Genocide Studies, ed. D. Bloxham and A. D. Moses (Oxford University Press, 2010), pp. 259–79, at p. 261 and p. 268. 13 Eric D. Weitz, A World Divided: The Global Struggle for Human Rights in the Age of NationStates (Princeton University Press, 2019), pp. 59–60, at p. 97. In the 1850s whites in California often used the term ‘extermination’ during their genocide of California Indians; see Benjamin Madley’s Chapter 17 in Volume I I of this series. See also Kiernan, ‘Is “genocide” an anachronistic concept?’, and Ben Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007), p. 160. 14 Christian Gerlach, ‘Extremely violent societies: an alternative to the concept of genocide’, Journal of Genocide Research 8:4 (2006), 455–71, at 464.

4

General Editor’s Introduction to the Series

demographic and environmental circumstances, with earlier technological and organisational limitations. However, these long-established terms, especially if their meanings changed over time, could lack agreed precision or conceptual rigour. The Norwegian term for genocide, folkemord, has been traced back to 1846. But in the 1920s the philologist Torleiv Hannaas used it in a broader sense, to protest at the destruction of the ‘national language’ of a minority ethnic group (the Sámi) under the policy of Norwegianisation in the schools in the Finnmark county of Norway. Hannaas wrote: ‘Is this not language coercion? I have called it a folkemord. The word is strong. But not at all too strong. If the language is the people, it is genocide [folkemord] to kill an old national language.’15 Today we might call this ‘cultural genocide’. However, not all would agree that to ‘kill a language’ in itself amounts to genocide. As important as cultural genocide is, that is not the focus of The Cambridge World History of Genocide. To take a different case from the other side of the world: no word for ‘genocide’ is found in Cambodian dictionaries until as late as 1978. Nor, more surprisingly, after Cambodia had ratified the UN Genocide Convention in 1950, is the term found in the 1956 Cambodian Penal Code.16 In 1978 a Khmer term for ‘genocide’ did appear, in the English–Khmer Dictionary published by Yale University Press in that year: ជនឃាត (chun khéat) – ‘destruction of a people’.17 The Yale Dictionary also included a term for ‘exterminate’, tveu aoy son (‘make into nothing’). But neither of these terms came into popular usage for ‘genocide’ after the 1975–9 Khmer Rouge era. A new Khmer term emerged in 1979, and it quickly became both the official and popular word for ‘genocide’: prolay pouc sas ( ល័យពូជសាសន៍), literally, ‘destruction of the seed of a race’. This three-volume project, The Cambridge World History of Genocide, is unified, in part, by an agreed conception of the phenomenon it addresses. 15 Torleiv Hannaas, ‘Maaltvangen i Finnmark’ (c.1923), translated and quoted in Even Sebastian Skallerud, ‘Acts shocking to the conscience of mankind: why Norway voted to delete cultural genocide from the 1948 Genocide Convention’ (MA thesis, Faculty of Law, University of Oslo, 2019), p. 17. Thanks to Bernt Hagtvet for drawing this work to my attention. 16 These include the first published Cambodian dictionary, Joseph Guesdon’s 1930 Dictionnaire Cambodgien–Français; the 1938 Khmer-Khmer dictionary; Chuon Nath’s authoritative 1968 Buddhist Institute Khmer-Khmer dictionary, Vacana¯nukram khmaer; Robert Headley’s Cambodian–English Dictionary published by the Catholic University of America Press in 1977; and Franklin E. Huffman and Im Proum, Cambodian–English Glossary, published by Yale University Press in 1977. Thanks to Thavro Phim for research assistance. 17 Franklin E. Huffman and Im Proum, English–Khmer Dictionary (New Haven: Yale University Press, 1978), p. 249. The word khéat is listed in Guesdon’s 1930 Dictionnaire as meaning both ‘kill’ and ‘destroy’ (p. 338). The 1977 Glossary translates khéatekam as ‘murder’ (p. 22).

5

ben kiernan

There are now over twenty different definitions of genocide, both legal and sociological.18 Nonetheless, there is only one international, legal definition. Article II of the 1948 United Nations Convention on the Prevention and Punishment of the Crime of Genocide defines that crime as any of the following acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such: (a) Killing members of the group; (b) Causing serious bodily or mental harm to members of the group; (c) Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part; (d) Imposing measures intended to prevent births within the group; (e) Forcibly transferring children of the group to another group.19

The UN Genocide Convention also specifies different crimes, such as conspiracy, incitement and attempt to commit genocide, as well as complicity in genocide, and it adds that ‘Persons committing genocide or any of the other acts enumerated’ may include ‘constitutionally responsible rulers, public officials, or private individuals’. There are three parts to the core UN Convention definition of genocide itself: the mental element, or the ‘intent to destroy’; the physical element, or the ‘acts’ of genocide listed in Article II (a)–(e), quoted above; and the kinds of groups ‘protected’ by the Convention, namely ‘national, ethnical, racial [and] religious’ groups. The mental element is the most controversial and perhaps the least understood. Contrary to popular understandings that racial hatred must be the motive for any genocide, the word ‘intent’ does not refer to the perpetrator’s motive for the act or acts, but to their deliberate nature, what the perpetrator sets out to do. This means first that recklessness or criminal negligence do not qualify as ‘intent’ to commit genocide; the acts must be intentional. On the other hand, as the sociologist Helen Fein has shown, the motives for the acts may be varied: racial hatred, theft of the group’s possessions, seizure of its territory, ideological conviction, retribution, economic development, and despotic ambitions are all possible motives for genocide.20 However, under the Genocide Convention, no motive need be proven. The words ‘intent to destroy’ a group ‘as such’ do set a high burden of proof, which 18 Adam Jones, Genocide: A Comprehensive Introduction, 3rd ed. (London: Routledge, 2017), pp. 23–7. 19 See Human Rights Web, www.hrweb.org/legal/genocide.html (accessed 3 May 2021). 20 See Helen Fein, Genocide: A Sociological Perspective (London: Sage, 1993), pp. 28–9, table 1, for a listing of various typologies of genocides, including by motive.

6

General Editor’s Introduction to the Series

lawyers refer to as ‘specific intent’. This is the requirement that the group be attacked as that group, in other words, specifically for its identity, because of who they are. There must be a ‘conscious desire’ to destroy the group, or a substantial ‘part’ of it. That of course neither assumes nor precludes any reasons, or motives. These are irrelevant to the legal definition of the crime. The fact that for genocide to be committed, a group must be targeted ‘as such’, does not mean that the group must be targeted solely because of who they are.21 They may be targeted for other reasons as well. The legal definition of genocide is in this sense broader than the popular sense of the term, which often assumes racist motives, even exclusively racist motives. However, for many scholars of genocide, this legal definition remains insufficiently broad. Some, as we shall see, prefer to include ‘acts’ that are not physically or biologically destructive, now often referred to as ‘cultural genocide’, a term Lemkin was using by the 1950s for a phenomenon he had earlier encompassed under the term ‘vandalism’ – assaults on a people’s culture and heritage,22 acts that he had included in his own 1943–4 definition of genocide.23 Certainly, acts of cultural eradication, both the physical destruction of sites (for instance those in Bosnia in the 1990s)24 and linguistic and educational suppression, often form part of or accompany genocidal campaigns. As late as May 1948, a UN Ad Hoc Committee proposed that Article III of the forthcoming Genocide Convention include the following statement: In this Convention genocide also means any deliberate act committed with the intent to destroy the language, religion, or culture of a national, racial or religious group on grounds of the national or racial origin or religious beliefs of its members such as: 1) prohibiting the use of the language of the group in daily intercourse or in schools, or the printing and circulation of publications in the language of the group;

21 See Berel Lang, Genocide: The Act as Idea (Philadelphia: University of Pennsylvania Press, 2017), pp. 99–100. 22 Philippe Sands, East West Street: On the Origins of Genocide and Crimes against Humanity (London: Weidenfeld & Nicolson, 2016), p. 157. Sands credits the Romanian scholar Vespasian V. Pella with pioneering this usage of the term ‘vandalism’. For Lemkin’s use of the term ‘cultural genocide’, see Lemkin, Lemkin on Genocide, e.g. pp. 41, 74, 388–90. 23 Lemkin, Axis Rule in Occupied Europe, pp. xi–xii. 24 Uğ ur Ümit Üngör, ‘Cultural genocide: destruction of material and non-material human culture’ in The Routledge History of Genocide, ed. Cathie Carmichael and Richard C. Maguire (London: Routledge, 2015), pp. 241–53, at p. 247.

7

ben kiernan

2) destroying or preventing the use of libraries, museums, schools, historical monuments, places of worship or other cultural institutions and objects of the group.25

But this proposed article, and therefore acts of ‘cultural genocide’, were ultimately excluded. The only possible exception is Article II(e), ‘Forcibly transferring children of the group to another group’, an act which could be categorised as ‘cultural genocide’ because even under coercion it does often (but as we shall see, far from always) lead to the children surviving though being deprived of their indigenous culture. As for those particular groups who can be potential victims of genocide, yet other scholars prefer to include not only ethnic, national, racial and religious groups but also political groups and possibly even wider, less distinct groups, such as socioeconomic classes. In The Cambridge World History of Genocide we follow the UN Genocide Convention definition in excluding most cases of ‘cultural genocide’. We extend our historical coverage beyond that legal definition of genocide in only two directions. First, we include those group-selective cases of mass killing which fall under another legal definition, that of the crime of ‘extermination’, which was included in the 1998 Rome Statute of the International Criminal Court. This is an older and separate legal term, coming under the category of crimes against humanity, but one which largely overlaps with most sociological definitions of genocide. Extermination is legally described by the Rome Statute as conduct that ‘constituted, or took place as part of, a mass killing of members of a civilian population’ and was ‘committed as part of a widespread or systematic attack directed against a civilian population’.26 Extermination is a crime against humanity which includes not only massacres but, like genocide, also covers ‘the intentional infliction of conditions of life, inter alia the deprivation of access to food and medicine, calculated to bring about the destruction of part of a population’.27 The UN-sponsored Truth Commission for East Timor, for instance, found in 2005 that Indonesian

25 Quoted in Skallerud, ‘Acts shocking’, 24–6. 26 Official Records of the Assembly of States Parties to the Rome Statute of the International Criminal Court, First session, New York, 3–10 September 2002 (United Nations publication, Sales no. E.03.V.2 and corrigendum), part II.B; Article 7(1) (b) Crime against humanity of extermination. See International Criminal Court, www.icc-cpi.int. 27 Rome Statute of the International Criminal Court, Part 2, Article 7, paragraph 2 (b), p. 4, www.icc-cpi.int/resource-library/documents/rs-eng.pdf (accessed 28 March 2022).

8

General Editor’s Introduction to the Series

forces there had perpetrated ‘extermination as a crime against humanity’ in the period 1975–99.28 Like genocide (under the UN Genocide Convention), in the case of a crime against humanity the intentionality of the crime is important, though again the purpose or motive of the extermination is not relevant to guilt. Yet unlike genocide, to meet the definition of extermination the population or the part of it that is targeted need not be an ethnic, national, racial or religious group. Thus, extermination may cover political and social groups, as do most sociological definitions of genocide. Nor do charges of crimes against humanity such as extermination require proof, as the UN definition of genocide does, of ‘specific intent’, that is, the ‘intent to destroy’ a group, in whole or in part, ‘as such’. That higher level of intent is not required for the crime of extermination, though it too is a crime committed intentionally, not accidentally or without foreknowledge, as is made clear by the deployment in its legal definition of the terms ‘widespread or systematic’, ‘intentional’ and ‘calculated’. The legal definition of extermination clearly applies to most of those cases of mass murder that are treated as genocide by many genocide scholars yet are not covered by the UN Genocide Convention. While crimes against humanity (e.g. murder) may be committed against individuals as well as groups, this three-volume project emphasises the nature and historical occurrence of crimes against groups, ‘collectivities’ or communities that are encompassed by the terms ‘genocide’ and ‘extermination’.29 The focus of this project is on cases that fit either of these terms in their legal definitions, or those of sociologists, such as Helen Fein’s: ‘Genocide is sustained purposeful action by a perpetrator to physically destroy a collectivity directly[,] or indirectly, through interdiction of the biological and social reproduction of group members, sustained regardless of the surrender or lack of threat offered by the victim.’30 Second, we also include in our coverage cases of extreme violence that the sociologist Leo Kuper, writing in 1981, and others since, have categorised as ‘genocidal massacres’. This category, too, is not covered by the UN 28 Chega! The Report of the Commission for Reception, Truth and Reconciliation in Timor-Leste (CAVR), October 2005. See also Ben Kiernan, Genocide and Resistance in Southeast Asia: Documentation, Denial and Justice in Cambodia and East Timor (New Brunswick: Transaction, 2007), pp. 137–99, and Geoffrey Robinson’s chapter in Volume I I I of the series. 29 The crime of ‘extermination must be collective in nature rather than directed towards singled out individuals’. David Scheffer, All the Missing Souls: A Personal History of the War Crimes Tribunals (Princeton University Press, 2012), p. 435. 30 Fein, Genocide: A Sociological Perspective, p. 24.

9

ben kiernan

Genocide Convention. It describes shorter, more restricted outbreaks of killing that target a specific local or regional community because of its membership in a larger group.31 Nevertheless in some cases, perhaps because of a limited range of accessible evidence, it may remain unclear whether or not the perpetrators possessed an intent to destroy a particular group ‘as such’, even in part. The Genocide Convention’s limitation of its protections to ethnic, national, racial and religious groups seems in one sense a justified focus on those groups to which individuals usually have little choice in belonging, as distinct from membership of political or even social groups, which are covered by international law pertaining to crimes against humanity and by sociological definitions of genocide and ‘genocidal massacres’ that overlap with the legal concept of ‘extermination’. This is not to say that extermination of political, economic or social groups is anything other than a heinous offence, but merely that it is an even more heinous crime to perpetrate genocide against groups whose members can have had no choice in joining them. Even so, in some cases the perpetrators may blur the boundaries between political and ethnic groups. In Argentina in May 1977, General Ibérico Saint Jean defined the enemies of that country’s military dictatorship (1976–83), and specified how they should be treated during the ‘Dirty War’ it was waging against them. Saint Jean became governor of the important Buenos Aires province, after making this statement: ‘First we will kill all subversives, then we will kill all of their collaborators, then those who sympathize with subversives, then we will kill those that remain indifferent, and finally we will kill the timid.’32 In Saint Jean’s project, it seems, only strong supporters of the Argentine dictatorship were to be spared. It is difficult to see such a statement as simply the targeting of a ‘political group’ when the victims so easily encompassed ‘those that remain indifferent’ and ‘the timid’. As Federico Finchelstein writes, ‘Anti-Semitism, anticommunism, and the idea of the internal enemy as a non-Argentine Other’ were key elements in the junta’s fascist ideology.33 Daniel Feierstein and Lucrecia Molinari give more attention to this particular case in Volume I I I of the series. Compounding the 31 Leo Kuper, Genocide: Its Political Use in the Twentieth Century (New Haven: Yale University Press, 1981), pp. 59ff.; Chalk and Jonassohn, History and Sociology of Genocide, p. 26. 32 Quoted in Federico Finchelstein, The Ideological Origins of the Dirty War: Fascism, Populism, and Dictatorship in Twentieth-Century Argentina (Oxford University Press, 2014), p. 127. 33 Ibid., p. 122.

10

General Editor’s Introduction to the Series

dictatorship’s antisemitism, the ‘non-Argentine Other’ seems to have served as a fabricated ‘ethnic’ group, like the very large group of Spanish civilians that General Franco’s regime targeted as ‘anti-Spain’ during and after the Spanish Civil War, as Antonio Míguez Macho shows, also in Volume I I I.

Causes of Genocide So much for definitions of genocide. How can it be explained historically? This is not a matter of the phenomenon as a crime and its punishment, which can apply only to cases that occur in the period since the UN Genocide Convention came into force as international law in 1951 (and later as national law in some states). Historical explanations of genocide must focus not on its criminality but on its causes. The historical factors that give rise to genocide may usefully be divided into long-term and immediate causes.34 Possible long-term causes include war, poverty, economic depression or dislocation, cultural divisions, social crisis, political instability, mass migration, drastic climatic change or other factors that may destabilise sizeable previously stable communities. A brutal war, for instance, may generate the ascendancy of a genocidal group. This happened in Cambodia in the early 1970s, when the spread of the Vietnam War there facilitated the rise of the Khmer Rouge regime of Pol Pot.35 Something similar might have occurred with the Taliban’s resurgence and return to power in Afghanistan in 2021. ‘We have shot an amazing number of people, but to my knowledge, none has ever proven to be a threat,’ said General Stanley McChrystal, the senior US and NATO commander in Afghanistan in 2010. According to the New York Times, McChrystal was answering ‘questions from troops in the field about civilian casualties’ and was ‘seeking to reduce the killings of innocents’. The newspaper added that the persistence of such ‘deadly convoy and checkpoint shootings’, known as ‘escalation of force’ episodes, ‘has led to growing resentment among Afghans fearful of Western troops and angry at what they see as the impunity with which the troops operate – a friction that has turned villages firmly against the occupation’. The beneficiaries included Taliban insurgents. Sergeant Major Michael Hall, the senior NATO enlisted man in Afghanistan, told the troops: ‘There are stories after stories about how these people are turned 34 See Ben Kiernan, ‘Long-term and immediate causes of genocide: a response to Yehuda Bauer’, Yad Vashem Studies (online) 49:2 (2021), 79–91. 35 Ben Kiernan, ‘The American bombardment of Kampuchea, 1969–1973’, Vietnam Generation 1:1 (Winter 1989), 4–41.

11

ben kiernan

into insurgents. Every time there is an escalation of force we are finding that innocents are being killed.’ Even so, the casualties from such shootings were ‘fewer in number than deaths from air strikes and Special Forces operations’.36 Social, economic, demographic, environmental, political or, in this case, military destabilisation may enable an otherwise marginalised political group with possibly secret genocidal goals to recruit numerous supporters, build an army, seize power, impose its policies and violently enforce its will on a population. The immediate causes of genocide, on the other hand, are precisely those decisions made by people wielding the power to launch a genocidal campaign. The historian of Indonesia, Geoffrey Robinson, put this issue well when he described what I would classify as the possible long-term factors that contribute to genocide, such as ‘ancient cultural proclivities, deep-seated religious differences, or underlying socioeconomic conditions’. These ‘tensions, conflicts and hatreds’, he wrote, do not, ‘no matter how deeply rooted they may be’, lead naturally or inevitably to ‘mass violence and genocide’. Such long-term social and historical factors are usually necessary, but always insufficient to explain the emergence of genocide in a particular place and time. That outcome also requires immediate causes, usually the decisions individuals or small groups make because they wish to capitalise on the historic opportunities they face, and choose to mobilise human institutions for genocide. In Robinson’s view, ‘The turn to mass killing . . . requires something more – specifically, agents who articulate the idea that conflict should be resolved through violence, and institutions with the inclination and capacity to do so.’37 It is here that analysis of motives is most useful in explaining outbreaks of genocide. Historian Frank Chalk and sociologist Kurt Jonassohn have offered a typology of genocides, classifying them according to the dominant motive of the perpetrator in each case: whether ‘to eliminate a real or potential threat’, or ‘to spread terror among real or potential enemies’, or ‘to acquire 36 ‘Tighter rules fail to stem deaths of innocent Afghans at checkpoints’, New York Times, 27 March 2010, p. A4. See also Ben Kiernan and Taylor Owen, ‘Roots of U.S. troubles in Afghanistan: civilian bombing casualties and the Cambodian precedent’, Asia-Pacific Journal 8:26, no. 5 (28 June 2010), http://japanfocus.org/-Ben-Kiernan/3380; and Ben Kiernan and Taylor Owen, ‘Making more enemies than we kill? Calculating U.S. bomb tonnages dropped on Laos and Cambodia, and weighing their implications’, Asia Pacific Journal 13:17, no. 3 (27 April 2015), https://apjjf.org/BenKiernan/4313.html. 37 Geoffrey Robinson, ‘A time to kill: the anti-communist violence in Indonesia, 1965–66’ in Political Violence in Southeast Asia since 1945: Case Studies from Six Countries, ed. Eve M. Zucker and Ben Kiernan (New York: Routledge, 2021), pp. 21–40, at p. 32.

12

General Editor’s Introduction to the Series

economic wealth’, or ‘to implement a belief, a theory, or an ideology’.38 In a chapter entitled ‘Why Genocides?’ in their book Why Not Kill Them All?, historical sociologist Daniel Chirot and psychologist Clark McCauley discuss ‘the four main motives leading to mass political murder’, which they analyse as ‘convenience’, ‘revenge’, ‘simple fear’ and ‘fear of pollution’.39 Historians and sociologists recognise genocide perpetrators’ array of motives as being widely variable. Similarly, in his book Revolution and Genocide: On the Origins of the Armenian Genocide and the Holocaust, the political scientist Robert Melson usefully distinguished, on the basis of the perpetrators’ motive, between ‘total genocide’ (or genocide ‘in whole’ under the UN Convention) and ‘partial genocide’ (genocide ‘in part’ under the Convention). Melson writes: ‘The attempted total genocides of Armenians, Jews, and Gypsies’ aimed ‘to transform the social structure by physically and socially eliminating a communal group or class from society’. These ‘total genocides’ were not ‘mass murders designed to terrorize a civilian population, or to turn it into a mass of helots, or to intimidate an upwardly mobile or recalcitrant ethnic group, or to repress a political movement’. The latter motives, Melson argues, more often lie behind ‘partial genocides’.40 Chirot and McCauley also assess how, having made their decision to unleash genocide, the perpetrators ‘get ordinary people to become butchers’. This process involves skilful organisation, the resort to emotional appeals, including fear, anger, hate, love, shame and humiliation, disgust and special techniques for ‘essentializing others’, including deliberate targeting of ‘the dangerous similar others’.41 Genocide involves careful planning. Well-documented modern cases allow historians to look closely at how a genocide immediately unfolds, but it is also useful to keep in mind the longterm factors that make genocide possible, or more likely, but not necessarily inevitable. For the sake of argument, let us take a prehistoric example: the extinction of the world’s Neanderthal population around 40,000 B C E. According to a recent study, ‘The flipping of the Earth’s magnetic poles together with a drop in solar activity 42,000 years ago could have generated an apocalyptic environment that may have played a role’ in major events 38 Chalk and Jonassohn, History and Sociology of Genocide, p. 29. 39 Daniel Chirot and Clark McCauley, Why Not Kill Them All? The Logic and Prevention of Mass Political Murder (Princeton University Press, 2006), pp. 19–45. 40 Robert Melson, Revolution and Genocide: On the Origins of the Armenian Genocide and the Holocaust (University of Chicago Press, 1992), p. 28. 41 Chirot and McCauley, Why Not Kill Them All?, pp. 52–89.

13

ben kiernan

including those that brought about ‘the end of the Neanderthals’. A large number of major environmental changes had taken place at around the same time. As one of the researchers, earth scientist Professor Chris Turney of the University of New South Wales, pointed out: ‘We see this massive growth of the ice sheet over North America . . . tropical rain belts in the west Pacific shifting dramatically at that point, and then also wind belts in the southern ocean and a drying out in Australia.’ Global climate change combined with a more immediate factor, that of human interaction and potentially violent competition with the Neanderthals: ‘the rise in the use of caves by our ancestors around this time, as well as the rise in cave art, might be down to the fact that underground spaces offered shelter from the harsh conditions. The situation may also have boosted competition, potentially contributing to the end of the Neanderthals’.42 No matter how difficult the background physical or social conditions, human (literally in this case) decisions and actions also seem to have played a significant role. On the other hand, we must be careful not to confuse competition, conflict or even war with genocide, even when they might lead to extinction. Genocide requires a level of premeditation and deliberation that is far from evident in our sources on many such cases. This is true not just of prehistoric but also of early historic and even some medieval societies, where we do have written sources and the evidence suggests that conflicts routinely failed to rise to the level even of warfare, let alone genocide. For instance, the Irish historian D. A. Binchy has warned us not to judge the wars in early medieval Ireland ‘by modern standards’. Meanings of the term ‘war’ may have changed over time, or may imply different processes in earlier or different societies. Binchy wrote some decades ago that surprisingly few of the so-called ‘wars’ that are regularly chronicled in the Annals, involved either conquest or occupation or dethronement of the old tribal dynasty. Some of them, in fact, were nothing more than punitive expeditions by an over-king to levy by force tribute which had been withheld by a subordinate. And the great majority of the remainder were just examples of the crech, the armed raid for booty, chiefly livestock. These tipand-run affairs were the small change of inter-tribal relations, a sort of competition between the aristocratic warriors on both sides of the frontier.43 42 Nicola Davis, ‘End of Neanderthals linked to flip of Earth’s magnetic poles, study suggests’, Guardian (UK edition), 18 February 2021, https://bit.ly/3r3hHlR (accessed 28 March 2022). 43 D. A. Binchy, ‘Secular institutions’ in Early Irish Society, ed. Myles Dillon (Dublin: Cultural Relations Committee of Ireland, 1954), pp. 52–65, at p. 65.

14

General Editor’s Introduction to the Series

Genocide, occasionally possible to detect even in the prehistoric era, as we see for instance in Helle Vandkilde’s chapter in this volume, and in the early historic era, would become more frequent only with territorial conquests and the rise and expansion of empires. Even then, the multicultural composition of most empires often militated against the targeting of ethnic, national, racial or religious groups. Nevertheless, ethnic prejudices did exist in some ancient societies, and the later rise of racial thinking contributed greatly to the frequency of genocide.

The Nature and Limits of Racism Racialist theories proclaim a special status for a group or groups they construct as superior, and they often propel violence against target groups. In practice, however, racists are frequently inconsistent and at times even surprisingly ineffective at advancing the interests of members of their supposedly superior group. Most importantly, racists rarely provide protection for all members of the supposedly protected race, even if it is supposedly superior. Racism as a primary cause of genocide more often demands unending racial conflict, usually at deadly cost to both sides. In the maelstrom of the Holocaust of the Jews, Hitler and his minions also targeted, with lesser intensity though in significant numbers, not only Romani or Sinti but other German victims such as homosexuals, the disabled, communists, Jehovah’s Witnesses, liberals, trade unionists and other German oppositionists. In the Nazi purge of German culture, books and paintings were burned, literary and film criticism abolished and modern music banned.44 The day after the 1938 Kristallnacht pogrom against Jews, for instance, Hitler also speculated that he ‘might one day exterminate the intellectual classes in Germany if they no longer proved to be of use’.45 It was always clear that large numbers of Germans would perish in the aggressive wars of conquest that the Nazis launched the next year. These murders and foreseeable deaths were not cases of genocide, but their causes were related to the racialist thinking that spurred the Nazi genocides. For example, during the Nazi invasions of Poland and the USSR, German forces removed over three-quarters of a million ethnic Germans from their homes and placed them in camps, according to the historian Richard Evans. 44 Jeffrey Richards, Visions of Yesterday (London: Routledge, 1973), p. 292. 45 Richard Evans, In Hitler’s Shadow (New York: Pantheon, 1989), p. 81, citing Ernst Nolte.

15

ben kiernan

The Schutzstaffel (SS) investigated and interrogated them. ‘Those who were deemed “racially unfit,” or who were opponents of Nazism, or [who were] disliked by the SS for some other reason, were handed over to the Gestapo to be killed.’ Thousands more were mistreated for months before the SS decided to send them either to Germany or to the east. The result, Evans wrote, was that ‘numerous villages and other areas of German settlement were emptied, destroyed, or turned over to the Poles’. He concluded that the Nazi definition of a ‘German’ or a ‘Slav’ was ‘arbitrary and elastic’.46 As Jeffrey Richards pointed out, this illustrates ‘the rejection of the individual in favour of the race’, by no means a privileging of individuals for their membership of the favoured race.47 Such privileging of the race over its members meant that German soldiers were expendable. On 20 December 1941, after being warned by General Heinz Guderian that digging in on the Russian front would lead to enormous German losses, Hitler responded in the presence of his entire military entourage that Frederick the Great’s troops had ‘wanted to live, too, but the king was right in asking them to sacrifice themselves. I believe that I, too, am entitled to ask any German soldier to lay down his life’.48 The German commander whom Hitler considered his best field marshal, Walter Model, also believed in Germany’s superior culture and racial supremacy in Europe, but, as with Hitler, that belief by no means extended to a loyalty to all German people. As late as 18 March 1945, as Nazi Germany’s defeat loomed, Model ordered ruthless punishment of ‘inferior elements in the civilian population’.49 On that very same date, the day before Hitler’s ‘scorched earth’ decree for the destruction of ‘all material assets within the Reich territory which the enemy can render usable’, Hitler told his minister of armaments and war production, Albert Speer: ‘If the war is lost, then the people too is lost. This fate is irreversible . . . the people had shown itself to be the weaker . . . What will remain after this struggle will be in any case only the inferior ones, since the good ones have fallen.’50 It is perhaps not surprising, then, that during World War Two, German military casualties alone rivalled even the enormous continental Jewish death toll of nearly 6 million.51 46 Ibid., p. 95. 47 Richards, Visions of Yesterday, p. 289. 48 Quoted in Ian Kershaw, Hitler 1936–45: Nemesis (New York: W. W. Norton, 2000), p. 454. 49 Quoted in Ian Kershaw, The End: The Defiance and Destruction of Hitler’s Germany (New York: Penguin, 2011), p. 304. 50 Quoted ibid., p. 290. 51 German historian Rüdiger Overmans has estimated that the German military casualties numbered 5.3 million, including 900,000 men conscripted from outside of Germany’s 1937 borders, in Austria and in east-central Europe: Wikipedia, https://en.wikipedia.org/wi

16

General Editor’s Introduction to the Series

This toll does not include the large number of non-Jewish German civilians whom the Nazis killed. According to the post-war West German government, 140,000 non-Jewish Germans (and 160,000 German Jews) died from Nazi racial, political and religious persecution. The Nazi eugenics programmes murdered 200,000 Germans with disabilities. The number of ethnic Germans killed after the war, in reprisals by states and populations that had been targeted by the Nazi regime during the war, is estimated by the postwar German government at 2.2 million.52 Many Germans seem to have been aware of the likelihood of Nazi racial violence turning back on them. After the severe Allied bombing of Hamburg in 1943, one German resident of that city speculated in a letter to another that Hamburg’s destruction was ‘retaliation for our treatment of the Jews’, adding that other ‘bombed out’ residents were also concerned that ‘if we hadn’t treated the Jews so badly we wouldn’t have had to suffer so much from the terror attacks’.53 In November 1944, after the Hitler regime highlighted Soviet atrocities against Germans on the Eastern Front, the Stuttgart office of the SD (the intelligence agency of the SS and the Nazi Party) reported that German people were describing the regime’s press stories as ‘shameless’. The Nazi leadership, the SD office commented, must surely know that every thinking person . . . will immediately contemplate the atrocities that we have perpetrated on enemy soil, and even in Germany. Have we not slaughtered Jews in their thousands? Don’t soldiers tell over and again that Jews in Poland had to dig their own graves? And what did we do with the Jews who were in the concentration camp in Alsace? The Jews are also human beings. By acting in this way, we have shown the enemy what they might do to us . . . We can’t accuse the Russians of behaving just as gruesomely towards other peoples as our own people have done against their own Germans . . . After all, what does human life amount to here in Germany[?]54

ki/German_casualties_in_World_War_II (accessed 20 February 2021). Another million German prisoners of war died in Soviet hands. Allied bombing of Germany killed nearly half a million people, mostly civilians. Kershaw, The End, pp. 375, 236. 52 Wikipedia, https://en.wikipedia.org/wiki/German_casualties_in_World_War_II (accessed 20 February 2021). 53 Quoted in Dietmar Süss, Death from the Skies (Oxford University Press, 2013), pp. 99–100; see Aaron William Moore, Bombing the City: Civilian Accounts of the Air War in Britain and Japan, 1939–1945 (Cambridge University Press, 2018), pp. 1–2. 54 Report to SD-Leitabschnitt Stuttgart, 6 November 1944, quoted in Kershaw, The End, p. 117.

17

ben kiernan

Racism is an ideology of preference for the idea of a racial group, but not always a practical programme to advance the interests of its members. It is often closely associated with idealist thinking rather than the delivery of material benefits for, or even the survival of, its proclaimed beneficiary race. SS leader and Holocaust architect Heinrich Himmler’s rallying cries to his divisional commanders, circulated among the German troops, specifically stressed idealism over materialism: ‘strong hearts triumph over mass and matériel’. Field Marshal Model demanded quick repression of German civilian groups ‘infected by Jewish and democratic poison of materialist ideas’. Some Nazi soldiers echoed these Nazi idealist themes: ‘Spirit against material. It has never yet happened that mere technology has conquered spirit.’55 A parallel with the Maoist preference for ‘red versus expert’ soon comes to mind. In this respect, the far left aligns with the far right. The political spectrum turns out to be more like a horseshoe.56 An equally illuminating example of idealism trumping materialism comes from Cambodia, in the form of the Pol Pot regime’s selective rendition of Stalin’s 1938 work, Dialectical and Historical Materialism. This text was republished in 1951 while the young Pol Pot was a student in Paris. A quarter of a century later, in 1976, his Communist Party of Kampuchea (CPK) journal Tung Padevat (Revolutionary Flags) published what it called a ‘Review of Dialectical Materialism’. The article summarised the first of the three parts of Stalin’s work, namely the section on the dialectical method, but it totally omitted the second and third parts, those on philosophical materialism and historical materialism. Unlike orthodox Marxism, then, the CPK’s version of communism, even when directly borrowed from Stalin, had no place for ‘materialist’ ideology. Like Hitler’s Nazis, it subsumed material conditions under the category of ‘technology’ and ranked them as inferior to human energy and willpower. ‘Technology is not a determining factor. The determining factors of the revolution are politics, revolutionary people and the revolutionary system.’ Pol Pot’s CPK espoused a voluntarism that emphasised human will struggling against the tide of material or ‘objective’ conditions: ‘We cannot just let something go its own way . . . Do not put the blame on the objective [conditions]’.57 A senior CPK cadre wrote in 1978: ‘Ever since 1957 our Party has stood on the standpoint of politics . . . We did not stand on 55 Quoted in Kershaw, The End, pp. 223, 304, 71. 56 I thank Ian D. Black for first acquainting me with this simile. 57 Tung Padevat, September–October 1976, quoted extensively in Ben Kiernan, ‘Kampuchea and Stalinism’ in Marxism in Asia, ed. Colin MacKerras and Nick Knight (London: Croom Helm, 1985), pp. 232–50.

18

General Editor’s Introduction to the Series

technology as was the experience [all] over the world, whether in the area of the imperialist Western countries or in the area of revolution.’58 This was CPK idealism in the sense of voluntarism. As in the Nazi case, such antimaterialist idealism proved very compatible with racism, as several chapters in Volume I I I of The Cambridge World History of Genocide demonstrate.

Gender Prejudice in Genocidal Ideology The ancient Roman senator Marcus Porcius Cato warned in 195 B C E: ‘There is the greatest danger from any class of people, once you allow meetings and conferences and secret consultations.’ Cato later became a major inspiration behind the genocidal Roman destruction of Carthage in 146 B C E. He would continually urge the Senate, ‘Delenda est Carthago’ (‘Carthage must be destroyed’), as the historian Tristan Taylor recounts in Chapter 10 of this volume. But when he warned against ‘secret consultations’ in 195 B C E, Cato was opposing the repeal of a wartime law denying women the right to ‘possess more than half an ounce of gold, or wear parti-coloured clothing, or ride in a horse-drawn vehicle in a city or town’. Increasing numbers of women clamoured for repeal of this law, the classical historian Livy tells us. Women ‘came in from the towns and rural centres’ and ‘beset all the streets of the city and all the approaches to the Forum’. Cato found himself asking them: ‘Are you in the habit of running out into the streets, blocking the roads, addressing other women’s husbands? . . . And yet, even at home . . . it would not become you to be concerned about the question of what laws should be passed or repealed in this place.’ Cato saw politicised women as a threat to the Roman republic. Our liberty, overthrown in the home by female indiscipline, is now being crushed and trodden underfoot here, too, in the Forum. It is because we have not kept them under control individually that we are now terrorised by them collectively . . . But we (heaven preserve us) are now allowing them even to take part in politics, and actually to appear in the Forum and to be present at our meetings and assemblies! . . . What they are longing for is complete liberty . . . The very moment they begin to be your equals, they will be your superiors.

He denounced the female throng as an ‘untamed animal’, a ‘secession of the women’. His misogyny was quite compatible with his later calls for the 58 Diary entry dated 18 August 1978, photographed by the author in Tuol Sleng prison in 1980.

19

ben kiernan

destruction of Carthage. But the phenomenon was bigger than Cato himself. Plutarch explained that the Senate even launched an invasion of Dalmatia in 156 largely ‘because they did not want the men of Italy to become womanish through too lengthy a spell of peace’.59 From another era altogether, a different kind of gendered violence is evident in the genocidal thinking of the English Protestant planter Sir Charles Coote, military governor of Dublin in 1641–2. In December 1641, Coote was twice alleged to have urged ‘a generall massacre’ of Catholics in Ireland.60 In addition, according to a manuscript account from the 1650s, ‘that human blood sucker Sir Charles Coote gave his opinion once in the Councell table in Dublin . . . that all the Irish women should be deprived of their papps, and the men gelded, to render the one incapable of future generation and the other of nurishinge’.61 In Hitler’s Germany, the historian Claudia Koonz wrote, ‘Women, who had won suffrage and equal rights in 1919, were relegated to a separate “feminine” sphere.’62 Wendy Lower, in her book Hitler’s Furies, quotes Hitler’s address to a 1934 Nazi Party rally in Nuremberg: ‘If the man’s world is said to be the State, his struggle, his readiness to devote his powers to the service of the community, then it may perhaps be said that the woman’s is a smaller world. For her world is her husband, her family, her children, and her home.’ In a speech to the National Socialist Women’s League that same year, Hitler reiterated that ‘neither sex should try to do that which belongs to the sphere of the other . . . Whereas previously the programs of the liberal, intellectualist women’s movements contained many points, the programme of our National Socialist Women’s movement has in reality but one single point, and that point is the child, that tiny creature which must be born and grow strong.’63 In her study Women and the Nazi East, 59 See Kiernan, Blood and Soil, pp. 55–6. 60 For more details on Coote’s words and deeds, see Ben Kiernan, ‘Settler colonies, ethno-religious violence, and historical documentation: comparative reflections on South-East Asia and Ireland’ in Ireland, 1641: Contexts and Reactions, ed. Micheál Ó Siochrú and Jane Ohlmeyer (Manchester University Press, 2013), pp. 254–73, at pp. 262–3, 271n.44. 61 Trinity College Dublin MS 846, ‘R. S.’, An Aphorismicall Discovery of Treasonable Faction, ch. 6, para. 52, p. 8. See also the printed version in John T. Gilbert, ed., A Contemporary History of Affairs in Ireland from 1641 to 1652, 3 vols. (Dublin: Irish Archaeological and Celtic Society, 1879), I, p. 32. 62 Claudia Koonz, The Nazi Conscience (Cambridge, MA: Harvard University Press, 2003), p. 71. 63 Wendy Lower, Hitler’s Furies: German Women in the Nazi Killing Fields (Boston: Houghton Mifflin Harcourt, 2013), p. 22; and ‘Hitler’s Speech to the National Socialist Women’s League (September 8, 1934)’, German History in Documents and Images, https://ghdi.ghi-dc.org/docpage.cfm?docpage_id=2330 (accessed 26 March 2021).

20

General Editor’s Introduction to the Series

the historian Elizabeth Harvey concurred that the Nazi leaders viewed women’s place in the public domain as one ‘subordinate to that of men’, adding: ‘the model household of Nazi fantasy was the peasant family centred upon a mother who was as competent in modern housekeeping as she was at cultivating folk tradition’.64 The head of the Nazi Party’s Office of Racial Politics, Walter Gross, lectured women on the theme of ‘You are nothing. Your Volk is everything’, and congratulated them on being ‘little drops in the mighty bloodstream of the Volk’.65 Gender as well as race also shaped patterns of mass killing decades later in Southeast Asia. For the first two years after the Pol Pot regime took over Cambodia in 1975, it targeted that country’s ethnic Vietnamese minority community for forcible expulsion to Vietnam. But ‘from April 1977’, the UNsponsored tribunal found in 2018, nearly all Cambodia’s remaining ethnic Vietnamese were murdered, designated ‘for destruction as such’. ‘In each case,’ the court found, ‘Vietnamese were targeted not as individuals but based on their membership of the group and their perceived ethnicity.’ These crimes rose to the level of genocide, extermination and the additional crime against humanity of persecution on racial grounds. Yet a gender prejudice was also in operation. It targeted in particular ethnic Vietnamese women, and their children, and it did so even more viciously than the deadly racial prejudice aimed at ethnic Vietnamese men. In cases of mixed-race families, the tribunal found that the Khmer Rouge had targeted ‘Vietnamese mothers and their children while sparing Khmer fathers’, but when the ethnicity of the spouses was reversed, the Khmer Rouge ‘targeted Vietnamese fathers while sparing Khmer mothers and children’.66 In other words, the children of Vietnamese men were spared, but the children of Vietnamese women were murdered with their mothers. This Khmer Rouge criterion of matrilineal descent is an example of the arbitrary application of gender as a way of determining who will live and who will die in a genocide. The tribunal convicted Pol Pot’s deputy, Nuon Chea, and their regime’s head of state, Khieu Samphan, of the genocide of Cambodia’s ethnic Vietnamese minority in 1977–9. But that was not the only case of persecution of women in which Nuon Chea was involved. Before his arrest and trial, 64 Elizabeth Harvey, Women and the Nazi East: Agents and Witnesses of Germanization (New Haven: Yale University Press, 2003), p. 9. 65 Quoted in Koonz, Nazi Conscience, pp. 114–15. 66 Extraordinary Chambers in the Courts of Cambodia (ECCC), Trial Chamber, Summary of Judgment, Case 002/02, 16 November 2018, 15; Case 002/02 Judgment, 27 March 2019, Case File/Dossier No. 002/19-09-2007/ECCC/TC, 581–92, 1699–1782, esp. 1740–2, 1782, 2164–5.

21

ben kiernan

Chea told a Cambodian interviewer who had won his confidence why the Pol Pot regime had instituted a nationwide system of forced marriage. ‘The man always wants to choose a beautiful girl, so that’s why we forced them to get married and Angkar [the revolutionary organisation] chose the wife.’ Note the up-front totalitarianism here. After critiquing the lack of choice for women in traditional Cambodian society, the Khmer Rouge regime denied both bride and groom any choice in their marriage partner. Its proclaimed egalitarianism involved ending rather than expanding individual freedom. As his interviewer reported, ‘Nuon Chea thought the purpose of marriage should be to serve the movement.’67 The cultural studies scholar Raya Morag has suggested that ‘the crimes of forced marriage and rape epitomize the perverse unconscious of the Cambodian revolution’.68 In Chapter 3 of this volume, political scientist Adam Jones and historian Wendy Lower provide a detailed analysis of the dynamics and consequences of the relationship between genocide and gender.

Perennial Themes of Genocide Another perennial theme of genocide is the inclination of its perpetrators to turn to ancient models to justify their actions, including mass murder in the pursuit of territorial expansionism. In his book Genocide: A World History, Norman Naimark describes ‘one of the most powerful stories of genocide in the Old Testament and, indeed, in human history’. This was the repeated biblical condemnation of the Amalekites, ‘a semi-nomadic people of the desert’. Even after Joshua and his army had fought off the Amalekites’ attacks and vanquished them, the Hebrew Bible recorded that God told Moses: ‘I will utterly put out the remembrance of Amalek from under heaven.’ And Moses duly reported: ‘The L O R D will have war with Amalek from generation to generation.’ Then, generations later, the prophet Samuel approached Saul, the designated future king of Israel, and explained God’s orders: ‘Now go and smite Amalek, and utterly destroy all that they have, and spare them not; but both man and woman, infant and suckling, ox and sheep, camel and ass.’ The book of Samuel states that Saul assembled an Israelite army, attacked the Amalekites in their capital, and ‘utterly destroyed all of the people with the edge of the sword’, sparing only their king, Agag, and allowing the 67 Gina Chon and Sambath Thet, Behind the Killing Fields: A Khmer Rouge Leader and One of His Victims (Philadelphia: University of Pennsylvania Press, 2010), p. 41. 68 Raya Morag, Perpetrator Cinema: Confronting Genocide in Cambodian Documentary (New York: Columbia University Press, 2020), p. 136.

22

General Editor’s Introduction to the Series

Israelites to keep the best of the enemy’s livestock. Yet God was angered that Saul had not fully obeyed him. So Samuel, in God’s presence, ‘hewed Agag in pieces’ and punished Saul by depriving him of the throne of Israel. As Naimark comments, ‘Saul had not followed God’s injunction to finish the job of genocide’.69 In Chapter 6 of this volume T. M. Lemos demonstrates the ‘abundance of literature depicting genocide’ in biblical sources, and shows how it must be carefully assessed. But the literary fate of biblical Amalek indeed proved a powerful model in human history. In 1643, several years after English settlers had conducted their genocide of Connecticut’s Pequot Indians (described in detail by historian Benjamin Madley in Volume I I of this series), ‘a group of Harvard graduates’ turned for a favourable precedent to the references to Amalek in the Book of Samuel. They boasted that ‘the name of the Pequots (as of Amaleck) is blotted out from under heaven, there being not one that is or (at least) dare call himself a Pequot’.70 Another perennial theme of genocide, related to this reverence for ancient models, is a romanticisation of agrarian life. Perpetrators often see farm cultivation as the original, pure, most appropriate lifestyle for their favoured populations, as opposed to nomadic, hunter-gathering or urban lifestyles. Thus, as Tristan Taylor points out in Chapter 11 of this volume, Julius Caesar criticised the ancient Germani peoples for their limited agriculture: ‘There is no private or separate use of land among them, nor do they remain for longer than a year in one place for cultivating land. They do not use much corn, but mostly they live off milk and cattle.’71 A more passionate, reactionary version of agrarianism formed an important part of Adolf Hitler’s racialist thinking. Hitler regarded the German peasantry as ‘the solid backbone of the nation’. By contrast, modern industrial cities were ‘abscesses on the body of the people (Volkskörper) – places where all evils, vices, and sicknesses appear to unite. They are above all hotbeds of blood-mixing and bastardization’.72 In a speech broadcast to the German nation on 18 October 1944, Himmler announced that Hitler had called on his people to defend the soil of their homeland. Preserving Germandom meant in particular a strong defence of ‘our woods’ and ‘our villages’, which Himmler saw as the essence of ‘our people’: ‘We have heard 69 Norman M. Naimark, Genocide: A World History (Oxford University Press, 2017), pp. 8–9. 70 Quoted in Geoffrey Parker, Global Crisis: War, Climate Change, and Catastrophe in the Seventeenth Century (New Haven: Yale University Press, 2013), p. 450. 71 Caes. BGall. 4.1. 72 Quoted in Kiernan, Blood and Soil, p. 431.

23

ben kiernan

from their own mouths,’ Himmler declared, ‘that we have to expect from our enemies the destruction of our country, the cutting down of our woods, the break-up of our economy, the destruction of our towns, the burning down of our villages and the extirpation of our people.’73 Similarly, in early 1999, before his imprisonment and trial for crimes against humanity and genocide, Nuon Chea visited his home village in northwest Cambodia for the first time since his regime had held power from 1975 to 1979. He was disappointed to find that new, denser settlement and new houses had modernised what he nostalgically recalled as once serene, lush pastoral scenes of fields and grass. ‘It was quiet then but the village looked beautiful because it was not dusty and there were no cars and motorbikes. Everything in the past reflected Cambodian tradition. I really loved it. But now it has disappeared and it’s too crowded . . . I don’t like it.’74 Chea preferred the rural idyll of the past, a country whose cities had been depopulated and their residents subjected to mass murder.

Connections between Genocidal Regimes The persistence of the biblical metaphor of the fate of Amalek is only one comparatively tenuous intellectual link between genocidal regimes through the centuries. But connections between various genocides also emerge from broader historical patterns, and in some cases the evidence reveals institutional and personal parallels, and even direct personal links. For the early modern period, the chapters by historians Ned Blackhawk and Rebe Taylor in Volume I I of the series illustrate many of the more material connections between the history of settler colonialism and outbreaks of genocide in both North America and Australia. In Southeast Asia, indirect if not personal links and lessons drawn connect the mass murders of the 1960s in Indonesia and those in Cambodia in the 1970s, and there is a possible political association between the genocide of Muslims in Cambodia and that which followed in Burma.75 Twentieth-century genocides in Africa, the Middle East and Europe are yet more closely linked by personal connections. In his article ‘From Africa to Auschwitz’, Benjamin Madley points out that the German terms Lebensraum (‘living space’) and Konzentrationslager (‘concentration camp’), well known 73 Himmler quoted in Kershaw, The End, p. 106. 74 Nuon Chea quoted in Chon and Thet, Behind the Killing Fields, p. 149. 75 Zucker and Kiernan, eds., ‘Introduction’ to Political Violence in Southeast Asia since 1945, pp. 1–18, at pp. 5–7.

24

General Editor’s Introduction to the Series

from their use by the Nazis, were in fact first ‘minted years earlier in reference to German South West Africa’, when German colonists ‘committed genocide against the local Herero and Nama peoples’ in the early twentieth century, the subject of a chapter by Leonor Faber-Jonker in Volume I I of the series. Ominously, Herero and Nama victims were ‘sent to concentration camps in cattle cars, tattooed and issued with numbers’.76 Moreover, Madley adds that the later Nazi figures Hermann Göring, Franz Ritter von Epp and Eugen Fischer had all ‘served as human conduits for the flow of ideas and methods between the colony and Nazi Germany’. Göring’s father, Heinrich Göring, had been the first Reichskommissar for German Southwest Africa in 1885–91, ‘suppressing African revolts, deploying German troops, signing treaties, working to occupy hundreds of thousands of square miles, and frequently flogging Africans’. At his post-war Nuremberg trial, Hermann Göring listed ‘the position of my father as first Governor of Southwest Africa’ as one of four key ‘points which are significant with relation to my later development’. For his part, Franz Ritter von Epp, who served in Southwest Africa in 1904–6 and participated as a company commander in the Herero genocide, later ‘played a crucial role in developing the Nazi Party’. Von Epp ‘helped launch Hitler’s political career’ and in 1933 Hitler appointed him governor of Bavaria, a post he held until 1945. He presided over the construction of Dachau concentration camp and the extermination of Bavarian Jews and Romani.77 The anthropologist Eugen Fischer arrived in German Southwest Africa in 1908, and conducted a racialist study of mixed-race children there. In its 1913 publication, Fischer wrote that people of mixed race were ‘inferior’ and should be protected ‘only for so long as they are useful to us. Otherwise free competition, which means in my opinion, their extinction!’ In the words of historian Henry Friedlander, Fischer’s work ‘influenced all subsequent German racial legislation, including the Nuremberg Laws’ passed by Hitler’s regime in 1935. Later, as Madley has shown, Fischer himself played ‘direct roles in programmatic Nazi racial violence’.78

76 Kavena Hambira and Miriam Gleckman-Krut, ‘Germany apologized for a genocide. It’s nowhere near enough’, New York Times, 8 July 2021. 77 Benjamin Madley, ‘From Africa to Auschwitz: how German South West Africa incubated ideas and methods adopted and developed by the Nazis in eastern Europe’, European History Quarterly 35:3 (2005), 429–64, at 429–30, 450–8. See also Jürgen Zimmerer, ‘Colonialism and the Holocaust’ in Genocide and Settler Society, ed. A. Dirk Moses (New York: Berghahn, 2004), pp. 49–76. 78 Madley, ‘From Africa to Auschwitz’, 453–6.

25

ben kiernan

Meanwhile, in the 1910s, a significant number of German officials stationed in the Ottoman Empire during the Armenian genocide (discussed in historian Taner Akçam’s chapter in Volume I I I of the series) also later rose to prominence in Nazi Germany. Franz von Papen, chief of the general staff of the Fourth Turkish Army during World War One, served in 1933–4 as Hitler’s vice chancellor and as president of Prussia, then became special ambassador to Austria and prepared its annexation by the German Reich in 1938. Konstantin Freiherr von Neurath was councillor at the German Embassy in Constantinople in 1915–16 charged with monitoring the Young Turk regime’s operations against Armenians. He later served as foreign minister in Hitler’s cabinet and then as protector of Bohemia and Moravia after the German occupation of Czechoslovakia. Other Nazi leaders who had also served in the Ottoman Empire during World War One included Karl Dönitz, who became supreme commander of the German Navy in World War Two and, briefly, Hitler’s successor; General Alexander von Falkenhausen, military governor in Belgium, 1940–4; and Rudolf Höss, commandant of Auschwitz in 1940–3.79 Colonial and wartime genocides leave long legacies. These few examples illustrate the importance of ideological, institutional, historical and personal relationships between the German Southwest African and Ottoman Armenian genocides and the later Nazi Holocaust, as well as the more general connections that it is possible to make between cases of colonialism and genocide.

Continuing Challenges Sadly, in recent decades not only has genocidal violence proven to have been more widespread still, but also, its accelerating recognition continues to meet resistance. Facing pages of the New York Times on a single day, 29 May 2021, contained stories from different continents. One headline ran, ‘Germany officially recognizes mass killings in colonial-era Namibia as genocide’. The previous day, Berlin had agreed to offer an apology for the crimes, mentioned above, committed in German Southwest Africa in 1904–8, when German soldiers ‘targeted people of two ethnic groups – the Herero and the Nama’.80 The other headline announced, ‘Mass grave may contain Canadian Indigenous youth’. That story began: ‘For decades, most Indigenous children 79 Vahakn N. Dadrian, German Responsibility in the Armenian Genocide (Cambridge, MA: Blue Crane Books, 1996), pp. 199–203. 80 ‘Germany officially recognizes mass killings in colonial-era Namibia as genocide’, New York Times, 29 May 2021, p. A6.

26

General Editor’s Introduction to the Series

in Canada were taken from their families and forced into boarding schools. A large number never returned home.’ A mass grave had been located on the grounds of one such former residential school in British Columbia that is believed to contain the remains of 215 children.81 Later, two more unmarked mass graves were found, reportedly containing the remains of 933 people near the sites of other former residential schools, in Saskatchewan and British Columbia.82 Over more than a century beginning in 1883, Canada had separated up to 150,000 Indigenous children from their families and sent them to government-funded, church-run schools in an effort to assimilate them. More than 4,000 of the children died.83 In 2015 a Canadian Truth and Reconciliation Commission found that these residential schools, sometimes using violence, had prohibited Indigenous languages and cultural practices, and that the programme constituted a case of ‘cultural genocide’. It might well have been more than that. Investigations continue, including at former so-called Indian Schools in the United States.84 Historian Preston McBride’s chapter in Volume I I of the series explores this issue further. The distinction between genocide and ‘cultural genocide’ remains fraught, though real. The field of Genocide Studies continues to evolve by re-examining and reevaluating the past. The distinction between genocide and extermination, also made in this series of volumes, is one that is rejected by the veteran genocide scholar, Israel Charny. He has long included ‘all types of mass murder’ in his broader conception of genocide.85 In his recent book, Israel’s Failed Response to the Armenian Genocide, Charny devoted a chapter to ‘Israel’s denial–concealment of cruelty, genocidal expulsion, and massacres of Arabs in the nonetheless entirely just war of independence’. There he forthrightly stated, while demonstrating his general admiration for and loyalty to Israel, that some of the massacres of Palestinians committed by Israeli forces in the 81 ‘Mass grave may contain Canadian Indigenous youth’, New York Times, 29 May 2021, p. A7. 82 ‘Hundreds of graves found at former residential school for Indigenous children in Canada’, Washington Post, 24 June 2021; ‘Remains of hundreds of Indigenous children are discovered in Canada’, New York Times, 25 June 2021; ‘Latest First Nations discovery reveals 182 unmarked graves at Canada school’, Guardian (UK edition), 30 June 2021, https://bit.ly/3uyD6UH (accessed 28 March 2022). 83 ‘Remains of 215 Indigenous children discovered at former Canadian residential school site’, Washington Post, 28 May 2021. 84 ‘Discovery of graves bolsters Indigenous push for reckoning in Canada’, New York Times, 27 June 2021; ‘Search for burial sites in U.S. is to be conducted at former “Indian schools”’, New York Times, 25 June 2021. 85 Israel Charny, ‘Towards a generic definition of genocide’ in Genocide: Conceptual and Historical Dimensions, ed. George J. Andreopoulos (Philadelphia: University of Pennsylvania Press, 1994), pp. 64–94, at p. 92.

27

ben kiernan

1948 conflict and afterwards were ‘genocidal’ massacres. According to Charny, ‘at various times, beginning especially with the War of Independence, Israel has engaged in acts of wanton cruelty and genocidal destruction, and has often adopted a policy of rank untruth and denial’. He stated especially in relation to the Deir Yassin case that ‘there is absolutely no doubt that in April 1948 a massacre did take place of the villagers there – who we now know were largely peaceful and that the population murdered was largely one of old people, women, and children’. Charny added that in 1981, ‘I was fully caught up in the propaganda of Israeli denial’. He also described, among other cases, ‘the genocidal massacre of unarmed civilian Arabs in Kfar Kassem that we executed in 1956’.86 It is also true that Arabs massacred Jews on several occasions in 1947–8, for example in Haifa and Kfar Etzion.87 Since 1981, Charny has become a leader in the field of Genocide Studies, as executive director of the Institute on the Holocaust and Genocide in Jerusalem. While the Cambridge World History of Genocide uses a more restrictive definition of genocide, and therefore does not include in these volumes coverage of Israeli massacres of Arabs or Palestinian massacres of Jews, Charny’s clear-sighted spirit is worthy of emulation. The threat of future genocides and the need for genocide studies remain. For instance, the 1980s and 1990s saw the republication in Malaysia and Indonesia of the notorious antisemitic forgery, The Protocols of the Elders of Zion.88 As recently as 2018, US President Donald Trump, on a visit to Europe for the 100th anniversary of the end of World War One, told his chief of staff

86 Israel Charny, Israel’s Failed Response to the Armenian Genocide: Denial, State Deception, Truth versus Politicization of History (Boston: Academic Studies Press, 2021), pp. 107–8, 125–8, 136, 139. See also Israel Charny, The Genocide Contagion: How We Commit and Confront Holocaust and Genocide (Lanham, MD: Rowman & Littlefield, 2016), pp. 73, 185–97. On Deir Yassin, see e.g. Noam Chomsky, The Fateful Triangle: The United States, Israel and the Palestinians (London: Pluto Press, 1983), pp. 95–6, 166–7, 236. 87 For instance, according to Peter Beinart: ‘After members of a right-wing Zionist militia, Etzel, threw grenades into a Palestinian crowd near an oil refinery in Haifa in December 1947, the crowd turned on nearby Jewish workers, killing 39 of them. In April of 1948, after Zionist forces killed more than 100 unarmed Palestinians in the village of Deir Yassin, Palestinian militiamen burned dozens of Jewish civilians to death in buses on the road to Jerusalem. In May of that year, Arab fighters vowing revenge for Deir Yassin killed 129 members of the kibbutz of Kfar Etzion, even though they were flying white flags.’ Peter Beinart, ‘A Jewish case for Palestinian refugee return’, Guardian (UK edition), 18 May 2021, https://bit.ly/35jQweM (accessed 29 March 2022). 88 See Anthony Reid, ‘Entrepreneurial minorities and the state’ in Essential Outsiders: Chinese and Jews in the Modern Transformation of Southeast Asia and Central Europe, ed. Daniel Chirot and Anthony Reid (Seattle: University of Washington Press, 1997), pp. 33–71, at pp. 63–4, 70n76.

28

General Editor’s Introduction to the Series

that ‘Hitler did a lot of good things’.89 Two years later, after Trump had lost the presidential election, General Mark Milley, US chief of the joint chiefs of staff, was shocked by Trump’s false claims of election fraud, and compared them to ‘Hitler’s insistence to his followers at the Nuremberg rallies that he was both a victim and their savior’. Milley saw Trump’s supporters, the Proud Boys, as ‘the same people we fought in World War II’.90 In their 2021 book I Alone Can Fix It: Donald J. Trump’s Catastrophic Final Year, Washington Post reporters Carol Leonnig and Philip Rucker recount that Milley saw Trump as ‘the classic authoritarian leader with nothing to lose’. ‘This is a Reichstag moment,’ Milley told aides, ‘The gospel of the Führer.’ Trump also told his White House chief of staff John Kelly that he wanted his generals to ‘be like the German generals … in World War II’ who ‘were totally loyal to’ Hitler.91 Also in 2020, a Louisville, Kentucky high school newspaper, The Manual Redeye, and the New York Times revealed that until as late as 2013 the Kentucky State Police department taught a training course for cadets that urged them to ‘always fight to the death’ and encouraged each trainee to become a ‘ruthless killer’. The course included statements made by Adolf Hitler, such as this: ‘The very first essential for success is a perpetually constant and regular employment of violence.’ The New York Times reported that Hitler ‘is quoted more than anyone in the training document. Some of the statements attributed to Hitler link to a website providing biographical information about him and listing books by and about him’.92 From 2015 to 2020 ‘Kentucky state troopers shot and killed at least 41 people . . . including 33 in rural areas’. A nationwide

89 ‘Trump told chief of staff “Hitler did a lot of good things”, book says’, Guardian (UK edition), 7 July 2021, www.theguardian.com/us-news/2021/jul/06/donald-trumphitler-michael-bender-book (accessed 16 July 2021). 90 Quoted in Dwight Garner, ‘Two accounts of Donald Trump’s final year in office’, New York Times, 15 July 2021. 91 ‘Watergate reporter hits out at Trump officials for not speaking sooner about “Hitlerian” administration’, Independent (UK edition), 15 July 2021, https://bit.ly/36z OLKT; ‘Trump smears America’s top general as “choking dog” after book reveals he compared ex-president to Hitler’, Independent (UK edition), 15 July 2021, https://bit.ly/ 35h4XjL (both accessed 29 March 2022). Leonnig and Rucker quote another senior Trump administration official as saying that Trump is a ‘guy who takes fuel, throws it on the fire, and makes you scared shitless’, then says ‘“I will protect you”’. ‘That’s what Hitler did to consolidate power in 1933’, Guardian (UK edition), 16 July 2021, https://bit.ly/3uFrCz0 (accessed 29 March 2022). ‘Trump wanted Pentagon generals to be like second world war Nazis, book says’, Guardian, 8 August 2022, www.theguardian.com/us-news/2022/aug/ 08/trump-pentagon-generals-nazis-second-world-war (accessed 23 August 2022). 92 Nicholas Bogel-Burroughs, ‘Kentucky police training quoted Hitler and urged “ruthless” violence’, New York Times, 31 October 2020, https://nyti.ms/35h4ZYV (accessed 30 March 2022).

29

ben kiernan

investigation found the Kentucky State Police to be ‘the agency with the largest number of such deaths in the six-year period’.93 In Poland in May 2021, the Warsaw District Court began to hear a civil suit brought by a Polish publishing house that distributes Ritual Murder: Historical Facts, a book which recycles medieval falsehoods about Jews. The book is co-authored by Grzegorz Braun, a member of the Polish parliament from the extreme right-wing Confederation Party. In the case, the publishing house alleged that calling this book antisemitic violated its good name and reputation.94 Meanwhile, in Germany, police officers in several states have been fired for participating in online groups that shared content like ‘swastikas, Hitler images, and anti-foreigner postings’. For instance, in June 2021 the state authorities in Frankfurt disbanded an entire elite police unit and suspended eighteen of its members after an investigation found them to have ‘exchanged racist messages and glorified the Nazis’.95 Germany’s defence ministry also recalled a tank platoon from a NATO mission in Lithuania, and moved to disband the platoon, after some of its members engaged in ‘singing a song to mark Adolf Hitler’s birthday, in violation of a sergeant’s order’, a breach that the sergeant then failed to report.96 Rising antisemitism in Europe, the United States and the Middle East and rising anti-Muslim prejudice in the United States, Europe and Asia both indicate widespread reluctance to learn from the past.97 In the face of all this, abiding the silence that often reigns after genocide is no option for historians. The task confronting scholars of genocide remains far from complete. 93 Alysia Santo and R. G. Dunlop, ‘Where police killings often meet with silence’, Indian Express, 15 August 2021, and ‘Shooting first and asking questions later’, Investigations, https://wfpl.org/kycir-rural-police-shootings-ksp/ (accessed 16 August 2021). 94 Never Again Association, ‘Is antisemitism antisemitic?’, 10 May 2021, www .nigdywiecej.org/en/our-news/195-articles-from-2021/4563-is-antisemitism-antisemitic (accessed 10 May 2021). 95 ‘Elite German police unit is ended over racist chat’, New York Times, 11 June 2021, p. A10. 96 ‘Germany recalls platoon on NATO mission’, New York Times, 18 June 2021, p. A9, citing Der Spiegel. 97 ‘Gaza conflict led to record rise in UK antisemitic attacks, charity says’, Guardian (UK edition), 5 August 2021, https://bit.ly/3wIAy96 (accessed 29 March 2022); ‘U.S. faces outbreak of anti-semitic threats and violence’, New York Times, 26 May 2021, www .nytimes.com/2021/05/26/us/anti-semitism-attacks-violence.html; Christine Ogan et al., ‘The rise of anti-Muslim prejudice: media and Islamophobia in Europe and the United States’, International Communication Gazette 76:27 (2014), https://doi.org/1177/ 1748048513504048; ‘Assaults against Muslims in U.S. surpass 2001 level’, Pew Research Center, 15 November 2017, www.pewresearch.org/fact-tank/2017/11/15/assaultsagainst-muslims-in-u-s-surpass-2001-level/ (accessed 3 August 2021); ‘Around the world, from China and India to the U.K. and U.S., reports of hate crimes against Muslims are increasing’, Newsweek, 3 August 2021, www.newsweek.com/islammuslims-islamopohobia-racism-u-s-u-k-1531333 (accessed 3 August 2021).

30

Introduction to Volume I t. m. lemos, tristan s. taylor and ben kiernan

Questions about Premodern Genocide Two preliminary questions must be confronted in approaching the history of genocide in the premodern and early modern worlds.1 The first, as discussed in the General Editor’s Introduction, is the foundational, definitional question: what exactly is genocide? The second is the question of possible anachronism. This stems from the fact that the term ‘genocide’ itself is a modern term, a neologism introduced by the Polish jurist Raphael Lemkin in his 1944 work Axis Rule in Occupied Europe to describe atrocities being perpetrated during World War Two, but also earlier cases, drawing on his own historical research that stretched back to antiquity. Here, then, we deal with two interrelated problems.2 The first is the question of whether genocide should be seen as a transhistorical phenomenon, or alternatively a peculiar product of modernity. While mass killing is clearly transhistorical, is there something particular about what is termed ‘genocide’ in the modern period? The second issue, which we might term the ‘value problem’, is that the term has carried with it, since its inception,3 an implicit moral judgement. David Bell clearly articulates the nature of this judgement when he questions the labelling of the massacres in the Vendée during the French Revolution as ‘genocide’: ‘the word genocide does not just describe the crime itself – it implicitly condemns everything about the perpetrators and what they were 1 For further exploration of these questions, see also T. S. Taylor, ‘Introduction’ to A Cultural History of Genocide in the Ancient World, ed. T. S. Taylor (London: Bloomsbury, 2021), pp. 1–29; F. Quesada-Sanz, ‘Genocide and mass murder in Second Iron Age Europe’ in The Routledge History of Genocide, ed. C. Carmichael and R. C. Maguire (London: Routledge, 2015), pp. 9–22, esp. pp. 9–12. 2 Ben Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007), p. 12. 3 R. Lemkin, Lemkin on Genocide, ed. S. L. Jacobs (Lanham, MD: Lexington Books, 2012), pp. 22–3, referring to Gabriela Mistral.

31

t. m. lemos, tristan s. taylor and ben kiernan

trying to accomplish’.4 In other words, it impugns their motives as well as their actions. Peter McPhee discusses that case in detail in Volume I I of the series. However, it is clear that – while mass violence in the ancient world did not occur in a moral vacuum, and could on occasion prompt anxiety, criticism and the perceived need for justification5 – mass violence of a scale that in the modern age might warrant the label of genocide and could often be subject to a cover-up on the part of the perpetrators, in the premodern world might on occasion be the cause of boasting, and even official, public celebration. In such a context, some have asked whether it is appropriate to use a term such as genocide for events in the premodern world, which may import a moral judgement. Let us explore these two questions in turn.

The Meaning of Genocide Lemkin’s original conception of genocide was broad. It referred to the destruction of a national or ethnic group through a co-ordinated plan that destroyed the essential foundations of the group – by social, cultural and economic as well as physical means. Lemkin envisaged this would take two phases – the first involving the destruction of the oppressed group’s national pattern, and the second, the imposition of that of the oppressor.6 Lemkin’s understanding recognises that human groups are social constructs, and can therefore be destroyed by more than the physical destruction of the group’s members. As such, his view is broader than many popular understandings of genocide, which tend to focus exclusively on mass killing. This popular conception is well represented by dictionary definitions of ‘genocide’ as ‘the deliberate murder of a racial or ethnic group’ (Penguin English Dictionary) or ‘the deliberate extermination of a people or nation’ (Concise Oxford English Dictionary) – often shaped by what is seen as the paradigmatic genocide, the Jewish Holocaust. Lemkin’s original understanding is also wider than the definition found in the international instrument that defined the ‘crime’ of genocide, drafted in the wake of the Holocaust, the United 4 D. A. Bell, ‘The French Revolution, the Vendée and genocide’, Journal of Genocide Research 22:1 (2020), 1–25, at 24. For an argument that the massacre of the Vendée represents the first ‘modern’ genocide, see M. Levene, Genocide in the Age of the Nation State, vol. I I: The Rise of the West and the Coming of Genocide (London: I. B. Tauris, 2013), pp. 103–61. 5 See D. Konstan, ‘Anger, hatred and genocide in ancient Greece’, Common Knowledge 13:1 (2007), 170–87. 6 R. Lemkin, Axis Rule in Occupied Europe: Laws of Occupation, Analysis of Government, Proposal for Redress (Washington, DC: Carnegie Endowment for International Peace, 1944), pp. xi–xii, 79.

32

Introduction to Volume I

Nations Convention on the Prevention and Punishment of the Crime of Genocide (otherwise known as the Genocide Convention).7 It is this Convention definition, set out above (see p. 6), rather than Lemkin’s more expansive conception, that forms the point of departure for most scholarly definitions of genocide.8 In contrast to Lemkin’s inclusion of the elimination of the sociocultural foundations of a group (now usually known as cultural genocide), the Genocide Convention embraces a much more limited range of actions focused on the physical or biological destruction of certain specific groups and the forcible transfer (or ‘removal’) of children. It also requires that to constitute genocide the actions must be committed with a specific intent, namely an intent ‘to destroy, in whole or in part, a national, ethnical, racial or religious group, as such’ (emphasis added). The Convention has drawn criticism in relation to all its definitional components: the nature of the intent required; the magnitude of killing required; the limitation of the Convention’s protection to a limited range of groups, excluding in particular political and other social groups; and the specific actions covered, especially the focus on biological destruction. This criticism has spawned a legion of alternative definitions of genocide,9 often sociological ones, leading to ‘something approximating definitional chaos’.10 Another response to the Convention definition has been the creation of additional neologisms. A fertile approach has been the coining of further ‘-cides’,11 such as ‘politicide’ for the destruction of political groups or – very pertinent to the premodern world – ‘urbicide’ or ‘the obliteration of urban living-space as a means of destroying the viability of an urban civilization and eroding its collective values’.12 In Chapter 8 of this volume, Paul Cartledge explores that concept

7 Convention on the Prevention and Punishment of the Crime of Genocide, Paris, 9 December 1948, United Nations, Treaty Series, vol. 78, no. 1021; www.hrweb.org/le gal/genocide.html. 8 S. Straus, ‘Contested meanings and conflicting imperatives: a conceptual analysis of genocide’, Journal of Genocide Research 3:3 (2001), 349–75, at 361. 9 A. Jones, Genocide: A Comprehensive Introduction, 3rd ed. (London: Routledge, 2017), pp. 22–7. 10 P. R. Bartrop, ‘General editor’s preface’ to Cultural History of Genocide in the Ancient World, ed. Taylor, pp. xi–xvi, at p. xv. 11 See Jones, Genocide, pp. 34–7. 12 Ibid., pp. 37–9. See also J. L. Wright, ‘Urbicide: the ritualized killing of cities in the ancient Near East’ in Ritual Violence in the Hebrew Bible: New Perspectives, ed. S. M. Oylan (Oxford University Press, 2015), pp. 147–66, for ‘urbicide’ in the ancient Near East, and H. van Wees, ‘Genocide in the ancient world’ in Oxford Handbook of Genocide Studies, ed. D. Bloxham and A. D. Moses (Oxford University Press, 2010), pp. 239–79, at pp. 243– 7, for the connection between the destruction of cities and genocide in the ancient world more generally.

33

t. m. lemos, tristan s. taylor and ben kiernan

and practice in classical Greece, and Tristan Taylor (Chapter 10) the specific examples of the Roman destructions of Carthage, Corinth and Numantia. There are other neologisms, but these two appear to us to make some sense. Given this plethora of different approaches to the very concept of genocide, and the issue of anachronism (see below), what is the approach to be adopted in this volume? The foundational premise will be the UN Genocide Convention definition, although the implications of alternative definitions will also be discussed. We do not ignore the various critiques of the Convention definition, but we also acknowledge that fruitful comparative study requires the consistent application of a definition across time,13 and that in this context the Convention definition is not only the sole legal definition but also the only internationally recognised definition, and thus ‘in the eyes of the international community, it is the only suitable means by which genocide might be described’.14 However, we also include in these volumes consideration of cases that fall within the different, but related, legal concept of extermination, a crime against humanity (see General Editor’s Introduction). Before applying the Convention definition of genocide, it is important to acknowledge both its boundaries and how these boundaries have been enlarged, or confined, by other scholarly approaches to the concept.15 Alternative definitions of genocide tend to be more expansive or restrictive than the Convention in relation to the intention of the perpetrators, and who might be considered agents or perpetrators of genocide, as well as which kinds of groups can be the victims of genocide, the scale of the violence, and the strategies or actions adopted.16

Genocidal Intent The Genocide Convention’s definition of genocide requires a ‘specific intent’ – to bring about the end, in whole or in part, of a specific group of people ‘as such’.17 Importantly, this has been interpreted as requiring an 13 Kiernan, Blood and Soil, p. 12. 14 Bartrop, ‘General editor’s preface’, xv. Other studies of the pre- and early modern periods that apply this definition include van Wees, ‘Ancient world’, and Kiernan, Blood and Soil. 15 On the legal boundaries of genocide, see C. Kreß, ‘The crime of genocide under international law’, International Criminal Law Review 6:4 (2006), 461–502; W. A. Schabas, Genocide in International Law: The Crime of Crimes, 2nd ed. (Cambridge University Press, 2009); and W. A. Schabas, ‘The law and genocide’ in Bloxham and Moses, Oxford Handbook of Genocide Studies, pp. 123–41. 16 Jones, Genocide, p. 28. For a slightly different, but overlapping, division of analysis, see Straus, ‘Contested meanings’, 361. 17 Schabas, ‘Law and genocide’, 137–8.

34

Introduction to Volume I

intention to bring about a particular result; it is not enough that there was deliberate killing, let alone recklessness or even criminal negligence. Even those scholarly definitions that depart from the Convention agree that genocide should be purposeful.18 While such ‘purpose-based’ intent has been generally favoured in jurisprudence surrounding the Convention, there has also been some support for the adequacy of ‘knowledge-based’ intent – where a person acted with the knowledge that his or her actions would further an overall genocidal goal.19 In addition to this overarching requirement of specific intent, some of the ‘acts’ of genocide listed in Article II involve further intent requirements – such as imposing measures ‘intended’ to prevent births (not just measures that practically do so) or conditions of life ‘calculated’ to bring about physical destruction.20 Importantly, while the Convention is concerned with imposing criminal liability, in historical analysis the attempt is, almost inevitably, to discern a collective ‘intent’ for mass violence, rather than the particular intent of each individual involved.21 As will be seen, most perpetrators of mass violence in the premodern and early modern worlds were part of an armed collective in the context of armed conflict, from a militia to an organised standing army, whether representing the fighting force of a tribe or that of a pluralistic empire. In this context, the problems that can exist in the modern world with establishing genocidal intent multiply for the premodern era,22 where surviving evidence of anything is much less copious and any direct testament by those ordering mass violence is extremely rare,23 while subjective accounts of the ordinary soldier are almost completely absent. Even so, though evidence of an official plan or policy might help in determining such collective intent,24 it is not a requirement for proof of genocide.25 The Convention’s words ‘as such’ require that the victims be proven to have been targeted for destruction as members or ‘because of’ their membership of a particular group. Such intent on the part of the perpetrators is distinct, in legal thought and in the Convention definition, from any ‘motive’, 18 Jones, Genocide, pp. 31–2. 19 Kreß, ‘Crime of genocide’, 492–8; Jones, Genocide, p. 50. 20 Schabas, ‘Law and genocide’, 138. 21 On the tension between individual intent and collective purpose in the Genocide Convention, and a possible resolution, see Kreß, ‘Crime of genocide’, 494–8. 22 See Quesada-Sanz, ‘Genocide’, 10, and Taylor, ‘Introduction’, 5–11, on source issues for the ancient world. 23 An exception here is Caesar’s account of his Gallic conquest in his Bellum Gallicum. See Chapter 11 in this volume. 24 Schabas, ‘Law and genocide’, 138–9. 25 Kiernan, Blood and Soil, pp. 19–20.

35

t. m. lemos, tristan s. taylor and ben kiernan

that is, the reason or reasons why the perpetrators targeted that particular group. The Convention requires no proof of any particular motive – its sole concern is whether or not there is an intention to destroy a group of people, in whole or in part, as a group.26 In the shadow of the Holocaust and the dystopian ideology pursued by the Third Reich, it is often mistakenly thought that genocide must be conducted in order to pursue a particular ideological, for example a racialist, goal – and this basis has been used to object, as we shall see, to the application of the concept of ‘genocide’ to premodern mass violence, where such a racialist or other ideological apparatus is often lacking.27 Certain non-legal scholarly definitions of genocide actually incorporate particular motives,28 and some, such as those of Helen Fein and of Chalk and Jonassohn, have created genocide typologies that are based on motives (see General Editor’s Introduction).29 This, however, is useful in historical analysis but unrelated to the legal definition of genocide.

Perpetrators The Genocide Convention itself calls for the punishment of ‘persons committing genocide . . . whether they are constitutionally responsible rulers, public officials, or private individuals’ (Article IV). It therefore makes no requirement that a perpetrator of genocide be some form of state or authority, although scholarly definitions – influenced by the experience of modernity – often restrict genocide to actions performed by such authorities.30 But there is no need for disputation so far as this volume is concerned. As we shall see, in the pre- and early modern worlds – at least in the historical periods – most perpetrators of mass violence were members of armed forces of greater or lesser degrees of organisation.31 Even if a narrower understanding of 26 Ibid., p. 18. 27 See e.g. N. Barrandon, Les massacres de la république romaine (Paris: Fayard, 2018), p. 336. 28 For example, Irving Louis Horowitz states that genocide occurs ‘to assure conformity and participation of the citizenry’ (quoted in Jones, Genocide, pp. 23–4); Levene argues that genocide occurs to preserve the integrity of a nation-state’s agenda (M. Levene, Genocide in the Age of the Nation State, vol. I: The Meaning of Genocide (London: I. B. Tauris, 2005), p. 35); and Daniel Chirot and Clark McCauley describe it as ‘politically motivated violence’ (quoted in Jones, Genocide, p. 27). 29 See for example the discussion in A. D. Moses, ‘Toward a theory of critical genocide studies’, Mass Violence & Resistance (online), 18 April 2008, http://bo-k2s.sciences-po.fr/ mass-violence-war-massacre-resistance/en/document/toward-theory-critical-genocidestudies (accessed 28 January 2021). 30 Jones, Genocide, pp. 28–9. 31 Taylor, ‘Introduction’, 14. On the evolution of armies and warfare in the ancient world, for example, see R. A. Gabriel, Empires at War, 3 vols. (London: Greenwood Press, 2005), and B. T. Carey, Warfare in the Ancient World (Barnsley, UK: Pen & Sword, 2005).

36

Introduction to Volume I

‘perpetrator’ were to be adopted than exists in the Convention, these perpetrators of the pre- and early modern world are readily seen as analogous to modern states or authorities, despite the many differences between earlier polities and modern nation-states.32

Victim Groups The Convention lists specific types of human collectives as potential victims of genocide: ‘national, ethnical, racial or religious’ groups. These groups were chosen because they were considered relatively stable groups into which people were born (although clearly membership of some, such as religious groups, can be susceptible to change).33 This specific listing has been the subject of controversy, in particular for its exclusion of political and other sociocultural groups.34 Thus, many scholarly definitions are broader than the Convention, and eschew its approach of listing specific categories of human group, instead tending to favour broad formulations such as ‘human collectivity’, ‘minority group’ or ‘social collectivity’ or even just ‘group’.35 Regardless of whether one adopts a broad conception of human group, or a specific list such as the Convention does, a key question arises as to how groups are identified or verified. Should the group exist as an ‘objective’ reality? Or should the determination be ‘subjective’ – and if so, is the perpetrators’ view determinative, or should it be necessary that the victims also consider themselves part of the group? The answer to this question can have a crucial impact where, as in the Albigensian crusade, as Mark Gregory Pegg argues in Chapter 18 of this volume, the victim group had no collective existence outside of the perpetrators’ imagination. Historically, genocide represents a power imbalance, and perpetrator labels and definitions have had little to do with victim self-identification – as also in the case of the Nazi race laws establishing who was Jewish.36 This power imbalance is recognised in those scholarly definitions of genocide that adopt the perpetrators’ identification of victim groups as determinative.37 In the context of the Convention, the establishment of agreed objective criteria for the four categories of protected groups has proved elusive, reflecting their nature as social 32 See e.g. van Wees, ‘Ancient world’, 243–4. 33 Kreß, ‘Crime of genocide’, 474. 34 For example L. Kuper, Genocide: Its Political Use in the Twentieth Century (New Haven: Yale University Press, 1981), p. 39. 35 See the definitions of Peter Drost, Vahakn Dadrian, John L. Thompson and Gail A. Quets, and Frank Chalk and Kurt Jonassohn respectively, in Jones, Genocide, pp. 23–5. 36 Schabas, ‘Law and genocide’, 133–4. 37 See Jones, Genocide, p. 29.

37

t. m. lemos, tristan s. taylor and ben kiernan

constructs rather than scientific categories.38 Some, non-determinative, criteria identified by Kreß include: • National: nationals of a state.39 • Ethnic: a group sharing a ‘common culture, history, way of living, language or religion’.40 • Racial: shared hereditary physical traits or characteristics.41 • Religious: ‘a transcendental point of reference in the form of belief in the existence of one or more divinities or spiritual powers’.42 The practice adopted in jurisprudence in relation to the Convention has generally been case by case, utilising a mixture of objective and subjective elements.43 Importantly, as William Schabas has identified, the meaning of these categories of groups has changed even in the short time since the Convention’s drafting in the late 1940s.44 So there is some difficulty in unproblematically identifying such groupings in the pre- or early modern period, eras that lacked both scientific racism45 and the concept of a modern nation-state. Nonetheless, inhabitants of the pre- and early modern worlds were quite capable of identifying particular, stable groups that are readily analogous to the core idea embodied in the Convention’s four categories. For example, as Paul Cartledge and Tristan Taylor show in Chapters 8, 10 and 11 of this volume, the premodern Mediterranean city-state performed a similar role in social, economic and political – even religious – organisation to that of a modern nation-state. The ‘urbicide’ or elimination of a city-state and/or its population can thus be readily compared to the destruction of a modern ‘national’ group.46 And for those definitions of genocide that adopt a broader 38 Schabas, ‘Law and genocide’, 134 and Genocide in International Law, pp. 124–9. This issue proved particularly acute in relation to Rwanda, where the common objective criteria for distinguishing ethnic or national groups did not produce a clear distinction between Hutu and Tutsi: Kreß, ‘Crime of genocide’, 477. 39 Kreß, ‘Crime of genocide’, 476–8; see also Schabas, Genocide in International Law, pp. 134–9. 40 Kreß, ‘Crime of genocide’, 476–8; see also Schabas, Genocide in International Law, pp. 134–7. 41 Kreß, ‘Crime of genocide’, 478; see also Schabas, Genocide in International Law, pp. 139–43. 42 Kreß, ‘Crime of genocide’, 478–9; see also Schabas, Genocide in International Law, pp. 147–50. 43 Schabas, ‘Law and genocide’, 134. 44 Schabas, Genocide in International Law, p. 137. 45 Though see B. Isaac, The Invention of Racism in Classical Antiquity (Princeton University Press, 2004). 46 Van Wees, ‘Ancient world’, 210; Taylor, ‘Introduction’, 14–15.

38

Introduction to Volume I

category of human collectivity or group, pre- and early modern human groupings present no particular conceptual challenges.

The Scale of Genocide What magnitude of a population must be targeted for destruction in order to qualify as a genocide? The Genocide Convention definition refers to the intended destruction of a group ‘in whole or in part’, but does not specify what is sufficient to constitute a ‘part’. Importantly, the phrase ‘in whole or in part’ attaches to the intent; it does not reflect any requirement of what the perpetrators may have actually achieved. The matter remains somewhat unsettled, and both qualitative and quantitative interpretations – and combinations of the two – have been advanced. Within a quantitative framework, it has been argued that the number of people targeted must be a ‘substantial’ proportion of the group.47 In addition, it has been suggested that a qualitative dimension exists – the targeting of particularly significant members of a group, such as an elite, may constitute genocide.48 Both approaches were used in judgments relating to the massacring of the adult male Bosnian Muslim population of Srebrenica in 1995. Such a gender-focused massacre is often also called ‘gendercide’. The population of Srebrenica was considered to represent a substantial proportion of Bosnia’s Muslim population, and its male population represented a ‘significant’ part of that community, especially given the patriarchal character of Bosnian Muslim society.49 The model of Srebrenica is pertinent to consider as an analogue for mass violence in the premodern world. In antiquity, the destruction of an urban community or city-state was often accompanied by the massacre of the male population and the enslavement of the women and children. The ancient world was also typically patriarchal (though not infrequently bilateral), with many communities primarily thought of as composed of men; as such, the destruction of the male population was considered the equivalent of the destruction of the community.50 In addition to gender-differentiated violence, the fact that the victims of mass violence were often the inhabitants of a single town or city-state raises a different question of scale as well. Often such people would be members of 47 Schabas, ‘Law and genocide’, 135–6. 48 Ibid., 136–7. 49 Ibid.; for a critique of this reasoning, see Kreß, ‘Crime of genocide’, 491–2. 50 H. van Wees, ‘Genocide in archaic and classical Greece’ in Our Ancient Wars: Rethinking War through the Classics, ed. V. Caston and S.-M. Weineck (Ann Arbor: University of Michigan Press, 2016), pp. 19–37, at p. 25; Taylor, ‘Introduction’, 16.

39

t. m. lemos, tristan s. taylor and ben kiernan

a larger ‘ethnic’ group and the destruction of a particular city would not be accompanied by an attack on that wider group. Some have argued that this distinction means that the destruction of cities, unaccompanied by a campaign against the wider ethnic group, ought not to be seen as genocide, such as the destruction of Carthage and allied towns in 146 B C E, which was unaccompanied by a campaign against all other Punic states (the ethnic group to which Carthage belonged).51 However, the better view, as argued above, is to see city-states as distinct groups of people, analogous to national groups. Competing scholarly definitions, when they address this issue, range in breadth from requiring total destruction through to a ‘substantial proportion’. As Adam Jones notes, even when not specified, the focus of analysis – as indeed it is in this volume – is usually to actions involving ‘mass’ or ‘substantial’ casualties.52 The separate concept of a ‘genocidal massacre’ was coined by Leo Kuper to embrace shorter, more limited episodes of mass violence aimed at a specific local or regional group because of its membership of a larger group – that is, a massacre that may not reach the ‘in part’ qualitative or quantitative threshold embodied in the Genocide Convention.53

Techniques of Genocide As already noted, the Genocide Convention narrows the range of actions that might inflict genocide to five ‘acts’ that, with the exception of the forcible transfer of children, involve physical or biological destruction of a group: killing members of a group; causing serious bodily or mental harm; inflicting conditions of life calculated to end the group; or preventing births within the group.54 While this is narrower than Lemkin’s original vision, some scholarly definitions draw the net even tighter, focusing exclusively on mass killing.55 In the pre- and early modern world three other kinds of mass violence require consideration on the margins of, or outside, the UN convention definition: deportation, mass enslavement and deliberate destruction of culture. Deportation, or the forced movement of a group of people from one territory to another, is in some senses analogous to what in the modern 51 For example Barrandon, Massacres, p. 328. 52 Jones, Genocide, p. 29. 53 See the discussion in Kiernan, Blood and Soil, pp. 13–15. 54 On each of these limbs, see the analysis in Kreß, ‘Crime of genocide’, 480–4; Kiernan, Blood and Soil, pp. 12–13. 55 Jones, Genocide, p. 31.

40

Introduction to Volume I

world is termed ‘ethnic cleansing’.56 Here, the premodern motive – usually retribution – differed from the destructive ethnic hatred often associated with the act in the modern world. Further, in the ancient Assyrian Empire, for example, deported groups were often kept together, not dispersed.57 Mass enslavement, by contrast, could result in the dispersal of a population, and the dissolution of a group as such. The dynamics and context of enslavement were, however, highly variable – enslaved people could be ransomed and released, or people could be enslaved and then kept together in one place to work, as the Spartan Helots were. In such circumstances, it is less likely that there was a genocidal intent, although the consequences could often be devastating to the enslaved or enserfed labourers, many of whom might perish through disease, overwork and appalling living conditions, as in Hispaniola after the Spanish conquest.58 In other contexts, however, such as the Roman destruction of Carthage, where the population was either killed or enslaved and dispersed, and the city demolished, with the land ritually cursed, it is difficult to conclude that the mass enslavement was not part of a wider plan to destroy the Carthaginians as a distinct group.59 But is mass enslavement genocide within the meaning of the Convention? Van Wees has argued such actions would fit under Article II(c): deliberately inflicting conditions of life calculated to bring about the group’s destruction.60 As noted, ‘cultural genocide’ – sometimes referred to as ‘ethnocide’ – was a key component of Lemkin’s conception of genocide, but is largely absent from the Convention. As in the modern era, the pre- and early modern worlds also saw attacks on culture, which included the deliberate destruction of important cultural locations within a city, for example its temples and shrines. Within the context of the Convention, such attacks on culture are seen as providing evidence that attacks on the population could be construed as part of a campaign to destroy the group as a group.61

56 On this see B. Lieberman, ‘“Ethnic cleansing” versus genocide?’ in Oxford Handbook of Genocide Studies, ed. Bloxham and Moses, pp. 42–60. 57 See e.g. B. Oded, Mass Deportations and Deportees in the Neo-Assyrian Empire (Wiesbaden: Reichert, 1979), pp. 21–5, 85. 58 See e.g. Kiernan, Blood and Soil, pp. 85–6; see also Chapter 23 in this volume. 59 On this see also B. Kiernan, ‘The first genocide: Carthage, 146 BC’, Diogenes 51:4 (2004), 27–39; Kiernan, Blood and Soil, pp. 49–58. 60 Van Wees, ‘Archaic and classical Greece’, 21. 61 Jones, Genocide, p. 42.

41

t. m. lemos, tristan s. taylor and ben kiernan

A Transhistorical Phenomenon? The preamble to the Genocide Convention recognises that ‘at all periods of history genocide has inflicted great losses on humanity’, adopting Lemkin’s transhistorical view that ‘[g]enocide is a new word, but the evil it describes is old’.62 However, a significant body of Genocide Studies scholarship focuses on the link between genocide and modernity,63 and in turn on concerns that the use of the term ‘genocide’ for pre- and early modern periods may be anachronistic. As noted above, this concern can take two forms: the first is phenomenological – that mass violence in the pre- and early modern worlds was fundamentally different. The second is moral: that mass violence in preand early modern eras may have lacked the moral opprobrium it does today, whereas the term ‘genocide’ itself inevitably imports a moral judgement. Let us turn first to the phenomenological question. As a preliminary, the issue must be addressed whether the pre- and early modern world understood the concept of genocide, even if there was no specific term for it.64 Here one can readily identify, even in ancient cases, both the capacity to identify groups analogous to those listed in the Convention, as discussed above, and an intent to destroy such groups ‘as such’. A clear example here is Caesar’s explicit desire to destroy the very ‘stock and name’ (stirps ac nomen) of a particular Germanic tribe, the Eburones, by causing neighbouring tribes to raid them until they could no longer survive.65 It is difficult not to agree that this expresses a desire to destroy an ethnic group as such by imposing conditions of life calculated to bring about their destruction in whole or in part: genocide according to the Convention definition. A different but equally illuminating example comes from India in the third century B C E. There, the Buddhist emperor Asoka left an inscription in which he initially reflected with remorse (anusocana) on the excessive violence of his ‘victory over the country of the Kalingas’. The inscription continued: ‘One hundred and fifty thousand [Kalingas] were captured and deported, one 62 Lemkin, Lemkin on Genocide, p. 20. 63 Eric Weitz stresses ‘the systematic, total character of twentieth-century genocides’ (Eric D. Weitz, ‘The modernity of genocides’ in The Specter of Genocide: Mass Murder in Historical Perspective, ed. R. Gellately and B. Kiernan (Cambridge University Press, 2003), pp. 53–73, at p. 73). See also A. Dirk Moses, ‘Genocide and modernity’ in Historiography of Genocide, ed. Dan Stone (London: Palgrave Macmillan, 2008), pp. 156–93, at p. 156. 64 For a discussion of pre- and early modern terms, see B. Kiernan, ‘Is “genocide” an anachronistic concept for the study of early modern mass killing?’, History 99:336 (2014), 530–48. 65 Caes. BGall. 6.34.

42

Introduction to Volume I

hundred thousand were killed there, and many times this number died . . . For the slaughter, death, and deportation of people that take place in the course of conquering an unconquered country is considered very painful and deplorable by [Asoka]’.66 Nevertheless, Asoka announced an exception to his pain and suffering: ‘And even the inhabitants of the forests which are included in the dominion of [Asoka], even those he pacifies and conciliates’ – but only if they desist from ‘crimes’ they had committed. The inscription continued: ‘And they are told [of] the power to punish them which [Asoka] possesses in spite of his repentance, in order that they may be ashamed of their crimes and may not be killed’. The historian Upinder Singh describes this as a ‘stern threat’. Despite Asoka’s Buddhist remorse, his edict ‘does not abjure the use of force to suppress recalcitrant forest people and/or forest chieftains (atavi), who, in fact, posed a serious impediment and challenge to the expansion of all premodern Indian states’.67 What distinguishes the pre- and early modern from the modern context is not so much the presence or absence of a concept of group destruction that accords with the Convention definition – although some conceptual difficulties can occur, as noted above. Rather, what differs is the ideological complex within which such violence happens. In particular, as will also be seen in later chapters, mass violence in the pre- and early modern world is often intensely pragmatic rather than coupled with particular ideologies, for example the racialist ideology of the Third Reich.68 To give one, recurring, example: mass violence is often associated with the actions of pre- and early modern empires. On the other hand, ideologies of genocidal violence were not absent in antiquity, as the h¯erem (annihilation) ideology of the ancient Levant, dis˙ cussed by T. M. Lemos in Chapter 6, clearly demonstrates. Ancient China may offer another example. When Raphael Lemkin died, in 1959, leaving unfinished his manuscript on the history of genocide, he had intended to ‘make several trips to the Library of Congress to examine certain edicts of Chinese Emperors relating to the destruction of Buddhist sects’. Lemkin wrote that he could read in twelve languages, but did ‘not understand 66 Quoted in Upinder Singh, Political Violence in Ancient India (Cambridge, MA: Harvard University Press, 2017), p. 268; and Odisha History, ‘The 13th Major Rock Edict of Asoka’, www.historyofodisha.in/the-13th-major-rock-edict-of-asoka/#more-36 (accessed 19 February 2021). 67 Odisha History, ‘13th Major Rock Edict of Asoka’; Singh, Political Violence in Ancient India, pp. 458, 462, 269–70. 68 On the Third Reich, see e.g. Kiernan, Blood and Soil, pp. 416–54; C. Browning, ‘The Nazi empire’ in Oxford Handbook of Genocide Studies, ed. Bloxham and Moses, pp. 407–25.

43

t. m. lemos, tristan s. taylor and ben kiernan

Chinese and will have to pay for translations of a limited number of Chinese texts already identified’.69 We do not know precisely which edicts, nor which Chinese emperors Lemkin had in mind. However, later in this volume Victoria Tin-bor Hui provides a comprehensive survey of ‘Genocide, Extermination and Mass Killing in Chinese History’. Premodern empires often lacked the capacity to deploy significant military force everywhere to control populations. They rather depended for their longevity on their reputation for deploying extreme violence – a failure to respond to rebellion, or resistance, was seen as a sign of weakness that could pose an existential threat to the empire.70 Such violence could both deter rebellion and encourage ready submission during periods of expansion. It was not necessarily accompanied by a process of ‘othering’ or ‘dehumanising’ of the victims,71 although ethnic prejudice could play a role in ‘lowering the threshold’72 for violence against particular groups (for example, as deployed by urbanised imperial powers against non-urbanised or ‘barbarian’ peoples). Thus, ethnic hatred was by no means absent from mass violence in the preor early modern worlds. Consider the massacre of Romans and Italians in Asia Minor (roughly modern-day Turkey) under the orders of the king of Pontus, Mithridates VI, in the first century B C E. During a war with Rome, Mithridates ordered that all Romans, Italians and freed slaves of Italian birth be massacred on a particular day – an order that was, in some cities, carried out so zealously and sadistically that the right of asylum (where people might seek refuge in temples) was ignored, and children were butchered in front of their parents.73 Both the ‘racial’ focus of the order – targeting even freed slaves of Italian birth – and the cruelty of the attack demonstrate the depth of ethnic hatred in that case.74 Nevertheless, in the pre- and early modern periods the more general absence of racialist ideology, and the sometimes highly pragmatic nature of mass violence, is sometimes encapsulated in the notion that people were 69 Lemkin, Lemkin on Genocide, p. 5. 70 See, on the Roman Empire, S. P. Mattern, Rome and the Enemy (Berkeley: University of California Press, 1999), and for a classic ancient expression, the so-called ‘Melian Dialogue’ in Thucydides 84–116, esp. 5.95–7; see also van Wees, ‘Ancient world’, 250–6. 71 Isaac, Invention of Racism, p. 222. T. M. Lemos in Violence and Personhood in Ancient Israel and Comparative Contexts (Oxford University Press, 2017), esp. ch. 2, however, discusses the dehumanisation of enemy groups by the Israelites, Assyrians and Egyptians. 72 Van Wees, ‘Archaic and classical Greece’, 33 uses this phrase in relation to different Greek ethne¯ (broader groupings than the typical Greek city-state or polis, such as Dorian or Ionian), but it sums up well the general impact of ethnic prejudice in ancient mass violence. 73 App. Mith. 22–3. 74 Taylor, ‘Introduction’, 21.

44

Introduction to Volume I

killed not for ‘who they were’, but for ‘what they did’.75 This perceived contrast perhaps represents the lurking ‘conceptual constraint’ of a preoccupation with the ideological racist violence of the Holocaust; yet it also oversimplifies many premodern experiences.76 Many of the victims of ancient mass violence – whether the Mithridatic massacre or Roman-incited violence against particular tribes or cities – were killed for no more than ‘who they were’ – individual responsibility for whatever perceived wrong had occurred was effaced, and instead attributed to a group as a whole, with children killed, or enslaved, for no reason other than their membership in the group.77 This brief discussion raises a question to which this volume – along with other transhistorical studies78 – may help construct an answer: is the phenomenon of group-based mass violence aimed at the destruction of groups in the pre- or early modern eras so different from the modern period that we are discussing quite different phenomena, rather than the evolution of a particular phenomenon? Our answer is the latter, given that, with some exceptions, most of the cases discussed in this and subsequent volumes of The Cambridge World History of Genocide do fit the Convention’s definition of genocide. The second question is the thorny ‘value problem’ – namely, whether mass violence in the premodern and early modern periods occurred in a different moral context, as opposed to the moral opprobrium which genocide incurs today as the ‘crime of crimes’. Is it therefore anachronistic to use that term to study mass violence in early history? It is certainly true that slavery, for its part, was universally accepted as a social form, and that even mass violence might be boasted of, such as in Caesar’s commentaries on his Gallic conquest, or become the subject of public praise and memorialisation, such as the execution of captives depicted on the column of Marcus Aurelius in Rome. Yet from antiquity through to the Spanish conquest of the New World, such violence also had its critics.79 Pliny the Elder termed Caesar’s huge death toll in his wars an injury on the 75 For example Barrandon, Massacres, p. 336; D. Konstan, ‘Motivations and justifications’ in Cultural History of Genocide in the Ancient World, ed. Taylor, pp. 51–71, at p. 71. 76 See D. Moshman, ‘Conceptual constraints on thinking about genocide’, Journal of Genocide Research 3:3 (2001), 431–50. 77 Taylor, ‘Introduction’, 26–9. 78 For example Kiernan, Blood and Soil; Taylor, ed., Cultural History of Genocide in the Ancient World; Bloxham and Moses, eds., Oxford Handbook of Genocide Studies. 79 See Taylor, ‘Introduction’, 25–7 for a brief discussion of ‘ethical constraints’ on ancient mass violence.

45

t. m. lemos, tristan s. taylor and ben kiernan

human race (Natural History 7.92) and Polybius recorded mixed, even critical reactions to Rome’s destruction of Carthage (36.9.3–17). As David Konstan has argued, it is clear that even in the harsh world of antiquity it was thought that such mass violence could provoke some anxiety, and required justification.80 Jumping ahead millennia, Las Casas would similarly critique Spanish actions of ‘extermination’ in the Americas.81 A key difference is that in the modern period, genocide cannot ever be seen as warranting justification (only cover-up or denial), whereas in the premodern world it could be, in certain contexts – in particular as retribution for a perceived wrong – considered justifiable.82 Interestingly, contrasting views on the First Crusade (1096–9), discussed in two chapters in this volume, serve to illustrate some of the scholarly differences that exist over both the phenomenological and moral questions. Thomas Fudge, without engaging in genocide denial, argues in Chapter 17 that the massacre perpetrated by the Crusaders when they took Jerusalem in 1099 has undergone exaggeration in the sources and was not exceptional for its time – it was driven in part by the difficulties they encountered ‘struggling in a foreign environment . . . after undertaking the long and arduous journey from Europe’, as well as strategic concerns – to remove the threat from the defenders in the city. As an unexceptional event, it thus would not have been subject to significant moral opprobrium – in contrast to today. Certainly, from antiquity to the Middle Ages, the sack of a city was often a scene of extensive violence, exacerbated by the difficulties encountered during a long siege, and mass violence in these contexts often had ‘pragmatic’ motives associated with a military campaign. However, Maya Soifer Irish suggests in Chapter 16 that there may have been more at work than anger at a difficult campaign, or pragmatic considerations, in Jerusalem in 1099. She shows that from the beginning of the crusade in 1096, even before setting out for Jerusalem, crusaders targeted Jews for mass killing in their homes in Germany and elsewhere in Europe, slaughtering up to 2,000 in Worms and Mainz, and others in Trier, Metz, Prague and possibly Rouen. Similarly, the historian Benjamin Kedar draws attention in particular to Albert of Aachen – now considered among the most reliable of the Christian sources – who records that on 17 July, the third day after the city’s fall, the crusaders in Jerusalem had resolved to conduct what Kedar calls an ‘organized, total massacre’ of Muslims remaining in the city, suggesting again that there was 80 Konstan, ‘Anger, hatred and genocide’, 187. 81 Kiernan, ‘Is “genocide” an anachronistic concept?’, 532–3. 82 See e.g. Konstan, ‘Anger, hatred and genocide’, 187; van Wees, ‘Ancient world’, 256–7.

46

Introduction to Volume I

more at work than anger created by the difficulties of the campaign. Indeed, Kedar goes further to argue that ‘the massacre in Jerusalem was considerably more extensive than in other towns of that time that were taken by storming’.83 Certainly, one of the earliest sources on the massacre, found in a cache of Jewish documents from the Geniza synagogue in Cairo, shows the shock waves that spread rapidly, and suggests that, despite the exaggeration of the sources, significant devastation had taken place. Thus, a letter which Kedar suggests was written only two weeks after the July 1099 crusader massacre of Jews and Muslims in Jerusalem, states: we received tidings of the great disaster and all-comprising visitation, which befell our brothers, the Jews living in the Holy City, may God restore it forever, the holy Torah scrolls, and the captives, suffering multiple vexations inflicted upon them by the enemies of God and haters of his people . . . Our kidneys are in flames, because of the burning of the house of our God, our glorious sanctuary, may God forgive – our sun has darkened and our stature is bent . . .84

While it is something to be mindful of, the clear moral difference between the modern world after the ‘human rights revolution’ that followed the Second World War and the premodern and early modern worlds should not serve as a barrier to analysis of the phenomenon of mass violence in these periods. Rather, this difference serves as one more contextual variable to bear in mind. As van Wees has pointed out, premodern mass violence that bears the hallmarks of what we might term ‘genocide’ was nonetheless relatively rare, despite this different moral context. That alone makes it worth asking in what contexts such violence was regarded as legitimate and, in any case, why it happened when it did.85

The Chapters in this Volume This volume, covering several millennia – spanning prehistory, history and multiple continents – predictably includes a diverse set of cases of genocidal and exterminationist violence, influenced by a wide range of factors and uncovering a complex set of patterns. These are shaped by time period and 83 Benjamin Z. Kedar, ‘The Jerusalem massacre of July 1099 in the Western historiography of the Crusades’, Crusades 3 (2004), 15–75, at 64, 40, 53, 66, 68, 74. 84 Ibid., 63; Edward Peters, ‘The siege and capture of Jerusalem (June–July 1099)’ in The First Crusade, 2nd ed. (Philadelphia: University of Pennsylvania Press, 1998), ch. 8, pp. 271–2. 85 Van Wees, ‘Archaic and classical Greece’, 19.

47

t. m. lemos, tristan s. taylor and ben kiernan

geography, as might be expected, but the most important patterns that emerge are arguably those that cross-cut historical periodisations and geographic regions. We will begin with these intersecting factors and trends. They fall into categories – material, political, sociocultural and ideological – that are at best imprecise and overlapping. We use them here largely for convenience.

Material Factors Different chapters in this volume present cases in which material factors in some way influence or serve as a catalyst for genocidal violence of different kinds – whether genocide of a group (in whole or in part), genocidal massacres, ‘genocidal moments’ or some other form of targeted mass violence, such as extermination. These material factors range from climatic or ecological changes to demographic pressures, from agricultural crises and shortages of land to competition over other scarce resources. While the thematic chapters by Bridget Conley and Alex de Waal on famine and genocide (Chapter 4), and by Francis Ludlow, Chris Morris and Conor Kostick on climate, violence and genocide (Chapter 5), of course focus on such material dimensions, the case studies by T. M. Lemos (Chapter 6), C. P. Lewis (Chapter 15), Maya Soifer Irish (Chapter 16), George Dutton (Chapter 20) and Danielle Kurin (Chapter 22) also address such material issues, making this one of the major themes of this volume. The fact that many chapters assess genocidal violence from such widely divergent time periods and geographic regions – ancient Israel, medieval Europe, fifteenthcentury Vietnam, pre-Columbian South America, and elsewhere – suggests that ecological, demographic and agricultural crises should be taken into account as some of the major long-range influences to consider in a multicausal approach to understanding genocide. For example, Lemos’ chapter on ancient Israel highlights a confluence of factors, with population increases and land shortages combining with highly dichotomised views of ethnicity and h¯erem ideologies to produce extreme interethnic violence and genocidal ˙ imaginaries. Dutton and Kurin, in their respective chapters, similarly address a combination of material factors, such as ecological conditions, population increases and land shortages, as well as cultural and/or political influences that contributed to the outbreak of genocidal violence. Soifer Irish writes of the impact of ‘famines, plague, wars, civil strife and an economic downturn’, and states that those who exacted genocidal violence against Jews in medieval Europe were ‘driven by a mix of economic grievances, resentment of royal authority and religious hatred’. Such a multi-causal approach has increasingly 48

Introduction to Volume I

become the norm in Genocide Studies and is what many if not most of the chapters in this volume utilise, with a consideration of long-term materialist dimensions forming an important part of the methods of their respective authors. These of course combine with what might be called the short-term factors, the decision-making of leaders who launch genocidal violence.

Political Factors Unsurprisingly, political factors emerge as another important catalyst of genocidal violence in many chapters of this volume. Leaving aside the question of whether or not genocide can occur without states, an issue that has been raised by some scholars of genocide, it is clear that the activities, ambitions and predations of states are linked to genocidal violence in a large number of cases. Imperialism and the expansionist violence of imperialist regimes is an important theme of this volume that is treated in many of its chapters (Lemos (Chapter 6), Lemos and Richardson (Chapter 7), Taylor (Chapters 10 and 11), Gambash (Chapter 12), May (Chapter 19), Dutton (Chapter 20), Aquil (Chapter 21)). One finds in cases of imperialist expansion a constellation of many kinds of violence, some of it genocidal and some not falling under that rubric. Still, it would be erroneous to conclude that genocidal violence and imperialism go hand in hand since this form of violence was in fact atypical under some major early historical imperialist regimes, including those of the Mesopotamians (as Lemos and Richardson argue).86 This variation in imperial practice is important for at least two reasons: examining particular cases of imperialist expansion occurring in specific historical contexts allows us to go beyond reductionism in positing imperialism as a cause of genocide; and the wide range of violent activities performed by so many imperialist regimes offers scholars of genocide an opportunity to better theorise the relationship between genocide and other forms of violence. The latter is arguably an area where the interdisciplinary field of Genocide Studies requires more extensive and more nuanced research, and while the chapters in this volume do not specifically engage in this theorising, they offer rich data sets for scholars who wish to do so. Genocide of course also occurs outside of imperial contexts, practised within national boundaries or between groups or small-scale polities more properly seen as competitors in regional contexts rather than as imperial states per se. Chapters such as those of Helle Vandkilde (Chapter 1), 86 See also Philip Spencer, ‘Imperialism, anti-imperialism and the problem of genocide, past and present’, History 98:4, no. 332 (October 2013), 606–22.

49

t. m. lemos, tristan s. taylor and ben kiernan

T. M. Lemos (Chapter 6) and Danielle Kurin (Chapter 22) offer cases that fall into these categories. In these cases, groups or small polities respond to various factors – demographic pressures or competition over resources – through political interventions that might make use of small-scale military forces or the marshalling of civilian groups for violence targeted against ethnic ‘others’ or different groups seen as rivals or as endangering the state in some way. Thus, connections between political structures and genocidal violence show a great deal of diversity in form.

Sociocultural Factors The sociocultural factors influencing the development of genocidal violence are also manifold, as the chapters in this volume amply demonstrate. One of the most prominent to emerge is gender. Various chapters here present examples of gendercide – the targeting of one gender in situations of genocidal violence (see especially Vandkilde (Chapter 1), Jones and Lower (Chapter 3), Cartledge (Chapter 8), Taylor (Chapters 10 and 11), May (Chapter 19), Aquil (Chapter 21) and Kurin (Chapter 22)). In many of the cases that emerge in this volume, it is males of specific groups who are targeted for slaughter while females and sometimes male infants and/or children are instead taken as slaves or forcibly assimilated into the perpetrator society (May and Dutton provide examples of the latter). Sexual violence against women also occurs in genocidal contexts, as is well known (see for example Jones and Lower (Chapter 3) and Aquil (Chapter 21)). The relationship between identity or identity formation and genocidal violence also emerges as a topic of discussion in different chapters (Vandkilde (Chapter 1), Jacobs (Chapter 2), Lemos (Chapter 6), Champion (Chapter 9), Tin-bor Hui (Chapter 14) and Pegg (Chapter 18)). These identities take different forms, whether political, ethnic or religious. It is not simply a matter of group identity leading one group to target another for exterminationist violence. Rather, the impact of genocidal violence can in fact shape – it can construct, reinforce or revise – identities for both perpetrators and victims. Thus, genocide can be a tool as well as a product of ethnic boundarymarking and identity formation. Indigenous conceptions of totalising violence are an interesting cultural aspect of the study of genocide that emerges in certain chapters in the volume. For example, one sees in the chapters by T. M. Lemos and Danielle Kurin indigenous terminology used for totalising violence. Lemos writes in detail of the Israelite term h¯erem, which refers to a form of violence ˙ in which Israelites are commanded by their deity, Yahweh, to completely 50

Introduction to Volume I

annihilate particular ethnic groups. These groups occupied the land that this deity had designated for possession by the Israelites alone. At least one other group in the ancient Levant, the Moabites, also thought in terms of this concept. Kurin, for her part, writes of huañuchiyta atipani, ‘a Quechua phrase that connotes the utter and absolute power to manufacture death’. While these terms and the conceptions of violence they lexicalise do not entirely overlap with our word ‘genocide’, the fact that some premodern cultures developed such terms militates against the critique that applying the term ‘genocide’ to premodern societies is ‘anachronistic’.87 On the other hand, however, to better understand violence in general and genocidal violence in particular, we must ask which social groups and formations are identifiable in different societies, which of them have greater meaning, and which play an organising role in the society in question. Arguably, these are not just or even primarily questions of historical era – which groupings were important in ancient societies – but of cultural and social difference. In one society, certain groupings might be clearly discernible and thus potentially become targets. In another society, the same groupings may exist but may not be discernible or prominent at all, while other groups might be at the forefront of social tensions and thus become clear targets. Of course, every society is comprised of social groups, and complex societies are comprised of very many, but it seems problematic to assume that these will always map from one society on to another. The chapters in this volume support treating such matters with caution. An important example for this failure to map, if you will, is the specific social and cultural grouping of the city. As noted, cities in many ancient societies were not merely residential structures or even merely political structures. They were in so many cases an identity group, and one imbued with symbolic and other cultural meanings, often including ethnic identity. Thus, it is not surprising that so many of the chapters in this volume that are devoted to ancient societies speak of ‘urbicide’ – the targeted destruction of a city as a political, social and cultural entity – as a form of genocide, or raise the important question of whether urbicidal violence constituted genocide in societies such as ancient Mesopotamia (Lemos and Richardson (Chapter 7)), the ancient Greek world (Cartledge (Chapter 8), Champion (Chapter 9)), or the ancient Roman world (Taylor (Chapters 10 and 11), Gambash (Chapter 12)). This, again, is another instance in which the chapters

87 Kiernan, ‘Is “genocide” an anachronistic concept?’

51

t. m. lemos, tristan s. taylor and ben kiernan

in this volume either directly address a topic of major importance to the theorising of genocide, or offer case studies that may aid in such theorising. Religion – religious identities, practices and ideologies – emerges as another important aspect of sociocultural analysis in various chapters. Steven Jacobs focuses on the overall theme of religion, but it is also a major topic in the case studies by Gil Gambash, Carl Rice, Maya Soifer Irish, Mark Pegg, George Dutton and Raziuddin Aquil. Interestingly, the one chapter in the volume that treats religious texts at length, that by T. M. Lemos, does not refer to Israelite ‘religion’ in any way. This is a reflection of the fact that the category of religion has become a highly contested one in the fields of Religious and Premodern Studies in the past ten to twenty years, with more and more scholars questioning whether or not ancient groups had a concept of religion, whether the concept of religion is an invention of modernity or purely a ‘western’ concept, and whether any terms in Hebrew, Greek or Latin – including the word religio itself – should even be translated as ‘religion’.88 The latter term has a range of meanings in ancient texts, with ‘worship practice’ or ‘rite’ or ‘obligation’ better encapsulating the sense of religio than the freighted English term ‘religion’ does, according to some scholars.89 This is not yet a settled question in the field of Religious Studies. Still, it is a matter with important implications for how genocidal violence – and particularly violence aimed at groups that are sometimes or most often seen as religious groups – should be understood and theorised. Our volume does not aim to settle these questions, but provides evocative data to ground future research in this area. Questions surrounding religion in some sense relate to the broader issue of target groups and what counts as genocide (for instance under the UN Genocide Convention definition), so to speak. If genocide is exterminationist violence performed against a social or cultural grouping with an identity obtrusive enough for the group to be seen or constructed as a target of genocidal violence, then one must ask which social groupings might fall under this rubric 88 See esp. Daniel Dubuisson, The Western Construction of Religion: Myths, Knowledge, and Ideology (French ed. 1998), trans. William Sayers (Baltimore: Johns Hopkins University Press, 2003); Hans G. Kippenberg, Discovering Religious History in the Modern Age (German ed. 1997), trans. Barbara Harshav (Princeton University Press, 2002); Guy G. Stroumsa, A New Science: The Discovery of Religion in the Age of Reason (Cambridge, MA: Harvard University Press, 2010); Brent Nongbri, Before Religion: A History of a Modern Concept (New Haven: Yale University Press, 2013); and Carlin A. Barton and Daniel Boyarin, Imagine No Religion: How Modern Abstractions Hide Ancient Realities (New York: Fordham University Press, 2016). 89 See e.g. Nongbri, Before Religion, esp. ch. 2 and conclusion; Barton and Boyarin, Imagine No Religion.

52

Introduction to Volume I

of potential targets. The question of whether ancient groups, or groups in the premodern era, or groups outside of ‘western’ contexts, thought of themselves in terms of a ‘religion’ raises the additional question of whether or not such groups were organised into religious groups or subgroups. Also, did those groups or subgroups cross-cut other social groupings, such as ethnicity, culture, socio-economic class or race? Presumably, a group could not be targeted qua religious group unless religion were a category that existed in that specific culture, region, or era. If early Christians were not members of a ‘religion’, then what were they – what kind of social or political grouping were they being targeted for being a member of? These questions of religion and other overlapping groupings relate in important ways to cases explored in several chapters of the volume (Jacobs (Chapter 2), Lemos (Chapter 6), Gambash (Chapter 12), Rice (Chapter 13), Pegg (Chapter 18), among others). Another important question related to the issue of which groups were targeted for genocidal violence is whether ‘national, ethnical, racial or religious’ groups – the Genocide Convention’s list of protected groups – must be modified depending on the culture or society one is examining. That is, in order for the category of genocide to be legitimately applied to ancient or non-‘western’ societies, it stands to reason that the list of possible target groups considered by scholars must be expanded or modified in order to align with the social groupings that play an organising role in that particular society. For example, while scholars may debate whether or not groups in the ancient world thought in terms of ‘religion’, certainly they thought in terms of ‘cult’ and ‘cultic’ associations – that is, worship practices and groups geared towards particular deities and temples. Should not one ask, then: if a ruler in antiquity attempted to, for example, completely wipe out the priesthood of a particular deity, should not this be classified as ‘genocide’ (of a ‘religious group’)? If a king sought to destroy the prophets, devotees and iconography of a particular cult, should this too be considered genocidal? It is certainly difficult to see how such destruction could not be construed as acts committed with intent to destroy a ‘religious group’, in whole or in part, under the UN convention’s definition of genocide. But these are among the many important questions that the chapters in this volume raise for scholars of genocide and religion alike.

Ideological Factors Each of the sections discussed above includes phenomena with elements that could be included under the heading of ‘ideology’. For example, religion – however one defines it or sees it as having emerged historically – involves 53

t. m. lemos, tristan s. taylor and ben kiernan

cultural and ideological elements in every case. Imperialism, especially in its more large-scale forms, has involved complex ideologies of power relations involving categorisations and hierachies of ethnicity, geography and – at least in the early modern and modern eras – quite often race. Racial thinking is a less important theme in this volume than it is in the two volumes of The Cambridge World History of Genocide treating later periods. While it would be wrong to conclude that racial categorisations and ideologies were absent in antiquity,90 race is not a category emphasised by many chapters in Volume I. Still, ideologies that dehumanise target groups or construct such groups in very dichotomising terms are described in certain chapters. Examples are the h¯erem and apocalyptic ˙ ideologies described by T. M. Lemos (Chapter 6), the animalising conceptions of the Cham by their Vietnamese conquerors described by George Dutton (Chapter 20) and the anti-Jewish ideologies detailed in the chapter by Maya Soifer Irish (Chapter 16). In addition to the various chapters that address imperialism, other chapters discussing ideological elements involved in genocidal violence are those by Michael Champion (Chapter 9) and Victoria Tin-bor Hui (Chapter 14), who writes of how a Confucian ideology of ‘unity’ was marshalled in the commission of genocidal violence: ‘Genocide, extermination, mass killing and human sufferings thus recurred with the drive to achieve “unity”, expand “unity” and maintain “unity”.’ Worth noting, too, is how the genocidal ideology of earlier texts in some cases inspired or was used to legitimate later genocidal violence (see Lemos (Chapter 6), Soifer Irish (Chapter 16)). This afterlife of genocidal ideologies, as it were, will become apparent in the subsequent volumes of The Cambridge World History of Genocide.

Changes over Time Considering the very long time span covered by the chapters in this volume – several millennia of prehistory and history – it is reasonable to consider whether notable shifts occurred over that time in the practice of genocide, in its conceptualisation, or in the sources available for studying 90 Arguably, the biblical books of Ezra and Nehemiah – particularly Ezra 9 – contain a very early example of racial thought, where Israelite holiness is located in its ‘seed’ (i.e. semen) and is profaned through mixture with foreigners, who are constructed in highly negative terms, whose status is immutable, and who must be ejected from Israelite society. Benjamin Isaac does not discuss this case but elaborates on many others in The Invention of Racism in Classical Antiquity.

54

Introduction to Volume I

it. In many ways, it is in the latter two areas that changes are clearest, with the last of those presenting perhaps the most marked shift. While urbicide seems more prominent as a form of genocidal violence in antiquity than in later eras, one could argue that this is as much due to societal and cultural differences as to any changes in perpetrator practices that may have occurred over time. This would contrast, to offer a clarifying example, with the technologies of genocide, which have certainly transformed in the last century as compared with earlier eras, when the use of poisonous gases to take myriad human lives would have been impossible. Similarly, racial ideologies became far more prominent in genocidal violence in the early modern era and following – as the later volumes of The Cambridge World History of Genocide will make clear. The pseudo-scientific racial ideologies of Nazism would have been impossible in antiquity, though that era was not devoid of views that one may characterise as racial or proto-racial, and marked differentiations between in-group and putative out-group(s) targeted for genocide are a commonplace of totalising violence, as so many chapters in this volume make clear. Shifts in the sources by which one can study genocide are more like technological changes in the clear-cut way in which they have created change. The primary shift in the nature of sources occurred with the advent of writing. The chapters in this volume that deal with prehistoric societies (Vandkilde (Chapter 1), Kurin (Chapter 22)) are naturally reliant on archaeological evidence to reconstruct genocidal violence, whereas those from literate societies rely very heavily on written sources. (Some chapters, e.g. Chapter 6 by T. M. Lemos, draw on both written and archaeological evidence.) This emphasis on literary evidence is not surprising but is still noteworthy in light of a major weakness of written evidence, which is that these sources for most periods of history have overwhelmingly been the product of elites. Archaeological evidence offers, one could argue, a more ground-level, egalitarian view. However, because of the way in which legal and political bodies and genocide scholars have all emphasised intentionality in determining whether mass violence should be deemed ‘genocide’, written sources have become especially useful or even, in some cases, indispensable. Archaeological evidence so often cannot reveal genocidal intent. By contrast, written sources may make perpetrator intent all too clear, but they can also shift scholars’ attentions too readily to elite interests and to ideological factors over other longer-term phenomena that may establish some of the preconditions for genocidal violence. It is for this reason, among others, that the multicausal approach to genocide that so many scholars utilise in this volume is all 55

t. m. lemos, tristan s. taylor and ben kiernan

the more valuable and noteworthy. Their approach reveals with great clarity the complex and multifaceted nature of genocidal violence, which bridges, manipulates and weaponises the material conditions that make life possible or impossible, the political and social structures that give life shape, and the cultural and ideological patterns that give life meaning.

56

part i *

THEMES OF GENOCIDE THROUGH HISTORY

1

Genocide before the State? helle vandkilde

On the Question of Non-state Genocide – Preamble Genocide before the emergence of states rarely figures in theoretical debates or in interpretations of concrete sets of prehistoric data. Unlike the term ‘genocide’, the terms ‘warfare’, ‘warriors’, ‘battlefields’, ‘trauma’, ‘massacres’ and ‘mass graves’ are common in archaeological research proceedings. This chapter focuses on European prehistory with non-state cases in other regions and eras as an analytical backdrop. It therefore relies primarily on archaeological and bioarchaeological sources on warfare, burial rites and ancestry. Although a difficult subject, there is no viable reason to avoid the important and currently relevant topic of genocide. This chapter develops three lines of reasoning in order to track early forms of genocide, namely warfare, abnormal graves and genetics in societies that may be described as tribal. The term ‘tribal’ envelops a wide array of societal forms, including mobile hunter-gatherers, pastoral herders and dispersed household-organised agriculturalists and communities with formalised hierarchies and central hubs. Since it is possible that genocidal behaviour changed with the coercion, taxation and bureaucracy of states,1 this chapter does not examine pristine states and cases of states conquering tribal areas. It is, however, relevant to draw on ethnohistorical records. Intertribal and clan wars at the fringe of state control can illuminate some of the mechanisms at work in prehistory. Military institutions, inventions and expansions are certainly not confined to state societies, which, in non-Mediterranean Europe, were only implemented in the centuries after the collapse of the Roman Empire. In the first globalisation-like eras of the Chalcolithic and the Bronze Age c.4000–700 B C E, clusters of urban societies and states developed in the 1 C. Tilly, Coercion, Capital, and European States, AD 990–1990 (Hoboken, NJ: WileyBlackwell, 1990).

59

helle vandkilde

southern zone of Afro-Eurasia.2 This urban zone contrasted with the changeable diversity of non-urban stateless societies inhabiting the north in the same periods. Considerable demographic change took place, with migratory movements as one of the key factors, and multi-scalar connectivity glued together large tracts of Afro-Eurasia, effectuated not only by people but also by innovations, ideas and commodities (notably metals) on the move. Genocide is a strong concept, coined after World War Two to synthesise the social phenomenon of intentional ideology-driven annihilation of other human groups by way of systemic and extreme construction of ‘otherness’. The demonising and dehumanising of other people are legitimised through notions of ethnicity, cultural tradition, national identity, social position or religion, quite often in a combined form. The culmination of these impulses is the ruthless killing of members of the other group, but precursors to this killing include removing their sense of belonging to their own group, causing bodily and mental harm and humiliation, destroying essential life conditions and preventing births and thus long-term social and biological reproduction. The genocide vocabulary somehow echoes what we know of the deep past, but it also poses a challenge when we are confronted with primarily archaeological sources, which are currently of insufficiently high resolution to reveal the scale, systematics, logic, intent and drivers behind the maltreatment of other groups. Some studies distinguish between genocide and massacres, claiming that genocide is dispassionate and instrumental (a mere obeying of orders) whereas massacres are passionate mass killings.3 Due to archaeology’s reliance on material traces of past actions, this distinction makes it more difficult to identify early forms of genocide. It could therefore be argued that it is more fruitful to follow a track that pursues the underlying intentionality of concrete actions. P. Willey specifies that a massacre is an event of purposeful violence causing the death of a substantial portion of a group, often combatants and non-combatants alike.4 This definition was developed from archaeological data, in particular the 300–400 skeletons at the precolonial site of Crow Creek in South Dakota. The numerous casualties of the so-called Bone 2 H. Vandkilde, ‘Bronzization: the Bronze Age as pre-modern globalization’, Prähistorische Zeitschrift 91:1 (2016), 103–23. 3 C. Alfsdotter and A. Kjellström, ‘The Sandby Borg massacre: interpersonal violence and the demography of the dead’, European Journal of Archaeology 22:2 (2019), 210–31. 4 P. Willey, ‘Bones in the village: fragmentary human bones and scattered contexts from the Crow Creek village’ in Massacres: Bioarchaeology and Forensic Anthropology Approaches, ed. C. P. Anderson and D. L. Martin (Gainesville: University Press of Florida, 2018), pp. 94–115.

60

Genocide before the State?

Bed and the Village Area clearly derived from one single successful attack led by intruders from the Central Plains in c.1330 C E. The existence of two opposing tribal groups – the ‘us against them’ theme – was clearly a motivating factor in this massacre. The Crow Creek massacre occurred in the midst of widespread violence fuelled by in-migration from the Central Plains. The fortified village at Crow Creek was subsequently resettled. The cultural form of the settlers was a hybrid between the indigenous Middle Missouri tradition and the intruding Central Plains tradition. Young adult women are distinctly under-represented in the Bone Bed skeletal debris and the emerging so-called Coalescent tradition may therefore be explained as the result of biocultural meetings between exogenous males and indigenous females. This pattern of forced inclusion in the victors’ society is recognisable across periods and cultures and is hence fairly universal. On a similiar line, D. Komar reveals substantial similarities between selected mass graves in the modern era and in the past, and relates them to genocide, which, in her analysis, is distinct from more socially acceptable forms of violence and warfare. Two main variations can be identified from modern data, and, as we shall see, these variations are also broadly identifiable in prehistoric societies: considerate burial occurs exclusively in multiple graves executed by friendly forces (self) while erratic commingling of multiple interments occurs exclusively from the agency of enemy forces (other). These two main groups can be summarised as follows: 1. order/self: burial by the agency of allies or the deceased’s own community or family; and 2. disorder/other: burial by the agency of the enemy.5 Three key elements may suggest genocide: the physical element of mass graves as acts of genocide; the victims’ shared identity in terms of, for example, ethnicity or religion; and the psychological or ideological element of the perpetrators’ intention to entirely or partly destroy the outsider group. Crow Creek fits the first two elements and also likely the third element. These three elements serve to demonstrate that mass graves, massacres and genocide all belong to the same culture of violence and oppression, and that they all result from concrete action. This aligns with the phenomenon of ‘genocidal massacres’ or, more broadly, ‘genocidal moments’. Rather than necessarily emanating from a universal state or master plan for genocide, genocidal massacres are brief episodes of killing that arise from local outbreaks of conflict between specific communities or groups of people, and 5 D. Komar, ‘Patterns of mortuary practice associated with genocide: implications for archaeological research’, Current Anthropology 41:1 (2008), 123–33.

61

helle vandkilde

may serve political ends as warnings for all parties involved. Theoretically, genocidal moments that aimed to annihilate a particular population could have occasionally formed a long series of such episodes and hence contributed to a general climate of genocide being repeatedly reproduced by a number of oppressive means. This explanation may fit the increasingly strong evidence in Europe and worldwide for the territorial expansion of exogenous people, who, in a prolonged process, engulfed the original inhabitants and their culture. Ethnohistorical sources notably describe the predatory expansion of certain tribes.6 Such expansions in some instances transformed into the emergence of vast and powerful tribal empires.7 Historically recorded genocides pertain to war,8 which is also the kind of social environment that surrounds prehistoric abnormal graves and mass fatality sites. Societies without the state are nevertheless neither innately violent and anarchic nor innately peaceful and civil. The prehistoric past, with its vast social and cultural diversity, thus resists Hobbesian or Rousseaunian preconceptions and instead suggests historical rhythms run by complexities of human and non-human factors. In alignment with the data, prehistoric realities were diverse and changeable over the millennia. In order to interpret them accurately, it is therefore important to consider the precise sociocultural context, including extrasocietal conditions such as climate change, migration and globalisation. This chapter will show that life conditions for our ancestors were sometimes harsh and included the victimisation, dehumanisation and annihilation of perceived others.

State of the Art, Data and Methods In 2008, Komar noted the complete absence of genocide from archaeological interpretations of violence, conflict and mass death.9 This absence could mean that genocide did not exist in prehistory or that archaeologists are still not able to recognise it. Although this absence largely persists today regarding prestate contexts, different data and approaches currently used suggest the latter explanation as more plausible. Interestingly, the Routledge Handbook of the Bioarchaeology of Human Conflict only applies the concept of 6 M. D. Sahlins, ‘The segmentary lineage: an organisation of predatory expansion’, American Anthropologist 63:2 (1961), 332–45; K. F. Otterbein, How War Began (College Station: Texas A&M University Press, 2004). 7 P. Hämäläinen, The Comanche Empire (New Haven: Yale University Press, 2008). 8 Komar, ‘Mortuary practice’, 123–33. 9 Ibid., 131.

62

Genocide before the State?

genocide to the chapters on modern conflict.10 In recent books on the bioarchaeology of conflict and massacres, the G-word is still largely lacking,11 yet, with regards to multiple graves, the underlying hatred towards and contempt for victims are now noted, together with the occasionally apparent systematic planning and meticulous execution of the violence. Data on Warfare This stems from weapons made for war and war-related blunt force or sharp force trauma on skeletal remains found mainly in (both normal and abnormal) burials.12 Weapon combinations, weapon wear trace, rock art imagery and fatality sites provide indications of how warfare was performed and organised. There is little clarity regarding the frequency of ancient warfare. Depending on period and place, there may have been frequent small-scale warfare, periodic larger-scale conflict or some other pattern of fluctuation over time and geographical space. Large fortified hubs known especially from the Bronze Age and later periods suggest the more or less successful control of warfare as a tool of military power, which may have been more difficult to obtain in societies with dispersed settlements. It was only in the wake of Lawrence Keeley’s seminal book in 1996 that evidence of multiple war-related trauma was discovered in prehistoric or non-state settings around the world.13 Today there are many studies of prehistoric warfare and interpersonal violence more generally. Data on Abnormal Graves This is substantial. Abnormal graves mostly contain more than one individual and can sometimes contain hundreds of individuals. They diverge in a number of respects from the norm of burying kin and family with love and care. Normative burials are by far the most frequent. Monumental stone graves built for community interments are a cultural characteristic of many first farming communities who settled in Europe from c.7000 B C E. Alongside these ancestral monuments with several 10 C. Knüsel and M. J. Smith, eds., The Routledge Handbook of the Bioarchaeology of Human Conflict (New York: Routledge, 2013). 11 For example M. Smith, Mortal Wounds: The Human Skeleton as Evidence for Conflict in the Past (Barnsley, UK: Pen & Sword, 2017); and Anderson and Martin, eds., Massacres. See however M. Teschler-Nicola, F. Gerold and M. Bujatti-Narbeshuber et al., ‘Evidence of genocide 7000 BP – neolithic paradigm and geo-climatic reality’, Collegium Antropologicum 23 (1999), 437–50. Genocide during the Roman expansion is mentioned in M. Fernández-Götz and N. Roymans, eds., Conflict Archaeology: Materialities of Collective Violence from Prehistory to Late Antiquity, Themes in Contemporary Archaeology 5, European Association of Archaeologists (London and New York: Routledge, 2018). 12 H. Vandkilde, ‘Conflict and war, archaeology of: weapons and artifacts’ in The International Encyclopedia of Social and Behavioral Sciences, ed. J. D. Wright, 2nd ed., (Oxford: Elsevier, 2015), I V, pp. 606–13. 13 L. H. Keeley, War before Civilization: The Myth of the Peaceful Savage (Oxford University Press, 1996).

63

helle vandkilde

burials, and particularly after c.2900 B C E with the so-called Steppe Influx, a clear preference exists for individual inhumation in earthen graves in flatgrave cemeteries or below burial mounds. From c.1400–1300 B C E, cremations are prevalent across Europe. Abnormal graves usually have a balanced demographic profile with all age groups and both sexes represented, hence meeting the definition of massacres. Modern bioarchaeology – which determines age, sex and ante- and perimortem traumata – takes the investigation and interpretation of abnormal graves a step forward. When available, the demographic profile provides an overview of bodies and the age profile in question can be compared with mortality profiles known for normal attrition, catastrophic events and preindustrial massacres, such as at Tell Brak, in Syria.14 Trauma patterns, absentees and increased mortality for some sex-age groups are obviously important. Abnormal graves have not been routinely examined for marks of violence, and the physical anthropology associated with early excavations in particular is today widely unreliable. Where visible traces of violence are absent, one should consider that skeletons are often badly preserved and that much physical violence does not leave visible traces on the skeleton. On the one hand, abnormal graves vary considerably and old excavations with outdated records tend to hamper interpretation. On the other hand, it is possible to identify recurring features and contextual similarities across time and place that aid interpretation. Many abnormal graves connect to warfare, but there are cases with no apparent reason for the untimely death and deviant interment. The phenomenon of abnormal burials encompasses bodies in all stages of decay from full articulation to complete disarticulation. Some alterations are clearly a by-product of the previous movement of bodies from the site of execution to a secondary place of interment. Such movement may occur more than once in a relatively complex manner, as is apparent from modern mass graves.15 Detailed forensics have come far in advancing our knowledge of secondary processes, which, in many archaeological studies (particularly in early studies), were confused with ritual behaviour such as cannibalism and sacrifice.16 This is not to deny that ritualised (repetitive) action formed part of 14 A. McMahon, A. Sołtysiak and J. Weber, ‘Late Chalcolithic mass graves at Tell Brak, Syria, and violent conflict during the growth of early city-states’, Journal of Field Archaeology 36:3 (2011), 201–20. 15 E. Jessee and M. Skinner, ‘A typology of mass grave-related sites’, Forensic Science International 152:1 (2005), 55–9. 16 K. Knüsel and A. Outram, ‘A comparison of human and animal depositions at VelimSkalka through an integrated approach’ in Velim: Violence and Death in Bronze Age Bohemia, ed. A. F. Harding et al. (Prague: Archeologický ústav AV Cˇ R, 2007), pp. 97–136.

64

Genocide before the State?

the pattern of violence, since it is clear that, in some cases, extreme violence formed part of the flow of actions leading to mass interment. A related concern is the question of whether interments of bodies occurred as a single event or over a period of time. Even with high-precision dating and separation layers, this is often difficult to determine with full certainty. Genomic Data Tracing the ancestry of individuals and groups has unravelled a complexity of population replacement and admixture of groups with different ancestry. Genetics-driven studies of population dynamics are currently popular within archaeology, but DNA sampling is still randomly applied to human remains and is not automatically coupled to, for example, abnormal graves. Even so, our deep history is being charted through ancient DNA with an increasing detail that has significance for answering the question of non-state genocide. In short, population affinity and biological kinship are extracted from genome-wide data. This method brings to light details of ancestry and kinship in the autosomes, which, together with the maternal side (mitochondrial genome) and the paternal side (Y-chromosome), reveals multigenerational pedigrees and admixture between groups of different ancestry. A comparison of the ancestry of the various chromosomes can reveal patterns of male and female behaviour, including sex bias during periods of population mixture.17 Present evidence for Europe discloses only patrilineal descent, patrilocality (couples living with the husband’s family) and exogamy. Connecting genetic ancestry data with material culture data can be politically delicate, with risks of abuse,18 and their coupling must follow analytical and ethical protocols. Nonetheless, archaeological interpretations are typically underscored today by a plethora of biocultural data.

Warfare without the State How and when warfare began is hotly debated, and, to some extent, the answer depends on how warfare is defined and synthesised sociologically. There are almost as many definitions of warfare as there are scholars, but many of these definitions align with the basic one proposed here: warfare is any violent encounter with other groups. More specifically, warfare is action co-ordinated within a group aiming to harm another group by means of 17 D. Reich, Who We Are and How We Got Here: Ancient DNA and the New Science of the Human Past (Oxford University Press, 2018), pp. 233–4. 18 S. E. Hakenbeck, ‘Genetics, archaeology and the far right: an unholy trinity’, World Archaeology 51:4 (2019), 1–12.

65

helle vandkilde

violence.19 This basic definition aligns well, but not flawlessly, with the archaeological record. The desire to harm may be mutual but confrontation does not necessarily ensue between equal parties in terms of preparedness, organisation, training, equipment or staffing. In fact, warfare sometimes consists of simple assaults on households with the aim of harming an opponent group as much as possible. Such assaults are essentially raiding, while feuding is reciprocated warfare that assumes a repetitive retaliating pattern. In this sense, warfare is a specific form of violent social interaction that differs from other forms of violence, such as interpersonal aggression like homicide, intra-family hostilities and wife beating, forceful sports, human sacrifice or political punishment. Furthermore, warfare is implicitly mobile, fundamentally social, and encompasses everything from improvised raiding to military campaigns framed by war-effective military institutions, which can range from small war bands with a few warriors to large armies with numerous fighters and warriors. Weapons may be multifunctional tools or made specifically for war. Lastly, warfare is not merely hostile interaction directed at other groups; it is also a flow of violent and non-violent actions that breaks into phases of persuasion, agreement, planning and co-operation. In addition to the actual violent confrontations, post-war rituals often take place as a finalising act. Several abnormal graves with multiple dead, with or without testimony for rituals, make sense as the finalising stage of warfare campaigns. It has recently been posited that warfare became a possibility when humans gained the capacity for symbolic thought, communication and sociality, which first emerged among developed hominids.20 This capacity, which, logically, also comprises the ability to negotiate peace, may not have initially been realised to any large extent, but it would accord with the above definition of warfare as violent social interaction among groups. The first evidence of violence-related trauma is very sparse, comprising only approximately twenty instances across the extensive time span of the Pleistocene age. Most of these early cases of trauma cluster in the Middle and Late Palaeolithic period, with Neanderthal and modern Homo sapiens skeletal remains. Yet, in these cases, it appears that some sort of interpersonal violence is involved, not necessarily warfare according to the above meaning. Most early cases of trauma are difficult to interpret and occur among small 19 Vandkilde, ‘Conflict and war’, 606–13. 20 M. Kissel and N. M. Kim, ‘The emergence of human warfare: current perspectives’, American Journal of Physical Anthropology 168:567 (2018), 141–63.

66

Genocide before the State?

mobile hunter-gatherer bands operating within spacious and thinly populated territories. The capacity for warfare was surely present, but potential opponent groups were often far away. It is interesting that the two convincing cases for warfare, Jebel Sahaba, in Sudan,21 and Nataruk, in Kenya,22 both date to the epi-Palaeolithic period, c.12000–8000 B C E. This marks the transition to the Holocene age when living conditions were rapidly changing and forager groups were likely under pressure, as they were retreating into attractive resource-rich refugia. These two sites can clearly be associated with warfare according to the above meaning and are the earliest known sites with mass fatalities. In the following millennia, evidence for war-related trauma becomes quite frequent, testifying to higher rates of warfare as population densities increased alongside the invention and spread of farming and other crucial innovations during the Neolithic, Chalcolithic and Bronze Ages. In many cases, migrants founded sedentary communities able to assimilate, expel and annihilate the indigenous populations who were themselves originally immigrants to the land. More organised forms of warfare developed in these societal settings, including weapons made for war (battleaxes, daggers, swords and spears) and well-staffed military institutions capable of waging war ferociously.23 Sociologically, with a footing in the thinking of Vivienne Jabri, warfare is always waged against outsiders.24 Violent conflict as such tends to arise from the construction of exclusionist identities, expediting ideas of ‘us against them’. This means that warfare embeds a potential for the annihilation of outsider others. The very instant warfare emerges on the social scene, there is the potential (which is not necessarily acted upon) for genocidal massacres and parallel means of annihilating other groups. The presence of warfare also contains the potential for peace negotiations as well as for the complexities of alliances and confederacies. Ethnohistorical sources reveal that the most ferocious tribal wars were fought against non-kin groups. A striking example of more general relevance is the patrilineal segmentary lineages of Tiv and 21 F. Wendorf, ed., The Prehistory of Nubia, 2 vols. (Dallas: Southern Methodist University Press, 1968). 22 M. M. Lahr, F. Rivera and R. K. Power et al., ‘Inter-group violence among early Holocene hunter-gatherers of West Turkana, Kenya’, Nature 529:7586 (2016), 394–8. 23 H. Vandkilde, ‘Warriors and warrior institutions in Copper Age Europe’ in Warfare and Society: Archaeological and Social Archaeological Perspectives, ed. T. Otto, H. Thrane and H. Vandkilde (Aarhus University Press, 2006), pp. 393–422. 24 V. Jabri, Discourses of Violence: Conflict Analysis Reconsidered (Manchester University Press, 1996).

67

helle vandkilde

Nuer, practising what Sahlins described as ‘predatory expansion’,25 a kinshipled impetus to move in all directions and intrude on landscapes occupied by non-kin groups. A constantly changing lineage structure prompts expansion effectuated by kinship-fostered warrior fraternities and war without constraints against other tribes. Population growth is a concomitant factor in the predatory success of segmentary descent groups: producing daughters to send away in marriage and sons to build strong war bands or found new patrilateral residences. Captive slaves are a further defining factor. A calculated outcome of warfare was often the capture of the defeated foe. In ethnographical accounts, warfare is almost always associated with the seizure of captives and their selective killing. Young women and adolescents were commonly taken captive to function as wives, labourers or cultural intermediaries, while adult males, children and the elderly were more often killed.26 The advent of institutionalised warfare may have escalated this practice, since, compared with ad hoc war parties, warrior fellowships and armies are efficient war machines. In non-Mediterranean Europe, the first military organisations, together with the first weapons made specifically for war, quite possibly coincide with the first Steppe influx of herder migrants in around 2900 B C E. Over the next millennia, a long chain of such signature events followed suit and eventually fuelled the creation of states. Prehistoric states are territorial and internally ranked polities maintained by political, military and administrative (taxation) institutions of power. The first pristine states were the city-states appearing in Mesopotamia in c.3500 B C E along with writing, while for example in northern Europe states did not emerge until the first millennium C E. Did warfare change character with the state? It is perhaps possible to answer both yes and no to this question. The state, and thus the combined oppression, law and order of Leviathan rulers,27 is often presented by scholars as either a benign peacemaker or a ruthless oppressor. State warfare was usually organised into armies, which were political and military instruments that served to keep their own unruly populations from rioting. Such actions would be prevented through the mere threat or, ultimately, exercise of institutionalised violence. In contrast, it is also well documented that, by securing the safety and prosperity of its citizens, the state apparatus had 25 Sahlins, ‘Segmentary lineage’, 332–45. 26 C. M. Cameron, Captives: How Stolen People Changed the World (Lincoln and London: University of Nebraska Press, 2016). 27 I. Morris, War! What Is it Good for? Conflict and the Progress of Civilization from Primates to Robots (London: Profile Books, 2014).

68

Genocide before the State?

a pacifying effect. States, however, tended to expand into the territory of other states or tribes. This obviously involved a significant amount of lethal violence and atrocities, as testified by mass graves (among other things). History is replete with examples of the founding of empires through systemic military warfare against resisting forces and civilians. Many of these elements, including territorial expansion and armies, emerge well before the state and in tribal areas at the margins of state authority. Extreme atrocities linked to genocidal moments and massacres are reported from tribal and clan zones and spring from local conflict rather than state interference. Many clan wars in early historic Scotland ended in massacres on both sides and included adult individuals of all ages and both sexes as well as children and young people.28 In Europe’s Iron Age, the various Germanic tribes, and similarly the British tribes, ferociously fought each other, resulting in multiple casualties, as reported by Roman sources (e.g. Tacitus) and verified by archaeological sources.29 There are similar examples from the early colonial tribal zone. Otterbein writes that the African Nuer expanded at the expense of the Dinka tribe and compares this to the Iroquois, who expanded into Huron territory in the North American Great Lake region and wiped out this neighbouring tribe. Survivors retreated into a small piece of land, which became an Iroquois hunting territory.30

Abnormal Graves and Fatality Sites A preliminary survey conducted for this chapter has identified approximately one hundred prehistoric abnormal graves and fatality sites, primarily from Europe. This number is far from comprehensive, but it may indicate the internal variability of this class of graves and their characteristics. Abnormal graves are different from normative burials in ways that lead one to infer that the underlying circumstances were highly particular. Although abnormal graves share some features with each other, they are also diverse and exhibit 28 J. Prebble, Glencoe: The Story of the Massacre (Harmondsworth: Penguin, 1969); www.scotsman.com/news-2-15012/6-of-scotland-s-bloodiest-clan-battles-1-4393383 (accessed 17 December 2019). 29 See Tacitus on Britain and Germany, trans. H. Mattingly (including the Agricola and the Germania) (Harmondsworth: Penguin, 1948); M. K. Holst et al., ‘Direct evidence of a large northern European Roman period martial event and post-battle corpse manipulation’, Proceedings of the National Academy of Sciences 115:23 (2018), 5920–5; and A. G. Western and J. D. Hurst, ‘“Soft heads” evidence for sexualised warfare during the Later Iron Age from Kemerton Camp, Bredon Hill’ in Bioarchaeology of Human Conflict, ed. Knüsel and Smith, pp. 161–84. 30 Otterbein, How War Began, pp. 98, 208.

69

helle vandkilde

18 Disordered abnormal grave Ordered abnormal grave Fatality site

22

19

15

11 16 17 7 10 6 14 3 13 5 20 8

4 9

21

12

1

2

Map 1.1 Selected prehistoric abnormal graves and mass fatality sites. 1. Jebel Sahaba, 2. Nataruk, 3. Ofnet, 4. Schöneck-Kilianstädten, 5. Talheim, 6. Asparn-Schletz, 7. Wiederstedt, 8. Halberstadt, 9. Bergheim, 10. Koszyce, 11. Eulau, 12. Tell Brak, 13. Schleinbach, 14. Nìžná Myšl’a, 15. Wassenaar, 16. Velim-Skalka, 17. Blucˇina Cerzavy, 18. Sund, 19. Cliffs End Farm, 20. Stillfried, 21. Jurilovca-Orgame, 22. Sandby Borg, 23. Crow Creek (South Dakota, not mapped). (© Helle Vandkilde)

Genocide before the State?

individual characteristics, which underlines the importance of scrutinising the context of each individual case. They can be found outside or sometimes on the periphery of the community burial ground. Most often, the final resting place for the untimely dead is on the periphery of settlements, where they are disposed of in pits or in the ditches that delineate and fortify a larger area of settlement character. They are therefore sometimes referred to as settlement graves, a widespread phenomenon that increases in frequency over time. To this category of settlement graves it is possible to add enclosures with activities whose character seems more in alignment with sacrificial rites taking place in a sacred precinct (wooden, earthen or stone), but possibly also with war. The strange site of Cliffs End Farm, on the Isle of Thanet, in the Thames Estuary, exemplifies this.31 This is the site of a grave pit containing five individuals who belonged to the Late Bronze Age, c.1100–800 B C E. The bodies were articulated but were lying in distorted or extreme positions, as though the victims had been bound. According to the excavator, a lethal sharp-force trauma to the head of an elderly man (sword cut) signifies the main act of sacrifice against a group of disposable members of the community. Lamb and cattle bones as well as pottery were also present in the pit. The precinct character of the site suggests this ritual was a sacred action, perhaps to mark the end of a conflict. It seems significant that adult females are absent from this structure, which contains two adult men and three youths. The frequency of such sites is unknown, but they do not differ significantly from the bulk of settlement graves in layout and composition. Abnormal graves and primary fatality sites usually contain individuals of all ages, from infants to the elderly, but there may be absentees. A. Hårde notably reports a very low number of juveniles and young adults (aged fifteen to twenty) in the substantial number of Early Bronze Age abnormal graves in Austria, the Czech Republic and Slovakia.32 Abnormal graves usually contain adults of both sexes, but the adult group can exhibit sex bias whereby one sex is either under-represented or over-represented. Young women are relatively under-represented in some abnormal graves. Bioarchaeological investigations often, but not always, report peri-mortem trauma. Could epidemics like plague (Y. pestis), known from prehistory, be 31 J. I. McKinley, M. Leivers, J. Schuster, P. Marshall, A. J. Barclay and N. Stoodley, Cliffs End Farm, Isle of Thanet, Kent: A Mortuary and Ritual Site of the Bronze Age, Iron Age and Anglo-Saxon Period with Evidence for Long-Distance Maritime Mobility (Salisbury, UK: Wessex Archaeology, 2014), pit 3666, pp. 67–70. 32 A. Hårde, ‘Funerary rituals and warfare in the Early Bronze Age Nitra culture of Slovakia and Moravia’ in Warfare and Society, ed. Otto, Thrane and Vandkilde, pp. 341–82.

71

helle vandkilde

the motivating reason behind some abnormal graves that do not reveal forced trauma? This is a possibility, but the very abnormal character of the burials often seems to tally with identities as outsiders, enemy others or lowstatus people. Moreover, grave goods are largely absent in the graves, and there is often an apparent disrespect for the people whose lives were ended in this aberrant manner. From around 1400–1300 B C E, at the onset of the Late Bronze Age, cremation is the new norm, but abnormal inhumation graves continue to exist and even increase in frequency, with bodies thrown unceremoniously into a pit or ditch. This may indicate a perceived fundamental difference between ‘us and them’, which has deep roots. In the same way that modern abnormal graves (see above) vary considerably, a fairly clear divide exists in prehistory between disordered and ordered abnormal graves. In a sample of seventy graves, 79 per cent were disordered and perhaps handled by enemies, 17 per cent were ordered and likely attributable to allies or relatives, while 4 per cent contained both ordered and disordered traits. Abnormal graves are summarised and exemplified in what follows using a selection of well-documented cases. Disordered Abnormal Graves This includes graves with multiple bodies lying twisted in a commingled mass filling the space of a pit, sometimes as a bone bed, in an advanced stage of disarticulation. Yet it can also include alternative spaces such as remote caves and rock shelters. Very frequently, however, one or more distorted bodies lie in discrete assemblages within a pit or ditch, the bodies of each assemblage in contact with one another, sometimes superimposed and with different orientations and body positions. Here, the grave space is often not wholly occupied, and the bodies are found in irregular body positions, such as face down or bent in frog and sitting positions. Supine and crouched positions commonly occur and may appear normal at first sight, but a closer look reveals deviations such as crossed erratic-looking limbs and extreme crouching of the body. This may show that these people were thrown haphazardly into the grave from different angles, at times with hands and legs tied and the body bound into a bundle. However, it is often unclear whether they were already dead when this final action occurred or whether they were executed in situ. The ditches and pits typically already existed and were reused when disposing of these individuals. Animal bones and carcasses and other kinds of refuse, like potsherds, in the earthen fill suggest that the bodies were seen as ‘waste’. On some bones, it is possible to identify the markings of animal gnaws, which shows how deposits were sometimes left temporarily uncovered, yet the intention behind this is unclear. The majority of abnormal graves fit the 72

Genocide before the State?

above descriptions and the vast majority of them date to the Bronze Age and early Iron Age, 2000–600 B C E. Some are distinctly earlier. A relation to war often seems present. Having described the more general characteristics of disordered abnormal graves, it is now relevant to examine particular examples (minimal number of individuals in all examples). We may start with the Ofnet Cave (Bavaria, Germany) in the Late Mesolithic period 6500–6200 B C E among late hunter-gatherer bands. Here, a whole community of thirty-six women, men and youngsters was stored in pits represented mainly by their skulls decorated with red ochre, shell and teeth ornaments. This may suggest burial and worship of these skulls, perhaps resulting from trophy hunting.33 Warfare is indicated by a very high level of blunt force trauma. A thousand years later in the Early Neolithic period c.5200–4800 B C E, disordered abnormal graves appear that call to mind those of later periods. Migrants of Middle Eastern ancestry and their descendants at this time colonised large parts of Europe north of the Alps. A sample of five abnormal graves is associated with the so-called Linear Pottery Culture (LBK) that expanded across the vast loess area of central Europe. We do not currently know who the buried outsiders were, whether colonists or indigenous hunters. A war relation is often apparent through signs of physical trauma,34 and unmistakeable ritual components are absent in these and most other abnormal graves. At Schöneck-Kilianstädten (Hessen, Germany), twenty-six individuals were identified lying in a ditch.35 Half of these individuals were adults and half were less than seventeen years old. Women were in the minority compared with the adult men. Uncontrolled blunt force cranial injuries are present and perimortem long-bone fractures show the smashing of lower legs as if to hinder escape. This may be interpreted as a deliberate extermination of a whole community, sparing the young women. The truly unique characteristic of the Kilianstädten site is the pattern of targeted peri-mortem demolition of the distal segment of the lower limbs. The Talheim site (middle Rhine valley) similarly testifies to apparently uncontrolled blunt force cranial injuries 33 J. Orschiedt, ‘The head burials from Ofnet cave: an example of warlike conflict in the Mesolithic’ in Warfare, Violence and Slavery in Prehistory: Proceedings of a Prehistoric Society Conference at Sheffield University, ed. M. Parker Pearson and I. J. N. Thorpe (Oxford: Archaeopress–BAR, 2005), pp. 67–73. 34 C. Meyer et al., ‘Patterns of collective violence in the Early Neolithic of central Europe’ in Prehistoric Warfare and Violence: Quantitative and Qualitative Approaches, ed. A. Dolfini et al. (Cham: Springer, 2018), pp. 21–38. 35 Ibid.

73

helle vandkilde

enforced by a stone axe from behind.36 Here, thirty-four individuals were found in a ditch. The grave seems to represent a whole community with adult women slightly under-represented. At Asparn-Schletz (Lower Austria), sixty-seven individuals were identified in the ditch of what is interpreted as a fortified enclosure.37 Again, uncontrolled blunt force cranial injuries were inflicted from behind using a stone axe. One individual had been killed by an arrow. It is likely that the victims were hit by a right-handed person while fleeing. The skeletal remains displayed gnawing marks, suggesting that the bodies had been left temporarily exposed before covering. The demographic profile indicates that a whole community ended up in the ditch. Adult women are under-represented. At Wiederstedt (Sachsen-Anhalt, Germany), ten individuals, mostly children and young people, had been thrown haphazardly and without grave goods into an LBK refuse pit.38 A woman among them had received a more normal burial. Since peri-mortem trauma could not be observed, other reasons than strictly war may have been at play. An epidemic disease – known to accompany war – would accord with the demographic profile. At Halberstadt (Sachsen-Anhalt, Germany), nine individuals were located in a pit near contemporaneous LBK longhouses.39 Their Sr isotopic profiles suggest they were outsiders to the area. The blunt force trauma seems controlled and mostly inflicted with precision from behind, like an execution. There were also breakages to the ribs. Adult women and sub-adults were markedly under-represented. This looks like the systematic killing of a selected group of non-local people, sparing women and young people. Carnivore gnawing shows that the pit was temporarily left open, perhaps to show indifference or contempt. A thousand years later, in the Late Neolithic period c.4000 B C E, the site of Bergheim (Alsace, France) exhibits the same features. The site may connect to the expansions of the Michelsberg culture, which constructed a multi-tiered settlement structure around hill forts. Here, fifteen adults and children were found piled in a special silo-like pit, lying tightly commingled yet with some apparent order in their deposition. The skeletons display severe blows to the frontal skull. They had clearly been executed and tortured as arms were amputated. Other similar pits with human remains occur in the Bergheim area and adjacent regions in the fourth millennium B C E.40

36 Ibid. 37 Ibid. 38 Ibid. 39 Ibid. 40 F. Chenal, B. Perrin, H. Barrand-Emam and B. Boulestin, ‘A farewell to arms: a deposit of human limbs and bodies at Bergheim, France c. 4000 BC’, Antiquity 89:348 (2015), 1313–30.

74

Genocide before the State?

Many abnormal graves on the perimeter of settlements are known to date back to between 2000 and 1500 B C E, in the Early Bronze Age and the transition to the Middle Bronze Age. Many of these graves link with hill forts in eastern central Europe. At the Early Bronze Age site of Schleinbach (Mistelbach, Lower Austria), four individuals lay in a storage pit. The adult male and the three children were killed on the same occasion. The extreme crouched position of the children suggests binding. The three- to four-yearold child died by a blow to the head. Placement of one body over the other indicates a deliberate act.41 It is unclear if the killing took place in the pit. At the Mad’arovce-Otomani hillfort of Nižná Myšlˇa (Nitra, Košice, Slovakia), a similar pit for refuse dates to the transition to the Middle Bronze Age, c.1700–1500 B C E. The bodies of two women and three children were found lying in a disordered manner with slit throats, bound bodies and, reportedly, having been partly boiled. Forensics suggest that killing took place in the pit, which also contained animal bones. It seems possible to identify three phases of killing, which may have spanned over a month.42 The Middle Bronze Age 1600–1350 B C E seems to lack abnormal graves, yet they reappear at the threshold to the Late Bronze Age, when culture and society changed quite markedly. Most such graves relate to central European enclosed sites and hill forts. Velim-Skalka (Bohemia, Czech Republic) is such a site dating to c.1400–1300 B C E, at the time of the decline of the Tumulus culture and the rise of the Urnfield culture.43 Here, around a hundred individuals lay in pits and ditches in a settlement-like enclosure. Adults of both sexes were among the dead, but women were under-represented. Children of all ages make up the largest group. Various peri-mortem trauma include incisions, blows and cuts – a total of eighty-five impacts, mostly to the skull. The site has been discussed as a place for sacrifice, cannibalism and rituals. However, new forensic investigations reveal that all the individuals were thrown into pits and ditches and then covered, and that it was subsequent digging activity that accounted for disarticulations, post-mortem cuts and the collection of skulls. At the contemporaneous and similar site of 41 K. Rebay-Salisbury, ‘Cold case Schleinbach: why it makes sense to preserve skeletons in situ’, Motherhood in Prehistory (online blog), https://motherhoodinprehistory .wordpress.com/2016/11/29/cold-case-schleinbach-why-it-makes-sense-to-preserveskeletons-in-situ/ (accessed 17 December 2019). 42 J. Jakab, L. Olexa and J. Vladar, ‘Ein Kultobjekt in der Otomani-Kultur in Nižná Myšlˇa’, Slovenská archeológia 47:1 (1999), 91–127; Hårde, ‘Funerary rituals and warfare’, 372–81. 43 A. F. Harding, ‘Interpretation and discussion’ in Velim, ed. Harding et al., pp. 144–53; A. F. Harding, ‘Velim and violence’, Cuadernos de Prehistoria y Arqueología de la Universidad de Granada 23 (2013), 165–82.

75

helle vandkilde

Blucˇina Cerzavy (Brno-Venkov, Moravia), at least 130 burial features with 200 individuals similarly defy a firm ritual explanation. Instead, the data speaks more in favour of warfare: the ditch contains several warlike features, including weapons and arrowheads.44 The human interments are relatively ordered in the ditch, where there are several burial groups, articulated bodies and some with grave goods. This relative order and care could be due to the agency of friendly forces. Velim, Blucˇina and other similar cases prompt the two bioarchaeologists, Christopher Knüsel and Alan Outram, to compare these cases with the mass graves resulting from multiple killings in modern ethnic violence.45 Despite the geographical distance, the abnormal grave at Sund (Inderøy, Trondheim, Norway), which dates to c.1300 B C E, shares the same features. Here, twenty-two individuals lay in a commingled mass, with the bones widely disarticulated. The bones exhibit extensive healed and unhealed sharp-force trauma, likely from a sword. A pit contained six concentrations of human bone with animal bones also present. There were adults and children in equal measure and ante- and peri-mortem injuries were visible. The human bones furthermore revealed signs of malnutrition and starvation. By these parameters, the hinterland abnormal grave at Sund contrasts with the prosperous mound burials at coastal Toldnes, where the skeletons did not reveal any traces of violence.46 This could well have been where the weaponwielding perpetrators were buried. There are many reported cases of abnormal graves that date to the Late Bronze Age and the transition to the Hallstatt Iron Age, 1300–700 B C E. Like the previous cases discussed, their context is often fortified sites. The research literature tends to favour ritual and funerary explanations for these sites, because traces of peri-mortem violence seem only variably present. Nevertheless, the generally disordered design, poor or absent grave goods, and the relatively normal demographic profile remains the same as in the previous cases. For eastern Europe, S. C. Ailinc˘ai notes that normal cemetery burials are individualised and rich and that signs of violence are rare.47 However, in abnormal graves, 44 Harding, ‘Interpretation and discussion’, 144–53; Knüsel and Outram, ‘Comparison of depositions’, 131–6. 45 Knüsel and Outram, ‘Comparison of depositions’, 133. 46 H. Fyllingen, ‘Society and the structure of violence: a story told by Middle Bronze Age human remains from central Norway’ in Warfare and Society, ed. Otto, Thrane and Vandkilde, pp. 319–29. 47 S. C. Ailincai, ˘ Living with the Dead: Burials in Early Iron Age Settlements between the Balkans, Tisza and Dnestr (Cluj-Napoca: Academia Româna˘ Centrul de Studii Transilvane, 2015), pp. 191–2.

76

Genocide before the State?

the interments are poorly equipped and violence is more frequent. Normal and abnormal graves have the demographic profile of an average population, and the ratio of abnormal to normal graves is 1:2.5. This means that a considerable portion of the population ended up in abnormal graves. At the hill fort of Stillfried (Weinviertel, Lower Austria), thirty individuals were found in two pits dating to c.900 B C E, a point when the Urnfield culture was rapidly declining. The site was burnt down in c.800 B C E. Although lethal trauma was not detected on the bones, the pits were located in the fortification rampart: there were three adult males, three adult females, and the rest were adolescents and children. There was evidence of much illness and malnutrition. The individuals lay articulated with unusual body positions.48 At the contemporaneous site of Jurilovca–Orgame (Tulcea, Romania), fifteen individuals, among them four adult females, seven adult males and three adolescents and children, were thrown into a deep pit. A sixteen-year-old girl had peri-mortem trauma to the skull. There were few young women. Some pottery and personal items were present.49 Ordered Abnormal Graves This is characterised by bodies placed in a certain order showing consideration for the deceased. Still, the burial design and circumstances deviate from normal intra-cultural procedures. At Koszyce (Proszowice, Lesser Poland), c.2800 B C E, an abnormal grave contained the remains of fifteen people belonging to Globular-Amphora herder-farmers of eastern central Europe (3300–2700 B C E). Men, women and children were lying commingled, and their bodies were irregularly positioned and orientated in a pit. Still, some order was apparent. Genomic analysis revealed that the deceased were ordered according to kinship and that they belonged to an extended family; they were obviously buried by someone knowledgeable about these family relations. There were three children, seven adult males and five adult females of whom three were elderly or mature. Adolescents and young women were under-represented and might have been abducted. All the dead had been killed by blows to the head, mostly from behind and in a systematic manner recalling execution. The DNA was further able to reveal an ancestry distinctly different from the neighbouring Corded Ware people, who are Steppe-related. The abnormal grave at Koszyce may therefore underpin a linkage between ancestry, kinship and violence within the context of Corded Ware territorial expansion.50 Among Corded Ware populations, skeletons often 48 I. Hellerschmid, ‘Mord oder Opferung? Die Niederlegung der “Sieben” in Grube V1141 am Kirchhügel von Stillfried’, Archaeologia Austriaca 99 (2015), 203–31. 49 Ailincai, ˘ Living with the Dead, pp. 82–8. 50 H. Schroeder et al., ‘Unraveling ancestry, kinship, and violence in a Late Neolithic mass grave’, Proceedings of the National Academy of Sciences 116:22 (2019), 10705–10.

77

helle vandkilde

show ante- and peri-mortem trauma, suggesting a high level of warfare.51 A case in point is the Corded Ware cemetery at Eulau (Naumburg, Germany) for the period 2700–2500 B C E.52 Most graves were of the single-grave normative type, but four were multiple graves with two to four interments of adults and children. Alongside trauma such as arrows in situ and cranial blows, the proximity of the graves can suggest a single event of warfare. Multiple burials were adapted to the sex-differentiated norm, notably with the adult male in a crouched position still holding the stone battleaxe ready for use, which can be expected from a warrior. At Eulau, like Corded Ware sites more generally, male DNA accords with Steppe ancestry whereas women are of mixed ancestry in a manner abiding with a patrilineal kinship system coupled to female exogamy.53 Much later, at the transition to the Middle Bronze Age c.1700 B C E, an ordered abnormal grave was made at Wassenaar (Haag, Netherlands) (Figure 1.1). Placed near a settlement, the pit contained twelve adults and children who had died by blows to the head and humerus. An arrow was still embedded in the rib of an adult male and another adult male had an arrow in the chest. Women were distinctly under-represented. Owing to the site’s proximity to the North Sea coast, it is possible that this was the result of a maritime raid.54 The death of the twelve individuals occurred at the same time. It is likely that the individuals were buried by a friendly party who decided to arrange the burials in two parallel rows. The adults were laid on their back in an orderly fashion and the children were put in a crouched position.55 Fatality Site Variant A variant involves sites where human bodies merely lie where they fell. This variant is rare, perhaps because, if only for practical reasons of resettling a site, dead bodies were usually removed for burial. Such primary fatality sites cannot really be called graves, since measures have not been taken to perform an interment of the dead. Some of the abnormally occurring interments in the ditches of fortified sites could also be described as 51 H. Peter-Röcher, Gewalt und Krieg im prähistorischen Europa: Beiträge zur Konfliktforschung auf der Grundlage archaologischer, anthropologischer und ethnologischer Quellen (Bonn: Verlag Dr Rudolf Habelt, 2007); H. Meller and M. Schefzik, eds., Krieg – eine archaologische Spurensuche (Darmstadt: Theiss, 2015). 52 C. Meyer et al., ‘The Eulau eulogy: bioarchaeological interpretation of lethal violence in Corded Ware multiple burials from Saxony-Anhalt, Germany’, Journal of Anthropological Archaeology 28:4 (2009), 412–13. 53 W. Haak et al., ‘Ancient DNA, strontium isotopes, and osteological analyses shed light on social and kinship organization of the Later Stone Age’, Proceedings of the National Academy of Sciences 105:47 (2008), 18226–31. 54 J. Ling, T. Earle and K. Kristiansen, ‘Maritime mode of production: raiding and trading in seafaring chiefdoms’, Current Anthropology 59:5 (2018), 488–524. 55 L. P. Louwe Kooijmans, ‘An Early/Middle Bronze Age multiple burial at Wassenaar, the Netherlands’, Analecta Praehistorica Leidensia 26 (1993), 1–20.

78

Genocide before the State?

Figure 1.1 The ordered abnormal grave from Wassenaar (Haag, Netherlands), c.1700 B C E. (© Leendert Louwe Kooijmans 1993. Reproduced by permission of Leendert Louwe Kooijmans)

non-burials, since the bodies lie in the area where they were slain. Nataruk, in Kenya, is not only one of the earliest known sites of warfare but also represents the earliest primary fatality site without any sign of burial.56 The twenty-seven bodies, primarily adults of both sexes, probably lay where they fell c.7500–8500 B C E in the epi-Palaeolithic period. The skeletons lie in erratic positions, so this is certainly not an ordinary burial ground. Cranial blunt force trauma occurs together with other lethal lesions, including imbedded projectiles. The connection to war is thus clear. The individuals had been left at the shore of an old lagoon near Lake Turkana under circumstances that 56 Lahr et al., ‘Inter-group violence’, 394–8.

79

helle vandkilde

may be described as social stress in the early Holocene age, when the first farming communities were about to arise in the Fertile Crescent. Despite being separated by thousands of years, Nataruk in Kenya is remarkably similar to Sandby Borg on the island of Öland, in the Baltic Sea, off the coast of eastern Sweden. Like Nataruk, Sandby Borg is a recently discovered site that dates to the Scandinavian Late Iron Age 400–500 C E. The twenty-six bodies were found lying in twisted positions in the houses and streets of a fortress; they had tried to escape the perpetrators. Sharp force, blunt force and punctuating trauma can be observed on all age groups, from infants to the elderly. Notably, women were absent or at least in the clear minority (excavation is incomplete), assumedly taken captive while the men and children were killed. There was no sign of burial and the fort was never resettled. The Scandinavian Migration period was turbulent, with many small-scale chiefdoms struggling for power during the instability that arose when the western Roman Empire fell.57 In partial conclusion, war-related abnormal graves and fatality sites occur throughout Holocene prehistory and in non-state settings more generally. It is probable that there are yet undiscovered fluctuations over time. The violence, hatred and contempt that continue to emanate from the abnormal graves fit the concept of genocidal massacres. It is most likely significant that the earliest certain cases pertain to the epi-Palaeolithic age, 12000–8000 B C E. When examining details of the contexts, it is striking that cultural transitions marked by stress and socio-economic change figure prominently alongside hill fort-dominated landscapes with quasi-political warfare targeting opponent ‘others’. The Chalcolithic Koszyce grave is the only case that indicates ethnic-motivated violence due to the genetic data available. According to present knowledge, the number of these abnormal sites increases markedly in the Bronze Age 2000–700 B C E, which boasts 77 per cent of the recorded sites. Many abnormal graves are associated with fortified settlements, which is a further connection to warfare and power, since these were sites of militarised societies in possession of lethal weapons and institutions of war. Bronze Age and early Iron Age hill forts are especially widespread across a large area from the British Isles in the west to central and eastern Europe in the east. They reflect the rise and fall of power hierarchies flaunting and exercising military muscle and any other political instrument at hand. The political effect of the killings and the abnormal graves should not be underestimated: as shown, warfare is a flow of action in need of a finalising act. 57 Alfsdotter and Kjellström, ‘Sandby Borg massacre’, 210–31.

80

Genocide before the State?

The taking of captives is a parallel issue. Starting in the Neolithic period, there are many cases where young women and adolescents form the minority of those found in the graves. This is presumably because more became captives of war. The increase in abnormal graves in the Late Bronze Age and early Iron Age suggests that unfree or low-class people constituted a large and possibly increasing part of the population – a bottom rung. Genetics may provide an insight into the roles they undertook in the captor society.

Genetics and Ancestry Recent genetics-led research has revealed early Europe’s population history to be a series of immigrations into landscapes already inhabited. The Neolithic farmers of Near Eastern ancestry were incomers to Europe and from c.7000 B C E they began to replace the indigenous hunter-gatherers. Complex scenarios of cultural exchange and some genetic admixture and cohabitation between farmers and hunters-gatherers have been revealed,58 but, except in peripheral retreats, the hunters-gatherers’ possibilities for biological and cultural reproduction were undoubtedly reduced.59 The influx of Steppe migrants then began c.2900 B C E. Over the next millennia, the socalled Indo-Europeanisation of huge tracts of Europe was completed. North of the Alps, Neolithic farmers were largely replaced by Chalcolithic Corded Ware and Bell Beaker groups. Evidence of Steppe ancestry is clearly present in this data, just as it is in data from the Bronze Age and later. This is testimony to the expansiveness of these Steppe-derived people. Like previously, the data reveals biocultural admixture with indigenous groups. What is truly remarkable is a distinct sex bias, with different patterns for males and females. Apparently, Britain and Iberia saw a replacement of 90–100 per cent of the Y-chromosomes by people of Steppe ancestry between 2500–2000 60 B C E. Biomolecular genetics repeatedly shows that Steppe descendants spread their Y-chromosomes effectively in Europe, while the X-chromosomes vary more widely, involving genetic admixture with indigenous females above all. Scientists talk about a Y-chromosome bottleneck 58 M. Lipson et al., ‘Parallel palaeogenomic transects reveal complex genetic history of early European farmers’, Nature 551:368 (2017), 368–72; T. Terberger et al., ‘Step by step – the Neolithisation of northern central Europe in the light of stable isotope analyses’, Journal of Archaeological Science 99 (2018), 66–86. 59 Reich, Who We Are, pp. 104–7. 60 I. Olalde et al., ‘The Beaker phenomenon and the genomic transformation of northwest Europe’, Nature 555:7695 (2018), 190–6; I. Olalde et al., ‘The genomic history of the Iberian Peninsula over the past 8000 years’, Science 363:6432 (2019), 1230–4.

81

helle vandkilde

around this time. This is a radical reduction in Y-chromosome diversity, which concretely means a decline in the male effective population size. The phenomenon of a limited number of remaining powerful males had a high impact on the next generations, and here may lie the roots of social inequality.61 Since the pattern is distinctly different for males and females, epidemics were not the sole factor but may well have caused demographic busts at an early stage. Organised lethal warfare may then have killed the surviving adult males while retaining young healthy individuals and especially the women as potential wives. Knüsel and Outram correctly identify that the massive loss of adults in a community would have affected the population structure more profoundly than an epidemic disease.62 Present recordings of trauma are inconclusive and it seems clear that these migrants also fought among themselves, which accords with tribal practices. Cemeteries in Germany’s Lech river valley, 2500–1500 B C E, suggest that the privilege of fathering children and carrying weapons resided with only a few senior men who ran the male lineage of Steppe type. Women were of more mixed ancestry. A poorly equipped male group of burials, also of genetically mixed ancestry, had no relatives; they were likely unfree members of the household. Remarkably, the male-biased operational sex ratio in evidence at Lech could have generated a pool of unmarried men who moved away to found new residences or join war bands (near or far) as a career strategy. We are here dealing with intra-household inequality, which has drastic biocultural and social consequences.63

Genocide before the State? The intentionality, time patterns and nexuses of warfare, abnormal graves and genetic data all guide this concluding discussion of the possibility of early genocide. This theme has previously been avoided in the existing literature. It should be remembered that current knowledge is fragmentary, especially regarding the scale and regularity of the practices elucidated by the above three lines of reasoning. The definitional requirements of genocide coupled 61 Reich, Who We Are, pp. 237–40; T. C. Zeng, A. J. Aw and M. W. Feldman, ‘Cultural hitchhiking and competition between patrilineal kin groups explain the post-Neolithic Y-chromosome bottleneck’, Nature Communications 9:2077 (2018), 1–12. 62 Knüsel and Outram, ‘Comparison of depositions’, 133. 63 A. Mittnik et al., ‘Kinship-based social inequality in Bronze Age Europe’, Science 366:6466 (2019), 731–4.

82

Genocide before the State?

with the limitations of the archaeological sources should certainly prompt caution. One might claim that, according to present knowledge, the preconditions for prehistoric genocide were clearly present, but that evidence for genocide itself is often circumstantial in character. Genocidal massacres and moments, however, stand out reasonably clearly in the data. A master plan for genocide is highly unlikely, though it is possible that genocidal moments at times formed a linked or repetitive pattern. Especially in this latter area, progress will rely on future high-resolution combinations and correlations of evidence for warfare, abnormal graves and ancestry. Warfare is always intentional, and the moment it is enacted as violent action against other groups, there is the potential for genocidal moments of unconstrained violence. This does appear to happen around the beginning of the Neolithic. There are further indications of raised warfare levels in the Chalcolithic linked to the Steppe influx and subsequent territorial expansions, which accord with the genetic evidence. There may here be a linkage to the innovations of weapons made for war and a tighter military organisation. Bronze Age data holds clearer components of effective warfare expedited by lethal weapons and probably military units, but certainly also peaceful interchange and thriving trade. Warfare is a fairly constant companion to the abnormal graves, which in terms of physical characteristics call to mind recent examples of genocidal massacres and mass graves. The victims in these graves clearly shared some form of cultural or social identity. In these cases, the perpetrators’ intention was undoubtedly also to partly or entirely destroy that slain outsider group. Hatred and contempt were undeniably present in this intention, which can be seen from the relative frequency and layout of disordered interments. The prevalent use of refuse pits as graves discloses the dehumanisation of these ‘others’. The intentionality is undeniable, yet it is less obvious who the victims and the perpetrators were, and this must have changed over time. From the advent of the Neolithic period, the pattern of abnormal graves assumes a recurring rhythm pointing forwards in time, but the number of abnormal graves was most likely low. In the Final Neolithic and Chalcolithic periods, there was an increase in the number of abnormal graves, yet this increase is not statistically significant. It was in the Bronze Age that the number of abnormal graves increased markedly, and this is clearly significant in the light of the increased militarisation and inequality in Bronze Age society. Several of the abnormal graves presented above are associated with periods of social, cultural and economic pressure and change, which suggests 83

helle vandkilde

‘genocidal moments’ more than planned perpetrator campaigns or long-term unrestrained hatred and violence towards certain others. The best cross-period suggestion is the persecution of ethnic others and socially marginalised people, which are certainly not mutually exclusive. The under-representation of young women and adolescents in abnormal graves from the Neolithic period onwards clearly supports the idea that these two groups were taken captive. An ethnic component is visible only in the Koszyce mass grave, but perhaps we can hypothesise that current data underestimates this feature. Genetic data provides complex new insights especially for the period 2900– 1500 B C E from the time of Steppe influx and into the early Middle Bronze Age. The sex-biased demography leads to questions about underlying actions and intentions. The missing men of non-Steppe ancestry in the genealogies indicate that these males were killed or otherwise removed from the effective population ratio. A limited number of Steppe-derived men ran the successful male lineages via genetic admixture with indigenous females in particular. The women could have been exchanged into the group from the outside, but some were undoubtedly abducted, likely together with adolescents, to be used as labourers. The data shows that these individuals adapted to their new group but also that new and geographically widespread forms of hybrid culture emerged, which we call Corded Ware culture and Bell Beaker culture. Was this an overriding purposeful annihilation of the others who originally settled the land, first the hunter-gatherer groups replaced by the First Farmer populations who were then replaced by people carrying Steppe ancestry? It is perhaps more likely that it was all embedded in a biocultural logic, which included territorial expansion and which over time resulted in population change. The early data can suggest a biocultural pressure that restricted these outsider groups’ possibility to continue reproducing themselves economically, biologically and culturally. The individuals required to produce the future were gone: dead from diseases, stolen, or deliberately killed. For the later Bronze Age and early Iron Age, genetics would be immensely helpful in clarifying who the multiple dead in the abnormal graves were, apart from the fairly obvious point that they were social outcasts or low-class. High-tech scientific techniques will continue to break new ground due to their ability to chart victims’ histories, ancestry and relatedness. In close connection with archaeological data, they have the potential to deliver revolutionary new insights into the subject area of genocidal massacres in the past. This will be relevant for us today as a background for understanding recent genocidal events. 84

Genocide before the State?

Bibliographic Note This chapter, first submitted in January 2020, draws on a substantial number of archaeological excavation reports, articles and books. Only key references are provided above. The study relies on a combination of archaeological evidence and bioarchaeological screenings of skeletal remains in terms of demographic profile, trauma and – when available – palaeogenetic and isotope signatures. Ethnographic insights have further enriched the archaeological interpretations. My long research engagement with prehistoric warfare and warriorhood has taken a new path thanks to colleagues at the Max Planck Institute for the Science of Human History in Jena and at my home Department of Archaeology and Heritage at Aarhus University, in Denmark.

85

2

The Religion–Genocide Nexus steven leonard jacobs

Introduction Is there something inherent in the very social construction we call ‘religion’ that lends itself, adapts itself all too easily to those communities, both nationstates and non-nation-state actors, that perpetrate genocide, either in actuality or in potential? This is a question that the literature on genocide has under-theorised, under-analysed, under-examined, under-explored and under-reported. For example, neither of the scholars Doris Bergen and Kate Temoney, nor the earlier two contributors to my own anthology Confronting Genocide1 – namely the late Leo Kuper in his ‘Theological warrants for genocide: Judaism, Islam, and Christianity’ and Leonard Glick in ‘Religion and genocide’ – address this fundamental and initial question while citing numerous examples throughout history. An earlier edited collection seemingly addressing the nexus between religion and genocide, In God’s Name, does not come to grips with this question either.2 Bergen, however, begins her important contribution to Dan Stone’s The Historiography of Genocide with the rather trenchant observation that ‘the first problem in discussing the historiography of religion and genocide is to define the parameters, a difficult task, because all the categories involved are open to interpretation’.3 Her observation is fully echoed by Kate Temoney in her chapter, ‘“Those who have the sin . . . go to this side”: genocide and religion’, 1 Steven Leonard Jacobs, ed., Confronting Genocide: Judaism, Christianity, Islam (Lanham, MD: Lexington Books, 2009). 2 Omer Bartov and Phyllis Mack, eds., In God’s Name: Religion and Genocide in the Twentieth Century (New York: Berghahn Books, 2001). However, the contributors to this volume were not scholars of religion, religious studies or Judaic Studies – my own subdiscipline – but, as has often been the case, historians, political scientists, legalists, sociologists and psychologists. My edited volume welcomed into the conversation scholars specifically trained in the academic study of religion. 3 Doris Bergen, ‘Religion and genocide: a historiographical survey’ in The Historiography of Genocide, ed. Dan Stone (New York: Palgrave Macmillan, 2008), pp. 194–227.

86

The Religion–Genocide Nexus

to wit, ‘there is no singular discussion of religion and genocide that is capable of being exhaustive’.4 Each scholar goes on to address four foci. For Bergen, these are: 1. ‘religion as a motivating and/or legitimating force for perpetrators of extreme violence; 2. ‘religion as a target of genocidal attack and identifier but also comforter of victims’; 3. ‘religion as a brake against mass brutality’; and 4. ‘religion as a healing or concealing factor in post-genocide situations’.5 And for Temoney, they are: 1. ‘theological warrants and frameworks, which make genocide not only thinkable but permissible’; 2. ‘religious alterity, which reinforces the distinctiveness of the group targeted for extermination and imbues the victims’ “otherness” with sacred import’; 3. ‘the technique of usurping religion’s potency through rhetoric, which simultaneously appropriates religious authority and crushes potential religious opposition’; and 4. ‘the role of religious figures and institutions in genocide perpetration, which symbolises the divine endorsement and diffuses the responsibility of perpetrators’.6 Thus, I begin with something of a theoretical look–see vis-à-vis that nexus between religion and genocide by suggesting applicable definitions for both and further outlining the constituent factors of each. (Note: there are, in truth, uncomfortable similarities between religious groups and genocidal perpetrator groups which, to my understanding, have never been addressed or explored.) To further bolster my overall argument – that religion, however defined and understood, is a ‘participating factor’ (my preferred term) in all genocides, both historical and contemporary – an enumeration of cases, using Raphael Lemkin’s tripartite division from his incomplete History of Genocide – Antiquity, Middle Ages, Modern Times – is then shown to determine whether my thesis holds.7 Finally, Bergen ends her own survey with the comment, ‘I am not optimistic that there will be a large amount of scholarship on issues of religion and genocide, broadly conceived. The stakes are too high, and the rewards are too few.’8 Temoney, however, counters this rather bleak 4 Kate Temoney, ‘“Those who have the sin . . . go to this side”: genocide and religion’ in The Routledge History of Genocide, ed. Cathie Carmichael and Richard C. Maguire (London: Routledge, 2015), p. 227. 5 Bergen, ‘Religion and genocide’, 196. 6 Temoney, ‘“Those who have the sin”’, 232. 7 See Raphael Lemkin, Lemkin on Genocide, ed. Steven Leonard Jacobs (Lanham, MD: Lexington Books, 2012). 8 Bergen, ‘Religion and genocide’, 218. Bergen’s use of the word ‘reward’ is troubling. Increasing knowledge of what factors contribute to genocidal perpetration is important. My ongoing argument is that, given the academic disciplines that have thus far looked for those factors, the omission of considering religion as one such source and exploring its own relationship to genocide would certainly constitute one further ‘reward’.

87

steven leonard jacobs

assessment by advocating that ‘it [religion] should be a major part of genocide studies discourse, both as a practiced phenomenon and as a method of inquiry’, and further addressing ‘how religion potentiates genocide’ as well as ‘how it can hinder its commission and potentially forecast its perpetration’, accompanying more attention to ‘the perspective of genocide perpetrators’ as to their motivations – ‘religious and otherwise’ – and rejecting the use of the Holocaust as the paradigmatic evaluative standard in examining genocides.9 My own position on these issues is clarified in the conclusion. * Scholars of the world’s religions; academic critics of how religion is studied both theoretically and methodologically in higher education; and/or Judaic Studies scholars in their own various sub-disciplines (i.e. history, philosophy, political science, psychology, sociology, religious studies) come late to the emerging field and ongoing conversations vis-à-vis Genocide Studies, an outgrowth of Holocaust Studies. Israel Charny, one of the doyens of the field, has suggested a possible rationale for this lateness, given his own difficulties in finding someone to write on the topic of ‘religion and genocide’ for his 1994 text The Widening Circle of Genocide,10 until Leonard Glick undertook that assignment: There is quite a story to be told about how many years it took before we succeeded in getting this important open treatment of the subject of religion as both setting the expressed moral direction of Thou Shalt Not Kill [sic]11 while in itself being responsible for so many genocidal killings over the centuries. Before Leonard Glick’s fine contribution, there had been several well-known scholars in the field of religion who had agreed to do the project and then dropped it at a very late stage, almost without explanation. To me, it seemed that what happened was that they were unable to tell the truth about the religious establishments with which they were variously connected.12 9 Temoney, ‘“Those who have the sin”’, 238. 10 Israel Charny, ed., The Widening Circle of Genocide (New Brunswick: Transaction, 1994). 11 A more accurate translation of the Hebrew Lo tirtzah would be ‘You shall not murder’. Thus, ‘murder’ would be defined as unsanctioned life-taking by either individuals or groups, and/or state agencies or institutions, whereas ‘killing’ would be statesanctioned life-taking in three appropriate contexts: legally legitimate military conflicts, police–perpetrator confrontations, and prison executions. Genocide would thus constitute murder; killings outside of those just enumerated by the state would also constitute murder. 12 Quoted in the introduction to Jacobs, ed., Confronting Genocide, p. ix (emphasis added). See also Israel Charny, ‘A passion for life and a rage at the wasting of life’ in Pioneers of Genocide Studies, ed. Samuel Totten and Steven Leonard Jacobs (New Brunswick: Transaction, 2002), pp. 429–78.

88

The Religion–Genocide Nexus

Even earlier, Henry Huttenbach of the City College, New York, challenged the Annual Scholars Conference on the Holocaust and the Christian Churches (founded in 1970 by Franklin Littell and Hubert Locke) to address issues of genocide and religion as well, writing in that critique: [R]eligion – meaning the faithful, the doctrine, the clergy, and their institutions – can easily be prompted to buttress genocidal thought and action in a wide array of capacities. The religion-genocide nexus . . . in particular, must be carefully monitored in times of social crisis . . . It is the task of scholars to expose and explore it, and for policy makers to dismantle the religion-genocide connection.13

Difficult as this challenge is and remains, it is high time to address it fully and frontally, to explore whatever relationships exist between these two seemingly disparate phenomena, how historically and contemporarily they have interacted, and what – if anything – can be done to bolster positive preventative connections and decrease their negative interactions.

Parallel and Intersecting Lines In the Journal of Hate Studies in 2010, I advanced definitions of both ‘religion’ and ‘genocide’, with sets of factors delineating each, and, thus, by comparison, a set of commonalities which appears, to me at least, self-evident in both.14 After admitting what is patently and painfully obvious in the academic study of religion – that there is no one universally accepted definition of what scholars and/or practitioners mean when we use the word ‘religion’ – to advance my argument, I defined it and here elaborate that definition somewhat, as ‘a system of communal beliefs and practices encouraging moral/ ethical and ritual/ceremonial behaviors for the [seeming] betterment of both the individual and community, and [oft-times] addressing questions of profound meaning, including [but not always] those of deity and deities’. Thus, a religion possesses the following five features: 1. A system, that is, an organised progression of thoughts and behaviours that continue to ‘make sense’ on intellectual, emotional, physical and spiritual 13 Quoted in the introduction to Jacobs, ed., Confronting Genocide, p. ix (emphases added). 14 Steven Leonard Jacobs, ‘Genocidal religion’, Journal of Hate Studies 9:1 (2010), 221–35. A concise summary of many of the arguments presented in this section can be found in the entry ‘Religion and genocide’ in Dictionary of Genocide, ed. Samuel Totten and Paul R. Bartrop, with contributions by Steven Leonard Jacobs (Westport, CT: Greenwood Press, 2007), I I, pp. 361–2.

89

steven leonard jacobs

2.

3.

4.

5.

levels to their adherents, no matter how much or little the system is understood by outsiders. A communal multi-generational enterprise for its very survival, and, therefore, characterised quite easily as ‘historical’ (i.e. that which has survived the lives of its founder or founders) or ‘new’ (i.e. that whose founder or founders are still alive). A set of behaviours – moral/ethical and ritual/ceremonial – and consists of a series of guidelines in three areas: (a) how members are to behave towards each other (almost always positive); (b) how members are to behave towards those who are not (or not yet) members of the community (usually a mix of positive and negative); and (c) how members are to behave towards their god or gods, and, if appropriate, towards other heavenly entities (almost always positive). An attempt to make meaningful sense of the human journey with all of its twists and turns, peaks and valleys, joys and sorrows, and concerned with perceived significant events in the past, the present, and the future. A system of communications usually achieved through authoritative spokespersons (more often male than female), sometimes secretive, sometimes not, and almost always understood by adherents to have a much closer relationship to the divine than most of the membership.

Secondly, while not fundamentally disagreeing with the 1948 Genocide Convention as the only internationally legally mandated definition of what constitutes genocide as a prosecutable and therefore punishable offence against the human community, building upon that understanding, I would also suggest that ‘genocide’ is ‘a systematic physical and/or cultural destruction of a victim group or groups, in war or peace, defined as such by a perpetrator group or groups and sanctioned by the state’. And therefore: 1. Genocides are systematic behaviours, rational and organised, the result of careful planning to achieve specific ends and that fully make sense to their perpetrators, though often far less so to their victims. 2. Genocides are not only the physical destruction – murder – of all or some of the members of victim groups; they are the destruction of the cultural output of the group as well – that is, everything the group has created by and through which it identifies itself to itself and by which others also identify its members. 3. Central to genocidal behaviours, then, is the necessity of defining victim groups, that is, ‘Who is the enemy who is to be destroyed?’, as well as 90

The Religion–Genocide Nexus

defining those who constitute members of the ‘in-group’, in this case perpetrators. 4. Genocides are usually, but not always, state-sanctioned exercises, that is, to achieve their desired ends, they go through a process of conversation, sometimes secretive, sometimes not, by all those responsible for such decision-making and decision-implementing before they can be carried out. Thus, bringing these two seemingly disparate entities together, I would, therefore, suggest that they exhibit four ‘commonalities’: 1. Tribalism or tribalistic superiority: the supposed ‘superiority’ of the ‘ingroup’ as it perceives itself, coupled with the very real and evident ‘inferiority’ of the so-called ‘out-group’ as it is perceived by both. 2. Exclusivism: the valuing of the ‘in-group’ and the devaluing of the ‘outgroup’. (This phenomenon is also called ‘othering’ in such contexts as gender bias, racial bias, etc.) 3. Privileged access to power: in the case of religionists, to god or gods or authoritative spokespersons; in the case of genocidaires, to political, economic, military or social power and those in authority, and sometimes even to the divine and/or religious leaderships. 4. Sacred texts: in the case of religionists, to so-called holy books which further grant them access to their god or gods (e.g. the Torah, New Testament and Quran); in the case of the genocidaires, to so-called leader documents (e.g. Hitler’s Mein Kampf or the Rwandan ‘Bahutu Ten Commandments’, or, less well known, the ‘Ten Commandments’ of the Young Turks’ Committee of Union and Progress). In addition to all of the above, there are certain ‘core elements’ usually associated with religions but equally applicable to perpetrator genocide groups as well: 1. Community: both religionists and genocidaires strive to affirm and/or reaffirm to those who are already their members and equally strive to convince others with whom they choose to identify and who choose to identify with them, that, together, they constitute a community of likeminded persons, and others – outside their circle for whatever are perceived to be valid reasons (racial, religious, ethnic, linguistic, intellectual, moral, social, etc.) – are objects of derision, and, taken to extreme, of necessary violence. 2. Doctrines and texts: central to both religionists and genocidaires are texts, both those written by their present leaders to inspire their followers and/ 91

steven leonard jacobs

3.

4.

5.

6.

or those sources of inspiration written prior to their group’s founding, to provide a certain historical legitimacy if needed. As both groups develop, they expand their literatures to include other texts which often tend to solidify which behaviours are permitted and encouraged and which are not. Ethics: within their own groups, both religionists and genocidaires develop behavioural standards that tend to segregate their communities, bolster their visions and understandings, and internalise their own senses of superiority, while, equally, all too often, denigrating the behaviours of those outside their groups as immoral, unethical, corruptible and, worst of all, threatening to the very society and/or nation-state wherein they reside, or, taken to conclusion, assaulting the very civilisation of which they see themselves as the only hope for salvation, either religious or secular. Myths: in the study of religion, myths are of two primary types: (a) stories told about god, gods and/or ancillary divine beings and/or human heroes, and (b) stories told about the origins or the group. Both are regarded as ‘true’ for the group and serve to bolster its own understanding of itself, regardless of how such stories are perceived by others, and oft-times disregarded and discarded as irrelevant, or, possibly, false by them. Genocidaires likewise tell stories about their own founding and superiority – sometimes divinely inspired according to their leaderships – recipients of a rich history even if presently and momentarily at a low point, and the overriding need to ‘right the wrongs’ presently inflicted on them by often supposedly lesser and inferior groups and thus return them to their supposed former greatness. Relevance: both religious and perpetrator communities share a common desire to ‘make sense’ of the world to their followers, to explain to their adherents causes for celebratory success and, even more importantly, how their failures were caused by others who operated insidiously and dishonestly. The staying power of those in leadership in both groups is thus directly contingent on their continuing ability to explain past, present and future to those who are their members. Rituals: outside of the moral/ethical realm, the other core set of behaviours in which religious communities engage is rituals, that is, repetitive behaviours which draw their adherents, or so they believe, closer to their god or gods, their religion, and to each other. Perpetrator communities, whether they are formal military groups or militia groups and/or the like, also regularly engage in such ongoing rituals (e.g. the ‘Horst Wessel’ lied 92

The Religion–Genocide Nexus

(song) of the Nazis, the hierarchical system of officers and enlisted, leaders and followers, and their ‘correct’ behaviours in each other’s presence, etc.). All such repetitive behaviours by both groups serve to further strengthen affiliation and commitment. In some instances, such practices tend to be secretive, away from the ‘prying eyes’ of outsiders and media or expressed in coded languages known only to the in-group after initiating rituals which bring welcomed outsiders into the group. 7. Sacrality: the very notion of the sacred is part and parcel of what German theologian, philosopher and comparative religionist Rudolf Otto (1869– 1937) characterised as das Heilige (‘the Holy’), that sense of awe and awesomeness, a profound emotional experience which he termed the ‘numinous’. It is this very sense of the emotionality of this awesomeness which is foundational to both religious groups and perpetrator groups – one is part of something far greater than one’s self – something which inspires ongoing commitment, wonder, mystery, excitement and the willingness to engage in behaviours which, under other or ordinary circumstances, would be neither conceivable nor doable. So-called religious wars and acts of genocide fall into this category. 8. Symbolisation: scholars of religion have long recognised the power of the symbol used to mobilise communities of faith (e.g. the Cross, the Crescent, etc.). Scholars of political science have long recognised the power of nation-state symbols to mobilise their citizenry (e.g. the flag, the American eagle, etc.). Genocidaires, too, have adapted and appropriated symbols, ‘positively’ in the formation of their own groups (e.g. Nazi Runes or the Totenkopf/‘Death’s Head’ of the SS, etc.), and negatively in their depictions of their enemies (e.g. hook-nosed swarthy Jews, overly muscled Blacks, etc.), through whatever media are at hand – television, radio, newspapers, the Internet, Facebook, Twitter, Instagram or other social media. Additionally, the wearing of such symbols (e.g. various medallions and/or military patches and insignias, including flags, special garments, etc.) ultimately serves to strengthen commitments and bind adherents to their communities, for good or ill. To summarise, historically and contemporarily, both communities – religious and perpetrator – in their own evolving trajectories of development as humanly crafted constructions draw upon the same flawed human systems out of our basic, biological need for socialisation. In their development over time, religious communities have engaged in the very same violent and justified behaviours as genocidal perpetrator communities.

93

steven leonard jacobs

Having thus laid out these paralleling intersections, even recognising this inherent uncomfortability in many quarters of human endeavour, it further opens the doors to newer and different directions in genocide research and further points to one equally uncomfortable conclusion heretofore little addressed, or expressed: that ‘religion’ may very well be a participating factor in all genocides, both historical and contemporary, one further legitimated source upon which perpetrators of genocide have drawn in the past, are drawing in the present, and may very well do so in the future. For example, in ‘The accountability of religion in genocide’, his contribution to Confronting Genocide: Judaism, Christianity, Islam, James Frazer Moore raises the uncomfortable and perhaps, even controversial question of what hymns were sung and what biblical passages from both the Old and New Testaments were read in Rwandan churches prior to their parishioner genocidaires going forth to perpetrate their murders. And, equally, he asks, what was the significance of the fact that the murders themselves occurred calendrically close to Easter? What impact did these associations have in the mindsets of the murderers themselves? And in those of their victims?15 Or, again, the fact that Serbian genocidal murderers were blessed by Serbian Orthodox priests and bishops prior to and during the slaughters they conducted; what impact did that supposed legitimation have upon them mentally? Or, even earlier, we may note that Johan Heinrich Ludwig Müller (1883–1945) was designated by the Nazi leadership as Reichsbischof (Reich Bishop) of the German Evangelical Church (the Deutsche Evangelische Kirche), and remained a committed Nazi until his suicide in 1945. What impact did his leadership have not only on committed Christians within Nazi Germany itself but on those in the military and quasimilitary organisations associated with Nazism (e.g. the Wehrmacht, Luftwaffe, SS, SA, etc.)? These and other cases are waiting to be explored.

‘Boots on the Ground’: What Does the Evidence Say? The phrase ‘participating factor’ is decidedly chosen to express a degree of neutrality when examining cases and causes of genocide. It is not intended here to argue that ‘religion’ serves as the primary, secondary or even tertiary rationale for acts of genocide, but, rather, that it remains one of several such factors – racial, biological, ethnic, political, geographic, economic, military, 15 James Frazer Moore, ‘The accountability of religion in genocide’ in Confronting Genocide, ed. Jacobs, pp. 65–76.

94

The Religion–Genocide Nexus

and others – all of which, taken together, constitute a constellation of variables, none of which can be isolated as exclusively pro-genocidal, but, when interwoven, provide foundational rationales for such behaviours. Thus, while Table 2.1 is not intended to be a full listing of every genocide committed throughout history, it serves as a starting point for examining Table 2.1 ‘Religion’ as a participating factor in genocide (Many of these cases are discussed in greater detail in other volumes in the series.) Case

Dates

Religious factor(s)

Punic genocide (battle of Carthage) Wu Hu genocide (Wie-Jie War) (China) Genocide of Cathars (Albigensian crusade) Zunghar (Dzungar) genocide (China) Black War genocide (Tasmanian Aboriginals) Maori genocide (New Zealand) Great Famine genocide (Ireland) California genocide (Spanish & US vs. Native Americans) Circassian genocide Selk’nam genocide, Tierra del Fuego, SA Herero & Namaqua genocide Armenian genocide (Medz Yeghern) Assyrian genocide (Seyfo) Greek & Pontic genocide (Megalis Catastrophi) Jewish genocide during Russian White Terror Soviet ethnic cleansing/ genocide Libyan genocide Kazakh genocide

149–146 B C E

No discernible religious factors

350–351 C E

No discernible religious factors

1209–29

Roman Catholicism vs Catharism

Confucianists vs. Buddhists & Muslims 1820s–1832 Christians vs Aboriginal religious traditions 1835–53 Christians vs indigenous religious traditions 1845–9 British Anglicanism vs Irish Catholicism 1846–73 Spanish & US Christians vs Native American religious traditions 1864–7 Orthodox Christians vs Islam Late 1800s/ Christians vs Aboriginal religious early 1900s traditions 1904–8 Christians vs tribal religious traditions 1914/15–22/3 Muslims & Secularists vs Christians 1755–8

1914/15–22/3 1914/15–22/3

Muslims & Secularists vs Christians Muslims & Secularists vs Christians

1918–23

Russian Orthodoxy vs Judaism

1920–51

Communists vs ethnic identities & religious traditions Muslims vs Christians Communists vs ethnic identities & religious traditions Communists vs Ukrainians/ Orthodox Christians

1923–32 1931–3

Great Soviet famine/genocide 1932–3 (Holodomor)

95

steven leonard jacobs

Table 2.1 (cont.) Case

Dates

Religious factor(s)

Romani genocide (Porajmos)

1935–45

Latvian genocide by NKVD

1937–8

Polish genocide by NKVD

1937–8

Jewish genocide by Polish collaborators with Nazis Holocaust (Shoah)

1939–45 1939–45

Belarusian genocide

1941

Nazis & Christians vs Romani religious traditions Communists vs ethnic identities & religious traditions Communists vs ethnic identities & religious traditions 2,000 years of antisemitic Christian animus 2,000 years of antisemitic Christian animus Communists vs ethnic identities & religious traditions 2,000 years of antisemitic Christian animus & ethnic & religious hatred Nazis vs Jews, Roman Catholics & Protestants Communists vs Muslims

Ustasˆe genocide of Jews, Serbs & others General Plan Ost, Genocide of Jews in Poland Aardakh genocide (Chechens & Vainakhs) Guatemalan genocide Bengali/Bangladeshi genocide in East Pakistan Burundian genocide of Hutus & Tutsis Cambodian genocide Ba’athist genocide against Kurds (Anfal) Isaaq genocide, Somalia Rwandan genocide Bambuti genocide Democratic Republic of Congo Darfur genocide ISIS genocide Rohingya genocide, Myanmar

1941–5 1941–5 1944–8

Roman Catholicism & Protestantism vs Mayan religious traditions Muslims vs other Muslims & Hindus

1962–96 1971

Belgian Catholic White Fathers’ division of Hutus & Tutsis Communists vs Buddhists & Muslims Sunni Muslims vs Shia Muslims

1972–93 1975–9 1986–9

No discernible religious factors Belgian Catholic White Fathers’ division of Hutus & Tutsis Belgian Catholic White Fathers’ division of Hutus & Tutsis

1988–91 1994 2002–3

2003–present Muslims vs Christians & animists 2014–present Islamists vs Yazidis, Christians & Shia Muslims 2017–present Buddhists vs Muslims & Roman Catholics

96

The Religion–Genocide Nexus

many genocides and those where, at least to this author’s way of thinking, religion does remain a ‘participating factor’.16 First off, in only two of these cases of genocide listed here does ‘religion’ not appear to be a significant factor: that of the Wu Hu genocide/Wie-Jie War of 350–351 C E, where military conquest appears to be its very rationale resulting in extermination; and the Isaaq genocide in Somalia of 1988–91, where tribal differences rather than religious discrimination resulting in annihilation appear to provide their own rationale. Also, the Third Punic War or battle of Carthage in 149–146 B C E may not have had a decidedly religious factor, but, rather, a significant agenda of annihilatory military conquest. Secondly, in the Soviet/Russian context and its Marxist ideology, we may label their genocidal work as ‘anti-religion’, consistent with their unsuccessful attempts to obliterate all expressions of religion in their nation-state. It should be noted here, however, that the Russian Orthodox Church during the Soviet era went largely underground and seemingly accommodated itself to its Soviet masters, and, only after the fall of Communism at the end of the twentieth century, did it reveal its own non-dormant vibrancy and flourishes today with an even greater impact upon its adherents, including its own manifestations of religio-theological antisemitism. Thirdly, in the cases of the secularist Turkish Committee of Union and Progress (CUP) and its genocidal agenda against Armenians, Assyrians and Greeks – all communities marked by their significant religious differences from their Muslim neighbours – it never completely separated itself from the Islamic tradition of its own past and present, very much embedded in the Ottoman Empire of which it claimed to be the quasi-legitimate successor. Fourthly, in examining this table in even more detail, it becomes obvious that the two most universalising faiths – Christianity and Islam – equally fall into the category of genocidal perpetrator. Their own evolving historical journeys not only in the West and Middle East but also in Africa have carried with them the seemingly sacrosanct idea of their own superiorities as ‘the one 16 Wikipedia’s comparable chart, Anon., ‘List of genocides by death toll’ (https://en .wikipedia.org/wiki/List_of_genocides_by_death_toll), does not address religion as a participating factor. Nor do the Twentieth-Century Atlas – Worldwide Statistics of Casualties, Massacres, Disasters and Atrocities on its website (www.rijnlandmodel.nl) – ‘Deaths by mass unpleasantness [sic]: estimated totals for the entire 20th century’; Rudolf Rummel, ‘20th century democide’, (www.hawaii.edu/powerkills/NOTE6 .HTMN); or Barbara Harff, ‘No lessons learned from the Holocaust? Assessing risks of genocide and political mass murder since 1955’, American Political Science Review 97:1 (2003), 57–73.

97

steven leonard jacobs

true faith’ to the denigration and worse not only of other religious traditions but, most particularly, of native and aboriginal religious traditions which had preceded them wherever they landed. And here, one must also mention the historical dominance of Roman Catholicism as well, not only on the European continent but in South/Latin America and on the African continent, and Sunni Islam’s own dominance over its Shi’a Islamic minority (today estimated at ~85% versus ~15%) throughout the Middle East. Both groups, coupled throughout history with nation-state political power, have used that power to advance their own theological agendas, and, in so doing, have legitimated genocidal violence against others understood to be both outliers and threats to their own staying power. Fifthly, in the specific case of the Jews, antisemitism – the world’s longest-lived hatred and ongoing prejudice – and reflected throughout its history, the ultimate expression of that contempt, at least in the West, would result in the genocidal violence orchestrated and harnessed by the Nazis and their allies during the Second World War. That outcome cannot be divorced from its long history of Christian denigration and violence. Today, given the ongoing tragedies of the Arab/Palestinian–Israeli conflict, the stated public desire of some in the Muslim world, such as former Iranian President Mahmoud Ahmadinejad (2005–13), to ‘wipe Israel off the map’, would constitute, were it ever to be realised, a second genocide against the Jews, driven by a radicalised Islamist reinterpretation of classical Islam.

Concluding Thoughts and Realistic Recommendations Not yet addressed, intentionally as well, is the ‘upside’ of the religion– genocide nexus. This is the healing power of religious leaderships, communities, traditions, sacred texts and rituals – not only for the victim populations but also for those perpetrators/genocidaires willing to acknowledge their own guilt and, perhaps in a limited way, attempt to make amends. Examples which come immediately to mind are the gacaca courts in Rwanda and the Truth and Reconciliation Commission of South Africa – one thinks of the role there of the Nobel Prize winner, Anglican Bishop Desmond Tutu – and that in Guatemala. Significantly, among the important roles the world’s religions play is that of ‘marking the moments’ of the life journeys of our common humanity: birth, growth, maturation, decay and death. This last, accompanied by 98

The Religion–Genocide Nexus

recognition of the traumatising loss of loved ones – family, friends, community – is the acceptance of the grieving process and the associated healing rituals (e.g. rending one’s garment, lighting candles, specific repetitive prayers). Scholars looking into the religion–genocide nexus should neither undervalue nor demean the psychic healing that those who see themselves as committed religionists experience and affirm. In an earlier (2004) article in the Journal of Hate Studies – ‘The last uncomfortable “religious” question: monotheistic exclusivism in Judaism, Christianity, and Islam’ – I advanced the idea of religiously committed scholars coming together and reading sacred texts in the presence of the other with an eye towards developing what I called, and based on traditional modes of Judaic textual understandings and interpretations, a ‘midrashic method’, and applying such a broader way of reading and thinking to both the New Testament and the Quran. (A somewhat similar method does already exist and has existed historically within the Roman Catholic tradition, but, within other expressions of Christianity – primarily more literalist Protestant evangelical and fundamentalist communities – the vaunted notion of sola Scriptura, a theological doctrine that the biblical text is the sole authority for faith and practice, has stifled such a development.)17 In a 2019 lecture at the International Conference on the Crime of Genocide, hosted by the Pan-Pontian Federation of Greece, Athens, Greece – ‘Preventions of genocide: rethinking the (1948) Genocide Convention in light of the Ruhashyankiko (1978) & Whitaker (1985) reports – what would Raphael Lemkin (1900–1959) have said?’ – I advanced two further recommendations. The first was to charge the United Nations through its secretary-general, supported by appropriate non-governmental organisations, with creating a ‘UN Anti-Genocide Education Task Force’ to develop a universal and international anti-genocide curriculum for all age levels, from elementary school through college and university and beyond, including seminaries and religious institutions (churches, synagogues, mosques, etc.), and that such a curriculum take as its starting point the preciousness – the sacrality – of human existence, drawing upon the work of generations of scholars in 17 This idea builds upon the work of Orthodox Jewish feminist Blu Greenberg, ‘The Holocaust and the gospel truth: Christian confrontations with the Holocaust’ (Holocaust and Genocide Studies 4:3 (1989), 273–82), as well as on Henry Knight, Confessing Christ in a Post-Holocaust World: A Midrashic Experiment (Westport, CT: Greenwood Press, 2000); and James Frazer Moore, Christian Theology after the Shoah (Lanham, MD: University Press of America, 1993).

99

steven leonard jacobs

history, literature, political science, philosophy, sociology and religious studies. Those responsible for the actual development and implementation of such curricula materials would be those persons, supported by the nation-states who send them and support them financially and administratively, who alone, by virtue of their training, are themselves schooled in the education of the young. And second, on the agenda of the meetings of the Parliament of the World’s Religions, now more than a century in existence and having met seven times in its past, there should be not only a formal resolution once and for all condemning genocide and building upon the work of Raphael Lemkin, author of the word ‘genocide’ and the motivating force behind the 1948 UN Convention on the Prevention and Punishment of the Crime of Genocide. The purpose of that Parliament is to bring persons of faith together to work for a more just, peaceful and sustainable world. Leading and high-ranking clergy representing the multiplicity of the world’s faiths, as well as scholars and practitioners, gather together, pray together, break bread together, and unite on issues of concern which bring them together rather than divide them. And because this parliament already has a UN Task Force, let us challenge them to challenge the United Nations and its secretary-general to institute the aforementioned UN Anti-Genocide Education Task Force.

Bibliographic Note Other than the four authors mentioned early on in this chapter – Bergen, Glick, Kuper and Temoney – and the two edited collections – Bartov and Mack’s In God’s Name and my own Confronting Genocide: Judaism, Christianity, Islam – there is yet very little ‘out there’ on the topic of religion and genocide from a more generalised and/or theoretical and/or methodological perspective. There are, to be sure, any number of texts which examine the ‘religious factor’ in particular genocides: see for example Branimir Anzulovic, Heavenly Serbia: From Myth to Genocide (New York University Press, 1999); Michael Sells, The Bridge Betrayed: Religion and Genocide in Bosnia (Berkeley: University of California Press, 1996); Carol Rittner and John K. Roth, eds., Genocide in Rwanda: Complicity of the Churches (St Paul: Paragon House, 2004). The place to start vis-à-vis the religion–genocide nexus, however, is the historical text by Dominican Friar Bartolomé de Las Casas (1484–1566), who accompanied the Spanish Christian conquistadores and witnessed at first hand the genocidal slaughter of indigenous peoples at their hands. His Brevísima relación de la destrucción de las Indias, written in 1542 but published in 1552, 100

The Religion–Genocide Nexus

translated by Nigel Griffin as A Short Account of the Destruction of the Indies (London: Penguin, 1992), graphically presents what transpired. Significantly germane to this discussion, however, is the growing body of literature addressing religion and violence among the world’s religious communities, both as committed perpetrators and post-violence healers. Important to this examination is the relevance of the sacred literatures of the world’s religions. Thus, for example, see John Renard, Fighting Words: Religion, Violence and the Interpretation of Sacred Texts (Berkeley: University of California Press, 2012), and Ra’anan S. Boustan, Alex P. Jassen and Calvin J. Roetzel eds., Violence, Scripture, and Textual Practice in Early Judaism and Christianity (Leiden and Boston: Brill, 2010). The one area where this nexus connects is that of the seemingly divine command in the Hebrew Bible/Old Testament Book of Joshua, whereby Joshua is instructed to annihilate/ exterminate the Canaanites. (Whether or not this commanded act actually took place partially or in totality or at all remains a source of scholarly dispute.) Interestingly enough, texts which address this biblical issue remain more so within the various iterations of Christianity, primarily fundamentalist or evangelical, rather than among scholars of Judaism, both religious and secular. The sole Judaic exception with which I am familiar is that of the former British Chief Rabbi Lord Jonathan Sachs, Not in God’s Name: Confronting Religious Violence (New York: Schocken Books, 2015), though, like Karen Armstrong in Fields of Blood: Religion and the History of Violence (New York and Toronto: Alfred A. Knopf, 2014), he argues that such violence is itself a perversion of the ‘true’ nature of religion, as does theologian William G. Cavanaugh in his The Myth of Religious Violence (Oxford University Press, 2009). An exception, however, is that of Christian Hofreiter, Making Sense of Old Testament Genocide: Christian Interpretations of Herem Passages (Oxford University Press, 2018). Recommended as well are J. Harold Ellens’ four-volume edited collection The Destructive Power of Religion (I : Sacred Scriptures, Ideology, and Violence; I I : Religion, Psychology, and Violence; I I I : Models and Causes of Violence in Religion; I V : Contemporary Views on Spirituality and Violence) (Westport, CT: Praeger, 2004); Andrew R. Murphy, ed., The Blackwell Companion to Religion and Violence (Chichester and Malden, MA: Wiley-Blackwell, 2011); Mark Juergensmeyer, Margo Kitts and Michael Jerryson, eds., The Oxford Handbook of Religion and Violence (Oxford University Press, 2013); and Jeffrey Ross’ edited three-volume Religion and Violence: An Encyclopedia of Faith and Conflict from Antiquity to the Present (Armonk, NY: M. E. Sharpe, 2011). Also important to note is the text by Mark Juergensmeyer, Terror in the 101

steven leonard jacobs

Mind of God: The Global Rise of Religious Violence, 4th ed. (Berkeley: University of California Press, 2017). (On a side note, given the historical reluctance of scholars of the American historical experience to address the tragic plights of genocidal violence against both Native American tribal communities and African American slave communities, the role of Christianity in both contexts as foundational to perpetrator violence receives a paucity of texts and attention, paralleling the overarching unwillingness of so many to address the role of religion as a participating factor in these genocidal events. An important exception is an older text, George E. Tinker, Missionary Conquest: The Gospel and Native American Cultural Genocide (Minneapolis: Fortress Press, 1993).) Thus, overviews remain important, even if not particularly focused on this nexus of religion and genocide. I would start with Adam Jones’ masterful textbook Genocide: A Comprehensive Introduction, 3rd ed. (London: Routledge, 2017); and Donald Bloxham and A. Dirk Moses, eds., The Oxford Handbook of Genocide Studies (Oxford University Press, 2010). Equally important are Daniel Jonah Goldhagen’s Worse than War: Genocide, Eliminationism, and the Ongoing Assault on Humanity (New York: Public Affairs, 2009), with an accompanying PBS video with the same title; Ben Kiernan’s Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007); and Norman M. Naimark’s Genocide: A World History (Oxford University Press, 2017). Three edited and equally important additions round out this overview: Paul R. Bartrop, ed., Modern Genocide: Analyzing the Controversies and Issues (Santa Barbara: ABC-CLIO, 2018); Robert Gellately and Ben Kiernan, eds., The Specter of Genocide: Mass Murder in Historical Perspective (Cambridge University Press, 2003); and Dan Stone, ed., The Historiography of Genocide (New York: Palgrave Macmillan, 2008). Finally, an important contribution to possibly rethinking the relationship between religion and genocide would be Andrew Bell-Fialkoff’s Ethnic Cleansing (New York: St Martin’s Griffin, 1996), reframing this entire conversation, as well as those texts which specifically address the question of cultural genocide, seeing religion as an aspect of human cultural productivity. Important here are Jeffrey S. Bachman, ed., Cultural Genocide: Law, Politics, and Global Manifestations (London and New York: Routledge, 2019); Lawrence Davidson, Cultural Genocide (New Brunswick and London: Rutgers University Press, 2012); and Elisa Novic, The Concept of Cultural Genocide: An International Law Perspective (Oxford University Press, 2016).

102

3

Genocide and Gender Dynamics and Consequences adam jones and wendy lower

Introduction Like all historical events and phenomena, genocide is gendered. It is driven by concepts of manhood and femininity, biological procreation and extinction, and is realised through the distinct targeting of men, women, infants of one sex and LGBTQ communities. Indeed, standard explanations of genocide, beginning with those offered by jurists like Raphael Lemkin, disregarded or glossed over the sexism and gendered concepts inherent in all the other variables and -isms (imperialism, nationalism, racism, militarism, sadism) that lawyers and scholars apply to explain how and why genocide happens. One could write a new history of the Holocaust, for example, that charts the escalating persecutory measures that led to sexualised mass murder. In 1933, Nazi leaders (all men) removed Jews from the civil service, which impacted mostly male heads of households and forced Jewish women into new roles, often public ones that made them more vulnerable to persecution. Similarly, in the cases of post-war genocide in Bangladesh, Burundi (1972), Guatemala, Bosnia-Herzegovina and Syria (explored below), the genocidaires first attacked the able-bodied men within their target group because the men were perceived as the greatest physical threat, and a campaign of male-on-male mass murder could be undertaken as if the genocide were a regular war. The transition from killing battle-age men first to communities of women, children and the elderly has long been a subject of interest in Holocaust Studies. Jewish men had been the targets of mass violence since the November 1938 Kristallnacht pogrom, and then during the Nazi military conquest of Poland in autumn 1939. In summer 1941, Hitler and Himmler argued that the women and children had to be killed, indeed whole families, in order to exterminate the entire race (what scholars refer to as a ‘root and

103

adam jones and wendy lower

branch’ genocide, discussed later in this chapter). It is seen as a pivotal moment in Hitler’s decision-making, and evidence has emerged that there was some initial reluctance among Himmler’s operatives to kill women and children. The gendered dimension of this moment and how it was overcome have become clearer, yet scholars have rarely explored this central feature of the normally accepted turning points of 1941. Gendered distinctions hold true even for the post-war period of justice in the treatment of male and female defendants and in sexualised perceptions of guilt, innocence and genocidal behaviour. Revisiting the history and interpretation of genocide as distinctively male or female, or as separate spheres of men’s and women’s history, often misses another significant dimension of agency: the interaction of men and women, and the dynamic it produces, can accelerate or slow down genocidal campaigns. This chapter provides a general overview of gender as a key concept that can illuminate new facets of perpetrator behaviour, the law, LGBTQ vulnerabilities, institutional violence and genocide prevention. It encourages readers to rethink major categories of analysis and themes, such as perpetrator, victim, collaborator, crime, violence, justice, reparations and memorialisation, as gendered phenomena.

Gender as a Concept Feminist scholars have long stressed the specificity of the female experience and of women’s history, illuminating the structural, patriarchal features and violent patterns (e.g. mass rape and forced sterilisation). Catharine MacKinnon illuminated rape as a form of torture and dehumanisation. But scholarship on sexual violence during war and genocide came much later, and did not develop as an outgrowth of Holocaust Studies. Rather, it came to the public’s attention and to Genocide Studies first in conjunction with the events and trials surrounding the former Yugoslavia and Rwanda. Among the first to grasp the importance of gender as a concept in Genocide Studies, Adam Jones located the roots of gender as a concept in Genocide Studies in the growing number of case studies of rape and violence against women and girls, sexualised enslavement, gender-selective killing and infanticide. For this chapter, we will adopt the concept of gender as both biological and cultural, rather than viewing these as mutually exclusive. International relations scholar Joshua Goldstein’s inclusive view of gender encompasses ‘masculine and feminine roles and bodies alike, in all their aspects, including the (biological and cultural) structures, dynamics, roles and scripts associated 104

Genocide and Gender

with each gender group’.1 Or in the instructive words of Genocide Studies scholar Elisa von Joeden-Forgey, gender is ‘a marker of biological sex and . . . a set of cultural practices and beliefs aimed at organizing relations of power between the sexes’.2 In binary genocidal thinking, the female gender was often projected as the positive, utopian ideal, often racialised as the embodiment of purity and procreation; and women and girls were seen as the carriers of the past into the future (as bearers of the next generation). In many cases, the presumed virtue vulnerability or biological value of the woman was contrasted with a criminal, dirty or threatening image of the male foe in the enemy group. Von Joeden-Forgey’s analysis of power relations between the sexes applies to the dynamics among perpetrators as well as to those between the perpetrators and their targeted victims. Thus, for example, women in the perpetrator society obtain greater power over the victims, both male and female, while they function within an often patriarchal, even misogynistic system of rule and social organisation. In fact, von Joeden-Forgey argues that the gendered ideas and practices of genocidal violence are founded in the socialisation and education experienced within the family or domestic sphere. She identifies extremely violent societies as those that nurture and enable violent behaviour at home. For example, a recent study of child-rearing practices in Germany found that 1930s paediatric physicians preached harsh methods that neglected the emotional needs of children, producing a generation of adults who still experience serious difficulty feeling compassion or love for others.3

Rethinking Women and Genocide Historically, very few women compared to men have wielded enormous power as monarchs and heads of state. Some, like Joan of Arc and the Trung sisters in early Vietnamese history, have waged wars and entered popular memory as warriors and martyrs. Yet rulers such as Queen Elizabeth I and Catherine the Great have not been the subject of deep critical analysis as far as their roles in promoting violent conflict and perhaps even genocide (for instance, in Elizabethan Ireland). And until recently, the fact that ordinary 1 Adam Jones, Genocide: A Comprehensive Introduction, 2nd ed. (London: Routledge, 2010), p. 462. 2 Elisa von Joeden-Forgey, ‘Gender and genocide’ in The Oxford Handbook of Genocide Studies, ed. D. Bloxham and A. D. Moses (Oxford University Press, 2010), p. 62. 3 Anne Kratzer, ‘Harsh Nazi parenting guidelines may still affect German children of today’, Scientific American, 4 January 2019, www.scientificamerican.com/article/harshnazi-parenting-guidelines-may-still-affect-german-children-of-today1.

105

adam jones and wendy lower

women also participated in mass violence and selective killing during ancient, medieval and early modern wars, modern revolutions, colonial conquests, pogroms and lynchings rarely led to further investigation of their historical and global role in genocide. Once under way, genocidal campaigns of mass persecution and violence require a mobilisation of resources similar to military and political campaigns and revolutionary movements that cut across public institutions and private domains. As such, genocidal systems draw upon the involvement of women and youth, who in the context of the modern state can wield a greater, often lethal influence. To be sure, men dominate as the leaders of such systems, and also as the perpetrators of the violence. No genocidal regime has ever been led by a woman, or by a group that included a significant number of women. However, the totality and relentlessness of genocide require the participation of the broader society to varying degrees. Women can provide sustenance and moral support, bolstering and provoking their mates, and recruiting the youth and others to take up arms; they have also reinforced notions of masculine power by shaming those men reluctant to participate in the persecution or killing. Independently of their male mates and comrades, female accomplices and perpetrators have robbed and abused the victims, and in some cases murdered them with their bare hands. Kurdish and Turkish female collaborators tormented and assaulted Armenian families while they were being forcibly marched out of their villages; then the women plundered their homes.4 Josephine Block, the wife of the Gestapo chief in Drohobycz, Ukraine, was witnessed brutalising Jewish women, men and children on the streets of the town, and escorting Roma to the woods to be shot. She was known to ‘wear the pants’ in her family, while her husband Hans developed a reputation for submitting to his domineering wife. Thousands of German women served as concentration camp guards in Nazi-occupied Europe, and tens of thousands worked in Nazi Party organisations and SS police offices administering campaigns of persecution and murder against Jews, Poles, Soviet POWs, the mentally and physically disabled, and others deemed inferior and/or enemies of the Reich. After the war, a few hundred women were prosecuted as camp guards, or as nurses who had killed people in the Nazi euthanasia program.5 4 John Minassian, Surviving the Forgotten Genocide: An Armenian Memoir (Lanham, MD: Rowman & Littlefield, 2020). 5 Wendy Lower, Hitler’s Furies: German Women in the Nazi Killing Fields (Boston: Houghton Mifflin Harcourt, 2013).

106

Genocide and Gender

Cambodian women joined the Khmer Rouge and became soldiers of Pol Pot. They participated in the torture of hundreds of prisoners at S-21 and other sites. The highest-ranking woman in the Pol Pot regime, Ieng Thirith, the minister of social affairs (1975–9), wife of Foreign Minister Ieng Sary, and sister of Pol Pot’s wife, helped to direct campaigns that inflicted a system of mass deprivation upon enemies as well as nationwide forced marriages and rape.6 In 2007, the UN-sponsored Cambodia Tribunal charged Thirith with crimes against humanity, charges that were dropped in 2011 only because she was by then judged mentally unfit to face trial. In Rwanda, Hutu women perpetrated violence in unprecedented ways by routinely ‘finishing off’ the wounded, killing women together with their children and urging Hutu men to commit murder and rape on a mass scale. Pauline Nyiramasuhuko (former Hutu minister for family welfare and the advancement of women) was the first woman to be convicted of rape by an international court. She was found to have incited her own son to commit rape and genocide in the ‘Butare’ case. Radicalised ISIS women in Syria and northern Iraq served as suicide bombers and members of the police force monitoring other women. Although the ISIS manifesto and guide for women of the Islamic State stresses that women and men are not equal, and that women should be largely absent from the public realm, it does state that women can wage jihad (holy war) and take up certain professions outside of the home (such as teaching and medicine).7 Tens of thousands of these female jihadists, now in refugee camps with their brutalised children, continue to promote violent extremism. Some have engaged in violence at the al-Hol camp, and incited acts of terror around the world.8 Besides direct participation in genocidal violence, women can and have played a much larger role as accomplices in organising all aspects of the campaign: the educational indoctrination at home, in schools and in propaganda ministries; the administration of persecutory measures as bureaucratic functionaries; the looting of victims’ property and personal belongings; and the suppression of crimes by destroying evidence and providing alibis for 6 Ben Kiernan, The Pol Pot Regime: Race, Power, and Genocide in Cambodia under the Khmer Rouge, 1975–1979 (New Haven: Yale University Press, 1996), esp. pp. 100, 160–2, 236–7; Raya Morag, ‘Gendered genocide: the female perpetrator, forced marriage, and rape’ in Perpetrator Cinema: Confronting Genocide in Cambodian Documentary (New York: Columbia University Press, 2020), pp. 132–79. 7 See Azdeh Niaveni, Guest House for Young Widows: Among the Women of ISIS (New York: Random House, 2019). 8 Lydia Khalil, ‘Behind the veil: women in Jihad after the Caliphate’, Lowy Institute, 25 June 2019, www.lowyinstitute.org/publications/behind-veil-women-jihad-aftercaliphate.

107

adam jones and wendy lower

their perpetrator colleagues and mates. Women perpetrators’ motivations differ little from men’s; they share similar emotions, ambitions and desires, such as fear, hate, greed, status and ideological convictions of nationalism, imperialism, racism, xenophobia, antisemitism. Von Joeden-Forgey has discerned patterns in female perpetration that occur in their domains, such as the killing of children and the abuse of forced labourers and other victims in or near their homes or other traditionally female professional settings, such as schools, hospitals and communal centres. Nazi perpetrator Erna Petri shot Jewish boys near her home. She explained after the war that she did it to prove herself to the men, and because she had been taught to hate Jews. As secretaries, welfare workers, nurses and wives, women have adapted conventional female roles to the needs of a genocidal state and system. Thus, women become active members of the society of perpetrators (often constituting a numerical majority), and can no longer be assumed to be the innocent, ‘gentler sex’. Early feminist scholarship on gender and genocide, which focused mostly on the Holocaust, brought women into the historical picture, restoring their agency and documenting their suffering. Yet, as Zoe Waxman pointed out, often these accounts perpetuated biased categories of analysis by stressing women’s roles as mothers and caregivers, and by overstating bonds of sisterhood as the reason for women’s heroic survival and resistance. While the narratives of female suffering, death or survival as victims restored their place in the narrative, this analysis left little room for representing experiences that did not conform to gendered norms and expectations. Similarly, the few studies of German female perpetrators depicted them as freaks of nature, hyper-sexualised camp guards whose sadism and more ‘masculine’ style of aggression betrayed their femininity. Their evil could be explained as a biological defect or as a response to male influence. In other words, a gendered approach requires excavating a few analytical layers. First, what were the prevailing notions and assumptions about male and female behaviour in the time and place being studied? Second, how were they changed by genocide? And third, do these notions and assumptions persist in the firstand second-hand narratives about what happened, perhaps obscuring other realities of genocide?9 Comparing the testimony of female and male victims reveals varied reactions to the persecution, both because and in spite of their gender. 9 Zoe Waxman, ‘Unheard testimony, untold stories: the representation of women’s Holocaust experiences’, Women’s History Review 12:4 (2007), 661–77.

108

Genocide and Gender

Genocidaires seek in all possible ways to humiliate, weaken and dehumanise their targets, and apply different gendered tools to these ends. For female victims, the erosion of the self often begins with assaults against their womanhood, their distinctive dress, their homes, their bodies and fertility. Ruth Kluger, who survived Auschwitz-Birkenau because of her mother’s bravery and acuity, experienced at first hand the awful challenges faced by Jewish women under these extreme circumstances: ‘We were raised to make gefilte fish, not to resist, [and not] to defy the men.’10 Women were taught to obey, to sacrifice, and certainly not to abandon family and children. The weight of this tradition, and how deeply it was internalised in gendered roles and expectations in the family unit, and in the minds of individual Jewish women, provides one explanation for why so many Jewish women went to their deaths with their children. They had already taken on the role of head of household when their husbands emigrated, were sent to camps, were murdered or fled.11 They cared for the remaining children and the elderly while they themselves feared for the future, and as they met horrific deaths together with them. Mothers helped children pack bundles and dressed them as they were corralled and driven to deportations, and held their hands as they were marched towards the sound of gunfire in the forest. Mothers comforted children, often not their own, even as they were filled with an unfathomable fear of suffering and death. Some might explain these acts as a reflection of a female trait: the ability and desire to nurture. Of course, women wanted to survive as individuals, and some abandoned or even killed their own children to survive. But most of the evidence shows that Jewish women tried to protect their children and sacrificed themselves. Perhaps beyond their maternal instincts, they also acted on another understanding of familial survival and duty, one to a community that was faced with extinction. Girls are also targeted in acts of mass state violence intended to reduce populations that burden the state. The history of euthanasia in Nazi Germany is one example. More recently, in China and India, the state and families have condoned the killing of newborn females and sex-selective abortions because of poverty, overpopulation and sociocultural taboos (pregnancies of unmarried women, dowry customs and so on).

10 Ruth Kluger, Still Alive: A Holocaust Girlhood Remembered (New York: Feminist Press, 2001), p. 30. 11 Marion Kaplan, Between Dignity and Despair: Jewish Life in Nazi Germany (Oxford University Press, 1996).

109

adam jones and wendy lower

In ‘root and branch’ genocides bolstered by racial theories and biopolitics, female victims experience assaults against their femininity and fertility. Women in the genocidaire community also surrender control over their bodies. Perpetrator societies seek to fortify their regimes through pro-natalist policies that effectively deprive women of choices over their own health, sexual orientation, reproduction and children. For example, in Cambodia the Khmer Rouge created a state policy of forced marriages and marital rape that was so extreme the Cambodia Tribunal judged that both forced marriage and rape had occurred at the level of crimes against humanity.12 When procreation becomes a state policy aimed at producing more children for the revolution, intercourse becomes a means to an end, downgrading human relationships and often leading to acts of rape. Sexual violence, mass rape and rape-murder of those deemed biologically worthless are more widespread in genocidal states, especially those engaged in wars of conquest and social engineering campaigns.

Rethinking Men and Genocide Gender theory has traditionally emphasised the role of men as perpetrators of genocide and crimes against humanity, especially against women and girls. The historical record demonstrates that the depiction of men as perpetrators is well founded. A vast majority of direct genocidal atrocities – likely over 95 per cent – have been inflicted by males; only in the realm of structural gendered violence is the ratio less skewed.13 Why this is the case, however, requires closer examination; so too does the class of victims and survivors. The former question can be boiled down to two causes: is it nature or is it nurture? Are males genetically and/or physiologically more predisposed to violence, particularly the kind of explosive and extreme violence that characterises genocide? Or are familial and social constructions of violent masculinity primarily responsible? Probably neither explanation is adequate in itself, and a more nuanced and dialectical understanding is called for. Males’ average greater physical size and stronger musculature, combined with their greater ‘disposability’ from a population-reproduction standpoint, likely meant that they would be on 12 Extraordinary Chambers in the Courts of Cambodia (ECCC), Trial Chamber, Final Judgement, Case 002/02, 16 November 2018, www.eccc.gov.kh/en/document/court/ case-00202-judgement. See e.g. pp. 1863–9, 2062–4, 2104–5, 2139–40, 2151, 2163. 13 For example, the direct perpetrators of female infanticide and female genital mutilation are typically female.

110

Genocide and Gender

the front line in defending the clan or community against animal or human predators.14 When a group became predatory in its turn, and adopted genocidal or proto-genocidal strategies against out-groups, males were likewise overdetermined as direct agents of the slaughter. Over millennia, these identifications became inculcated, institutionalised and increasingly embedded in broader societal strategies of patriarchal rule, economic production/ reproduction and cultural identity. We have already seen that attention to gendered forms of victimisation of females has tended to submerge the diverse ways in which women may participate in and contribute to genocides. So, too, has the concentration on men as genocidaires served to obscure the striking extent to which males are also the most direct and immediate targets of both pre-genocidal and genocidal violence. Indeed, men and boys are so commonly and selectively attacked first in genocides that there are grounds for dividing historical and contemporary genocides into two broad categories. In a case of gendercide, the genocide commences and ends with a selective exterminatory assault on a community’s males – virtually never a selective exterminatory assault on females – and this tends to bound the strictly murderous dimension of the genocide. In the broader case of a ‘root and branch’ genocide, on the other hand, an initial targeting of males is followed by, and enables, the extermination of most or all sectors of the community. Women and children (that is, the ‘root’ and the ‘branch’) are swept up in the slaughter, along with the elderly, the disabled and other groups that may under other circumstances be exempted from physical killing. In complex ways, but to a significant degree, three genocides of the twentieth century – the Armenian genocide, the Jewish Holocaust and the Rwandan genocide – fit the ‘root and branch’ category. In all three, though only briefly in the Holocaust and in a somewhat more muted form in Rwanda, men and boys were a special focus of the initial or early killing campaign. In perhaps only one such case, the Khmer Rouge genocide of Cambodia’s ethnic Vietnamese minority, nearly all the males and females, adults and children, were murdered in a short period, in 1977 and 1978. In numerous other instances, whether ancient or modern, the mass murder of a community’s males has largely bounded and restricted the directly exterminatory element of the genocide. Examples from the post-World War

14 See the discussion in Barbara Ehrenreich, Blood Rites: Origins and History of the Passions of War (New York: Henry Holt, 1997).

111

adam jones and wendy lower

Two period include genocides in Bangladesh, Burundi (1972), Guatemala, Bosnia-Herzegovina and Syria. Operative considerations in these cases include an association of males with the political, economic and military potency of the targeted group; and a sense that genocidaires are best acclimatised to their task by first murdering adult males of an imputed ‘battle age’ (roughly 15–55), who can more easily if not convincingly be depicted as mortal threats to the perpetrators. Once killers are so acclimatised, however, it is an open question whether they will turn and be turned to exterminatory violence against women, children and other sectors of a community now emasculated of its leaders and defenders. This serves as a reminder that few if any genocidal acts are inflicted according to a single variable (in this case, gender). Beyond core identifications and imputations of national, ethnic, racial, religious and political identity, perpetrators assign a range of specific associations to men and masculinities. They include community prominence, military prowess and often a high degree of mortal/sexual/existential threat. Regarding the last of these: the perception that men and boys pose a genealogical threat to perpetrators under patriarchy – that is, they cannot be forcibly incorporated into the patrilineal bloodline of the perpetrator population as readily as women and girls can – accounts for the regular pattern in classical warfare of murdering males while preserving alive females for enslavement and exploitation. Space does not permit an extensive sampling of such gendercidal atrocities against out-group males throughout history and around the world today. Classical exemplars, some surely legendary, can be found in the ‘western’ tradition’s most iconic work of literature, the Bible, as well as in Homer and Thucydides. In the contemporary era, state agents have perpetrated this kind of gendercide from Syria to Brazil to the Philippines, while sub-state formations like the Islamic State and Boko Haram are equally selective in their slaughter.15 Significant challenges exist in separating – for analytical purposes – the operation of the gender variable in these cases from those variables that accompany it. A sophisticated approach to genocidal dynamics should, however, preserve gender as a significant thread in the weave, as indeed is true of genocide and other mass atrocities against women and girls. This bears upon the question of whether males can be targeted ‘as such’ in 15 Adam Jones has compiled an incomplete catalogue of twentieth- and 21st-century examples. See for example the writings collected in his Gender Inclusive: Essays on Violence, Men, and Feminist International Relations (London: Routledge, 2009).

112

Genocide and Gender

genocide, as women are frequently claimed to be. We believe men and boys can be attacked ‘as such’ – and as other things besides, that is, with other variables intervening and overlapping. A notion of misandry (hatred of men, at least those of a certain group) paralleling misogyny should not be dismissed. According to the perceptions already discussed, it is standard for adult males especially to be cast as violent, terroristic, subversive, dirty/corrupt, sexually predatory and other gender stereotypes. These feed into the prima facie assumptions that generally mark out males as the initial, and usually predominant, victims of political-military forms of genocide, even though it is conducted mainly by other males. Genocidal forms of structural and institutional violence are less straightforward, as will be discussed. Two further thoughts can be offered, one focused on victims of genocide, the other on perpetrators. First, recent scholarship has emphasised that campaigns of mass repression and extermination against males very often have a significant sexual component, of the kind that has more readily been recognised as targeting females. Classical warfare, for example, featured attacks upon male genitalia, including the severing of the penis/testicles as ‘trophies’ – a practice not unknown in the contemporary period as well. Passages in the Hebrew Bible and other ancient sources describe rituals in which targeted men are emasculated in violent, humiliating acts that are intended to reduce them to animals – such as flaying their skin like a lamb, chaining them up like livestock and forcing them to crawl on the ground.16 Sexual violence against out-group males, especially in conditions of detention and incarceration, is pervasive and well documented in modern cases. Abu Ghraib, Sri Lanka, Congo and Syria offer relatively well-documented examples.17 The cultivation of masculine sensibilities and anxieties under patriarchy also serves to ‘prime’ men for a dominant role in genocide. We saw earlier that female archetypes are standardly deployed to mobilise women for their 16 T. M. Lemos, Violence and Personhood in Ancient Israel and Comparative Contexts (Oxford University Press, 2017). 17 On Sri Lanka, see Human Rights Watch (New York), ‘“We will teach you a lesson”: sexual violence against Tamils by Sri Lankan security forces’, 26 February 2013, https:// tinyurl.com/2fwavakb (accessed 5 April 2022). On Congo, see the ground-breaking documentary by the Refugee Law Project in Uganda, ‘Gender against Men’, YouTube, 21 May 2012, https://youtube/mJSl99HQYXc. On Syria, see Sarah Chynoweth, ‘“We keep it in our heart”: sexual violence against men and boys in the Syria crisis’, report for the United Nations High Commissioner for Refugees, October 2017. On the phenomenon generally, see E. A. P. Gorris, ‘Invisible victims? Where are male victims of conflict-related sexual violence in international law and policy?’, European Journal of Women’s Studies 22:4 (2015), 412–27.

113

adam jones and wendy lower

direct and supporting roles in genocidal enterprises. For men, culturally designated roles of provider and protector are fertile ground for a genocide’s planners and propagandists. Conditions of economic crisis, especially combined with military crisis or defeat, can provoke an existential crisis for males, as during the Depression era in Germany. In Rwanda in the early 1990s, a collapse in coffee prices, uprooting of populations by the RPF invasion, and downsizing of military and security forces under the Arusha Peace Accords threw thousands of men, women and families into penury and uncertainty. For younger Hutu men in particular, the loss of employment and subsistence foreclosed marriage possibilities and represented a specific kind of gendered humiliation. The lure of seizing Tutsis’ homes, possessions and jobs – and in many cases women – was powerful indeed.

Gender in the Law and Comparative Study of Genocide In the contemporary period, from a previously marginal status, gender has moved close to the centre of international law and humanitarian agendas. This is true, however, only with regard to forms of victimisation that disproportionately target women and girls. And insofar as it concentrates on females as a gender category, the legal articulations and advances have focused heavily on rape and sexual violence. The concept of ‘genocidal rape’ is the most notable contribution to international-legal terminology and jurisprudence. Among the emerging laws of war, ‘rape by soldiers has . . . been prohibited . . . for centuries, and violators have been subjected to capital punishment under national military codes’. Theodor Meron traces such bans to the fourteenth century in medieval Europe, and stresses the Lieber Code adopted during the American Civil War, under which soldiers could be and were executed for rape. Acts of rape and sexual violence, if widespread or systematic, were outlawed as crimes against humanity under the Rome Statute of 1998. In various cases throughout this period and up to the present, however, ‘rape has [also] been given license, either as an encouragement for soldiers or as an instrument of policy’.18 Its resurgence in the genocides of the early 1990s – in the former Yugoslavia and Rwanda – coincided with the creation of ad hoc tribunals for each case, and with global feminist-inspired activism around 18 Theodor Meron, ‘Rape as a crime under international humanitarian law’, American Journal of International Law 37 (1993), 425.

114

Genocide and Gender

sexual violence against women. The International Criminal Tribunal for the Former Yugoslavia (ICTY) and the International Criminal Tribunal for Rwanda (ICTR) accordingly became arenas of attention to gender, and of innovation regarding conceptualising sexual violence in crimes against humanity and genocide. The ICTY was notable not only for foregrounding sexual violence against women, especially Bosnian Muslim women, but for the Kunarac et al. case, in which the Trial Chamber supplied a detailed and influential definition of rape.19 The ICTR, for its part, is most renowned for its Akayesu judgment in 1998, upheld on appeal, that for the first time articulated rape as a strategy of the crime of genocide. Apparently led by the lone female judge, Navanethem Pillay, the Akayesu judges declared that acts of rape and sexual violence constitute genocide in the same way as any other act as long as they were committed with the specific intent to destroy, in whole or in part, a particular group, targeted as such. Indeed, rape and sexual violence certainly constitute infliction of serious bodily and mental harm on the victims [Article 2(b) of the 1948 Genocide Convention] and are even, according to the Chamber, one of the worst ways of inflicting harm on the victim as he or she suffers both bodily and mental harm. The judges also cited Article 2(d) of the Convention, in that acts of ‘sexual mutilation, the practice of sterilization, forced birth control, separation of the sexes, and prohibition of marriages’ could be strategies aimed at ‘preventing births within the group’.20

Despite widespread support in feminist and humanitarian ranks for Akayesu’s genocidal-rape finding, some criticism was also raised. In Rape Loot Pillage: The Political Economy of Sexual Violence in Armed Conflict, Sara Meger argued that ‘the focus of the ICTR on sexual violence as an instrument of genocide has effectively entrenched an understanding of the use of rape based on the “rape as genocide” narrative’. For Meger, this obscured the fact that ‘rape was something perpetrated against individual women by individual 19 The Kunarac definition reads as follows: ‘In light of the above considerations, the Trial Chamber understands that the actus reus of the crime of rape in international law is constituted by: the sexual penetration, however slight: (a) of the vagina or anus of the victim by the penis of the perpetrator or any other object used by the perpetrator; or (b) of the mouth of the victim by the penis of the perpetrator; where such sexual penetration occurs without the consent of the victim. Consent for this purpose must be consent given voluntarily, as a result of the victim’s free will, assessed in the context of the surrounding circumstances.’ See United Nations, International Residual Mechanism for Criminal Tribunals, Case Law Database: rape, https://cld.irmct.org/notions/show/742/rape. 20 Mark Ellis, ‘Breaking the silence: rape as an international crime’, Case Western Reserve Journal of International Law 225 (2007), 232–3, https://scholarlycommons.law.case.edu/ jil/vol38/iss2/3.

115

adam jones and wendy lower

men’, transforming the sexual attacks ‘into rape of the ungendered “Tutsi” by the ungendered “Hutu”’. It also promoted the ‘securitisation’ of sexual violence – its designation as a ‘crisis’ when it threatens ‘states themselves’, but not otherwise.21 The quotidian and institutionalised character of rape under ‘normal’ societal conditions was thereby submerged. It is worth noting, in closing, that gendered attention to crimes against males has been virtually absent from the international-legal sphere. (It is far better represented, albeit only recently, in the humanitarian/NGO sphere.) However, the issue of sexual violence against male detainees did surface in the ICTY proceedings, and the definition of rape crafted by the tribunal was gender-neutral and inclusively framed.22 As for the somewhat amorphous field of comparative Genocide Studies, the presence of gender has been limited and uneven, but still significant. A field as such is only evident from the early 1990s, by which time ‘gender’ (introduced by feminist scholars and activists, and usually understood as women and girls) was well established as a variable in mass violence, especially sexual violence. Males were incorporated as perpetrators, but rarely as victims and survivors. Early works published by Zed Books, like the Gender and Catastrophe and States of Conflict collections, tentatively broached the subject of gender and genocide.23 But they were positioned in disciplines such as International Relations, Development Studies and Women’s/Gender Studies, and have been little cited by genocide scholars. In Holocaust Studies, the gender-specific experiences of women and girls – including Nazi female perpetrators and accomplices – have been relatively well explored.24 Treatments in comparative Genocide Studies have been 21 Sara Meger, Rape Loot Pillage: The Political Economy of Sexual Violence in Armed Conflict (Oxford University Press, 2016), pp. 127, 136. 22 See International Criminal Tribunal for the Former Yugoslavia, Landmark Cases, www.icty.org/en/features/crimes-sexual-violence/landmark-cases, which notes: ‘The Trial Chamber found how after taking over the area of Prijedor, in northwestern of BiH [Bosnia-Herzegovina], Serb forces confined thousands of Muslims and Croats in camps. In a horrific incident in the Omarska Camp, one of the detainees was forced by uniformed men, including Duško Tadić, to bite off the testicles of another detainee. In May 1997 the Trial Chamber found Tadić guilty of cruel treatment (violation of the laws and customs of war) and inhumane acts (crime against humanity) for the part he played in this and other incidents.’ 23 Ronit Lentin, ed., Gender and Catastrophe (London: Zed Books, 1997); Susie Jacobs, Ruth Jacobson and Jennifer Marchbank, eds., States of Conflict: Gender, Violence and Resistance (London: Zed Books, 2000). 24 See Sarah Horowitz, ‘Gender, genocide, and Jewish memory’, Prooftexts 20 (Winter/ Spring 2000), 158–90; Zoe Waxman, Women in the Holocaust: A Feminist History (Oxford University Press, 2017); Sarah Helm, If This Is a Woman: Inside Ravensbruck: Hitler’s Concentration Camp for Women (London: Little, Brown, 2015); Lower, Hitler’s Furies.

116

Genocide and Gender

limited, but have served to establish gender as a viable and strengthening subfield. They range from seminal journal articles to case-specific monographs like Patricia Daley’s Gender & Genocide in Burundi, and to edited collections emphasising the comparative and global dimensions.25 The focus remains heavily on women and girls, especially on their vulnerability to sexual violence and genocidal rape. But perhaps because they are relatively recent, these works display some of the diversity and inclusivity that increasingly characterises Gender Studies as a whole, including the feminist literature on conflict and mass violence. Key figures include R. Charli Carpenter, who undertook clear-eyed investigations of gendered humanitarian framings, and Elisa von Joeden-Forgey, whose analysis of ‘life force atrocities’, targeting gendered bodies and identities, is one of the most insightful contributions of the 2010s.26 Von Joeden-Forgey has also developed a set of ‘Patterns of Genocide’ as a prevention tool that is notable for incorporating the gender variable in a nuanced and inclusive way.27 Adam Jones’ work on the genderselective mass killing, torture and incarceration of males in the former Yugoslavia dates to the mid-1990s. It was expanded to a study of ‘gendercide’, including its anti-male variant, in the 2000s.28 The targeting of LGBTQ populations has also begun to be explored, a subject to which we turn next.

LGBTQ Vulnerabilities The cultural roles and expectations associated with ‘gender’ are prominent in explaining the vulnerabilities of LGBTQ populations worldwide to violent victimisation. In an important recent article, Matthew Waites contended that the whole ‘discourse of genocide has been shaped by patriarchal power and heterosexuality’, for example in its foundational concept of the genos, which ‘shows a tendency to biologizing associations and racialization’, emphasising in heteronormative fashion the biological reproducibility of groups as such.29 25 Patricia Daley, Gender & Genocide in Burundi: The Search for Spaces of Peace in the Great Lakes Region (Oxford: James Currey, 2008). 26 R. Charli Carpenter, ‘Innocent Women and Children’: Gender, Norms and the Protection of Civilians (London: Routledge, 2016); Elisa von Joeden-Forgey, ‘The devil in the details: “life force atrocities” and the assault on the family in times of conflict’, Genocide Studies and Prevention 5:1–2 (2010), https://scholarcommons.usf.edu/gsp/vol5/iss1/2. 27 Elisa von Joeden-Forgey, ‘Patterns of genocide: a critical tool for genocide prevention’, 6 December 2019, https://genderinggenocide.files.wordpress.com/2019/12/patternsof-genocide-2019-1.pdf. 28 See Adam Jones, ed., Gendercide and Genocide (Nashville: Vanderbilt University Press, 2004). 29 Matthew Waites, ‘Genocide and global queer politics’, Journal of Genocide Research 20:1 (2017), 48.

117

adam jones and wendy lower

In many countries, cultures and historical periods – extending back to the ‘kill them all’ injunctions of the biblical Book of Leviticus,30 and manifested in contemporary death-squad activities in Iraq, Brazil, Colombia and elsewhere – LGBTQ individuals have been subjected to persecution, torture and targeted killing. Especially vulnerable are gay men and trans women. Overwhelmingly, the perpetrators are cisgender men, seemingly fuelled by gendered anxieties and hatreds similar to those discussed earlier. A contempt for gay men’s ‘failure’ or ‘refusal’ to play their assigned progenitor role for the Volk underpinned the Nazis’ incarceration in concentration camps of tens of thousands of gay males (lesbian women were not systematically targeted) between 1933 and 1945. Driven by the homophobic zeal of Heinrich Himmler, ‘whose loathing of homosexuals knew no bounds’,31 some 5,000 to 15,000 incarcerated gay men were ‘quickly and systematically exterminated by the SS’.32 It may still be the case that the Nazi campaign against gay men represents the only reliably documented instance of the genocidal targeting of LGBTQ populations. More recent death-squad killings of gay men and trans women in Iraq are at least proto-genocidal.33 Similarly, legal proscriptions against homosexuals, most notoriously in Uganda, fit well with the first four of Gregory Stanton’s Ten Stages of Genocide: classification, symbolisation, discrimination and dehumanisation.34 From a Genocide Studies perspective, we are hampered by the absence of gender and sexuality among the core identities explicitly protected by the 1948 Genocide Convention, though the UN Whitaker Report of 1985 did 30 Leviticus 20:13: ‘If a man lies with a male as with a woman, both of them have committed an abomination; they shall be put to death, their blood is upon them’ (New Revised Standard Version). 31 Richard Plant, The Pink Triangle: The Nazi War against Homosexuals (New York: Owl Books, 1988), p. 62. Plant’s book is the best-known work on the subject. See also Jack Nusan Porter, Sexual Politics in Nazi Germany: The Persecution of the Homosexuals during the Holocaust (Newton, MA: Spencer Press, 1991); Geoffrey J. Giles, ‘The denial of homosexuality: same-sex incidents in Himmler’s SS and police’, Journal of the History of Sexuality 11:1–2 (2002), 256–90; and Dagmar Herzog, ed., Sexuality and German Fascism (New York: Berghahn, 2004). 32 Saul Friedländer, Nazi Germany and the Jews, volume 1: The Years of Persecution, 1933–1939 (New York: HarperCollins, 1997), p. 206. 33 See the powerful BBC documentary Gay Witch Hunt in Iraq (2012), YouTube, https:// youtu.be/1vvMCXI1Z14 (accessed 5 April 2022). 34 See the discussion and documentation in Adam Jones, ‘Gendercide and vigilantism against gay and trans people’ in Genocide: A Comprehensive Introduction, 3rd ed. (London: Routledge, 2017), pp. 637–9 and the sources cited on pp. 655–6; Gregory H. Stanton, ‘The ten stages of genocide’, Genocide Watch, 2016, http://genocidewatch.net/geno cide-2/8-stages-of-genocide.

118

Genocide and Gender

propose protecting ‘sexual’ groups (‘such as women, men, or homosexuals’) in a revised convention. The phrasing was notable for including both males and LGBTQ people, and could have represented an important advance in international legal thinking about genocide. Unfortunately, the revisions proposed in the Whitaker Report were never taken up by the United Nations.35

Gender and Structural/Institutional Violence The Norwegian peace researcher Johan Galtung famously distinguished ‘between violence that hits human beings as a direct result of . . . actions of others, and violence that hits them indirectly because repressive structures . . . are upheld by the summated and concerted action of human beings’.36 The latter, for Galtung, constituted ‘structural violence’. The concept seems highly relevant to the study of gender and genocide, in part by serving as an important corrective to political-military framings that downplay the mortality caused by institutionalised, grassroots and less ‘event’-specific forms of mass violence. Given that the killing inflicted by gendered structural violence likely surpasses even that of genocide and war (as traditionally understood), it surely deserves serious consideration and concern. The structural dimension seems especially relevant to an understanding of women’s vulnerabilities and liabilities to violence. Institutions such as female infanticide, domestic violence and the targeted denial of nutrition and health care have exacted a gigantic toll on female lives throughout history. They continue to do so today, with the victims of maternal mortality annually alone numbering in the hundreds of thousands – though this figure has been dramatically reduced in the twenty-first century, suggesting how amenable the crisis is to intervention, if the political and collective will is present. The problem, of course, is that most authorities are still reluctant to extend their understanding of genocide to forms of gendercidal violence against females that challenge the top-down, plan-oriented, temporally restricted framing of the crime. Female infanticide, for example, is spawned by amorphous but pervasive cultural beliefs and expectations that devalue female offspring and render female suffering marginal or invisible. The death toll 35 See Benjamin Whitaker et al., Revised and Updated Report on the Question of the Prevention and Punishment of the Crime of Genocide, E/CN.4/Sub.2/1985/6, 2 July 1985, quoted passage at www.preventgenocide.org/prevent/UNdocs/whitaker/section6.htm. 36 Johan Galtung, ‘Violence, peace, and peace research’, in Peace: Research – Education – Action (Copenhagen: Christian Ejlers, 1975), p. 111.

119

adam jones and wendy lower

is heavily concentrated among the poor (with ultrasound and sex-selective abortion serving a similar function for the wealthy); governments and leaders are divorced from the decision chain, and may indeed be seeking to suppress the phenomenon; and women/mothers themselves are usually the designated agents of the killing. None of this fits easily with conventional or legal understandings of genocide, but on grounds of moral and humanitarian concern, it arguably merits inclusion. The difficulty with these forms of structural violence is locating the legally required ‘specific intent’ to destroy a group ‘as such’ on the part of a perpetrator. Gendercidal forms of structural violence against males are more closely linked to state-level agents and top-down decisions, proof of which is often closer to demonstrating specific intent. Perhaps the most destructive of all human institutions is corvée/forced labour, which as a global-historical and modern institution is heavily gendered male, at least in its most dangerous and onerous forms (e.g. quarry labour from ancient times to the Nazis’ Mauthausen camp; mining in Spanish America and the Soviet Gulag system; rubber tapping in the forests of the Belgian Congo), and has probably killed more men than political-military genocides. Quasi-enslavement by military conscription, and deployment as ‘cannon fodder’ in wars, deserves mention, as do incarceration and the death penalty. Neither conscription nor incarceration is inherently gendercidal, in the way that female infanticide and maternal mortality are. But at many times and in many places, they have served to orchestrate the mass destruction of predominantly male lives. One might note also the institution of blood feuds and associated forms of gang-based killing, which more closely mirror the local/grassroots character of most gendercidal violence against females. However, the limited scale and episodic character of most such killing could militate against a genocide framing.

Gender, Prevention and Intervention In the realms of intervention and prevention, does gender matter as a response and policy consideration? Upholding human rights is often feminised in international relations as a ‘softer’ side of diplomacy contrasted to the masculinised priorities of national security and war-making. Human rights and relief work, like charity, appear among the ‘caring’ professions. Furthermore, the history of rescue in the Holocaust and other genocides shows that shelter and aid often occur in private settings involving families and children where women are present and become involved, whether for altruistic reasons or to exploit the vulnerability of victims who are fleeing 120

Genocide and Gender

violence. To our detriment, rescue will be minimised as an open policy of national intervention as long as it is relegated to isolated acts of individuals willing to take risks, and done ‘behind the scenes’, often secretly in the private realm, carried out by husbands and wives such as the missionaries Carl and Teresa Wilkens in Rwanda.37 Some female leaders, including former US Ambassador to the UN Samantha Power, have attempted to make prevention and intervention a national and international priority. Female leadership in these areas also exists in the academy as scholars and teachers, and in advocacy work in international professional associations and NGOs, for example the scholarship and activism of Helen Fein and Alison Des Forges. Women who escaped Nazi Germany or survived the camps, such as Hannah Arendt and Simone Weil, prominently shaped our understanding of genocide and its redress, and some victims of recent cases have broken taboos by shining a light on the sexual violence and sexism within genocide, such as Nobel Peace Prize winner Nadia Murad. In many ways Murad’s story links with von Joeden-Forgey’s scholarship by establishing the significance of sociocultural contexts. Before the open, systemic acts of killing occur, and then simultaneously with them, ‘extremely violent societies’ develop where gender apartheid exists and women and girls are subjected to daily degradation, sexual slavery, rape, public beatings and executions.38 Recent work on the plight of women subjected to sexualised torture during the Greek military dictatorship (1967–74), the Guatemalan genocide, the Indonesian genocide (1965–6), the al-Anfal campaign in Iraq (1987–91), the Kenyan elections of 2007–8 and the Congo (1995 to the present), among other cases, reveals the central role of the state in using sexualised violence as a formal strategy for waging war and genocide.39 Attention to the gendered vulnerabilities of men and boys40 also shows considerable promise as a route to more effective prevention and intervention. Campaigns of mass repression against men of ‘battle age’ is an especially reliable ‘canary in the coal mine’, indicating that out-groups are being selectively targeted for persecution and crimes against humanity (such as 37 Carl Wilkens, I’m Not Leaving (Spokane, WA: World Outside My Shoes, 2011); Sara E. Brown, Gender and Genocide in Rwanda: Women as Rescuers and Perpetrators (London: Routledge, 2017). 38 Christian Gerlach, Extremely Violent Societies: Mass Violence in the Twentieth-Century World (Cambridge University Press, 2010). 39 Victoria Sanford, Katerina Stefatos and Cecilia M. Salvi, eds., Gender Violence in Peace and War: States of Complicity (New Brunswick: Rutgers University Press, 2016). 40 See R. Charli Carpenter, ‘Recognizing gender-based violence against civilian men and boys in conflict situations’, Security Dialogue 37:1 (2006), 83–103.

121

adam jones and wendy lower

torture, forcible disappearance and arbitrary imprisonment). Genocide may well be impending. The direct slaughter of males, as we have seen, serves as something of a tripwire in genocidal processes, either bounding their exterminatory dimension or presaging a full-scale ‘root and branch’ genocide. In any case, the emasculation of the targeted group facilitates a host of other atrocities, including gender-selective violence against women and girls. This chapter has also pointed to the role of economic crisis in ‘priming’ a society for genocide, and the particularly volatile and explosive crisis of identity and masculinity that it triggers among men, especially younger adult males. To the extent that outside agents exacerbate such crises (for example, through imposed ‘austerity’ measures and wild swings in resource and export prices on international markets), far greater caution, support and sensitivity are called for. Structural and institutional mechanisms of gender-selective violence that disproportionately target males also merit notice for their genocidal potential and, sometimes, impact. They include military conscription, corvée labour, lynching/vigilantism, the death penalty, torture, forced disappearance and inhumane or arbitrary forms of imprisonment. Another warning sign is the prevalence of domestic violence, marital rape and the murder of women, as well as regimes that introduce laws that control women’s reproductive rights and bodies (forced abortions/infanticide, marriages and sterilisation). It should be stressed also that when the perpetrator regime is no longer in power, its ideas and practices can perpetuate the violence of rape and sexual enslavement. In 2013, a United Nations report surveying 10,000 men in Asia and the Pacific determined that rape was committed in the region at high rates and with impunity. Sexual violence continued long after the genocide in Cambodia, where more than 20 per cent of men confessed to the crime of rape, the occurrence of gang rapes was higher than in neighbouring countries, and close to half of the men who admitted to raping women and girls had not been prosecuted.41 In order to break the cycle of violence in the aftermath of a genocide, scholar James Waller stresses, upstream prevention must include policies to promote gender equality and rights. Securing women’s physical security and livelihoods can go a long way towards achieving stability and peace, since ‘there is a strong correlation between levels of gender-based violence and conflict’.42 41 United Nations Development Programme, Asia-Pacific, ‘Why do some men use violence against women and how can we prevent it?’, 10 September 2013, https://tin yurl.com/2c9p8hy4 (accessed 5 April 2022). 42 James Waller, Confronting Evil: Engaging Our Responsibility to Prevent Genocide (Oxford University Press, 2016), p. 189.

122

Genocide and Gender

Overall, genocide prevention – and Genocide Studies – benefit from a nuanced and inclusive approach to studying the operations of gender in mass violence and conflict. A growing awareness of the distinctive vulnerabilities across the gender spectrum has both sharpened our understanding of the range of gendered experiences in genocide and provided important tools for intervention. These advances in knowledge have come from the victims who chose not to remain silent, breaking taboos against speaking about rape and speaking openly and publicly to prosecutors, judges and journalists.43 As academics and national/international authorities demonstrate increased interest in the subject, a more sophisticated understanding of gendered strategies and vulnerabilities should evolve. It is to be hoped, and perhaps expected, that more wide-ranging and effective strategies of countering genocidal violence will result.

Conclusion In this chapter we have presented a critical evaluation of the role of the gender variable in genocide, and of research into gender in the field of comparative Genocide Studies and beyond. We have sought to destabilise easy assumptions about gendered identities and experiences, and to examine how these shape the strategies and temporal stages of victimisation, perpetration and collaboration in genocidal events and processes. We have also attempted to move beyond the event-specific and temporally bounded framings of genocide that predominate in the Genocide Studies field and public/civic discourse. Attention to structural forms of violence, including genocidal and ‘gendercidal’ ones, seems particularly relevant when exploring the operations of the gender variable. It is important to stress, in closing, that we do not advance gender as a ‘magic key’ for understanding and unpacking genocidal processes. The recent ‘intersectional’ turn in the social sciences and humanities emphasises ‘how identities are related to each other . . . and how the social structures of race, class, gender, sexuality, age, and ability intersect for 43 Nicola Henry, ‘Witness to rape: the limits and potential of international war crimes trials for victims of wartime sexual violence’, International Journal of Transitional Justice 3:1 (2009), 114–34. The Cambodia Tribunal allows for greater direct participation of the victims, which has advanced knowledge there of the Khmer Rouge’s campaigns of marital and gang rape by state authorities, and brought to the screen. See Raya Morag’s ‘Gendered genocide: new Cambodian cinema and the case of forced marriage and rape’, Camera Obscura 35:1 (2020), 77–107.

123

adam jones and wendy lower

everyone’. Such ‘an intersectional approach develops a more sophisticated understanding of the world and how individuals in differently situated social groups experience differential access to both material and symbolic resources’.44 In the present context, this suggests that gender should be understood as overlapping and intersecting with a range of other variables and identities, whether self-asserted or imputed. For example, it is difficult to arrive at an adequate understanding of the systematic mass mortality inflicted upon women and girls without recognising how (feminine) gender overlaps with social class to mark out females as marginal or ‘disposable’, especially in conditions of bare subsistence and scarcity. A similar argument can be made with regard to males, who may to some extent be victimized ‘as such’ (that is, on the basis of misandrist assumptions paralleling misogynist ones) but who are also disproportionately targeted in instances of ‘eliticide’, since under patriarchy men generally occupy roles of social and symbolic power and authority. Variables of race and ethnicity, themselves typically intersecting with social class, crucially shape how gender identities are imposed, interpreted and experienced. We have seen how prevailing models and hierarchies of sexual identity can produce atrocious outcomes, including genocidal ones, against LGBTQ populations. The interaction of gender with an age variable has also been noted. In many contexts, infant girls are especially vulnerable to institutions of infanticide, a phenomenon that also reflects social class (as elites with access to ultrasound technologies often opt for sex-selective abortion). Likewise, relative youth marks out certain women and girls as special targets of sexual assault and exploitation, while men and boys of an imputed ‘battle age’ are targeted for the military threat that they are held to represent. We call, accordingly, for an inclusive and multifaceted approach to narratives of gender and genocide – one that incorporates diverse sexes, genders and sexualities, and maps gendered activities and experiences onto a broad spectrum of genocidal roles, identities and experiences. Such an approach, we hope, will bolster the utility of gender investigations in both understanding genocides of the past and present and when intervening to suppress and forestall genocides in future. 44 ‘Intersectionality’, in Miliann Kang, Donovan Lessard, Laura Heston and Sonny Nordmarken, Introduction to Women, Gender, Sexuality Studies, Open Library, 2017 (unpaginated), https://openbooks.library.umass.edu/introwgss.

124

Genocide and Gender

Bibliographic Note For concise overviews of gender and genocide, see the literature review in Adam Jones, ‘Gender and genocide’ in The Historiography of Genocide, ed. Dan Stone (London: Palgrave Macmillan, 2008), pp. 228–52; Joeden-Forgey, ‘Gender and genocide’; and Sareta Ashraph, ‘Acts of annihilation: the role of gender in the commission of the crime of genocide’, Confluences Méditerranée 103 (2017), 15–29. Major works include Amy E. Randall, ed., Genocide and Gender in the Twentieth Century: A Comparative Survey, 2nd ed. (London: Bloomsbury Academic, 2020); Joshua Goldstein, War and Gender: How Gender Shapes the War System and Vice Versa (Cambridge University Press, 2001); Jean Bethke Elshtain, Women and War (New York: Basic Books, 1988); and Ehrenreich, Blood Rites. Plural ‘masculinities’, from violent to victimised, were first theorised in 1993 by R. W. Connell in Masculinities (2nd ed.; Berkeley: University of California Press, 2005). The alleged roots of male/masculine violence in the primate realm are examined in Dale Peterson and Richard Wrangham, Demonic Males: Apes and the Origins of Human Violence (Boston: Mariner, 2003). Seminal studies of male perpetrators of genocide and other mass violence are Klaus Theweleit’s psychoanalytical investigation of the German Freikorps, entitled Male Fantasies, volume I, Women, Floods, Bodies, History, translated by Stephen Conway (Minneapolis: University of Minnesota Press, 1987); Christopher R. Browning, Ordinary Men: Reserve Police Battalion 101 and the Final Solution in Poland (New York: HarperCollins, 1992); and Waitman Wade Beorn, Marching into Darkness: The Wehrmacht and the Holocaust in Belarus (Cambridge, MA: Harvard University Press, 2014). Jean Hatzfeld’s Machete Season: The Killers in Rwanda Speak, translated by Linda Coverdale (New York: Farrar, Straus & Giroux, 2005), draws extensively on the recollections of Hutu male perpetrators of the genocide of Tutsis in 1994. See also Scott Straus’ The Order of Genocide: Race, Power, and War in Rwanda (Ithaca, NY: Cornell University Press, 2006), sampling testimonies by over two hundred jailed perpetrators. For studies of women as agents and accomplices of mass atrocity, see Lower, Hitler’s Furies; Hilary Matfess, Women and the War on Boko Haram: Wives, Weapons, Witnesses (London: Zed Books, 2017); Azadeh Moaveni, Guest House for Young Widows: Among the Women of ISIS (New York: Random House, 2019); and Laura Sjoberg, Women as Wartime Rapists: Beyond Sensation and Stereotyping (New York University Press, 2016).

125

adam jones and wendy lower

A pivotal human rights report is African Rights, Rwanda: Not So Innocent – When Women Become Killers (London: African Rights, 1995). Core treatments of sexual violence in war and genocide include Carol Rittner and John K. Roth, eds., Rape: Weapon of War and Genocide (St Paul: Paragon House, 2012); Sonja M. Hedgepeth and Rochelle G. Saidel, eds, Sexual Violence against Jewish Women during the Holocaust (Waltham, MA: Brandeis University Press, 2010); Alexandra Stiglmayer, ed., Mass Rape: The War against Women in Bosnia-Herzegovina (Lincoln: University of Nebraska Press, 1995); Human Rights Watch, Shattered Lives: Sexual Violence during the Rwanda Genocide and Its Aftermath (New York: Human Rights Watch, 1996); and Sara Meger, Rape Loot Pillage: The Political Economy of Sexual Violence in Armed Conflict (Oxford University Press, 2016). On sexual violence against men and boys, see Amalendu Misra, The Landscape of Silence: Sexual Violence against Men in War (London: Hurst, 2015), and Marysia Zalewski et al., eds., Sexual Violence against Men in Global Politics (London: Routledge, 2018). ‘Life force atrocities’, with their gendered essence, are examined in von Joeden-Forgey, ‘Devil in the details’. The term ‘gendercide’ was coined by the American scholar Mary Anne Warren in her book Gendercide: The Implications of Sex Selection (Totowa: Rowman & Allanheld, 1985). As the title suggests, Warren focused heavily on women and reproduction, but her theoretical framing in the opening chapter is gender inclusive, and essential. See also Marie Vlachová and Lea Biason, eds., Women in an Insecure World: Violence against Women: Facts, Figures and Analysis (Geneva: DCAF, 2005); chapter 1 addresses ‘Gendercidal institutions against women and girls’, while strategies of humanitarian intervention are well explored throughout the volume. On genocide/gendercide and crimes against humanity against men and boys, see Carpenter, ‘Recognizing genderbased violence’, and Jones, ed., Gendercide and Genocide, which also includes Stefanie Rixecker’s seminal essay ‘Genetic engineering and queer biotechnology’. Matthew Waites explores ‘Genocide and global queer politics’ in the Journal of Genocide Research 20:1 (2018), 44–67. The standard source on the Nazis’ persecution and extermination of gay men is Plant, Pink Triangle. Intersections of gender and race are studied in Tommy J. Curry, The ManNot: Race, Class, Gender, and the Dilemmas of Black Manhood (Philadelphia: Temple University Press, 2017). On gender, genocide prevention and humanitarian intervention, see Carpenter, ‘Innocent Women and Children’, and Mary Michele Connellan and Christiane Frölich, eds., A Gendered Lens for Genocide Prevention (London: Palgrave Macmillan, 2018). 126

4

Genocide, Starvation and Famine bridget conley and alex de waal

Starvation and Genocide: An Entangled History Hunger, Primo Levi wrote in If This Is a Man, does not describe the Auschwitz experience, rather ‘the Lager is hunger: we ourselves are hunger, living hunger’. Bread was everything: meagre nourishment, the material of prisoners’ dreams, currency, and ‘everything that the law of the place allows us to possess’.1 In the camp, all that did not belong to death, belonged to hunger, exposure to cold and damp, disease and exhaustion. These are the characteristic traits of mass starvation, which are not limited to lack of food, but include the intentional deprivation of objects essential to sustain life. By different means, mass starvation meant death, just as Auschwitz did. It would be difficult to find a case of genocide or mass atrocities in which starvation – systematic denial of access to the basic means for sustaining life – did not figure as a major instrument of killing. Throughout history, starvation has been a central experience of victims and a tool for perpetrators – sometimes casual and incidental, often carefully honed. The examples below are emblematic. The ancient Roman destruction of Carthage began with genocidal intent succinctly captured in the oft-repeated demand by Roman senator Marcus Porcius Cato, that ‘Carthage must be destroyed!’ The attack included not only killing, destroying the city and enslaving the occupants, culminating, according to myth, in sowing salt into the earth to render the area permanently uninhabitable. This dramatic story became a signature detail in modern narratives of the effort to erase Carthage from the map. However, it is apocryphal, of relatively recent vintage.2 What historians and chroniclers of Carthage noted at the time was the more pervasive use of deprivation as 1 P. Levi, If This Is a Man / The Truce (London: Abacus, 1987), pp. 80, 69. 2 R. T. Ridley, ‘To be taken with a pinch of salt: the destruction of Carthage’, Classical Philology 81 (1986), 140–6.

127

bridget conley and alex de waal

a means to destroy an entire people: a siege without distinction between military and civilian suffering. In 149 B C E, Roman forces surrounded the city and blocked all supplies with a rigour that increased over nearly three years, as a means to force capitulation. As historian Richard Miles writes, many of the city’s inhabitants began to starve in the last desperate months: During the last difficult years of the city, the only waste that seems to have been regularly removed was the corpses of the many who died as starvation and disease took hold. Now in the last terrible months of the city’s existence, in contrast to the care that had traditionally been taken of the dead, the corpses of both rich and poor were unceremoniously dumped in a number of mass graves just a short distance away from where they had lived.3

Finally, when the walls were breached, Romans stormed the city-state, murdering by sword or flame those who had survived the siege. Out of a population of an estimated 200,000 to 400,000, perhaps 145,000 perished, and the survivors were sold into slavery.4 Salt was hardly necessary. Conquistadors and colonial settlers deployed starvation against indigenous populations over the centuries; in some instances to such extremes as to provide an unequivocal case for genocide. Bartolomé de Las Casas’ description of Spanish colonisation in Hispaniola (present-day Haiti and Dominican Republic) provides exemplary detail of a process through which the Taino indigenous population was worked to death or killed. After describing brutal killings, Las Casas described the final deprivations imposed on the population: [T]hey sent the Males to the Mines to dig and bring away the Gold, which is an intollerable [sic] labor; but the Women they made use of to Manure and Till the ground, which is a toil most irksome even to Men of the strongest and most robust constitutions, allowing them no other food but Herbage, and such kind of unsubstantial nutriment, so that the Nursing Womens Milk was exsiccated [sic] and so dryed up, that the young Infants lately brought forth, all perished, and females being separated from and debarred cohabitation with Men, there was no Prolification or raising up issue among them. The Men died in Mines, hunger starved and oppressed with labor, and the Women perished in the Fields, harrassed and broken with the like Evils and Calamities: Thus an infinite number of Inhabitants that formerly peopled this

3 R. Miles, Carthage Must Be Destroyed: The Rise and Fall of an Ancient Civilization (London: Allen Lane, 2010). 4 B. Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007), p. 51.

128

Genocide, Starvation and Famine

Island were exterminated and dwindled away to nothing by such Consumptions.5

The Taino population, estimated at several hundred thousand to possibly even 1 million before 1492, fell to 32,000 by 1514.6 Mass starvation was a tool of conquest and subjugation in every continent and at every stage of imperialism. As European empires reached their zenith at the turn of the twentieth century, mass starvation was a favoured tool of genocide. In two key cases, that of the Herero and Nama in German Southwest Africa, and the Armenian genocide, perpetrators were aided by proximity to a desert. Even after largely defeating a Herero uprising against the European invaders, the newly appointed commander of the German colonial territories,7 Lieutenant General Lothar von Trotha, decided not to permit surrender. He ordered his forces to drive the Herero into the Kalahari desert and seal off access to water, sentencing them to almost certain death. People who tried to flee out of the desert were shot, beaten, hanged, starved and raped. An estimated 40,000 Herero and 10,000 Nama died.8 In 1915–16, Ottoman Turkish forces attacked thousands of Armenian villages, massacring, torturing and assaulting the inhabitants. Those who survived the initial onslaught, many of them women and children, were brutally deported in a forced march southeast into the arid lands of today’s Syria and Iraq. There, they were again attacked violently, but most deaths were due to hunger and thirst.9 Mortality estimates do not distinguish cause of death, but indicate that, in total, over 600,000 Armenians died in that way.10 Mass starvation in other twentieth-century genocides will be examined below. Prominent but controversial are cases where Communist policies created famines across the twentieth century. Soviet collectivisation, the socalled Great Leap Forward in China and Cambodia under the Khmer Rouge are the leading contemporary examples of forcible socio-economic 5 B. de Las Casas, A Brief Account of the Destruction of the Indies (1542), Project Gutenberg, www-personal.umich.edu/~twod/latam-s2010/read/las_casasb2032120321-8.pdf, p. 8. 6 W. F. Keegan, ‘Destruction of the Taino’, Archaeology 45 (January/February 1992), 51–6. 7 D. Schaller, ‘From conquest to genocide: colonial rule in German Southwest Africa and German East Africa’ in Empire, Colony, Genocide: Conquest, Occupation and Subaltern Resistance in World History, ed. A. D. Moses (New York: Berghahn, 2010), pp. 296–324, at p. 303. 8 A. de Waal, Mass Starvation: The History and Future of Famine (Cambridge: Polity Press, 2017), p. 73. 9 R. Suny, ‘They Can Live in the Desert but Nowhere Else’: A History of the Armenian Genocide (Princeton University Press, 2015), p. 309. 10 Ibid., pp. 354–5.

129

bridget conley and alex de waal

transformation that produced mass starvation. These examples, analysed alongside colonial restructuring of economies, sharpen a key challenge for studying famine and genocide together: intent. Too frequently, the discussion of intentions behind famine has produced polarised views: either famine was an unintended calamity at the intersection of nature and bad economic policy, or it resulted from genocidal intent. Countering this polarity is a recognition that famine, like genocide, unfolds as a process. A recent, near exception to the rule that genocides kill by starvation as well as direct violence, deserves special mention. In Rwanda in 1994, mass killing was carried out with devastating speed using firearms, machetes and fire. The 100 days of organised mass murder was not long enough for famine to develop before the regime fell. Nonetheless, severe hunger often forced Tutsi out of hiding places and into killers’ hands where they were shot or hacked to death. After the genocidal regime was militarily defeated by the Rwandese Patriotic Front (RPF), the government, army and militia fled in a massive exodus with civilians into neighbouring Zaire. There, deprivation again became a tool for the resurgent genocidaires to control the refugee population. As Human Rights Watch documented, The authorities responsible for the genocide rapidly re-established their rule over the refugees, using control over the humanitarian supplies of food, water, and medicine to force compliance with their orders. By the end of the year, the guilty authorities were proclaiming their intention to return to Rwanda ‘to finish the work’ of killing Tutsi, and their army and militia, nourished for months by the international community, were preparing for incursions into Rwanda.11

In one of history’s most brutal tales of revenge for a genocide, the RPF invaded to end the use of refugee camps as safe havens for genocidaires, and then chased soldiers and civilians alike through the Zairean forest, committing what a UN report described as a second ‘genocide’ in 1997.12 The desert reprised its genocidal role in late twentieth- and early 21stcentury atrocity campaigns in Sudan, when military officers drove civilians into uninhabitable areas to perish of hunger and thirst. In all of these cases, 11 Human Rights Watch, World Report 1995, www.hrw.org/reports/1995/WR95/AFRICA-08 .htm (accessed 12 December 2019). 12 United Nations Office of the High Commissioner for Human Rights, Democratic Republic of the Congo, 1993–2003 Report of the Mapping Exercise documenting the most serious violations of human rights and international humanitarian law committed within the territory of the Democratic Republic of the Congo between March 1993 and June 2003, Geneva, August 2010, para. 31.

130

Genocide, Starvation and Famine

the perpetrators knew full well what devastation would await civilian populations bereft of the means to survive in such harsh conditions. The Sudanese government had for long conducted counter-insurgency ‘on the cheap’ by utilising the services of tribal militia whose members paid themselves in kind by looting and burning, and seizing land.13 This is a kind of warfare guaranteed to cause massive deprivation, and the military and political leaders in Khartoum were well aware of this when they unleashed their firestorm on Darfur. One of the ironies of such delegated counter-insurgency is that a community’s neighbours are well informed about where to locate and steal key resources, and thus are more comprehensive and devastating in their looting than army units originating from elsewhere in the country.14 The height of the Darfur war was characterised by massacres of civilians in which an estimated 30,000 people were killed, followed with relentless inevitability by a humanitarian crisis in which perhaps 200,000 more people died of hunger, disease and exposure. As much as a third of the region’s population were driven from their homes to vast displaced camps on the outskirts of major towns, or became refugees. The death toll would have been much higher were it not for the vast international aid effort which, remarkably rapidly, brought down malnutrition and mortality rates among the displaced and refugees. There is no doubt that, at the height of the military campaigns in 2003–4, government forces committed starvation crimes. However, the subsequent claims by the prosecutor of the International Criminal Court, that there was an ongoing genocidal programme targeting the people in displaced camps, is not supported by any evidence.15 There has been no systematic study of how the fields of Genocide Studies and Famine Studies might speak to each other. Genocide, along with ‘mass atrocities’, has been primarily used in reference to intentional killing of entire groups. Famine describes an outcome of people lacking sustenance, to such a degree that it elevates mortality rates, whether through disease or actual starvation. As the above short history of the uses of starvation in genocide makes clear, both through the cases included and those excluded, there are

13 J. Flint and A. de Waal, Darfur: A New History of a Long War (London: Zed, 2008). 14 L. B. Deng, ‘Social capital and civil war: the Dinka communities in Sudan’s civil war’, African Affairs 109 (2011), 231–50. 15 A. de Waal, ‘Reflections on the difficulties of defining Darfur’s crisis as genocide’, Harvard Journal of Human Rights 20 (Spring 2007), 25–33; R. Brauman, ‘The ICC’s Bashir indictment: law against peace’, World Politics Review (online), 23 July 2008, www .worldpoliticsreview.com/articles/2471/the-iccs-bashir-indictment-law-against-peace.

131

bridget conley and alex de waal

a complex set of reasons why famine and genocide have been largely studied in separate silos. One reason is the development of the legal prohibitions against genocide and, separately, against starvation. In theory, international humanitarian law and the laws of war provide ample space for thinking about the two together. But in practice, political interests and prosecutorial strategies have crafted delineations between them. Famine is not a legal term; its conceptual trajectory has followed a different arc. A second reason is the ambivalence about attributing criminal intent to famine: too often hunger is seen as the inevitable corollary of disaster, natural or otherwise. This has notably been the case for both imperial-era famines and those inflicted by Communist regimes’ titanic social engineering. Finally, and perhaps most notable in our short description of Darfur, is the perceived divergent imperatives around endings. Humanitarian aid is the vehicle for saving the lives of those at risk of dying from starvation, but, and for good reason, it has been criticised as a mere fig leaf in contexts of genocide. The challenge is to discern a common logic underpinning killing and starving, and to craft response mechanisms that halt both. If we shift our focus to the experiences of victims and perpetrators then the salience of starvation as a weapon comes into focus. From this vantage point, it becomes clear that mass starvation has provided an insidious tool of atrocity – and at times, genocide – with profound consequences for victim groups. There is much to be learned by studying mass atrocities and famine at their points of overlap. We therefore use the term ‘starvation’ to denote deprivations conducted with intent, and ‘mass starvation’ to capture when this occurs at scale. As this chapter will show, a focus on the deliberate destruction or withholding of objects indispensable to the survival of the civilian population allows us to see more precisely the intent and outcome of the crimes committed.

Crafting Legal Prohibitions: Hungerpolitik and its Denials During World War Two, in both Europe and Asia, hunger was as great a killer as acts of direct violence.16 The Axis powers’ starvation policies – Hungerpolitik – were instruments of war and genocide favoured for their 16 L. Collingham, The Taste of War: World War II and the Battle for Food (London: Penguin, 2013).

132

Genocide, Starvation and Famine

cheapness and efficiency. The most terrible episode was the Nazi Hungerplan that accompanied Operation Barbarossa, which, had it succeeded in meeting its quota of 30 million starvation deaths in the occupied Soviet Union and Poland, would have become the biggest atrocity of the war in terms of its simple human toll. In the event, between 5 and 6 million perished of starvation, from a mixture of genocidal deprivation (withholding food from Ukrainian cities, the use of hunger in the Warsaw ghetto and death camps, and the deprivation of food, shelter and medicine from prisoners of war), the forced extraction of agricultural produce (from occupied Ukraine and southern Russia) and siege (Leningrad).17 In the East Asian theatre, Japan’s war also imposed starvation on its occupied peoples, on prisoners of war and even on its own troops – whom it dispatched to Indonesia and New Guinea with just a few days’ rations.18 The Allied powers’ version of Hungerpolitik involved denial – initially rubbing British and American starvation crimes out of the historical record, and then obfuscating the fundamental inhumanity of Nazi-imposed starvation in the rejuvenated post-war legal regime. In the aftermath of the war, the Allies frustrated even modest efforts to put starvation crimes in their rightful place in the post-war international legal settlement. The Nuremberg tribunal and its immediate sequelae forged many innovations in international criminal law, but on the issue of starvation they chose to be conservative. Accommodating starvation as a weapon of war was particularly striking at the High Command Trial, in which Field Marshal von Leeb was charged, inter alia, with enforcing the siege of Leningrad, in which a million people perished. On the grounds that the Lieber Code, the basic set of rules informing the US army’s conduct, permitted starvation as a means of compelling an enemy to submit, the prosecutors limited the charges to incidents in which German forces fired on civilians trying to escape the besieged city. But the Lieber Code permitted that too, and the judges concluded, ‘We might wish the law were otherwise, but we must administer it as we find it. Consequently, we hold no criminality attaches on this charge.’19 The basic reason for this conservatism was that the Allies had used starvation too: Britain blockaded occupied Europe, contributing to famine in Greece; wartime economic policies created starvation in British-ruled Bengal; and the United States mounted ‘Operation Starvation’ against Japan in 1945. Indeed, the maritime powers of the previous century – 17 Ibid., pp. 180–218. 18 Ibid. 19 United Nations War Crimes Commission, Law Reports of Trials of War Criminals, vol. X I I: The German High Command Trial (London: HMSO, 1949), p. 84.

133

bridget conley and alex de waal

Britain, France and the United States – regarded economic warfare as their preferred mode of conducting hostilities, including reducing their enemies to death by hunger. Even after the 1918 Armistice that brought the fighting of World War One to an end, Britain had sustained its blockade of Germany with the goal of forcing it to submit to the onerous conditions of the Treaty of Paris. Those who had taken the lead in drafting international law were unsurprisingly reluctant to prohibit their favoured instrument, the blockade. They did not explicitly prohibit siege, blockade or starvation; civilian starvation was regarded as an unavoidable collateral effect of using legitimate means of conflict. International law did not prohibit starvation, but rather codified how hunger could be lawfully used, to the advantage of those in possession of the naval capacity to enforce the measure.20 Even the modest criminal framework was limited in its application to starvation. A second opportunity for the high-profile prosecution of acts of starvation arose with the so-called Ministries Trial of 1947–9. Richard Walther Darré, Hitler’s minister of food and agriculture from 1939 to 1942, was a defendant. While he did not himself draw up the Hungerplan, his agricultural and economic policies were exploitative and discriminatory and had paved the way for his deputy and successor, Herbert Backe, to advocate openly for genocidal starvation. (Backe committed suicide before he could be charged.) In describing the crimes committed by Darré and his subordinates, the prosecutors and judges were obliged to face the reality that mass starvation was purposeful, systematic and vast in scale and consequence. But the story that emerged from the Nuremberg convictions was one in which starvation was not a crime in its own right, but merely an instrument for other crimes against humanity. Hunger was treated as an incidental outcome of other criminal acts. In its judgment, the tribunal observed that ‘[s]ome of [Darré’s] ideas were novel and somewhat bizarre, but it is not a crime to evolve and advocate new or even unsound social and economic theories’.21 He was exonerated of discrimination in providing rations to Jews on the grounds that the decrees ‘were not in themselves so severe or their effects so harsh as to cause sickness or exposure to sickness or death’. The judges noted, however, that Darré’s

20 N. Mulder and B. van Dijk, ‘Why did starvation not become the paradigmatic war crime in international law?’ in Contingency and the Course of International Law, ed. K. J. Heller and I. Venzke (Oxford University Press, 2020). 21 Military Tribunal IV, ‘Judgment of the Tribunal (Part 3)’, Trial 11 – Ministries Case. 54 (1949), https://digitalcommons.law.uga.edu/nmt11/54, p. 380.

134

Genocide, Starvation and Famine

rations were the prelude to later and ‘more drastic cuts which finally led to the denial of foodstuffs necessary in life’.22 The charges against Darré also included the war crime and crime against humanity of plunder and spoliation, notably through land expropriation and extraction of food quotas in occupied territories, and forced labour.23 The judges found that the economic demands made on Poland were far in excess of the resources of the country, that it amounted to ‘a program of ruthless spoliation and plunder of food and agricultural products’, and that ‘The evidence of widespread starvation among the Polish people in the Government General indicates the ruthlessness and severity with which the policy of exploitation was carried out.’24 Darré was found guilty. Nonetheless, it is hard to escape the conclusion that the tribunal members sought to avoid focusing on starvation as a criminal act in itself and found ways to interpret the law and evidence in a conservative manner. In the immediately following years, when the International Committee of the Red Cross, supported by smaller countries, sought to introduce prohibitions on starvation into the revised and expanded Geneva Conventions, they faced concerted diplomatic sabotage from the US and UK. As Nicholas Mulder and Boyd van Dijk observe, ‘The weapon of blockade was effectively legalised by Allied judges and treaty-drafters after the Second World War. Ultimately, they did not so much reduce or limit its strength, but rather rendered it legal by creating ever more blockade-approving jurisprudence.’25 The Geneva Conventions of 1949 offered limited protection for humanitarian supplies to civilians in occupied territories,26 but accepted the Allies’ starvation policies as inhumane rather than criminal and genocidal, unlike the German Hungerplan. In foreclosing legal sanction against the former, the Atlantic powers ended up providing cover for the latter. As international law developed over subsequent decades, multiple pathways for addressing starvation slowly emerged under four different but related legal regimes: crimes against humanity, international humanitarian law, the Genocide Convention, and human rights law. Elsewhere, we have 22 Ibid., p. 384. 23 T. Taylor, ‘Final prosecution brief on the criminality of the defendant Richard Walter Darré’, Trial 11 – Ministries Case. 7 (1948), https://digitalcommons.law.uga.edu/nm t11/7. 24 Military Tribunal IV, ‘Judgment of the Tribunal (Part 5)’, Trial 11 – Ministries Case. 7 (1949), https://digitalcommons.law.uga.edu/nmt11/55, p. 603. 25 Mulder and van Dijk, ‘Why did starvation?’, 18. 26 Convention (IV) Relative to the Protection of Civilian Persons in Time of War, Geneva, 12 August 1949.

135

bridget conley and alex de waal

proposed the term ‘starvation crimes’ as a chapeau for starvation-related acts that are prohibited under one or more legal regimes. The term is influenced by the coinage of ‘atrocity crimes’ by David Scheffer, which was intended to circumvent the constraints of focusing on genocide. The legal definition of genocide, found in the 1948 Convention on the Prevention and Punishment of the Crime of Genocide, altered, but did not erase, the place of starvation within Raphael Lemkin’s 1944 book, Axis Rule in Occupied Europe. Lemkin’s sociopolitical analysis had enumerated the ways in which food deprivation and starvation could be used as part of a co-ordinated plan to destroy the ‘essential foundations of the life of national groups’,27 namely racialised discrimination in feeding (specific caloric allotments as part of Nazi occupation guidelines that discriminated against some groups and favoured others), and endangerment of health (i.e. overcrowding in ghettos, withholding medicine and heating fuel during winter). His book contains extremely detailed data on how the Nazi authorities calibrated the calorific provision allocated to different groups. For carbohydrates, Germans were allocated 100 per cent of need, with 76 to 77 per cent for Poles, 38 per cent for Greeks and 27 per cent for Jews.28 This was precision biomedical engineering, at population scale, with genocidal intent. The 1948 Genocide Convention opened the door for considering starvation as a genocidal act. Its Article II(c) prohibits acts, when committed with an intent to destroy a protected group, that ‘deliberately inflict on the group conditions of life calculated to bring about its physical destruction in whole or in part’. In theory, one could also find language to address starvation conditions within Crimes against Humanity, as developed to prosecute crimes committed during the period of World War Two, and later incorporated into the Rome Statute of the International Criminal Court. For instance, if denial of food could be proven to be part of a widespread and systematic plan to ‘persecute’ a civilian population. However, this does not foreground the criminality of imposed starvation, per se. The most explicit legal prohibition against imposing starvation can be found in the 1977 Additional Protocols to the Geneva Conventions (Article 54(1) of Additional Protocol I and Article 14 of Additional Protocol II), which specifically provided a legal definition of starvation and prohibited it. This 27 R. Lemkin, Axis Rule in Occupied Europe: Laws of Occupation, Analysis of Government, Proposal for Redress (Washington, DC: Carnegie Endowment for International Peace, 1944), p. 79. 28 Ibid., pp. 87–8.

136

Genocide, Starvation and Famine

definition was reaffirmed in the 1998 Rome Statute and its 2019 amendment.29 The legal definition displays considerable nuance in defining what constitutes starvation in war and as a crime against humanity. Article 8(2)(b)(xxv) criminalises ‘Intentionally using starvation of civilians as a method of warfare by depriving them of objects indispensable to their survival, including wilfully impeding relief supplies as provided for under the Geneva Conventions’. Many different acts would qualify as ‘starvation’ by this definition. The clearest instances are acts of destroying, removing or rendering useless objects indispensable to the survival (OIS) of a civilian population, such as food stores, water supplies or farms. Other OIS include sanitation, shelter and health facilities. Although not specified in this manner in the relevant law, preventing or impeding activities indispensable to survival can have the same impact. Examples would be impeding maternal care for young children, halting labour migration or preventing people from foraging for wild foods in forests. There is no threshold of scale for the war crime of starvation: it can be perpetrated through deprivation alone, even if not to the level of causing death. In this sense, it resonates with genocide, which likewise does not require a single death, as long as the requisite acts conducted with ‘intent to destroy’ a protected group could be proven. However, unlike genocide, not once have starvation crimes been prosecuted on the international level. Despite these developments within the law, the political concerns of Allied forces in World War Two set the pattern for subsequent divergence of policies regarding imposed famine versus those addressing direct violence against civilians. The pre-war years had amply demonstrated that mass deprivation was politically explosive – the Great Depression had paved the way for Fascism and Communism. Post-war reconstruction began with immense humanitarian efforts to relieve hunger in Europe and East Asia. Punishing the defeated was seen as not only unethical but misguided. The new international financial order, centred on the Bretton Woods Institutions, was constructed with the goal (inter alia) of ensuring that mass unemployment did not again destabilise industrial nations. Among the specialised agencies set up under the United Nations was the Food and Agriculture Organisation, the first director general of which, David Boyd-Orr, saw his task as being one of social justice as much as agricultural science. As Josué de 29 ICC, Assembly of States Parties, Annex I, ‘Draft Resolution on Amendments to Article 8 of the Rome Statute of the International Criminal Court’, 3 December 2019, ICC-ASP/ 18/32, https://asp.icc-cpi.int/iccdocs/asp_docs/ASP18/ICC-ASP-18-32-ENG.pdf.

137

bridget conley and alex de waal

Castro noted in his influential book The Geography of Hunger,30 the return of famine to Europe had also opened an agenda of tackling hunger and its associated sociopolitical maladies across the world. However impressive and commendable these efforts may be, they confirmed the separation between the nutritional and the legal-political framings of hunger, and thereby provided an alibi for the Western powers’ inaction over legal prohibitions of mass starvation. Despite the subsequent efforts of some nutritionists to emphasise the sociopolitical factors behind humanitarian crises, emergency response has increasingly become a field dominated by a more narrowly biomedical approach: feeding the hungry and not attending to the causes of hunger.31 Famine prevention and relief became identified as a non-political charitable enterprise. This conceptual blockage has thwarted discussion of human agency in producing (or imposing) famine. Famine has instead been operationally defined with reference to metrics related to nutrition and food availability. Famines are defined by the gravity of outcome, regardless of cause, and the imperatives of humanitarian response. The current international standard for diagnosing and defining famine, the Integrated food security Phase Classification (IPC), is based on metrics for malnutrition, mortality and food availability. Indicators are derived from survey data, interpreted through an expert process that involves operational humanitarian agencies and the host government – a process guaranteed to ensure that controversial political dimensions are omitted. Conflict data and information about actors’ political and military intent are not included.32 For comparative analysis of historical famines, only thresholds of total numbers of excess deaths are used, for example 100,000 deaths for a ‘great famine’ and 1 million for a ‘catastrophic famine’.33 Causes are downgraded in favour of outcomes.

30 J. de Castro, The Geography of Hunger (Boston: Little, Brown, 1952). 31 S. Jaspars, ‘A role for social nutrition in strengthening accountability for mass starvation?’, World Peace Foundation, occasional paper 18, 24 June 2019, https://sites .tufts.edu/wpf/files/2019/09/A-Role-for-Social-Nutrition-Susanne-Jaspers.pdf. 32 D. Maxwell, ‘Famine early warning and information systems in conflict settings: challenges for humanitarian metrics and response’, Conflict Research Programme, November 2019, https://sites.tufts.edu/reinventingpeace/files/2019/11/CRP-FamineEarly-Warning-and-Information-Systems-in-Conflict-Settings-D-Maxwell-final-2019111 9.pdf. 33 P. Howe and S. Devereux, ‘Famine intensity and magnitude scales: a proposal for an instrumental definition of famine’, Disasters 28:4 (2004), 353–72; de Waal, Mass Starvation.

138

Genocide, Starvation and Famine

Faminogenesis The most severe famine in modern Europe was the Irish famine of 1845–51. With a million dead and approximately 2 million who emigrated during and after the so-called Great Hunger, Ireland is the sole European nation with a smaller population today than 175 years ago. The immediate trigger of the famine was ‘natural’, the potato blight, but the underlying cause of the calamity was British policies that had impoverished the peasantry, and a major reason for such exceptionally high mortality was extreme parsimony in providing relief, combined with opportunistic profiteering from distress to transform the Irish rural economy from smallholdings into larger commercial farms. For many, flight – often to the Americas – represented the best option for staying alive. Although some have argued that the famine should be considered an act of genocide,34 the most eminent historian of the era counters that an attempt to attribute genocidal intent would fail to capture the complexities of greed, ideological rigidity, callous indifference and fatalism that drove England’s great famine in Ireland.35 The famine was not intended as such, but the starvation that ensued was a perfectly foreseeable outcome of the policies pursued. Ireland became a model for British faminogenic policies in India. The inhuman rigour with which the British rulers of India calculated rations for relief works prefigured Nazi nutritional science in occupied Europe. The intent and aetiology of starvation were very different in imperial India to those in occupied eastern Europe, but the outcome was often similar. British rule in India reduced the subcontinent to penury, bringing recurrent famine to formerly self-sufficient lands. In 1876, rains failed across large parts of India, devastating harvests, and colonial British relief efforts – having proven successful but expensive during a previous drought – were abandoned. Sir Richard Temple, who had fallen under harsh criticism for ‘extravagant’ aid that saved lives in 1873–4, set himself to the task of making sure it would not happen again. India’s starving poor had to demonstrate their value as labourers, treading long distances, with many dying along the way, to prove they could earn their meagre rations of a pound of rice per day. Temple established rules for how much the starving would be fed, known as the Temple wage. Its nutritional scale ‘provided less sustenance for hard labour than the diet inside the later 34 A. J. P. Taylor, Essays in English History (London: Book Club Associates, 1977). 35 C. Ó Gráda, ‘Black ’47 and beyond: the Great Irish Famine in history, economy, and memory’, Medical History 45:1 (2000), 136–7.

139

bridget conley and alex de waal

infamous Buchenwald concentration camp’.36 In some areas a quarter or even a third of the population died, sparking protests across India and even among some in the colonial service itself. The return of rain brought malaria, and also, finally, relief from that bout of famine. The multiple causal factors contributing to famine and authorities’ mixed motives and multiple responses provide the substance for much debate about whether numerous historical cases of famine constitute genocide. Foremost among these are twentieth-century famines produced by Communist policies. Soviet agrarian collectivisation, the Great Leap Forward in China, and Cambodia under the Khmer Rouge are the leading contemporary examples of forcible socio-economic transformation that have produced mass starvation. These cases, examined in detail in Volume I I I of The Cambridge World History of Genocide, are briefly summarised here. The famines of 1930–4 in the USSR, affecting southern Russia, Kazakhstan and Ukraine, began with forcible collectivisation of peasant farming – certainly an inhumane policy implemented with great callousness, arguably a crime against humanity, but not initially aimed at destroying an ethnic or national group as such. However, when the Soviet administration learned of the starvation that its policies had unleashed, policies diverged. In southern Russia, collectivisation was eased and the famine was relieved. In Ukraine, by contrast, Stalin intensified the harsh conditions, imposing policies that prevented movement. His regime removed food and agricultural supplies, blacklisted villages and restricted trade.37 Given that the starvation policy specifically targeted selected national populations, the case for genocide can plausibly be made. Ukrainians know this terrible episode as the Holodomor. Lemkin himself described the starvation in Ukraine as an exemplary case of genocide, in which the Soviets violently targeted the ‘soul’ and ‘brain’ of the nation, the Church and the intelligentsia, and used famine against the ‘body’ – the peasantry.38 Estimates today suggest that 3.3 million excess deaths resulted in Ukraine.39 The starvation of Kazakhstan in 1930–3 is another egregious case. For a previously semi-nomadic population, forced sedentarisation into planned villages, along with collective farming and onerous production quotas, was 36 M. Davis, Late Victorian Holocausts (New York: Verso, 2001), p. 38. 37 A. Applebaum, Red Famine: Stalin’s War on Ukraine (New York: Doubleday, 2017), p. 347. 38 R. Serbyn, ‘Lemkin on the Ukrainian genocide’, Journal of International Criminal Justice 7:1 (2009), 123–30, at 126–8. 39 T. Snyder, Bloodlands: Europe between Hitler and Stalin (New York: Basic Books, 2012), p. 53.

140

Genocide, Starvation and Famine

an exceptionally brutal form of ‘modernisation’, which was sustained even as its calamitous impacts became clear. As much as 40 per cent of the Kazakh population died.40 The highest recorded fatalities in any famine occurred as a result of Communist economic transformation in China under Mao Zedong. During the Great Leap Forward (1958–62), an estimated 25–45 million people died as a direct result of the dismantling of peasant agriculture and its replacement with co-operatives and village steel production. Widespread suffering and death were met with Mao’s stubborn refusal to alter his policies.41 While calamitous this was not often classified as genocidal because there was no intent to destroy a protected group as defined by the Genocide Convention. The policies of the Khmer Rouge in Cambodia were possibly even more extreme than those of Mao, although they impacted a smaller population. Under Pol Pot’s efforts to remake Cambodian society and transform its peasant farming to a collective agricultural system, there was immense upheaval and deprivation. Of the estimated 1.75 million people killed under the Khmer Rouge, 69 per cent died of starvation and related causes.42 Widespread starvation in Cambodia was, as Randle DeFalco demonstrates, ‘a direct, foreseeable and avoidable consequence of Khmer Rouge policies . . . [O]ver time the regime’s leaders became aware of their policies’ disastrous effects, but nevertheless continued to enforce them strictly’.43 Insofar as the Khmer Rouge atrocities against the general Cambodian population are assimilated to the corpus of genocides, starvation is central to its perpetration. There is little research that brings the study of famine (defined as an outcome), into the kind of legal-political framework that allows for a dialogue with Genocide Studies. A significant exception is David Marcus’ framework of famine crimes. Marcus has developed a classification of famines, dividing them into four categories depending on the level of human intentionality in their causation.44 His concept of ‘famine crime’ is a hybrid between the definition of famine based on gravity (a certain threshold has to be met for an event to qualify as a ‘famine’) and one based on criminal actions

40 S. Cameron, The Hungry Steppe: Famine, Violence, and the Making of Soviet Kazakhstan (Ithaca, NY: Cornell University Press, 2018), p. 5. 41 De Waal, Mass Starvation, p. 77. 42 Ibid., p. 78. 43 R. DeFalco, ‘Justice and starvation in Cambodia: the Khmer Rouge famine’, Cambodia Law and Policy Journal 3 (2014), 45–84, at 49. 44 D. Marcus, ‘Famine crimes in international law’, American Journal of International Law 97:2 (2003), 245–81.

141

bridget conley and alex de waal

(and by implication, intent). This framework serves as a useful heuristic for analysing famine and genocide together. A first-degree famine crime is committed by a person or persons ‘determined to exterminate a population through famine’.45 This can be mass starvation as genocide, or extermination as a crime against humanity. The most straightforward cases are penal starvation:46 examples include Soviet prisoners of war held by the Nazis on the Eastern Front in 194147 and the widespread and systematic deprivation of food from political prisoners in Syria during the civil war.48 A second-degree famine crime is committed by a person ‘recklessly ignoring evidence that his or her policies are creating, inflicting or prolonging the starvation of a significant number of persons. [It] is committed by an official recklessly pursuing policies that have already proven their faminogenic tendencies’.49 These cases are much more common and more complex: famines in which political and military actors have caused mass starvation in pursuit of other goals, such as consolidating power, repressing insurgency, forcing an adversary to surrender or managing a coalition of diverse belligerent forces on modest resources. Other causal or exacerbating factors, such as climatic disaster, economic adversity or turbulence in food markets, or historic legacies of poverty and poor public health services, can also contribute to the occurrence of such famines and to their level of lethality. The third degree of culpable famine causation is when public authorities are indifferent: their policies may not be the principal cause of famine, but they do little or nothing to alleviate hunger. Their actions are irresponsible or malign but will not qualify as criminal. In the fourth category of famine, there is no official culpability: authorities are incapable or incapacitated, overwhelmed by food crises caused by external factors. These are rare and becoming vanishingly rare, as improved governmental capacities and global humanitarian response mitigate the impacts of food security disruptions. Over the last 150 years, 110 million people are estimated to have died in fifty-eight major famines. About 6 per cent of these victims perished in category 1 famines, 63 per cent in category 2, 18 per cent in category 3, and 13 per cent in category 4.50 Ibid., 247. R. Howard-Hassmann, State Food Crimes (Cambridge University Press, 2016). Snyder, Bloodlands, pp. 177–84. Human Rights Watch, ‘If the dead could speak: mass deaths and torture in Syria’s detention facilities’, December 2015, www.hrw.org/report/2015/12/16/if-dead-couldspeak/mass-deaths-and-torture-syrias-detention-facilities (accessed 4 May 2020). 49 Marcus, ‘Famine crimes’, 247. 50 De Waal, Mass Starvation, pp. 106–8.

45 46 47 48

142

Genocide, Starvation and Famine

First- and second-degree famine crimes are those of concern to genocide scholars. Analysis of famines from 1870 to 201551 reveals that some first-degree famine crimes are regularly studied as mass atrocities. However, considerably more variance occurs for events classified as second-degree famine crimes. These are cases where genocidal intent may be difficult to substantiate, yet they bear similarities to contexts described in terms of mass atrocity or crimes against humanity. Among such cases are the Syria–Lebanon famine during World War One and the British starvation of Germany at the end of that war; famines that accompanied the Russian civil war and civil wars in China (1921–30); numerous episodes in World War Two, including the blockade of Greece, the 1943 Bengal famine, and mass starvation in Southeast Asia under Japanese occupation; deprivations caused by sanctions in Iraq in the 1990s; the North Korean famine of 1995–7; and recurrent famines in Ethiopia, Somalia, South Sudan and Sudan. In these contexts, it is helpful to understand famine as one potential outcome of a process whereby starvation, as defined in the previous section, is used as a weapon against a civilian population. Like direct killing, such mass starvation is intentional, even if not always demonstrating genocidal intent. Case-by-case analysis of the complex interaction of multiple factors and actors can illuminate the precise balance in any given instance. Starvation crimes almost invariably intersect with other causes of deprivation, such as generalised food insecurity, poverty, inequality, poor health infrastructure or environmental degradation. There may be other intersecting triggers of disaster, such as climatic or environmental calamities including drought and pest infestations, economic crisis or international factors such as an increase in the price of foods on the global market. Political leaders prefer to blame famine on these factors – especially the weather – and the staff of relief agencies also like to depoliticise famine for reasons of fundraising from a donating public that finds the victims of ‘natural’ disasters more deserving of their charity, and because averting their gaze from the politics behind the starvation makes it easier to work in the affected country. Nonetheless, if a country is already highly vulnerable to food crisis and famine, this makes the perpetration of starvation more blameworthy, not less. For example, an army commander who destroys food crops in a poor, food-insecure region or country does so in the knowledge that this is more likely to create intense hunger there. Additionally, however, starvation crimes typically take a long time to unfold and take full effect. While 51 Ibid.

143

bridget conley and alex de waal

a massacre may be perpetrated within a single day, the infliction of starvation takes at least several weeks and often a whole year or longer. This has several implications. It complicates tracing cause and effect, as many intervening factors intrude, and different contributory causes can wax and wane over time. The long process of deprivation prior to mass starvation means that actions that caused or set in motion the process of starvation may not initially have been taken with that purpose, or even with an awareness of that outcome. But if they were, the time lag also makes it easier for the perpetrator to evade or disclaim responsibility. The duration also means that as events unfolded and the reality or imminence of starvation became clear, the perpetrator will have had opportunities to halt or reverse harmful policies, or to intervene to resolve the crisis. In assessing the responsibility of a political or military authority who has presided over starvation, it would therefore be necessary to ascertain what information was – or should have been – available about the impact of deleterious policies, and what actions could have been taken to halt or remedy those impacts.

Protection: The Complications and Necessities of Thinking Genocide and Starvation Together Divergent prohibitions and convoluted trails of intent and outcome are not the only factors that have caused the intertwined histories of genocide, famine and mass starvation to part ways. Perceived incongruities between the imperatives of protection of victims and punishment of perpetrators have also created firewalls between responding to famine as a ‘complex of humanitarian emergencies’, and to genocide. Humanitarian aid can become a fig leaf for failing to confront perpetrators. The war in Bosnia-Herzegovina (1992–5) provides an illustration. There, the priority of providing humanitarian aid – including mandating UN peacekeeping forces to protect the delivery of the aid but not the people under attack – failed to halt the ethnic cleansing and gave rise to what was derisively described as ‘the well-fed dead’.52 Bosnian Serbs, in particular, manipulated international aid throughout the conflict, bartering humanitarian ‘access’ for political and military concessions.53 Mass starvation was kept at bay by the 52 Editorial, ‘The well-fed dead in Bosnia’, New York Times, 15 July 1992, p. A20. 53 M. Cutts, ‘The humanitarian operation in Bosnia, 1992–1995: dilemmas of negotiating humanitarian access’, Policy Research Unit, UNHCR Working Paper no. 8, May 1999.

144

Genocide, Starvation and Famine

large-scale international aid operation, even while the threat of starvation continued to be used as a weapon of war. Bosnia demonstrates how starvation can be weaponised in pursuit of genocidal aims, even without itself causing widespread death. Carefully calibrated to serve as leverage, humanitarian supplies can assume a potent place in a perpetrator’s arsenal. In Bosnia, the policies of starvation reached an apex in the lead up to the Bosnian Serbs’ final push to capture Srebrenica in spring 1995. Bosnian Serb forces not only starved the civilian population sheltering there, but also interrupted supplies to UN peacekeepers stationed in the enclave. Humanitarian supplies were obstructed, as they had been throughout the conflict, as a means to advance the goal of forceful displacement. What differed in this case was the degree of lethality that followed. On 8 March 1995, Bosnian Serb President Radovan Karadžić issued Operational Directive 7, which commanded his VRS forces to tighten access to the enclave, directing: ‘By planned and well-thought-out combat operations, create an unbearable situation of total insecurity, with no hope of further survival or life for the inhabitants of Srebrenica or Žepa.’54 No clearer statement of intent to cause deprivation as a weapon of war could have been made. Would more robust measures to resupply the people in Srebrenica have warded off genocide? As a counterfactual question, we cannot know the answer. However, it is clear that the central issue did not concern supply chains, but rather the imperative to counter a genocidal politico-military calculus to destroy Muslim communities in areas claimed by the Bosnian Serbs. At the same time, it must be acknowledged that preventing deaths by starvation is a positive outcome of humanitarian interventions. Recognising the benefits of the one does not preclude the other; the challenge is to consider both. Thinking about famine and genocide together is best done, as we have argued, through the concept of mass starvation or starvation crimes, and requires grasping the anti-civilian logic embedded in both famine and genocide. The point is not to blur the categories together, treating them as synonymous. There are complications that adhere to each, exemplified by the limits of responding to genocide as a complex humanitarian crisis. A robust vision of protection for civilians at risk of genocide and mass atrocities requires addressing the full range of harms that threaten them, how they are produced, and how they can be mitigated. 54 Judgment, Krstić (IT-98-33-T), Trial Chamber, 2 August 2001, §199.

145

bridget conley and alex de waal

Starvation has long been a tool of warfare and oppression, including in policies designed to target specific groups. This feature of genocide continues into the contemporary era; in all of the major twentieth-century genocides, perpetrators deployed mass starvation or the threat of it as a weapon of choice. The convergence of famine and genocide, or mass atrocities, becomes clearer as one examines modern-day assaults: improvements in the technologies for food distribution, medical relief and nutritional support have rendered environmental hazards and economic crises less determinative of famine. But as the twentieth century gave way to the twenty-first, and with greater international attention paid to systematic killing of civilian groups, hunger and deprivation became signature tools for regimes that tried to escape the international condemnation that has increasingly accompanied mass killing. In Syria, South Sudan, Nigeria and Yemen, regimes intent on gaining or maintaining control over civilians have resorted to mass starvation, anticipating that policy and media frameworks that viewed famine as a technical malfunction in need of the ministrations of nutritional science would provide an alibi. In none of these cases was famine a ‘natural’ phenomenon, and indeed such famines had largely been banished from the world in the later twentieth century. Only when governments responded to or actively created shortages in ways that exacerbated a population’s vulnerability did mass starvation manifest. In 2017, the United Nations raised a clear red warning flag: famine had returned.55 While each of these cases emerged from a distinct configuration of causes, there are troubling global trends that may increase or alter vulnerabilities to mass starvation. One is an unintended consequence of recently increased international concern over the direct killing of civilians in conflict, including efforts to count those killed and attempts to achieve political and legal accountability for atrocities. This welcome advance may inadvertently lead perpetrators of violence to resort to acts of deprivation in an effort to mitigate their perceived responsibility for such internationally salient forms of suffering while still achieving the outcome of destroying targeted groups. Globalisation of food markets and urbanisation have also altered 55 S. O’Brien, ‘Under-Secretary-General for Humanitarian Affairs and Emergency Relief Coordinator, Statement to the Security Council on Missions to Yemen, South Sudan, Somalia and Kenya and an Update on the Oslo Conference on Nigeria and the Lake Chad Region’, United Nations Security Council, 10 March 2017, https://docs .unocha.org/sites/dms/Documents/ERC_USG_Stephen_OBrien_Statement_to_th e_SecCo_on_Missions_to_Yemen_South_Sudan_Somalia_and_Kenya_and_update_ on_Oslo.pdf.

146

Genocide, Starvation and Famine

vulnerabilities: as populations increasingly rely on integrated and tradedependent systems of food production and distribution, rather than local production, the disruptions of war have the potential to quickly create pockets of profound need. This is not just a matter of food availability: for populations reliant on purchasing food, economic collapse through extended warfare, as has been seen in Yemen and in pockets of South Sudan, creates conditions in which food is available but priced out of people’s means to access it. The ever-expanding interconnections of global economies increases exposure to outside shocks: during 2011, in Somalia, a spike in global food prices compounded by other factors, including the actions of conflict participants and US-imposed restrictions on aid to groups designated as ‘terrorists’, created famine.56 In pre-war Yemen, what little agricultural production was possible had been steadily shifting towards an export market rather than local consumption. In such circumstances, the use of starvation in warfare can and did cause a food economy to collapse. Finally, the consequences of climate change introduce new challenges, risks and sites of exposure. Famine has throughout history often been miscategorised as a natural phenomenon or as an unfortunate side-effect of conflict and political oppression. The numbers and names of its victims fade into the background, blurred traces of the horrors of history. This is both an inaccurate understanding of how starvation crimes are perpetrated and an injustice to victims. Starvation has been a principal tool of genocide; today’s famines are the product of policies and should be considered as mass atrocities.

Bibliographic Note Recent scholarly writing on famine and genocide fall into two principal categories: case studies that include accounts of the use of starvation as a tool of war, extermination and/or genocide; and legal analysis of crimes of hunger and starvation. Perhaps the most influential early case study of starvation as a tool of mass atrocity is David Keen’s The Benefits of Famine: A Political Economy of Famine and Relief in Southwestern Sudan, 1983–1989 (Princeton University Press, 1994). This is a detailed study of how famine was inflicted on the southern Sudanese by the counter-insurgency and predatory-commercial policies of the government of Sudan, and how 56 D. Maxwell and N. Majid, Famine in Somalia: Competing Imperatives, Collectives Failures 2011–2012 (Oxford University Press, 2016).

147

bridget conley and alex de waal

international humanitarian agencies and donors failed to engage with that destructive dynamic. Alex de Waal’s Famine Crimes: Politics and the Disaster Relief Industry in Africa (London: James Currey, 1997) is a comparative analysis of the modern history of famine and humanitarian action in Africa, focusing on the failures of political accountability and the deliberate political and military policies that have caused modern famines, and the failure of the ‘humanitarian international’ to grapple with these dynamics to the extent of intermittent complicity in the political causes of famine. It includes detailed case studies of famines in Ethiopia, Somalia and Sudan. A compelling historical case study is Mike Davis’ Late Victorian Holocausts: El Niño Famines and the Making of the Third World (London: Verso, 2002). This is an account of the combined impact of European imperial conquest and subjugation on Asia (with occasional forays into other continents) and climatic anomalies in the last decade of the nineteenth century, and the hardhearted response of those same imperial powers to the mass starvation that ensued. Collingham’s Taste of War is a comprehensive account of food and starvation during World War Two, framed by the observation that hunger killed as many people as direct violence, and including accounts of the blockade-famine in Greece, the Hungerplan and associated forced starvation on the Eastern Front, the Bengal famine and starvation throughout the East and Southeast Asian theatres of war. A more contemporary overview is the volume edited by Stephen Devereux, The New Famines: Why Famines Persist in an Era of Globalization (London: Routledge, 2007). This examines why famines have changed and persisted into the twenty-first century, including chapters on definitional and measurement issues, and on Bosnia-Hercegovina, Iraq, Madagascar, Malawi and the Horn of Africa, along with thematic analyses by many of the leading researchers on the topic. Turning to studies based on law, Marcus’ article ‘Famine crimes in international law’ is a seminal analysis of the relationship between human agency and famine. It informs a framework of four degrees of government conduct in relation to famine, discussed within the context of how existing bodies of international law address criminal culpability. Rhoda Howard-Hassmann’s State Food Crimes (Cambridge University Press, 2016) provides a comprehensive account of the policies of deprivation pursued in historical and contemporary cases of mass starvation, leading to an examination of the different legal regimes under which starvation should be prohibited, concluding with advocacy for a new international treaty on the right to food. Simone Hutter’s Starvation as a Weapon: Domestic Policies of Deliberate Starvation as a Means to an End under International Law (Leiden: Brill, 2015) 148

Genocide, Starvation and Famine

uses the framework of the several strands of international law to provide a thorough account and analysis of the illegality of deliberate starvation, focusing on cases in which states deploy starvation against their own people. Alex de Waal’s Mass Starvation is a recent wide-ranging account of the decline and resurgence of famine over the last 150 years, focusing on the commonalities between starvation and mass atrocity, concluding that famine can readily be ended by making the imposition of starvation politically and morally toxic.

149

5

Climate, Violence and Ethnic Conflict in the Ancient World francis ludlow, chris morris and conor kostick

Introduction Enquiry into the role of climate in human violence and conflict has seen increased attention in the past two decades. Research has focused on the reality and nature of ‘pathways’ linking climate, violence and conflict.1 Multiple parameters, from temperature to precipitation (plus the rapidity, persistence and severity of changes therein), have been assessed for associations with the timing, frequency, severity and typology (individual versus collective, organised versus spontaneous) of violence and conflict. Understandings are still, however, impeded by limitations in available data, dispute over methodologies, and the sheer complexity of the pathways that may connect climate to violence and conflict.2 A clear consensus is yet to emerge,3 and the growing number of studies that find some role for climate4 have sometimes become subject to notable contention.5

Work on this chapter was supported by an Irish Research Council Starting Laureate Award (CLICAB project, IRCLA/2017/303) and by US NSF Award #1824770. T. F. Homer-Dixon, Environment, Scarcity, and Violence (Princeton University Press, 1999), pp. 1–11; Hanne Seter, ‘Connecting climate variability and conflict: implications for empirical testing’, Political Geography 53 (July 2016), 1–9. 2 J. Scheffran et al., ‘Climate change and violent conflict’, Science 336:6083 (2012), 869–71; E. Meierding, ‘Climate change and conflict: avoiding small talk about the weather’, International Studies Review 15:2 (2013), 185–203; M. Brzoska, ‘Weather extremes, disasters, and collective violence: conditions, mechanisms, and disaster-related policies in recent research’, Current Climate Change Reports 4 (2018), 320–9; J. Selby, ‘Positivist climate conflict research: a critique’, Geopolitics 19:4 (2014), 829–56. 3 H. Fjelde and N. von Uexkull, ‘Climate triggers: rainfall anomalies, vulnerability and communal conflict in sub-Saharan Africa’, Political Geography 31:7 (2012), 444–53. 4 Brzoska, ‘Weather extremes’, 320; S. M. Hsiang and M. Burke, ‘Climate, conflict and social stability: what does the evidence say?’, Climatic Change 123 (2014), 39–55. 5 For example S. M. Hsiang et al., ‘Reconciling climate–conflict meta-analyses: reply to Buhaug et al.’, Climatic Change 127 (2014), 399–405. 1

150

Climate, Violence and Ethnic Conflict in the Ancient World

Debate over the role of a multi-year (2006–10) drought in the 2011 Syrian uprising offers insight into wider tensions in the climate–conflict literature, from which underlying or implicit stances (on how this research should be conducted and what factors are truly important) can be inferred. A potential role for drought in the onset of the conflict was proposed by scholars such as Peter H. Gleick in 20146 and Caitlin E. Werrell and colleagues in 2015,7 in each case emphasising a role for other political and socio-economic factors. Also in 2015, Colin P. Kelley and colleagues contextualised the drought against available instrumental weather data for the region, identifying it as the worst on record and suggesting its ‘catalytic’ role, ‘contributing to political unrest’ in ‘a country marked by poor governance and unsustainable agricultural and environmental policies’.8 Others minimised the drought’s role and even the appropriateness of considering this. Francesca De Châtel thus argued in 2014 that its impacts should be framed ‘in the context of rapid economic liberalization and longstanding resource mismanagement’, and concluded that ‘focusing on external factors like drought and climate change . . . is counterproductive as it diverts attention from more fundamental political and economic motives behind the protests and shifts responsibility away from the Syrian government’.9 Jan Selby and colleagues queried (in 2017) the basis of an oftcited climate–conflict pathway that posited internal migration of droughtimpacted rural dwellers to urban areas as heightening tensions, questioning the scale of migration and evidence for migrant participation in protests.10 Their paper provoked multiple responses,11 with a rejoinder by Selby and colleagues that cast further doubt upon the scale and role of internal 6

7

8

9

10

11

P. H. Gleick, ‘Water, drought, climate change, and conflict in Syria’, Weather, Climate, and Society 6 (2014), 331–40. C. E. Werrell et al., ‘Did we see it coming? State fragility, climate vulnerability, and the uprisings in Syria and Egypt’, SAIS Review of International Affairs 35:1 (2015), 29–46. C. P. Kelley et al., ‘Climate change in the Fertile Crescent and implications of the recent Syrian drought’, Proceedings of the National Academy of Sciences 112:11 (2015), 3241–6, at 3241. For the drought’s severity and regional variability, see S. Mathbout et al., ‘Spatial and temporal analysis of drought variability at several timescales in Syria during 1961– 2012’, Atmospheric Research 200 (2018), 153–68. Quote from online abstract accompanying Francesca De Châtel, ‘The role of drought and climate change in the Syrian uprising: untangling the triggers of the revolution’, Middle Eastern Studies 50:4 (2014), 521–35. J. Selby et al., ‘Climate change and the Syrian Civil War revisited’, Political Geography 60 (2017), 232–44. C. Kelley et al., ‘Commentary on the Syria case: climate as a contributing factor’, Political Geography 60 (2017), 245–7; P. H. Gleick, ‘Climate, water, and conflict: commentary on Selby et al. 2017’, Political Geography 60 (2017), 248–50; C. S. Hendrix, ‘A comment on “Climate change and the Syrian Civil War revisited”’, Political Geography 60 (2017), 251–2.

151

francis ludlow, chris morris and conor kostick

migration.12 Debate continues, including by those still finding a role for internal migration.13 Valuable observations are made by all these contributions, and scholars across the social and natural sciences routinely attempt to avoid monocausality when hypothesising on environmental agency in human affairs,14 usually framing climate as an indirect or background actor in violence and conflict, ranging from catalyst or trigger to contributor or exacerbator.15 But in the case of Syria, as elsewhere, a predisposition exists to relegate environmental forces to tangential roles, perhaps partly from concern to avoid accusations of environmental determinism and a degree of anthropocentrism. Yet all social phenomena occur within an environmental context. The very lack of an environmental stressor in a given period is the (nontrivial) environmental context to any conflict (origin, course and outcome) in that period. There is no contesting that the situation in Syria is tremendously complex and of multiple origins. Dictatorial governance with marked inequality is not likely conducive to long-term stability. Recognising multiple (anthropocentric) causal factors and pathways does not, however, preclude a meaningful climatic contribution. Indeed, in their diminishing of coping strategies and increasing of vulnerabilities to extreme weather, institutional and socio-economic factors may create conditions in which an uprising becomes a possible response to climatic stressors. Natural hazards are doubtless used to obfuscate failings, as De Châtel illustrates with Bashar al-Assad’s citation of the 2006–10 drought as ‘beyond our powers’.16 Indeed the similar framing of natural hazards as ‘acts of God’ has often shielded the socio-economic and political circumstances that engender vulnerability but often favour elites.17 But natural disasters can 12

13

14

15

16 17

J. Selby et al., ‘Climate change and the Syrian Civil War revisited: a rejoinder’, Political Geography 60 (2017), 253–5. For example J. Selby, ‘Climate change and the Syrian Civil War, part II: the Jazira’s agrarian crisis’, Geoforum 101 (2019), 260–74. Supporting a role for internal migration is K. Ash and N. Obradovich, ‘Climatic stress, internal migration, and Syrian Civil War onset’, Journal of Conflict Resolution 64 (2020), 3–31. Efforts to improve practices are ongoing: B. J. P. van Bavel et al., ‘Climate and society in long-term perspective: opportunities and pitfalls in the use of historical datasets’, WIRE’S Climate Change 10:6 (2019), e611. Of the Rwandan genocide, ‘environmental and population pressures’ are given ‘at most a limited, aggravating role’ among the ‘many factors [that] were operating in this conflict’: V. Percival and T. Homer-Dixon, ‘The case of Rwanda’ in Ecoviolence: Links among Environment, Population, and Security, ed. T. Homer-Dixon and J. Blitt (Lanham, MD: Rowman & Littlefield, 1998), pp. 201–22, at p. 201. De Châtel, ‘Role of drought’, 532, 535. D. Dickson, Arctic Ireland: The Extraordinary Story of the Great Frost and Forgotten Famine of 1740–1741 (Belfast: White Horse Press, 1997), pp. 70–4; T. H. Dixon, Curbing

152

Climate, Violence and Ethnic Conflict in the Ancient World

simultaneously act as ‘revelatory crises’ that expose circumstances that promote vulnerability and might otherwise remain unremarked.18 Care must indeed be taken in framing research and findings. Disaster Studies scholars thus repeatedly problematise the term ‘natural’, emphasising the human factors that turn natural hazards into natural disasters.19 Some governments may blame ‘global warming’ (as caused by others and beyond their control) for failure to prepare for extreme weather,20 but we cannot agree that this merits something akin to self-censorship over the role of climate (or anthropogenic climate change) in conflict, as argued in De Châtel’s conclusion that the ‘possible role of climate change . . . is not only irrelevant; it is also an unhelpful distraction’.21 This is particularly so when the question of climatic influences is far from settled. Not all droughts are born equal; their impacts vary not only by background societal conditions but interactively with these depending upon seasonality, severity, and the time since previous droughts. Fuller knowledge of the origins of violence and conflict must be based upon an understanding that all social activities are fundamentally ‘coproductions’22 of entangled or coupled human–natural systems.23 Impacts of (and responses to) extreme weather are always thereby mediated by socioeconomic, political and cultural circumstances, and vice versa. Comparative studies spanning different states and societies, identifying common and contrasting impacts and responses to climatic stress, can indeed help to isolate those human contexts often privileged by scholars seeking origins of violence and conflict.24 Comparisons can also be made through time, with differing impacts and responses revealing societal characteristics and

18

19

20

21 22

23

24

Catastrophe: Natural Hazards and Risk Reduction in the Modern World (Cambridge University Press, 2017), pp. 14–17. J. S. Solway, ‘Drought as a revelatory crisis: an exploration of shifting entitlements and hierarchies in the Kalahari, Botswana’, Development and Change 25:3 (1994), 471–95. For example T. Steinberg, Acts of God: The Unnatural History of Natural Disaster in America (Oxford University Press, 2000); C. Hartman and G. D. Squires, eds., There Is No Such Thing as a Natural Disaster: Race, Class, and Hurricane Katrina (London: Routledge, 2006). M. R. Dove and M. H. Khan, ‘Competing constructions of calamity: the April 1991 Bangladesh cyclone’, Population and Environment 16 (1995), 445–71. De Châtel, ‘Role of drought’, 532. M. Kervyn et al., ‘Hazards and disasters: learning, teaching, communication and knowledge exchange’, Belgeo 1 (2015), https://doi.org/10.4000/belgeo.16397. L. H. Gunderson and C. S. Holling, Panarchy: Understanding Transformations in Human and Natural Systems, 2nd ed. (Washington, DC: Island Press, 2001); J. Liu et al., ‘Complexity of coupled human and natural systems’, Science 317 (2007), 1513–16. Again for Syria: L. Eklund and D. Thompson, ‘Differences in resource management affects drought vulnerability across the borders between Iraq, Syria, and Turkey’, Ecology and Society 22:4 (2017), article 9; G. Soffiantini, ‘Food insecurity and political instability during the Arab Spring’, Global Food Security 26 (2020), 100400.

153

francis ludlow, chris morris and conor kostick

circumstances contributing to the origins, course and outcomes of violence and conflict.25 Drawing from history (including through archaeology) can thus advance understandings, with violence and conflict well documented from the ancient era onwards.26 Much work on historical violence of course exists, but progress has suffered from a failure to recognise the complementarity of studies exploring the role of climate and those intent on human factors. Stephen Vlastos’ classic study of peasant revolt (often following weather-related crop failure) in Tokugawa Japan is ‘less interested in what set of conditions “caused” peasants to protest than in the nature, form, and content of the movements. The structure of conflict and what it can tell us about class relations is the primary concern, events that acted as triggers are secondary.’27 Yet illuminating one surely illuminates the other. Potential now exists to study a broader temporal and geographical span than ever before, given the increasingly rapid development of natural archives (‘palaeoclimatic proxies’) providing climatic information of ever-greater scope (atmospheric and ocean temperatures, precipitation, circulation), temporal and geographical precision and accuracy.28 Studies of past or possible future climatic contributions to genocide are rare,29 perhaps partly due to the terrible scale and relatedly lesser incidence of the phenomenon relative to other violence, and an imperative to unravel its more explicitly human causes. Relevant, too, are doubts over the status of premodern cases as genocides. This chapter concentrates largely on climate and state violence (sometimes arguably amounting to genocide) in the ancient Near East, particularly involving the well-documented Neo-Assyrian Empire 25

26

27

28

29

F. Ludlow and C. Travis, ‘STEAM approaches to climate change, extreme weather and social-political conflict’ in The STEAM Revolution: Transdisciplinary Approaches to Science, Technology, Engineering, Arts, Humanities and Mathematics, ed. A. de la Garza and C. Travis (New York: Springer, 2018), pp. 33–65. D. L. Martin et al., eds., The Bioarchaeology of Violence (Gainesville: University of Florida Press, 2012); C. Knüsel and M. J. Smith, eds., The Routledge Handbook of the Bioarchaeology of Human Conflict (New York: Routledge, 2013). S. Vlastos, Peasant Protests and Uprisings in Tokugawa Japan (Berkeley: University of California Press, 1986), pp. 4–5. R. S. Bradley, Paleoclimatology: Reconstructing Climates of the Quaternary, 3rd ed. (Amsterdam: Elsevier, 2015). For example Percival and Homer-Dixon, ‘Case of Rwanda’; A. Alvarez, ‘Borderlands, climate change, and the genocidal impulse’, Genocide Studies International 10:1 (2016), 27–36; A. Alvarez, Unstable Ground: Climate Change, Conflict, and Genocide (Lanham, MD: Rowman & Littlefield, 2017); A. Exenberger and A. Pondorfer, ‘Genocidal risk and climate change: Africa in the twenty-first century’, International Journal of Human Rights 18:3 (2014), 350–68; J. Zimmerer, ‘Climate change, environmental violence and genocide’, International Journal of Human Rights 18:3 (2014), 265–80.

154

Climate, Violence and Ethnic Conflict in the Ancient World

and internal revolt in Egypt’s Ptolemaic state. Dates are hereafter in B C E unless noted.

Ancient Era Climate and Violence Ancient Near Eastern sources feature frequent intersections between crises of weather, famine, disease and violence. Some, like the second-millennium ‘city lamentations’ of Sumer,30 are particularly evocative, depicting a divinely sanctioned (or divinely permitted) storm that begins a process in which cities (Ur, Nippur and others) are so ‘utterly destroyed [that], all forms of life are crushed’.31 Atrocities occur in which none are spared, regardless of age or gender, even up to ‘the slashing of mothers’ wombs’.32 These poems have been read as reflecting real events, with the storm a metaphor for foreign invasion. For some, a ‘purely military interpretation . . . falls short’ and the storm represents ‘a combination of drought, famine and . . . epidemic, leading up to . . . foreign invasions’ that perhaps opportunistically attacked cities weakened by climatic adversity.33 In the late-third-millennium Sumerian epic The Matter of Aratta, drought and famine are plot devices that introduce power asymmetries underlying interstate or intercity dynamics.34 Adverse weather could also disrupt (or so be claimed) the delicate diplomatic network of gifting and exchange between ‘brother kings’ in the second-millennium ancient Near East and Egypt.35 Later, a letter dated c.668–664 from King Shamash-shumu-ukin of Babylon to his brother (and superior), Ashurbanipal, king of Assyria, excuses selling grain to Elam (then suffering famine) without permission. Here, Shamash-shumuukin likely exploited weather-related stress for his own ends, increasing tensions in what has been deemed a catalyst for the Neo-Assyrian Empire’s 30

31

32

33 34

35

J. Jacobs, ‘The city lament genre in the ancient Near East’ in The Fall of Cities in the Mediterranean, ed. M. R. Bachvarova et al. (Cambridge University Press, 2016), pp. 13–35. H. L. J. Vanstiphout, ‘The death of an era: The Great Mortality in the Sumerian city laments’ in Death in Mesopotamia, ed. B. Alster (Copenhagen: Akademisk Forlag, 1980), pp. 83–9, at p. 83. For military exploitation of weather elsewhere, see T. Jun and R. Sethi, ‘Extreme weather events and military conflict over seven centuries in ancient Korea’, Proceedings of the National Academy of Sciences 118:12 (2021), e2021976118. S. Bertman, Handbook to Life in Ancient Mesopotamia (New York: Facts on File, 2003), p. 174. Vanstiphout, ‘Death of an era’, 86. H. Vanstiphout, Epics of the Sumerian Kings: The Matter of Aratta (Atlanta: Society for Biblical Literature, 2003), pp. 2, 4, 23, 50, 53, 77, 83. See the fourteenth-century letter of Burna-Buriaš of Karduniaš (centred on Babylonia) to Amenophis IV of Egypt: W. L. Moran, The Amarna Letters (Baltimore: Johns Hopkins University Press, 1992), pp. 13–14.

155

francis ludlow, chris morris and conor kostick

downfall.36 When conflict occurred, Elam indeed fought on Shamash-shumuukin’s side, but its cities and people suffered devastating reprisals by a victorious Ashurbanipal. His annals describe campaigning in c.647: Susa [capital of Elam], the great holy city . . . I conquered. I entered its palaces, I opened their treasuries where . . . goods and wealth were amassed . . . I destroyed the ziggurat of Susa . . . I reduced the temples of Elam to naught; their gods and goddesses I scattered to the winds. The tombs of their ancient and recent kings I devastated . . . I devastated the provinces of Elam and on their lands I sowed salt.37

Here we see (at least in concept) environmental warfare which, alongside targeting of irrigation networks and diversion of watercourses to desiccate or flood lands and cities, are techniques that deliberately impact non-combatants by diminishing landscape carrying capacities and buffers against climatic variability.38 Ashurbanipal thus ‘boasted that henceforth no human cry would be heard throughout Elam for the land had reverted to wilderness’.39 In the Graeco-Roman world, weather was referenced by historians such as Polybius, who noted its tactical military importance; for example, it was snowing and the Trebbia was swollen by rainfall, or the day was clear and bright, or the Etesian winds had begun blowing.40 Livy also noted events that might be weather-related (triggered or compounded) and from 507 to 292 he recorded eleven instances of famine and pestilence (at roughly twenty-year intervals on average). These he often linked to political events when hunger and disease promoted social tensions and conflict, including the exile of Coriolanus (493) and the appointment of Cincinnatus as dictator (446).41 Vulnerability to extreme weather and climatic change is always context-dependent, and Livy thereafter 36

37

38

39

40 41

G. Brereton, ‘I am Ashurbanipal, king of the world, king of Assyria’ in I Am Ashurbanipal, King of the World, King of Assyria, ed. G. Brereton (London: Thames & Hudson and the British Museum, 2018), pp. 10–19, at p. 18. Annals of Ashurbanipal, Edition F; J. M. Aynard, Le Prisme de Louvre AO 19.939, Bibliothèque de l’École des Hautes Études 309 (Paris, 1957), pp. 48–61. K. Van Lerberghe et al., ‘Water deprivation as military strategy in the Middle East, 3.700 years ago’, Méditerranée: Journal of Mediterranean Geography 128 (2017), http://journals .openedition.org/mediterranee/8000. A. K. Grayson, ‘Assyria 668–635 B.C.: the reign of Ashurbanipal’ in The Cambridge Ancient History, ed. J. Boardman et al., vol. I I I, part I I, 2nd ed. (Cambridge University Press, 1992), pp. 142–61, at p. 153. Polyb. 3.72, 4.78 and 5.5. Livy, The Early History of Rome, trans. Aubrey de Sélincourt (Harmondsworth: Penguin, 1971), Cincinnatus: pp. 87–91, Coriolanus: pp. 148–76; see also Livy, Rome and Italy, trans. Betty Radice (Harmondsworth: Penguin, 1982); Livy, The War with Hannibal, trans. Aubrey de Sélincourt (Harmondsworth: Penguin, 1972); Livy, Rome and the Mediterranean, trans. Henry Bettenson (Harmondsworth: Penguin, 1976).

156

Climate, Violence and Ethnic Conflict in the Ancient World

records fewer such events. While there are many possible reasons, not least the changing nature of the sources drawn upon by Livy, it may be that Rome was increasingly able to access supplies from a wider area, including the ‘bread basket’ regions of Sicily and North Africa, with scarcity less commonly a trigger for trouble. Herodotus in his Histories relates how the Lydians suffered an eighteenyear famine such that half the population migrated from Asia Minor to Italy. He offers insufficient context for precise dating, but this may imply a drought in western Asia Minor in the period 950–850 (assumption-dependent).42 Regardless of veracity, this and similar tales signal contemporary awareness of mass migration as one outcome not only of violence but also of climatic precarity.43 Temporary climate-related migrations are also documented. For c.667, Ashurbanipal’s annals note: ‘When hard times arose in Elam and there was famine’, there were people who ‘fled before the hard time and settled [in Assyria] until rain fell in his land [in Elam] and there was a harvest’.44 For a normally sedentary population, large-scale movement into well-inhabited (or intensively utilised) areas has been a potential trigger for conflict,45 and the combined population may be exposed to increased risk of famine. Unless newcomers are accompanied by sufficient gains in food production via technological advance, the opening up of new land or expansion into (presumably) less-populated marginal lands (depending upon how successfully these are adapted to),46 options may be limited. Someone must move or face reduction.47 Feeding a combined population from existing land may be less 42 43

44

45

46

47

Hdt. 1.93–6. For further examples see G. Guarducci, ‘Empire of conflict, empire of compromise: the Middle and Neo-Assyrian landscape and interaction with the local communities of the Upper Tigris borderland’ in Imperial Peripheries in the Neo-Assyrian Period, ed. C. W. Tyson and V. R. Hermann (Louisville: University Press of Colorado, 2018), pp. 65–96, at p. 70. Annals of Ashurbanipal, Edition B; A. C. Piepkorn (ed.), Historical Prism Inscriptions of Ashurbanipal I (University of Chicago Press, 1933), pp. 57–9. For example in Gaelic Ireland: B. M. S. Campbell and F. Ludlow, ‘Climate, disease and society in late-medieval Ireland’, Proceedings of the Royal Irish Academy 120C (2020), 159–252, at 179, 196, 200, 216. Even with adaptation, vulnerability to climatic change and extreme weather (plus wider economic changes), may be heightened: T. J. Wilkinson, ‘Settlement and land use in the zone of uncertainty in Upper Mesopotamia’ in Rainfall and Agriculture in Northern Mesopotamia, ed. R. M. Jas (Istanbul: Nederlands Historisch-Archaeologisch Instituut, 2000), p. 20. See also early modern dislocation of Gaelic populations to marginal uplands: F. Ludlow and A. Crampsie, ‘Environmental history of Ireland, 1550–1730’ in Cambridge History of Ireland, vol. I I: Early Modern Ireland, 1550–1730, ed. J. Ohlmeyer (Cambridge University Press, 2018), pp. 608–37. Thus, the fourth century C E movement of Gothic peoples south and west into the Roman Empire ahead of Hunnic advances.

157

francis ludlow, chris morris and conor kostick

destabilising when nominal invaders insert themselves as a new aristocracy,48 or when a movement is carefully planned (targeting suitably resourced destinations), when a transplanted population comprises mainly women and children or is accorded lower status (e.g. slaves) that does not overtly compete with incumbents. Movement is comparatively easy for nomadic or semi-nomadic peoples, and migration around a characteristic circuit is a useful subsistence adaptation in marginal environments.49 Departing from this can be an effective short-term response to unanticipated extremes of weather or conflict, though doing so may engender further conflict. Tribes exploiting marginal (usually drier) Mesopotamian lands were rarely purely nomadic, and on longer timescales exhibited an adaptive ‘equilibrium between . . . settled and pastoral components . . . affected by climatic fluctuations, wetter . . . favouring farming and urbanism, drier . . . inducing urban contraction and pastoralist expansion’.50 Such movements have been associated with conflict and declining Assyrian and Babylonian power in the eleventh to tenth centuries,51 and management of these peoples was of ongoing concern to more permanently sedentary Mesopotamian states. Here is a letter from the governor of Assur, Tab-sil-Esarra, to Neo-Assyrian king Sargon II (r. 722–705) in Nineveh: As to the Arabs about whom the king, my lord wrote to me: ‘Why do they graze their sheep and camels in the desert where they must resort to plundering when hungry?’ Rains have been scarce this year; they had to settle in [the desert]. As to what the king, my lord, wrote to me: ‘Now, go to Hinzanu, and let them go and graze with you! . . .’ I will now go to Hinzanu, but they (are sure to) leave the territory I am assigning to them, move further downstream and plunder; they pay absolutely no heed to the chief scout I have appointed . . .52 48

49

50

51

52

English nobility found employment with the Varangian Guard in Constantinople after the Norman Conquest of 1066 C E; lowland British lords became pensioners in courts in the Welsh hills after the Saxon destruction of their realms. K. Lyublyanovics, New Home, New Herds: Cuman Integration and Animal Husbandry in Medieval Hungary from an Archaeozoological Perspective (Oxford: Archaeopress, 2017), p. 18. F. A. M. Wiggermann, ‘Agriculture as civilization: sages, farmers, and barbarians’ in The Oxford Handbook of Cuneiform Culture, ed. K. Radner and E. Robson (Oxford University Press, 2011), pp. 663–89, at p. 667; K. Radner, Ancient Assyria: A Very Short Introduction (Oxford University Press, 2015), pp. 69–70. J. Neumann and S. Parpola, ‘Climatic change and the eleventh–tenth-century eclipse of Assyria and Babylonia’, Journal of Near Eastern Studies 46:3 (1987), 161–82; J. N. Postgate, The Land of Assur and the Yoke of Assur: Studies on Assyria, 1971–2005 (Oxford: Oxbow Books, 2007), pp. 52–3. K. Radner, ‘Royal pen pals: the kings of Assyria in correspondence with officials, clients, and total strangers (8th and 7th centuries BC)’ in Official Epistolography and the

158

Climate, Violence and Ethnic Conflict in the Ancient World

In the letter the governor proposes relocation of the Arabs ‘to the north’ in the already ‘harvested agricultural zone between the confluence of the Tigris and the Upper Zab’, in the province of Kalhu, ‘presumably because he knows the grazing conditions’ in the king’s proposed location ‘to be inadequate’.53 The new location should constrain movement between tributaries of the Tigris, but if grazing elsewhere did happen, the Arabs’ camps must at least remain in Kalhu, providing hostages should misbehaviour occur.54 Natural phenomena also influenced societies via overlapping psychologicalideological-religious pathways dependent upon prevailing belief systems, acting synergistically with material climate impacts. Even phenomena without physical impacts (e.g. eclipses) could act thus. In the ancient Near East, ‘astral’ omens, including extreme weather, earthquakes and celestial phenomena, portended difficulties for king and state. These might influence decision-making, potentially sparking or exacerbating internal and external conflict (though sometimes suppressing it). Professional astronomers were consulted for interpretation, often referencing similar past events and their apparent consequences. Divinators might provide more refined answers by directly questioning the gods,55 including over military decisions.56 Weather could play a further, indirect, role here, with divination by animal organs potentially reactive to nutritional condition and weather-related pathologies (e.g. liver fluke).57 While divination also followed a canon of interpretations, some flexibility existed to turn apparently negative findings positive.58 Nonetheless, kings and those who might rebel followed omens and divination closely,59 with subjects expected to report omens

53 55

56

57

58

59

Language(s) of Power, ed. S. Prochazka et al. (Vienna: Österreichische Akademie der Wissenschaften, 2015), pp. 61–72, at p. 69. Ibid., 63. 54 Ibid., 68–9. S. M. Maul, ‘Divination culture and the handling of the future’ and D. Brown, ‘Mesopotamian astral science’ in The Babylonian World, ed. G. Leick (London: Routledge, 2007), pp. 361–72 and 460–72 respectively. S. Kang, Divine War in the Old Testament and in the Ancient Near East (Berlin: De Gruyter, 1989); F. H. Cryer, Divination in Ancient Israel and its Near Eastern Environment (Sheffield Academic Press, 1994), pp. 124–86; K. Ulanowski, Neo-Assyrian and Greek Divination in War (Leiden: Brill, 2021). U. K. W. Holz, The Babylonian Liver Omens: The Chapters Manzazu, Padanu and Pan Takalti of the Babylonian Extispicy Series Mainly from Assurbanipal’s Library (Charlottenlund: Museum Tusculanum Press, 2000). F. M. Fales and G. B. Lanfranchi, ‘The impact of oracular material on the political utterances and political action in the royal inscriptions of the Sargonid dynasty’ in Oracles et prophéties dans l’antiquité, ed. J-G. Heintz (Paris: De Boccard, 1997), pp. 99–114. State Archives of Assyria (SAA) 10.179, http://oracc.org/saao/P237270 (accessed 11 April 2022); K. Radner, ‘The trials of Esarhaddon: the conspiracy of 670 BC’, Isimu 6 (2003), 165–83.

159

francis ludlow, chris morris and conor kostick

from all corners.60 Extreme weather, earthquakes and eclipses could be injurious and/or widely observed, and thereby hard to overlook or interpret favourably and hence politically sensitive.61 The king might thus resort to apotropaic rituals, as in c.675 when the exorcist Bel-le’i advised King Esarhaddon on offerings to alleviate drought and ensure an abundant harvest.62 The political sensitivity of omens arose not least because of the king’s divine mandate as expressed in early Mesopotamian creation literature and texts such as the prologue to King Hammurabi of Babylon’s (eighteenth-century) law codes. Here the king’s physical and mental prowess is deemed essential to capitalise upon the earthly bounty provided by the gods, achieved by building canals, suppressing banditry, and mitigating weather and conflict-driven food shortages by regulating prices: in sum, ensuring an orderly universe for his people.63 This kingly behaviour would be reciprocated through divinely ensured stability of seasons and clement weather. First-millennium (Neo-Assyrian) palace reliefs advertise the king’s claimed attainment of an orderly universe, depicting abundance and control through aqueducts and irrigation, with fruit groves and zoological parks hosting captured lions (Figure 5.1).64 As noted by Gebhard Selz, ‘every ruler . . . is bound to the images they propagate or create’,65 and we suggest that failure to maintain this kingly ideal, evidenced by the very fact (and material impacts) of abrupt climatic change and extreme weather, may have periodically generated what Jacqueline S. Solway terms ‘revelatory crises’. In these, ‘the interruption of socio-economic patterns [occurs] to a degree sufficient to lay bare contradictions in the existing order that had been latent or contained’,66 potentially creating a ‘state of exception’ facilitative of political transformation, being: ‘a state of such disruption of conventional routine that, as Solway writes, “actors are given license to innovate with social and moral ideological and 60

61

62 63

64

65

66

Radner, ‘Royal pen pals’, 70; F. M. Fales, ‘Massartu: the observation of astronomical phenomena in Assyria (7th century BC)’ in The Inspiration of Astronomical Phenomena VI, ed. E. M. Corsini (San Francisco: Astronomical Society of the Pacific, 2011), pp. 361–70, at pp. 367–8. In premodern China portents were politically sensitive, and being linked to the ‘mandate of heaven’ that sanctioned a dynasty’s rule contributed to justifications for revolt: D. W. Pankenier, ‘Parallel planetary astrologies in medieval China and inner Asia’, International Journal of Divination and Prognostication 1:2 (2020), 157–203. SAA 8.461, http://oracc.org/saao/P237874/ (accessed 11 April 2022). D. D. Luckenbill, The Annals of Sennacherib (Chicago: Oriental Institute, 1924), p. 5; T. Jacobsen, ‘The Eridu genesis’, Journal of Biblical Literature 100:4 (1981), 513–29; D. Frayne, ‘Šulgi, the runner’, Journal of the American Oriental Society, 103:4 (1983), 739–48. P. Collins, ‘Life at court’ and J. E. Reade, ‘The Assyrian royal hunt’ in I Am Ashurbanipal, ed. Brereton, pp. 34–51 and 52–79 respectively. G. J. Selz, ‘Power, economy and social organization in Babylonia’ in Babylonian World, ed. Leick, pp. 276–87, at p. 277. Solway, ‘Drought’, 471–2.

160

Climate, Violence and Ethnic Conflict in the Ancient World

Figure 5.1 Top panels of a carved gypsum relief from the North Palace in Nineveh, Iraq, 645–640 B C E, depicting irrigation of lands with an aqueduct. (Photo: Francis Ludlow)

behavioral codes” . . . sanctioning what would have previously been . . . antisocial behavior. This exception applies not only to the local community but also to its relations with the . . . state’.67 These concepts provide useful framing for understanding the climate–conflict pathways at play in the Ptolemaic Egypt and the Neo-Assyrian Empire, and for their transformative potential. The Assyrian king’s mandate also required imperial expansion to bring order to a chaotic periphery,68 frequently via conquest that royal accounts can describe in genocide-like terms, especially for nominal undesirables such as the Arameans and Suti.69 They, in words ascribed to Neo-Assyrian king Sargon II (r. 722–705), were ‘tent-dwellers, fugitives, treacherous, a plundering race’ that ‘I slew with arms’ to reverse a state in which ‘their rich meadows had become like a wilderness, their cultivated grounds . . . forlorn of the sweet harvest song’.70 Here is a sentiment similar to Ben Kiernan’s characterisation of the agrarian ideal or romanticism underlying much genocidal thinking, not least English 67 68

69

70

M. R. Dove, Anthropology of Climate Change (Chichester: Wiley, 2014), p. 18. G. Brereton, ‘The Neo-Assyrian Empire’, in I Am Ashurbanipal, ed. Brereton, pp. 100–17, at p. 105. Babylonians may be differentiated from the Assyrians and Akkadians in preferring peaceful achievements as legitimation: Selz, ‘Power, economy’, 279. Frayne, ‘Šulgi, the runner’, 742.

161

francis ludlow, chris morris and conor kostick

justifications for early modern plantations in Ireland that cited the immorality of wasteful Gaelic agriculture.71 Parallels exist here with Egyptian world views. Texts and battle reliefs from the reign of Ramses II (r. 1279–1213) and earlier in the Middle Kingdom depict lands beyond Egypt as chaotic and disordered, describing their peoples in dehumanised terms consistent with a belief in ‘Egyptian supremacy in a bipartite world constructed as “here” vs. “there”’.72 Even if not providing the same ideological drive for imperial expansion, this made brutal treatment of others potentially more permissible. It also informed perceptions of peoples migrating into Egypt to escape climatic adversity. The Prophecy of Neferti, dating from the early Middle Kingdom (around 2000) but purportedly describing an Old Kingdom prophecy from the reign of Sneferu (2613–2589), predicted ‘a catastrophe when the Asiatics will overrun Egypt’, after which ‘the nation will be saved by a king from the south . . . who will drive the Asiatics out’:73 An alien bird will breed in the . . . Delta, having made its nest upon its neighbors because the people allowed them to approach through want [hunger]. Destroyed indeed are those things of happiness – the fish pools which were full of people gutting fish, which overflowed full of fish and fowl. All happiness has fled, and the land is laid low with pain, because of these feeding Asiatics who roam the land.74

Other texts are less flattering still, and the perception of outsiders has particular relevance to revolt in Ptolemaic-era Egypt. Assyria often appears among possible ancient-era perpetrators of genocide (or genocide-like) behaviour,75 but the earlier-cited letter from Tab-sil-Esarra reveals an empire pragmatic and sometimes (at least tactically) tolerant. Evidence of administrative texts from thirteenth-century Egypt also display greater tolerance of pastoralists, allowing them passage to water ‘to keep them alive’.76 Of Assyria, 71

72

73

74 75

76

Ben Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007), pp. 177–87. G. Moers, ‘The world and the geography of otherness in Pharaonic Egypt’ in Geography and Ethnography: Perceptions of the World in Pre-modern Societies, ed. K. A. Raaflaub and R. J. A. Talbert (Chichester: Blackwell, 2010), pp. 169–81, at pp. 174–5. G. S. Holland, Gods in the Desert: Religions of the Ancient Near East (Lanham, MD: Rowman & Littlefield, 2009), p. 92. Moers, ‘World and geography’, 177. A. Jones, Genocide: A Comprehensive Introduction, 3rd ed. (London: Routledge, 2017); F. Chalk and K. Jonassohn, The History and Sociology of Genocide: Analyses and Case Studies (New Haven: Yale University Press, 1990); D. M. Crowe, War Crimes, Genocide, and Justice: A Global History (New York: Palgrave Macmillan, 2014). J. B. Pritchard, The Ancient Near East: An Anthology of Texts and Pictures, 2nd ed. (Princeton University Press, 2011), pp. 235–6.

162

Climate, Violence and Ethnic Conflict in the Ancient World

Figure 5.2 Bottom two panels of a carved gypsum relief from the North Palace in Nineveh, Iraq, 645–640 B C E, showing a Neo-Assyrian attack on an Arab encampment. (Photo: Francis Ludlow)

concern also exists over the veracity of quasi-propagandistic palace reliefs portraying brutal treatment of opponents.77 Figure 5.2 thus portrays Ashurbanipal’s reprisals against Arabs supporting Elam in the 650s–640s.78 Yet this does not preclude the reality of Assyrian state violence and mass slaughter.79 Tolerance may have been less likely (more costly) when climatic or other stress reduced resources and (simultaneously) an ideological framework existed that visually celebrated (even normalised) violence against others. Important nuance also exists in these depictions. Violence against women (e.g. Figure 5.2) is rare, such that ‘only in the wars against the Arabs are Assyrian soldiers shown actually manhandling women’.80 Royal annals and inscriptions feature (as oft-remarked) formulaic repetition and aggrandisement in charting Assyrian imperial 77

78

79

80

P. M. Taylor, Munitions of the Mind: A History of Propaganda from the Ancient World to the Present Era, 3rd ed. (Manchester University Press, 2003), pp. 19–24. J. Novotny, ‘Ashurbanipal’s campaigns’, in I Am Ashurbanipal, ed. Brereton, pp. 196–211, at pp. 206–11. Taking the southern Levant, a comparison of the fortunes of regions in the north (e.g. Northern Kingdom of Israel) that were annexed in the late eight century, versus those that maintained semi-independence in the south (e.g. Kingdom of Judah), using archaeological evidence, suggests the devastation of Assyrian conquest: A. Faust, ‘The southern Levant under the Neo-Assyrian Empire’ in Imperial Peripheries, ed. Tyson and Hermann, pp. 97–127, at pp. 112–15. G. van Driel, ‘The Mesopotamian north: land use, an attempt’, in Rainfall and Agriculture, ed. Jas, pp. 265–99, at p. 281.

163

francis ludlow, chris morris and conor kostick

expansion, but nuance exists here too that may reflect the reality of severity of treatment according to resistance offered.81 Some (accepting Assyrian dominion) are spared. Others have cities burned or turned into mounds and heaps with soldiers tortured before death. More infrequently there is mass killing of noncombatants, as in Khalzidipkha, whose nobility had not only revolted but marched against Ashurnasirpal II (883–859). For them: ‘three thousand captives I burned with fire; I left not a single one . . . alive to serve as a hostage . . . Their corpses I piled into heaps; their young men and maidens I burned in the fire’.82 The expanding Neo-Assyrian Empire was notably multi-ethnic, even within core provinces,83 one reason that ethnicity is largely disavowed in explaining the divergent treatment discussed above.84 Yet distinct ethnicities can (even as commonalities exist) exhibit distinct cultures, the Assyrian perception of which may have influenced decision-making. Different cultures may in return have found the outcomes (to the extent foreseeable)85 of Assyrian domination more or less culturally palatable (or economically and environmentally tenable), informing decisions over resistance. But whether the Assyrians enacted effective and comprehensive policies (and whether aimed mainly at elites) to enforce new cultural norms is debated, as is the nature of ‘Assyrian’ identity. In its interactions with external peoples, some scholars see compromise, toleration and co-acculturation,86 while others see ‘de-culturation’,87 even amounting to ‘cultural genocide’.88 Assyrian mass transplantation (and replacement) of subject populations is particularly notable (Figure 5.3). Estimates run as high as 4.5 million (approx.) transplanted over several centuries,89 despite some likely 81

82

83

84

85

86

87

88

89

M. Karlsson, Relations of Power in Early Neo-Assyrian State Ideology (Berlin: De Gruyter, 2016), p. 158. Annals of Ashurnasirpal II; E. A. Wallis Budge, Annals of the Kings of Assyria (1902; London: Routledge, 2005), pp. 291–2 (also pp. 269–70 for civilian slaughter). H. Tadmor, ‘The Aramaization of Assyria: aspects of western impact’ in Mesopotamien und seine Nachbarn, ed. H. J. Nissen and J. R. Renger (Berlin: Reimer, 1982), pp. 449–70. Karlsson, Relations of Power, p. 158: for whom Neo-Assyrian atrocities are not ‘ethnoculturally motivated’ and hence ‘annihilation’ of populations is not ‘genocide’. Probable, given the Neo-Assyrian reputation in external sources: P. Machinist, ‘Assyria and its image in the First Isaiah’, Journal of the American Oriental Society 103:4 (1983), 719–37. Guarducci, ‘Empire’, 87–9; Geoff Emberling, ‘Ethnicity in empire: Assyrians and others’ in A Companion to Ethnicity in the Ancient Mediterranean, ed. J. McInerney (Chichester: Wiley Blackwell, 2014), pp. 158–74. M. Liverani, The Ancient Near East: History, Society and Economy (London: Routledge, 2014), pp. 412–13, 417, 506. M. Freeman, ‘Genocide, civilization and modernity’, British Journal of Sociology 46:2 (1995), 207–23, at 218–22. B. Oded, Mass Deportation and Deportees in the Neo-Assyrian Empire (Wiesbaden: Reichert, 1979), pp. 20–1.

164

Climate, Violence and Ethnic Conflict in the Ancient World

Figure 5.3 Gypsum relief from the North Palace in Nineveh, Iraq, 645–640 B C E, showing the sacking of an Egyptian stronghold by Neo-Assyrians under King Ashurbanipal. (Photo: Francis Ludlow)

165

francis ludlow, chris morris and conor kostick

aggrandisement by the sources.90 The goals and consequences of these transportations are complex, most overtly punishment for refusal to accept subject status. Economic imperatives are also certain. Deportees participated in likely arduous infrastructure and monumental construction projects, and were critical to agriculture and economy even in core provinces.91 This was a situation that for Nicholas Postgate ‘can hardly have contributed to the stability of the empire’,92 but was required not least because of the almost annual mobilisation of the army as comprised of ‘Assyrians’ proper (namely, ‘the longer established sectors of the population’).93 Even if well treated en route (evidence on conditions is scarce)94 and resettled in relatively intact groups, the severance of tangible and intangible connections with one’s homeland was at least potentially disruptive of community and identity, making deportees easier to control and (deracinated) more readily integrated into Assyrian culture over several generations.95 We might read deliberate intent to accomplish this in Sargon II’s campaigning in Syria–Palestine, of which he (purportedly) proclaimed that ‘because of the sin which they committed [opposing Assyria] I tore them away from their homes’.96 For Simo Parpola, the very act of mass transportation ‘from one part of their empire to another . . . [represented] an effort to destroy any sense of ethnic identity other than that of being an Assyrian’.97 The initial and intergenerational status (socio-economic, political) of deportees is critical to further understanding, but sources limit us here;98 so too the finer-scale geography of displacement and 90

91 92

93 94

95

96 97

98

J. A. Brinkman, ‘Babylonia in the shadow of Assyria (747–626 B.C.)’ in Cambridge Ancient History, ed. Boardman, vol. I I I, part I I , pp. 1–69, at p. 65. Postgate, Land of Assur, pp. 202, 210. W. N. Postgate, ‘Employer, employee and employment in the Neo-Assyrian Empire’ in Labor in the Ancient Near East, ed. M. A. Powell (New Haven: American Oriental Society, 1987), pp. 257–70, at p. 265. Ibid. In a report to Sargon II, we read of deportees arriving: ‘The people are ve[ry] we[ak]; weather has eaten up [their] loo[ks] and the mountains have crushed them. They are coming ague-stricken.’ The fact of such reports does, however, evidence concern over the (physical) well-being of deportees in transport (G. B. Lanfranchi and S. Parpola, The Correspondence of Sargon II, Part II (Helsinki University Press, 1990), p. 117.) Liverani, Ancient Near East, p. 413; G. Beckman, ‘Foreigners in the ancient Near East’, Journal of the American Oriental Society 133:2 (2013), 203–15, at 210. Machinist, ‘Assyria’, 723. S. Parpola, ‘National and ethnic identity in the Neo-Assyrian Empire and Assyrian identity in post-empire times’, Journal of Assyrian Academic Studies 18:2 (2004), 5–22, at 6. Postgate, ‘Employer, employee’, 264, assumes for some initial period and in central provinces that they remained ‘unfree’, but would integrate over several generations (Beckman, ‘Foreigners in ancient Near East’, 210).

166

Climate, Violence and Ethnic Conflict in the Ancient World

resettlement. Ultimately it should be remembered that economic, political and cultural imperatives driving mass transportations need not be mutually exclusive, and other related questions persist. These include the impact of transplantations on the climatic vulnerability of those transported and incumbents, and on the sustainability of the expanding (potentially overexpanding) Neo-Assyrian Empire itself. Deportees were certainly used to drive imperial expansion into marginal (or productive but still environmentally unfamiliar) lands.99 As a consequence, climate–conflict pathways that merit further study involve a greater risk of weather-related harvest failure here, knock-on conflict with pastoralists, and motivation (or fewer apparent alternatives but) to revolt. Whether the expanding empire substantially incorporated environmental knowledge from transported or incumbent (also pastoralist) peoples is similarly an open question.100 The value of doing this is suggested by peasants on northern Iraqi smallholdings who, in the latter half of the twentieth century, were sufficiently resilient to cope with ‘one failed harvest out of four because of the integration of animal husbandry into the economic system’.101 The strain of managing incorporated territories and deportees could also generate tension.102 Bel-liqbi, governor of Aram-Zobah, a former Aramean state that likely incorporated the more arid (agriculturally challenging) northern part of the Beqaa valley (Lebanon), plus parts of western Syria,103 thus wrote to Sargon II, describing his interaction with an official who had taken grain without permission: ‘I went and remonstrated him, saying: “Why did you selfwilledly, [with]out the permission of the deputy (governor) open the king’s granaries?” He would not look me in the eye [but said]: “My (supply of) grass diminished . . . and horses keep coming to me; I c[an]not [cope]”.’104

99

100

101

102

103

104

K. Radner, ‘How did the Neo-Assyrian king perceive his land and its resources?’, in Rainfall and Agriculture, pp. 233–46, at pp. 238–9. Guarducci, ‘Empire’, for Assyrian policy towards local incumbents, and pp. 72–3 for environmental adaptations. M. D. Ellis, Agriculture and the State in Ancient Mesopotamia (Philadelphia: Babylonian Fund, 1976), p. 73. S. Parpola, ed., The Correspondence of Sargon II, Part I (Helsinki University Press, 1987), pp. 82–4, 140–3, 152–4, 199. D. V. Edelman, ‘Solomon’s adversaries Hadad, Rezon and Jeroboam: a trio of “bad guy” characters illustrating the theology of immediate retribution’ in The Pitcher Is Broken: Memorial Essays for Gösta W. Ahlström, ed. S. W. Holloway and L. K. Handy (Sheffield Academic Press, 1995), pp. 166–91, at p. 185. Parpola, Correspondence of Sargon II, Part I, pp. 142–3.

167

francis ludlow, chris morris and conor kostick

Using Natural Archives Available Archives Understanding of climatic contributions to ancient violence can benefit from independent weather-sensitive natural archives, though challenges include variable accuracy and precision (resolution) in their dating and spatial coverage, coverage of certain seasons and climatic parameters only, or lack of clarity over the seasonality of climate information they have preserved.105 Efforts since the 1970s to explore environmental roles in ancient-era societal transitions or discontinuities thus faced notable challenges in establishing correspondences between environmental and societal changes.106 However, the past two decades have seen dramatic improvement in the quality and coverage of natural archives. Sedimentary archives and speleothems (stalagmites, stalactites) have been exploited to identify periods of social change and major events (e.g. the Late Bronze Age Collapse107 or the fall of the NeoAssyrian Empire)108 apparently coincident with climatic changes deemed impactful on water resources, landcover and subsistence in ways potentially generative of conflict and migration. Many studies of premodern societal or state collapse can be critiqued for simplifying and compressing (often due to limitations in the resolution of evidence) composite episodes in history that in reality comprised an interplay of many events and processes, including human decisions and acts.109 This critique may resonate with those who (seeking to understand past genocides and prevent future ones) would focus on ‘the human agency behind . . . allegedly immutable forces of history’.110 Such forces (including climate) for us offer vital context to understanding social phenomena, mass violence included, but we agree that evidence must be studied on as fine a temporal 105 106

107

108

109

110

Bradley, Paleoclimatology. For example B. Bell, ‘Climate and the history of Egypt: the Middle Kingdom’, American Journal of Archaeology 79:3 (1975), 223–69. B. L. Drake, ‘The influence of climatic change on the Late Bronze Age collapse and the Greek Dark Ages’, Journal of Archaeological Science 39:6 (2012), 1862–70; A. B. Knapp, and S. W. Manning, ‘Crisis in context: the end of the Late Bronze Age in the eastern Mediterranean’, American Journal of Archaeology 120 (2016), 99–149. A. W. Schneider and S. F. Adali, ‘“No harvest was reaped”: demographic and climatic factors in the decline of the Neo-Assyrian Empire’, Climatic Change 127:1 (2014), 435–46; A. Sołtysiak, ‘Drought and the fall of Assyria: quite another story’, Climatic Change 136 (2016), 389–94. J. Haldon et al., ‘Demystifying collapse: climate, environment, and social agency in pre-modern societies’, Millennium 17:1 (2020), 1–33. K. Roth, ‘Foreword’ to Annihilating Difference: The Anthropology of Genocide, ed. A. L. Hinton (Berkeley: University of California Press, 2002), pp. ix–xi, at p. x.

168

Climate, Violence and Ethnic Conflict in the Ancient World

scale as possible if causality and human agency is to be convincingly assessed. While much of the Mediterranean and Near East lacks ultra-high-resolution archives such as tree rings, the study of polar ice cores has now achieved sufficient precision (sub-annual resolution) to identify individual climate ‘forcing’ events, as well as the accuracy to date these within known uncertainties ranges. A recent 2,500-year history of explosive volcanism has thus corrected long-standing chronological errors that prevented recognition of linkages between explosive eruptions and events of the medieval period and earlier,111 including famine, disease, warfare and collapse.112

Hydroclimate and Revolt in Ptolemaic Egypt Explosive eruptions that inject large volumes of sulphate aerosols into the northern hemispheric stratosphere can diminish the annual summer flooding of the Nile, mainly by altering the hemispheric summertime heat imbalance that drives the northward migration of the Intertropical Convergence Zone and associated monsoon winds that carry spring–summer rainfall to the Ethiopian Highlands, where the Blue Nile rises.113 With Egyptian flood recession agriculture driven by this summer flooding, Egypt is an excellent laboratory for historical human–environmental entanglements, as demonstrated by Egypt’s well-documented Ptolemaic era, 305–30, when it fell under the control of a dynasty founded by Ptolemy I Soter, formerly a key general in Alexander the Great’s empire. Egypt at this time achieved much in material and intellectual terms, including the founding of Alexandria (one of the largest cities of the ancient world), boasting the great Lighthouse and Library, but also waged costly episodic wars with other part-inheritors of Alexander’s empire.114 Also characteristic of the period are multiple internal revolts against Ptolemaic rule, but these remain obscure in motivation, chronology, scale and geographical scope.115 Scholars of genocide have often focused on ethnicity as an element in the extreme ‘othering’ underlying genocides. Whether ethnicity was critical to the Ptolemaic-era revolts has similarly been a focus, with an 111

112

113

114 115

M. Sigl et al., ‘Timing and climate forcing of volcanic eruptions for the past 2,500 years’, Nature 523 (2015), 543–9. For example C. Gao et al., ‘Volcanic climate impacts can act as ultimate and proximate causes of Chinese dynastic collapse’, Communications Earth & Environment 2 (2022), article 234, https://doi.org/10.1038/s43247-021-00284-7. L. Oman et al., ‘High-latitude eruptions cast shadow over the African monsoon and the flow of the Nile’, Geophysical Research Letters 33:18 (2006), L18711. J. D. Grainger, The Syrian Wars (Leiden: Brill, 2010). For an overview see A.-E. Veïsse, Les ‘révoltes égyptiennes’: recherches sur les troubles intérieurs en Égypte du règne de Ptolémée III à la conquête romaine (Leuven: Peeters, 2004).

169

francis ludlow, chris morris and conor kostick

hypothesised ideological (even proto-nationalistic) resentment against an imposed Greek (Hellenic) elite attributed varying importance.116 One of the most significant cases is the Great Theban Revolt, c.207 – c.186, when Ptolemy IV Philopater lost control of large parts of Upper (southern) Egypt. These events are summarised from a pro-Ptolemaic standpoint in a multilingual priestly decree from the Temple of Isis at Philae: The rebel against the gods . . . he who had made war in Egypt, gathering insolent people from all districts on account of their crimes, they did terrible things . . . they desecrated [?] the temples . . ., they molested [?] the priests and suppressed [?] the offerings on the altars and in the shrines. They sacked [?] the towns and their population, women and children included, committing all kinds of crimes in the time of anarchy. They stole the taxes of the nomes, they damaged the irrigation works.117

Was this an ethnically charged crisis sparked by the hydroclimatic ‘shock’ from a major tropical eruption just two years earlier, in 209? Evidence suggests the revolt was led by Egyptians,118 Herwennefer and Ankhwennefer, who in turn took the title of Pharaoh, consistent with ethnicity as one causal context. The scale and manner of the violence is difficult to ascertain, but texts describe notable land abandonment in some areas (accredited to killing of a majority of farmers, though by whom is not explicit), and imply a government policy of enslaving and auctioning surviving rebels and their families.119 As a key general, Ptolemy I had been at least complicit in brutal treatment (deemed genocidal by some)120 of certain peoples who resisted Alexander’s rule. But claims that the Ptolemies created in Egypt what amounted to an apartheid state that wrought ‘cultural genocide’ in favour of Greek custom have been discounted,121 with considerable 116

117

118

119 120

121

For example W. D. Barry, ‘The crowd of Ptolemaic Alexandria and the riot of 203 B.C.’, Echos du monde classique 37:3 (1993), 415–31; B. C. McGing, ‘Revolt Egyptian style: internal opposition to Ptolemaic rule’, Archiv fur Papyrusforschung und Verwandte Gebiete 43:2 (1997), 273–314; C. Fischer-Bovet, ‘Social unrest and ethnic coexistence in Ptolemaic Egypt and the Seleucid Empire’, Past & Present 229:1 (2015), 3–45. Translation: W. Clarysse, ‘The great revolt of the Egyptians (205–186 BC)’ (Berkeley: Center for the Tebtunis Papyri, 2004), pp. 8–9, www.lib.berkeley.edu/sites/default/ files/files/TheGreatRevoltoftheEgyptians.pdf (accessed 11 July 2021). P. W. Pestman, ‘Harronophris and Chaonnophris: two indigenous pharaohs in Ptolemaic Egypt (205–186 B.C.)’ in Hundred-Gated Thebes: Acts of a Colloquium on Thebes and the Theban Area in the Graeco-Roman Period, ed. S. P. Vleeming (Leiden: Brill, 1995), pp. 101–38. Clarysse, ‘Great revolt’, 4, 12. I. Worthington, Ptolemy I: King and Pharaoh of Egypt (Oxford University Press, 2016), p. 47. J. G. Manning, The Last Pharaohs: Egypt under the Ptolemies, 305–30 BC (Princeton University Press, 2010), p. 64.

170

Climate, Violence and Ethnic Conflict in the Ancient World

continuities of pre-Ptolemaic practice (including the Ptolemaic adoption of the divine mantle of Pharaoh, taking credit for fertility brought by the Nile flood, but also responsibility for years of insufficiency).122 Unsurprisingly, therefore, the Ptolemies attempted to prevent subsistence crises. An example well advertised in notable collaboration with the Egyptian priestly caste through the multilingual (Greek, hieroglyphic and demotic Egyptian) Canopus decree of 238, relates how, on one occasion the rise of the river was insufficient and all the inhabitants of the country were terrified at what had happened and remembered the disaster that occurred under some of the previous kings, under whom it happened that all the people living in the land suffered from a drought, they [Ptolemy III and his wife, Berenice] showed their care for the residents in the temples and the other inhabitants of the country . . . and sacrificed a large part of their revenues for the salvation of the population, and by importing corn into the country . . . at great expense, they saved the inhabitants of Egypt . . .123

Yet the Ptolemies did enact changes that heightened stress. These included emphasising cultivation of drought-intolerant (versus customary barley) freethreshing wheat and a more intensive tax collection regime. Such policies likely reduced buffers against insufficient flooding, supporting arguments that revolts were driven by economic and subsistence concerns.124 But ethnicity, in being accompanied by preferential taxation for Greeks and (later) others involved ‘with the teaching and practice of Greek culture, with Greek education and literacy’,125 may have been a potent ingredient in revolts. Egyptians were certainly aware of their customary dominion over the Nile valley, reflected in sources such as the second section of the Prophecy of Neferti. Likewise, the famous Oracle of the Potter suggests the mindset of some of the Egyptian priestly caste in the Ptolemaic era, even if the Ptolemies frequently had their support, as in decrees such as Canopus.126 The text’s 122

123

124

125

126

Ptolemaic policy and the interactions between new and old elites is more fairly deemed ‘pragmatic’ and ‘transactional’: I. S. Moyer, Egypt and the Limits of Hellenism (Cambridge University Press, 2011), pp. 2, 35, 40, 275, 281. M. M. Austin, The Hellenistic World from Alexander to the Roman Conquest: A Selection of Ancient Sources in Translation, 2nd ed. (Cambridge University Press, 2006), 471. F. Ludlow and J. G. Manning, ‘Revolts under the Ptolemies: a paleoclimatological perspective’ in Revolt and Resistance in the Ancient Classical World and the Near East, ed. J. J. Collins and J. G. Manning (Leiden: Brill, 2016), pp. 154–71. Manning, Last Pharaohs, p. 143; W. Clarysse and D. J. Thompson, Counting the People in Hellenistic Egypt, vol. I I: Historical Studies (Cambridge University Press, 2006), p. 124 (who also note (p. 125) that the description of ‘Hellenes is not always ethnically straightforward’). Temples in turn benefitted from reduced tax and more: Clarysse and Thompson, Counting the People; Manning, Last Pharaohs.

171

francis ludlow, chris morris and conor kostick

original composition may be placed in the 130s (or perhaps the third century) and expresses anti-Greek sentiment within an ancient prophetic framework (similar to the Prophecy of Neferti) that proceeds to foretell the coming of a legitimate (i.e. Egyptian) king.127 The excerpt that follows details the impacts of illegitimate rule, with injustice and a disordered cosmos. ‘Typhonians’ and ‘Girdleweavers’ refer to Greeks and Graeco-Macedonian military dress, respectively.128 The river, since it will not have sufficient water, will flood, but only a little so that scorched will be the land . . . but unnaturally. For in the time of the Typhonians people will say ‘wretched Egypt you have been maltreated by the terrible malefactors who have committed evil against you’. And the sun will darken as it will not be willing to observe the evils in Egypt. The earth will not respond to seeds . . . The farmer will be dunned for taxes for what he did not plant. There will be fighting in Egypt because people will be in need of food. What one plants, another will . . . carry off. When this happens, there will be war and slaughter which will kill brothers and wives . . . For in the time of the Typhonians the sun will darken to highlight the character of the evils and to reveal the greed of the Girdleweavers . . . Speech of the potter to king Amenophis [Hellenisation of Amenhotep, Eighteenth Dynasty, 1550/49–1292] . . . concerning what will happen in Egypt.129

In assessing the contribution of climate in a context of ethnic tension, we might ask whether indigenous Egyptian dissatisfaction with this new elite (and their own relative status) was alone sufficient to drive revolts, was necessary but insufficient (thereby requiring additional contributions, plausibly also from hydroclimatic stress), or not always or ever necessary (implying a constellation of other causal contributions, acting either in isolation, in synergy, or antagonistically). But the above evidence already suggests the interdependence of ethnic and cultural dimensions and tensions, socioeconomic change and stress, and impacts of hydroclimatic shock, making this framing somewhat reductive. Ptolemaic-era socio-economic changes promoted vulnerability to Nile failure in a way that spread the burden unevenly along ethnic (and class) lines. The Pharaoh’s responsibility was ‘to oversee the cycle of the Nile’s ebb and flood and . . . ensure . . . the 127

128

129

Or ‘anti-Greek propaganda’: P. Green, Alexander the Great and the Hellenistic Age (London: Weidenfeld & Nicolson, 2007), p. xxxv. F. Ludlow and J. G. Manning, ‘Volcanic eruptions, veiled suns, and Nile failure in Egyptian history’ in Climate Change and Ancient Societies in Europe and the Near East, ed. P. Erdkamp et al. (London: Palgrave Macmillan, 2021). Normalised excerpt from S. M. Burstein, The Hellenistic Age from the Battle of Ipsos to the Death of Kleopatra VII (Cambridge University Press, 1985), p. 137.

172

Climate, Violence and Ethnic Conflict in the Ancient World

rhythms of nature necessary for the well-being of the land’.130 The fact of insufficient Nile flooding may have been particularly problematic for nonindigenous pharaohs whose legitimacy was already more tenuous, even with the mitigating measures claimed in the Canopus Decree. When driven by volcanism the potency of poor Nile flooding as an ethnically charged reflection of divine displeasure (and potential ‘revelatory crisis’) may have been compounded by ominous phenomena such as a darkened solar disc resulting from light interacting with high-altitude volcanic aerosols (‘dust veils’), as famously observed after the 1883 eruption of Krakatoa, Indonesia.131 Ice cores now reveal the Ptolemaic era as coterminous with a notably volcanically active era, certainly relative to the volcanically ‘quiescent’ Roman era of the next two centuries (roughly).132 At least twenty-four notable tropical and extratropical northern hemispheric eruptions are identifiable through the sulphate deposited in polar ice between 305 and 30, with four estimated to have surpassed the global radiative forcing of the major 1991 Pinatubo (Philippines) eruption, including one of the largest eruptions of the past 2,500 years in circa 43.133 Ptolemaic-era environmental conditions may thus have informed the Oracle of the Potter, including the conspicuously darkened sun. Indeed, court astronomers in Babylon record multiple instances wherein the ‘disk of the sun looked like that of the moon’ that align with ice-core-based eruption dates for the period.134 Joseph G. Manning and colleagues have studied the coincidence in timing between ice-core-based dates of explosive volcanism and Egyptian internal revolt onset.135 Given the preceding discussion, it is perhaps unsurprising that revolts frequently started in eruption years, or shortly thereafter. Of ten onset dates studied, three occurred in eruption years and a further five within two years following. This includes the Great Theban Revolt, preceded by the notable tropical eruption of 209, while the Canopus Decree of 238 followed the volcanically perturbed 240s with major tropical and extratropical (northern hemispheric) eruptions in 247 and 244, respectively. With the Ptolemaic era spanning almost twenty-seven decades, revolt onsets and eruptions occur 130 132 133

134 135

Holland, Gods in the Desert, p. 39. 131 Ludlow, ‘Volcanic eruptions’. Sigl et al., ‘Timing’. Third largest in polar sulphate deposition and hence potential climatic impact. Sigl et al., ‘Timing’; J. R. McConnell et al., ‘Extreme climate after massive eruption of Alaska’s Okmok volcano in 43 and effects on the late Roman Republic and Ptolemaic kingdom’, Proceedings of the National Academy of Sciences 117:27 (2020), 15443–9. Sigl et al., ‘Timing’. J. G. Manning et al., ‘Volcanic suppression of Nile summer flooding triggers revolt and constrains interstate conflict in ancient Egypt’, Nature Communications 8 (2017), article 900.

173

francis ludlow, chris morris and conor kostick

in only a fraction of those years. Their coincidence in timing thus feels intuitively unlikely to be random. Indeed, the formal probability of so many revolts beginning so close in time to eruptions was found to be less than 5 per cent (if occurring randomly).136 On this evidence, hydroclimatic shocks played a repeated role in revolt in Ptolemaic Egypt. The precise nature of the pathways along which these shocks propagated remains to be delineated, but appears likely to have involved socio-economic stresses sharpened with an ethno-religious edge.

Volcanic Hydroclimatic Shocks and the Neo-Assyrian Empire One historian has called the ancient Assyrians ‘butchers of subject nations’.137 Available evidence suggests many pathways by which extreme weather and ‘ominous’ phenomena might have promoted violence in the ancient Near East (see above). We thus follow Manning and colleagues in using explosive eruptions as ‘tests’ of societal sensitivity and response to sudden hydroclimatic shocks. Compared to Egypt, results are less certain because of greater uncertainties regarding volcanic hydroclimatic impacts here, and in the available human and natural archives, but are still highly suggestive. Much less work has been undertaken on responses of the Tigris and Euphrates than of the Nile, with access to streamflow data complicated by regional hydropolitics. The global average response to tropical eruptions is decreased streamflow, with reduced temperatures leading to drying because of lesser evaporation and rainfall. Atmospheric circulation and associated moisture-bearing winds are, however, also impacted by volcanism in ways that can regionally reinforce (as per the summer monsoon and Nile flooding) or counteract this drying.138 The Tigris and Euphrates have their source in the Anatolian highlands (Turkey) northwest of Mesopotamia (Syria, Iraq). The volume and timing of floodwater is controlled by spring rainfall and melting of accumulated winter snow. Much of this accumulation is governed by the strength and positioning of the winter westerlies that sweep moisture-laden air east from the Atlantic and Mediterranean.139 This process is influenced by 136

137

138

139

Simplifying but capturing the essentials of Manning, ‘Volcanic suppression’. Their testing assumes little more than independence of volcanic and revolt dates. Alternative counts of revolts are possible, but the observed non-random association is thus far insensitive: Ludlow and Manning, ‘Revolts under the Ptolemies’. R. I. Rotberg, ‘Genocide and ethnic cleansing: our global past’, Journal of Interdisciplinary History 48:1 (2017), 71–8, at 72. C. E. Iles and G. C. Hegerl, ‘Systematic change in global patterns of streamflow following volcanic eruptions’, Nature Geoscience 8 (2015), 838–42. H. M. Cullen and P. B. deMenocal, ‘North Atlantic influence on Tigris–Euphrates streamflow’, International Journal of Climatology 20:8 (2000), 853–63.

174

Climate, Violence and Ethnic Conflict in the Ancient World

volcanism, and the Near East (including Anatolia) has seen increased precipitation after several modern-era eruptions.140 This response will depend upon eruption location, however, with reduced flooding also conceivable. Coping with variability (timing and magnitude) of the spring–summer flood has always been challenging, such that extensive networks of dykes, storage basins and canals have been required,141 and flooding too late, little or voluminous could destroy crops. Indeed, in the late thirteenth century B C E the threat of the Tigris altering course, possible with exceptional streamflow, purportedly required Assyrian king Ashur-nadin-apli to use his special connection to the divine to avert disaster.142 We can (at the least) posit that explosive volcanism will have compounded the hydroclimatic variability that Mesopotamian societies experienced, both in more northern regions where rain-fed agriculture (dry farming) was possible, and in more southerly regions where irrigation was essential. Volcanic atmospheric optical phenomena were also likely observed (see earlier), and climate–conflict pathways involving hydroclimatic shocks were likely reinforced by interrelated religious– psychological–ideological dimensions. One of the best-documented periods of ancient Near Eastern history is that of the Neo-Assyrian Empire (911–609), with royal annals, stone reliefs and thousands of clay tablets surviving from palaces at Nineveh. Many are accessible in translation,143 offering a data set sufficiently rich to discern patterns of response to climatic variability. Such a data set is in preparation, available now as an index of conflict and other societal stressors for the wellcovered 750–650 period.144 It comprises an annual tally (when dating is sufficient) of major wars; minor wars; civil wars; famine; epidemics; locust outbreaks; drought; cold; harmfully high or unseasonable flooding; diminished sunlight; forced migrations; and two proxies for sociopolitical disruption: indications that no New Year ceremony occurred, and years in which the king availed of a ‘substitute’. These total 115 events or instances over the 101 years surveyed. While less comprehensive than the past 2,500 years when multiple Antarctic and Greenland ice cores allow a combined volcanic history, our

140 142

143 144

Iles, ‘Systematic change’, 2. 141 Holland, Gods in the Desert, p. 99. S. Bertman, Handbook to Life in Ancient Mesopotamia (New York: Facts on File, 2003), pp. 79, 202–3. See the 21-volume State Archives of Assyria, http://oracc.museum.upenn.edu/saao/. Part of the Irish Research Council (IRCLA/2017/303) Climates of Conflict in Ancient Babylonia (CLICAB) project. This data set (beta, v.0.9) is available by request.

175

francis ludlow, chris morris and conor kostick

knowledge of the full Holocene era (past c.11,000 years) has now been advanced by sulphate measurements from Antarctica using a chronology that appears accurate to within plus or minus two years for our period.145 These measurements identify notable sulphate deposition from eruptions dated 750, 723, 703 and 676. The 723 event in particular exceeds the (presumed) deposition from the great 1600 (Huaynaputina, Peru), 1884 (Krakatoa, Indonesia) and 1991 (Pinatubo, Philippines) tropical eruptions in the same core. All are known for environmental and human impacts. Without equivalent data from Greenland, we cannot determine whether all these events are tropical (able to deposit sulphate into both polar regions for a more global impact) or occurred in the extratropical southern hemisphere (depositing sulphate only in Antarctica). Preliminary Greenland data does suggest the 723 eruption was tropical.146 Regardless, even eruptions with aerosols confined to a single hemisphere are increasingly recognised as exerting an indirect ‘dynamical’ impact on the opposing hemisphere’s climate, and all four eruptions may have impacted Mesopotamia.147 To examine whether eruptions provoked any response visible in the available Neo-Assyrian sources, in Figure 5.4 we employ a ‘superposed epoch analysis’ that averages the number of conflicts and other documented stresses across the four eruption years (top panel), and each of the twenty years before and after, to provide a sense of stress frequencies before any volcanic-related shock, and the timing and persistence of post-eruption responses. Frequencies vary without sustained pattern (as expected) before our eruption dates, but increase notably from the first post-eruption year, remaining generally elevated for almost a decade. Statistical ‘significance’ testing suggests the likelihood that such elevated values occurred by chance is less than 10 per cent for the second, fifth and sixth post-eruption years (exceeding the solid horizontal line in Figure 5.4), with the first, fourth and seventh to ninth years being only marginally less unlikely (less statistically significant).148 145

146 147

148

J. Cole-Dai et al., ‘Comprehensive record of volcanic eruptions in the Holocene (11,000 years) from the WAIS divide, Antarctica ice core’, Journal of Geophysical Research: Atmospheres 126:7 (2021), e2020JD032855. Provisional determination by M. Sigl, with thanks. Z. Zhuo et al., ‘Proxy evidence for China’s monsoon precipitation response to volcanic aerosols over the past seven centuries’, Journal of Geophysical Research: Atmospheres 119:11 (2014), 6638–52. Statistical significance values are estimated by assessing the size of the observed (normalised) average frequency values relative to a randomly generated (Monte Carlo) reference distribution. Values in Figure 5.4 (top) exceeding the line are thus larger than those observed in 90 per cent of the 10,000 randomly generated averages (when drawing from the 101-year normalised time series of event frequencies). For more on this approach see Campbell, ‘Climate’, 250–1.

176

Climate, Violence and Ethnic Conflict in the Ancient World

Figure 5.4 Levels of conflict and sociopolitical stress recorded in Neo-Assyrian sources in the years before and after the dates of four volcanic eruptions, 750–650 B C E. Normalisation compensates for potential bias arising from changing record availability and allows more direct comparison of responses between eruptions. Event frequencies are hence given as Z-scores and can vary above and below zero (i.e. a value below zero in any given year means the average frequency in that year was below the combined average frequency of events in the preceding thirty years).

177

francis ludlow, chris morris and conor kostick

Examined individually (Figure 5.4, bottom), the timing and scale of response is variable after each eruption, a feature lost when viewing only the overall average. Some variability is certainly expected given remaining (small) uncertainties in our eruption dates and the conflict and stress dates in the surviving sources. The exact hydroclimatic impact of each eruption requires additional study and will also likely have influenced the timing and scale of response, as might the stressors and mitigation mechanisms prevailing before each eruption.149 Longer-term trends in climate (on multi-decadal and greater timescales) may also amplify or dampen the impact of any given volcanic eruption. Nonetheless, the consistent increase in post-eruption frequencies suggests ongoing vulnerability to hydroclimatic shocks and the existence of active climate–conflict pathways.150 We may gain some further insight by examining events following the largest eruption in our period, that in 723. These included the fall of the Northern Kingdom of Israel (c.720) with the Neo-Assyrian conquest of Samaria. This has seen considerable study, but with the challenge of reconstructing a narrative from limited and somewhat contradictory sources, no firm consensus exists regarding the course of events.151 What is likely is that throughout much of his short reign (727–722), Neo-Assyrian king Shalmaneser V campaigned in the region, with the Babylonian Chronicle (I 27–28) stating that he ‘ravaged Samaria’. Then, in 722, a revolution erupted in Assyria itself, unseating Shalmaneser. Upon his death, Sargon II (r. 722–705) ascended. After consolidating authority internally, he sent troops to Samaria where a coalition of Westland powers had rallied in rebellion behind Yau-bi’di of Hamath. The resulting NeoAssyrian victory led to the Northern Kingdom of Israel’s collapse and mass deportations of the Jewish population.152 Until now these events – revolution in Assyria, fall of the Northern Kingdom of Israel – have not been thought to bear any relationship to climate. Bob Becking writes that ‘Climate in the Iron Age II–III period remained stable in ancient Israel. We can therefore assume that no specific impulses from a (sudden) change in climate would have influenced the course of events 149 150

151

152

Gao et al., ‘Volcanic climate impacts’. Future data set versions will cover a longer period and differentiate phenomena (including violence by scale and character, e.g. battles or one-sided mass killings) to trace distinct climate–conflict pathways and determine whether the intensity of violence scaled according to the magnitude of the hydroclimatic shock, pre-existing stressors and other contexts. Recap of debates: S. Hasegawa, ‘The last days of the northern Kingdom of Israel’ in The Last Days of the Kingdom of Israel, ed. S. Hasegawa et al. (Berlin: De Gruyter, 2019), pp. 1–16. E. Frahm, ‘Samaria, Hamath, and Assyria’s conquests in the Levant in the late 720s: the testimony of Sargon II’s inscriptions’ in Last Days, ed. Hasegawa et al., pp. 55–86, at p. 71.

178

Climate, Violence and Ethnic Conflict in the Ancient World

leading to the end of the kingdom.’153 A focus on lower-resolution archives that capture mainly longer-term climatic changes can certainly suggest stability, but ice core evidence now reveals years of likely significant volcanic climate disruption. Severe cold is certainly recorded for Sargon II’s reign, though dated imprecisely. Evidence includes Urzana, king of the east Kurdistan realm of Mus¸as¸ir, excusing his failure to bring tribute to the Assyrian king because ‘snow has blocked the roads’,154 and an unidentified servant of the king in the company of a royal bodyguard explaining that they would be unable to send saplings to Dur-Šarruken, as asked, until the month of Adar (roughly March) as ‘there is much snow and ice’.155 Such conditions are consistent with the multiyear effects of a major tropical eruption on the apparent scale of that in 723. Of conflict, the Assur Charter (a pro-Sargon source) suggests internal revolt arose from Shalmaneser’s imposition of tax and corvée on the cities of Ashur and Harran. If new taxes coincided with extreme weather and corresponding agricultural difficulties, then their efficacy as a motive for a revolt sufficient to overthrow the king may have been enhanced. Viewed from the west, an Assyria weakened by civil war and harsh weather could partly explain why Yau-bi’di’s rebellion might have gained greater support than at less challenging times. This would regardless prove a miscalculation: with consolidation of his position and the easing of the weather, Sargon II was soon able to devastate the Westland.

Conclusions Every conflict, act of violence, of any scale, in any period or region, occurs within an environmental context. This includes nominally straightforward (usually unremarked) ‘fair weather’ or stable conditions that passively facilitated certain actions, or could at least be overcome given prevailing technologies and environmental adaptations. Military historians have long recognised the importance of the reverse, in which interactions with adverse weather, or environments poorly planned for or adapted to, actively contributed to the outcome of military operations.156 Stating more, that climate plays a causal role in violence and conflict, remains controversial. Relegating climate (and environment) to at best 153

154 156

B. Becking, ‘How to encounter an historical problem? – “722–720” as a case study’, in Last Days, ed. Hasegawa et al., pp. 17–34, at p. 23. SAA 5.146. 155 SAA 5.105, http://oracc.org/saao/P334543/ (accessed 11 April 2022). H. A. Winters et al., eds., Battling the Elements: Weather and Terrain in the Conduct of War (Baltimore: Johns Hopkins University Press, 1988), http://oracc.org/saao/P334369/ (accessed 11 April 2022).

179

francis ludlow, chris morris and conor kostick

factors of secondary importance fails to recognise the inseparability of humans and their environment,157 and therefore the complementarity of assessing both climatic and human origins of violence and conflict. Of genocide, recognising a role for climate in no way diminishes the role of human agency and individual volition (or culpability), but can instead help to better understand how such violence occurs and may perhaps even be foreseen, in a way that will only become more important as anthropogenic climate change and its impacts intensify. This chapter has sought to demonstrate a role for climate (abrupt changes, extreme weather) in state-enacted violence in the ancient world, focusing mainly upon activities in the Near East, as well as in violent internal revolt in Ptolemaic Egypt. Material agricultural impacts of extreme weather are most often cited as triggers for violence of both sorts. States are pressed into (at times violent) territorial expansion by resource needs, or opportunistically attack weakened rivals, while citizens or subjects facing starvation migrate to urban areas and/or revolt against their rulers. We additionally emphasise how the perception of natural phenomena plays a key role in both Egypt and the Near East, part of an ideological framework in which climatic variability and human actions were in dialogue. Here, environmental crises might communicate something ‘revelatory’ about current systems and circumstances to elites and even commoners, and promote a ‘state of exception’ whereby these systems and circumstances might be questioned and transformed (including by violent action). The increasing availability of high-quality natural archives allows us to better glimpse this dialogue and to identify and characterise underlying climate–conflict pathways. Revolts in Ptolemaic Egypt no longer represent static event dates, but are now dynamic, seen occurring relative to dates of explosive volcanism that likely induced materially, ethnically and symbolically impactful diminishment of the Nile flood. Of the Neo-Assyrians, the newly available dates of major explosive volcanism similarly reveal a society responding to climatic stress, including through violence. Many potential pathways are in operation here that await further illumination.

Bibliographic Note The study of climate–conflict linkages is inherently interdisciplinary, drawing upon multiple literatures in the sciences and social sciences. When focusing 157

For example R. Gillespie, ‘Climate, weather and social change in seventeenth-century Ireland’, Proceedings of the Royal Irish Academy 120C (2020), 253–71, at 271.

180

Climate, Violence and Ethnic Conflict in the Ancient World

upon historical cases, this requires additional period and region-specific literatures. Up-to-date work on climatology and palaeoclimatology is mainly journal-focused, but Bradley’s 2015 edition of Paleoclimatology: Reconstructing Climates of the Quaternary provides an accessible overview of human and natural archives globally. Consulting earlier editions (1999, 1985) allows one to trace developments. The climate–conflict literature is also journal-centric, but reviews appear periodically. Valuable is Homer-Dixon’s 1999 monograph, Environment, Scarcity, and Violence. Hsiang and Burke’s ‘Climate, conflict and social stability’ attempts to distil firm conclusions from a fractious literature. Studies linking climate to past or possible future genocides are rare, Alvarez’s Unstable Ground being a notable exception. A valuable recent treatment of climatic change and human response across the longue durée is John L. Brooke’s Climate Change and the Course of Global History: A Rough Journey (Cambridge University Press, 2014). More specific to the ancient Mediterranean environment and its Roman, Greek, Egyptian and Near Eastern intersections is Joseph G. Manning’s The Open Sea: The Economic Life of the Ancient Mediterranean World from the Iron Age to the Rise of Rome (Princeton University Press, 2018). Dove’s Anthropology of Climate Change is an excellent collection of articles that have advanced understandings of human– environmental relations. Chapters are introduced by Dove, and here he provides a novel interweaving of Jacqueline Solway’s articulation of the ‘revelatory crisis’ with Giorgio Agamben’s ‘state of exception’, together offering a conceptual framework by which to understand how crises precipitated by extreme weather and other disasters catalyse and make room for societal transformation that at times proves violent.

181

part ii *

THE ANCIENT WORLD

6

Genocide in Ancient Israelite and Early Jewish Sources t. m. lemos Introduction Narratives of mass violence not only frame the biblical corpus but also recur through it as a coda of destruction. In Deuteronomy 7 and 20, the Israelites are commanded to perform genocide, and this is precisely what they do in the book of Joshua, a narrative in which the chosen people of Yahweh enter the promised land of Canaan and slaughter its indigenous inhabitants wholesale. Yet this narrative is in fact eclipsed in the scale of its violence by the flood story in Genesis and the cosmic, eschatological destruction of some early Jewish literature. This chapter will explore the pervasiveness of narratives of genocide and mass violence in Israelite and early Jewish sources and relate this genocidal impulse to the interethnic violence that occurred in the ancient Levant in different periods. It will argue that various historical factors contributed to the prevalence of extreme violence in these sources: the brutality of imperialistic aggression carried out against the Israelites by several ancient Near Eastern empires; the demographic shifts that contributed to material scarcities in the Iron Age II and the Hellenistic-Roman era; and the dichotomised processes of ethnic boundary-marking that characterise so many Israelite and early Jewish texts.1 The merging of these factors generated an abundance of literature depicting genocide and other forms of mass violence.

Evidence from Israelite and Moabite Sources: H¯erem Texts ˙

The list of biblical sources that depict or command genocide is surprisingly long, and genocidal descriptions and imaginings form a fairly large portion of the corpus of Israelite texts, particularly when compared with the corpora 1 For simplicity’s sake, I use Iron Age II to refer to the period spanning 1000 to 586 B C E, rather than using the Iron II–III periodisation that some archaeologists prefer.

185

Jor dan

R.

t. m. lemos

Mediterranean Sea

IL

IS

TI

A

Jericho Jerusalem

Jordan R.

AMMON

AD

GILE

ISRAEL

PH

ARAM

Sea of Galilee

Heshbon

Mt. Nebo ‘Ataroth

JUDAH Dead Sea

Dibon

MOAB Bozrah

EDOM Map 6.1 Ancient Israel and surrounding regions in the Iron Age. (Cartography by Shane Kelley)

of other ancient Near Eastern groups. While violence is extremely common in, for example, ancient Mesopotamian or ancient Egyptian sources, descriptions of actual genocide are not. Yet, of the three-dozen books that form the anthology of Israelite literature called the Hebrew Bible (or, in Christian theological circles, the Old Testament), at least a dozen books refer to violence that could be characterised as genocidal. One even finds descriptions of anthropocidal violence – that is, violence leading to the destruction of the human race. Post-biblical early Jewish literature in some ways only furthers this centring of mass violence with the development of apocalyptic literature, which was apparently quite popular in Hellenisticand early Roman-era Judaism, if the number of apocalyptic texts produced then is any indication. It is most productive to begin with texts in the Hebrew Bible that describe genocide. No doubt the most glaringly genocidal texts in the Hebrew Bible – and arguably, in all of ancient West Asian literature – are those found in the biblical books of Deuteronomy and Joshua. Deuteronomy 2:31–34, which relates how the Israelites defeated King Sihon of Heshbon in the Transjordan,

186

Genocide in Ancient Israelite and Early Jewish Sources

is just the first of eight or more passages in these two books describing or commanding violence that virtually any scholar of genocide would label as genocidal.2 The text reads: Yahweh said to me, ‘See, I have begun to give to you Sihon and his land. Begin now to take possession of his land.’ And when Sihon came forth against us, he and all his people for battle at Jahaz, Yahweh our God gave him over to us; and we struck him and his children and his whole people down. We captured all his cities at that time, and on each city we performed h¯erem against men, women, and children. We left no ˙ survivors.3

The term h¯erem is a technical one that refers to totalising, exterminationist ˙ killing carried out against a group of people, in some cases including even the slaughter of their livestock.4 This word, as well as verbs from the same root, occur throughout the books of Deuteronomy and Joshua, with the violence aimed at the indigenous inhabitants of the land that Yahweh has promised to the Israelites. The books of Deuteronomy and Joshua present a narrative description of the Israelites not only dispossessing but slaughtering the peoples who inconveniently occupied land meant for the Israelites and the Israelites alone. Deuteronomy 7:1–2 states: When Yahweh your God brings you into the land you are entering to possess, and he casts out many nations before you – the Hittite, and the Girgashite, and the Amorite, and the Canaanite, and the Perizzite, and the Hivite, and the Jebusite, seven nations greater and mightier than you – and when Yahweh your God delivers them up before you and you strike them down, you will certainly perform h¯erem against them. Make no covenant ˙ with them and show them no mercy.

In Deuteronomy 20:16–17, the Israelites are again commanded to annihilate virtually the same list of peoples: But as for the towns that Yahweh your God is giving you as inheritance, you must not let anything that breathes remain alive. You shall annihilate them – the Hittite, and the Amorite, the Canaanite, and the Perizzite, the Hivite, and the Jebusite – just as Yahweh your God has commanded, so that they may not teach you to do any of the abhorrent things that they do for their gods, and you thus sin against Yahweh your God. 2 A longer discussion of definitions will occur later in the chapter. 3 All translations from ancient texts are my own unless otherwise noted. 4 While there are a handful of exceptions to this usage, those occur in later, specialised cultic texts and are not relevant here.

187

t. m. lemos

The book of Joshua provides an extended account of the putative destruction of the autochthonous inhabitants of what had been known as Canaan. Not even the domesticated animals or infants of the Canaanites are spared, as the Israelites dutifully execute the commands of their god. In the infamous tale of the felling of walled Jericho, we read: ‘By the edge of the sword, they performed h¯erem upon everyone in the city, both men and women, ˙ young and old, cattle, sheep, and donkeys.’5 In the very next chapter of Joshua, we come across the account of the Israelite Achan, who took for himself a beautiful mantle, gold and silver that should have been devoted to Yahweh. For his covetous heart – and for not scrupulously following the precepts of h¯erem – not just Achan but his entire household down to oxen, ˙ sheep and donkeys are made h¯erem, stoned to death, and burned with fire. ˙ The Israelite god tolerates not the slightest of deviations from totalising h¯erem ˙ war. With their ledger set aright through the killing of the house of Achan, the Israelites are then able to capture and destroy the city of ‘Ai just as they had destroyed Jericho. This time, they are allowed to take the livestock of ‘Ai as booty, but women and men they slaughter, and the king of ‘Ai they hang from a tree. Joshua 10:40–42 epitomises the violence of the book: So Joshua smote the whole land, the hill country and the Negeb and the lowland and the slopes, and all their kings; he left no survivor, but performed h¯erem upon all that breathed, just as Yahweh God of Israel had ˙commanded . . . Joshua captured all these kings and their land at one time, because Yahweh God of Israel fought for Israel.

Texts in Deuteronomy and Joshua that connect the genocidal killing of collectivities of people with land possession and dispossession include Deuteronomy 2:31–35; 3:1–7; 7:1–2; and 20:16–18; and Joshua 6; 8; 10; and 11. While historians and biblical scholars do not regard these texts as historically accurate accounts and the genocides they describe cannot be corroborated through other sources and very likely did not occur, it is nonetheless striking that Deuteronomy and Joshua are not the only Israelite sources describing h¯erem, and specifically h¯erem as genocidal killing. We ˙ ˙ find this usage also in Numbers 21:2–3, again referring to the slaughter of Canaanites; in 1 Samuel 15, where Saul slaughters the Amalekites; and in 2 Chronicles 32:14, which refers to the conquests of Israel’s earlier history – that is, those described in the Torah and Joshua – as h¯erem. In Jeremiah ˙ 50:26–7, the book calls for h¯erem against Babylon, including even her ˙ 5 Joshua 6:21.

188

Genocide in Ancient Israelite and Early Jewish Sources

bullocks, and the term is used again in Jeremiah 51:3, with the object still the imperial brutaliser Babylon. In 1 Chronicles 4:24–43, which relates some of the activities of the tribe of Simeon, the term h¯erem again appears: ‘These, ˙ registered by name, came in the days of King Hezekiah of Judah, and struck down their tents and the Me‘unim who were found there, and performed h¯erem against them to this day, and settled in their place, because there was ˙ pasture for their flocks there.’ Interestingly, like Deuteronomy and Joshua, this text explicitly ties genocidal behaviour to the desire for territory. While the root h-r-m is not used in Numbers 31, the text is similar to h¯erem texts in ˙ ˙ that Moses insists there that non-virgin Midianite females must be killed along with the males, and baby boys together with adult men, with only virgin girls being spared. There is not space in this chapter to treat every Israelite text that speaks of genocidal behaviour, exterminationist violence or even h¯erem, but there are ˙ a few other biblical sources worth noting. One of these is 2 Chronicles 20:23, which refers to an attempt by the Ammonites and Moabites to perform h¯erem ˙ against the Edomites. Another is Amos 1:13, where the prophet Amos also describes genocidal violence between Levantine groups: ‘Thus says Yahweh: for three transgressions of the Ammonites, and for four, I will not revoke [the punishment], because they have ripped open the pregnant women of Gilead in order to enlarge their territory.’ Notably, this passage mentions an act of violence also committed much later, during the 1994 Rwanda genocide, and situates it in a context of struggles over territory.6 Another important text is Isaiah 34, one of the most interesting h¯erem ˙ passages in the Hebrew Bible but also one of the most oblique. This text shares intriguing similarities with a Moabite text called the Mesha Inscription, to which I shall turn below. Isaiah 34 is striking for its portrayal of divine violence against the Edomites as a sacrifice and its juxtaposition of human and animal imagery. Verses 5–7 of this chapter read: For it is satiated (riwwĕ tâ), my sword in the heavens. / Behold, upon Edom it descends and upon a people my h¯erem descends for judgement. / The sword of Yahweh is full of blood; it is˙ engorged with fat, / From the blood of lambs and he-goats, from the fat of the kidneys of rams. / Because there is a sacrifice for Yahweh in Bozrah, a great slaughter in the land of Edom. / And wild oxen have descended with them, and steers with bull warriors (’abbîrîm). 6 See T. M. Lemos, ‘Dispossessing nations: population growth, scarcity, and genocide in ancient Israel and twentieth-century Rwanda’, in Ritual Violence in the Hebrew Bible: New Perspectives, ed. S. M. Olyan (Oxford University Press, 2015), pp. 35, 40, 43, 47.

189

t. m. lemos

The totalising violence of h¯erem is associated here with sacrifice and involves ˙ extreme violence being inflicted on the Edomites, one of the Israelites’ Levantine neighbours, by the Israelite deity. Israelite texts are not the only texts from ancient West Asia that use the term h¯erem, and with the same meaning of totalising violence that it holds in ˙ Israelite materials. The Mesha Inscription, a Moabite stele erected by King Mesha in the ninth century B C E, does so as well and in fact likely predates many of the biblical texts that refer to h¯erem (Figure 6.1). This text, like many ˙ of the Israelite sources just discussed, describes extreme violence between the small ethnic groups living in the ancient Levant and, more specifically, between the Moabites and Israelites. The most relevant section reads: I am Mesha, son of Kemosh-yat, king of Moab, the Dibonite . . . Omri was king of Israel, and he oppressed Moab for many days because Kemosh was angry with his land. His son succeeded him, and he also said, ‘I will oppress Moab.’ In my days he said this. But I prevailed against him and against his house, and Israel has utterly perished forever! . . . Now Kemosh said to me, ‘Go, seize Nebo from Israel.’ So I went at night and fought against it from the break of dawn until noon. I seized it, and I killed everyone in it – seven thousand adult men and male whelps (grn), adult women, female whelps (grt), and pregnant women (rhmt)7 – for I performed h¯erem (there) for ‘Ashtar˙ ˙ Khemosh’.8

In this text, it is the Moabite deity Kemosh rather than Yahweh who orders Mesha to seize Nebo from Israel, and the slaughter of the people of ‘Ataroth is apparently sacrificial in character – it is a ryt, a ‘satiation’ or ‘offering’ for Kemosh. In addition, Mesha boasts of the killing of the 7,000 residents of Nebo, applying a term to male and female Israelite infants that is typically used for the offspring of predatory animals such as lions or jackals. Like Amos 1:13, this text attests to the killing of pregnant women during wartime contexts.9 The slaughters of infants and pregnant women are actions that convey with gruesome clarity a desire to eliminate a group from existence. As a whole, the text demonstrates that genocidal longings occupied the minds of 7 On how to translate the terms used in this list of victims, see T. M. Lemos, Violence and Personhood in Ancient Israel and Comparative Contexts (Oxford University Press, 2017), pp. 52–3. 8 Lines 1–2, 4–7, 14–17. This translation is based on that of Kent P. Jackson, ‘The language of the Mesha‘ inscription’ in Studies in the Mesha Inscription and Moab, ed. J. A. Dearman (Atlanta: Scholars Press, 1989), pp. 97–8, with emendations based upon the reconstructions of Jackson; and J. A. Dearman, ‘The text of the Mesha‘ Inscription’, pp. 92–5 in the same volume, as well as proposals regarding readings of individual words by other scholars. See Lemos, Violence and Personhood, pp. 52–3, for a longer discussion. 9 There are two other biblical texts that also refer to this practice: 2 Kings 8 and 15:16.

190

Genocide in Ancient Israelite and Early Jewish Sources

Figure 6.1 The Moabite Stone, discovered in 1868, bearing an inscription left by the ninthcentury B C E Moabite king Mesha. (Print Collector / Contributor / Getty Images)

191

t. m. lemos

not just Israelites but also Moabites in the Iron Age II period and corroborates the many references in Israelite texts that point to extreme violence having occurred between Levantine groups in this period.

Evidence from Israelite and Early Jewish Sources: Anthropocidal Texts A review of biblical texts related to genocidal violence would arguably not be complete without at least a brief discussion of the biblical and post-biblical Jewish texts that describe the one form of violence that is even more extreme – what one can call anthropocide or omnicide. Although the Israelites are not the group that invented this concept, it is through Israelite sources that most people are familiar with it. The first case of anthropocidal violence in Israelite texts is the flood narrative in Genesis 6–9. This narrative is in fact heavily dependent on earlier Mesopotamian flood narratives, the two most prominent versions of which are found in Atrahasis and the Epic of Gilgamesh. In both Mesopotamian and biblical accounts, a deity destroys virtually the entire human race, sparing only one righteous man and his family. The Mesopotamian accounts precede the biblical version by roughly a millennium, so there is no question as to the origin or direction of literary dependency. There is not the space here to speculate on why this narrative of omnicidal killing held the popularity it seems to have had in Mesopotamian contexts. What is more important for our purposes, which are to understand the historical and cultural background of the prominent place of mass killing in the Israelite literary corpus, is the fact that the Israelites adopted and chose to include this piece of Mesopotamian mythology into their canon of texts, and that this text was situated in the earliest of archaic times soon after the creation of humanity itself. The text – whether or not this was the intention of its Israelite redactors – then provided a template of incredibly large-scale killing that was performed by the deity and thus of a legitimate form of violence. In the Hebrew Bible, we see this template at work in Exodus 15, where flood imagery is used to describe Yahweh’s slaying of Pharaoh’s armies. In Judges 5, Canaanite armies are swept away by a raging torrent in a battle in which even the stars fought from their courses for the victory of the Israelites. In Samuel 5:20, a reference to the flood is used in the context of violence against Philistines. In Jeremiah 47:2–4, too, flood imagery is central in an image of the destruction of ‘all of the Philistines’. In Habakkuk 3, flood images are used in the context of Yahweh displaying his furious anger against the 192

Genocide in Ancient Israelite and Early Jewish Sources

nations. Flood imagery is also used in the proto-apocalyptic violence against Gog of Magog, chief prince of Meshech and Tubal, in Ezekiel 38 (especially verse 22), in the apocalyptic visions of Daniel 9 and 11 (verses 10 and 40), and in the anthropocidal violence of Revelation 12:15. What we see in these cases is that references to the flood narrative in Genesis, a narrative in which all of humanity suffers God’s wrath, are used in the depiction of violence against specific foreign groups, with the violence typically carried out by Yahweh and on behalf of his people the Israelites. Thus, genocidal or other mass violence carried out against particular ethnic groups is connected with a mythological event from the distant past, sometimes in constructing a new mythological metastory – apocalypse – in which large-scale violence plays an equally important role. Despite this, in some proto- and early apocalyptic literature, descriptions of violence can be somewhat vague. This is the case in Isaiah 24 and Daniel 7–12, for instance. By contrast, the post-biblical War Scroll from Qumran, written about a hundred years after the apocalypse in Daniel, is much clearer in expressing the genocidal and anthropocidal intentions of eschatological violence. While at times specific foreign groups are mentioned (e.g. the ‘seven nations’ in column X I, in a clear reference to Deuteronomy 7:1), the more striking references are to all-encompassing violence against ‘the nations’ or ‘all of the nations’ (columns X I I and X V), ‘for destruction without any remnant’ (X I V), ‘to destroy all the sons of darkness’ (X I I I). The deity will not rest, column X V I I tells us, until ‘all the wicked nations are destroyed’. In this text, most of humanity stand among the wicked ‘sons of darkness’ who will be destroyed by the Judean god, which gives the text a general anthropocidal character. However, the specifically xenophobic, genocidal character of earlier biblical texts such as Deuteronomy and Joshua is maintained through repeated references to ‘the nations’ and to specific groups such as the ‘sons of Japheth’ (i.e. the Greeks, column X V I I I) or ‘the Kittim’ (likely the Romans, column X V I I I and elsewhere). This combination of universalising anthropocidal violence and xenophobic genocidal violence is present also in that most famous of apocalyptic texts, the book of Revelation. While Revelation is the very late, final book of the Christian New Testament and not typically categorised as an early Jewish text, it quotes or makes allusion to Hebrew Bible passages hundreds of times and was likely written by a Jewish Christian.10 This, and its 10 S. Moyise, The Old Testament in Book of Revelation, Journal for the Study of the New Testament supplement series 115 (Sheffield Academic Press, 1995), p. 14; C. R. Koester,

193

t. m. lemos

clear anthropocidal tendencies, make it worth a brief analysis here. Anthropocidal passages may be found in Revelation 6; 8 and 16 (especially verse 3), where great ecological destruction occurs; 9, where a third of humanity is killed (verse 15); 11; and 14, where blood flows ‘as high as a horse’s bridle’. Revelation 7 details how there will be 144,000 ‘sealed’ from the twelve tribes of Israel, which is often interpreted to mean that this is the number of human beings who would survive the eschatological war,11 though 7:9 implies that there were multitudes from all nations who lauded God and would be part of the flock of the Lamb.12 Revelation 17–19, which discuss violence against the Whore of Babylon, are seemingly more particularistic in their violence. Despite the reputation of the New Testament as contrasting with its message of a loving God the vindictive rage and pronounced violence of the Hebrew Bible’s depiction of God, in this final book of Christian scriptures the deity apparently means to destroy most of the human race and with them large portions of the earth itself and the nonhuman creatures dwelling in it, including every single creature living in the sea.13

Historical Context: The Levant in the Iron Age It is necessary at this point to turn in more concerted fashion to discussing the connection between the many biblical texts that speak of genocidal or other forms of exterminationist violence and actual historical events and processes. Do these texts serve as evidence that genocidal violence transpired in the ancient Levant in the Iron Age? If so, why did such violence transpire, and if not, what spurred such an intense genocidal imaginary among Israelites, Judeans and apparently Moabites as well? Finally, did biblical texts, whether or not they are historically accurate, inspire genocidal violence, either in antiquity or in more modern times? The consensus that exists today among scholars and has existed now for several decades is that the conquest narrative presented in Deuteronomy and Joshua is quite simply not historically accurate. The emergence of the Israelites as an ethnic group in the highlands of Canaan that occurred during the transition of the late Bronze Age and Iron Age was a complex phenomenon bearing little similarity to the events presented in these two biblical Revelation: A New Translation with Introduction and Commentary, Ancor Yale Bible Commentary series 38A (New Haven: Yale University Press, 2014), p. 65. 11 See also Revelation 14. 12 That is, Christ; see 7:17. 13 Revelation 16:3.

194

Genocide in Ancient Israelite and Early Jewish Sources

books.14 Still, as will become clear, one should not conclude from this that there was little or no interethnic violence between groups living in the ancient Levant in the Iron Age, or that this violence never took on genocidal dimensions, especially considering that both Israelite and Moabite sources describe violence with such dimensions. Deuteronomy and Joshua, rather than describing events occurring in the late second millennium B C E, relate more to the historical and social realities of the Iron Age II era (1000–586 B C E). Many sources from this era attest to complex social and cultural historical factors influencing or even directly engendering genocidal imaginaries on the part of the peoples of ancient West Asia, particularly the Israelites and Moabites. These factors include notable demographic changes, resource scarcities, competition between groups over resources, repeated cases of imperialistic violence, and imperialistic policies exacerbating resource challenges in the Levantine region. They also include cultural emphases on authoritarianism, rigidly and starkly demarcated boundaries between ethnic groups that were in reality quite similar to each other, and sacrificial aspects being attributed to violence against humans. It may seem that this is a lengthy list indeed of factors underlying genocidal violence in this region, but violence is after all a complex phenomenon, and since genocidal violence is if anything only more complex than most other forms of violence, one should expect the social and cultural factors giving rise to it also to be multifaceted and intricate, rather than facile. Despite the complexity of the matter at hand, I shall nonetheless endeavour to provide a succinct account of these historical factors. The account must begin in the early Iron Age (1200–1000 B C E, also known as the Iron I period), the era in which the Israelites emerged as an ethnic polity in ancient West Asia along with other small groups such as the Moabites, Ammonites and Arameans. One must begin there because the demographic shifts that would foster interethnic violence later on in the Iron II period began in that earlier era. After the collapse of the Late Bronze Age empires in the thirteenth to twelfth centuries B C E, the area of the Levant was more scarcely populated than it had been previously. The region of Palestine is estimated to have had

14 For summaries of the evidence and scholarly positions, see I. Finkelstein and N. A. Silberman, The Bible Unearthed: Archaeology’s New Vision of Ancient Israel and the Origin of Its Sacred Texts (New York: Free Press, 2001), pp. 72–96, 329–39; and M. Bishop Moore and B. E. Kelle, Biblical History and Israel’s Past: The Changing Study of the Bible and History (Grand Rapids, MI: Eerdmans, 2011), esp. pp. 77–142.

195

t. m. lemos

a population of just 60–70,000 at the beginning of the Iron I period.15 By the year 1000 B C E, however, the population of western Palestine had risen to roughly 150,000.16 Moving into the Iron II era, we see that in the mid-eighth century the population of the Northern and Southern Kingdoms of Israel was approximately 460,000, or even higher.17 Other regions of the Levant also experienced marked population increases in this era.18 Later, forced migrations of populations both into and out of the Northern and Southern Kingdoms of Israel in the eighth and sixth centuries by the Assyrian and Babylonian empires would affect population numbers, as would military assaults by these imperialist powers in 722–720, 701, 597 and 587–586 B C E. The latter assaults left widespread starvation in their wake, as biblical texts evocatively describe.19 With all of this, the population of Palestine was reduced and that of Judah dropped precipitously. The number of inhabitants of Judah fell to about 30,000 in the Persian era, contrasting with 110,000 just before the fall of Jerusalem in 587–586.20 These population shifts are relevant because biblical texts themselves describe the sometimes violent competition over resources that occurred as the number of occupants of this region increased markedly over the course of the Iron Age, seemingly hitting its apex in the eighth century B C E (and again in the Roman era after a period of decline). This competition occurred both between groups, as not just Israelite but also Moabite sources attest, and within groups, as elites practised what social scientists refer to as ‘resource capture’, accumulating and even seizing lands and other resources from nonelites. These seizures led to hardship, debt slavery and the weakening of social units.21 Biblical texts dating to the eighth century, the very century where population levels peaked in the Iron Age Levant, in fact speak of land 15 M. Broshi, ‘The population of Iron Age Palestine’ in Biblical Archaeology Today, 1990: Proceedings of the Second International Congress on Biblical Archaeology, ed. A. Biran and J. Aviram (Jerusalem: Israel Exploration Society, 1993), pp. 14–18. 16 M. Broshi and I. Finkelstein, ‘The population of Palestine in Iron Age II’, Bulletin of the American Schools of Oriental Research 287 (1992), 47–60, esp. 55. 17 See ibid., 54; and Oded Lipschits, The Fall and Rise of Jerusalem: Judah under Babylonian Rule (Winona Lake, IN: Eisenbrauns, 2005), p. 248, who feels that Broshi and Finkelstein’s estimates are a bit low, arguing that they undercount the number of settled dunams at the end of Iron II. Lipschits (259–61) also lists and briefly reviews various population estimates of Palestine for different eras. 18 Lemos, ‘Dispossessing nations’, 38 and 60, summarises relevant research. 19 See e.g. Lamentations 1:11; 2:12, 19; 4:9; 5:6, 9; Ezekiel 7:19, 34:29. 20 O. Lipschits, ‘Demographic changes in Judah between the seventh and the fifth centuries B.C.E.’ in Judah and Judeans in the Neo-Babylonian Period, ed. O. Lipschits and J. Blenkinsopp (Winona Lake, IN: Eisenbrauns, 2003), pp. 359–63. 21 See T. M. Lemos, Marriage Gifts and Social Change in Ancient Palestine: 1200 BCE to 200 CE (Cambridge University Press, 2010), ch. 4, for a detailed discussion of changes that

196

Genocide in Ancient Israelite and Early Jewish Sources

seizures, competition over land and other resources, debt slavery and other related social problems. In addition to Amos 1:13, quoted above, which explicitly connects genocidal violence between Levantine groups with a desire to increase territory, we have Micah 2:1–2, which states: ‘Alas for those who devise wickedness and evil deeds on their beds! When the morning dawns, they perform it, because it is in their power. They covet fields, and seize them; houses, and take them away; they oppress householder and house, people and their inheritance.’22 This text describes resource capture by elites, as does Isaiah 5:7–8: ‘For the vineyard of Yahweh of hosts is the house of Israel, and the people of Judah are his pleasant planting; he expected justice, but saw bloodshed; righteousness, but heard a cry! Ah, you who join house to house, who add field to field, until there is room for no one but you.’ 1 Kings 21 provides another example of the same phenomenon, and other relevant texts are Isaiah 3:14–15; 10:1–2; Amos 2:6–7; 4:1; 5:11; 8:4–6; and Micah 3:9–11, among many other examples from biblical prophetic texts for social abuses in this period. Both archaeological and literary evidence attest to a marked increase in wealth-based stratification in this era. These phenomena were occurring in Iron II just as the rise of the Neo-Assyrian Empire placed an increasing amount of pressure on the small polities of the ancient Levant. It would be unsurprising for increasing population and decreasing resources to lead to tensions, competition and even violence between groups, and biblical texts and other evidence indicate that this was indeed the case. The Mesha Inscription makes clear that Moabite and Israelite groups were competing over territory, and this is not the only piece of evidence we have that conflict was centred particularly on the land resources of the Moabites. For example, Jeremiah 49:3 and some archaeological and inscriptional evidence seem to indicate that the Moabite town of Heshbon was seized by Ammonites.23 Some biblical texts, however, include Heshbon among the inheritance of Israelite tribes.24 Also relevant are the biblical texts relating the story of the defeat of King Sihon the Amorite, seemingly with the intention of ‘justifying an Israelite claim to this territory’, as occurred in the social structure of ancient Israel in the pre-exilic period, including increases in social stratification and debt slavery and a weakening of kinship structures. 22 NRSV translation. 23 B. Routledge, Moab in the Iron Age: Hegemony, Polity, Archaeology (Philadelphia: University of Pennsylvania Press, 2004), pp. 46–7. 24 See Numbers 32:37 and Joshua 13:17, which attribute the city to Reuben; and Joshua 13:26, which assigns it to Gad; as well as J. A. Dearman, ‘Moab and Ammon: some observations on their relationship in light of a new Moabite inscription’ in Studies on Iron Age Moab and Neighboring Areas in Honour of Michèle Daviau, ed. P. Bienkowski (Leuven: Peeters, 2009), pp. 103–4.

197

t. m. lemos

B. Routledge writes.25 The combined evidence suggests, then, that scarcities of land in Israel and Ammon led those groups to try to claim and seize territory in Moab, and other sources indicate that the Ammonites encroached upon the Israelite territory of Gilead and that the Moabites possibly sought to take possession of Edomite territory.26 Archaeologists have also argued for violent assaults carried out by the Aramean king Hazael against Israelite and other sites in the ninth century B C E.27 That competition between Levantine groups in this era involved violence is clear, but whether or not this violence was specifically genocidal is impossible to discern archaeologically. If the writing produced during this period is any indication, however, it is likely that the violence at times took on extreme forms, involving the killing and disfigurement of pregnant women, the killing of children and attempts to eliminate particular groups from contested areas. These interethnic conflicts changed form considerably when the NeoAssyrian Empire grew dominant in the ninth and especially the eighth century B C E. Fear of the expansion of Neo-Assyrian power initially led many of the small polities of the Levant to join together in military coalitions to fend off the Mesopotamians. This occurs at the battle of Qarqar (Kurkh) in 853, where eleven different Levantine groups allied against Assyria, with some success. These groups included the Israelites, Ammonites and Arameans. Levantine groups joined together again in the 730s, when Aram and the Northern Kingdom of Israel formed an alliance to attempt to throw off the yoke of Assyrian vassalage. Judah in the south, however, refused to cooperate in this revolt against their mutual oppressor and suffered retaliation for it. Thus, the Arameans and Northern Kingdom of Israel and the Southern Kingdom of Judah became involved in hostilities against each other in what is known as the Syro-Ephraimite War. As all this indicates, on the whole the rising power of Assyria in no way led to more ethnic tranquillity in this region, despite occasional cases of Levantine groups coming together to resist this new, larger and more dangerous enemy. While military assaults and forced population movements 25 Routledge, Moab in the Iron Age, p. 45. Biblical passages that relate or refer to this story are, e.g., Numbers 21; Deuteronomy 2; Joshua 2:10; 9:10; 12:2; 13:9–10; Judges 11:19–21; Psalms 135:11–12; 136:19–22; Nehemiah 9:22. 26 Amos 1:13; 2:1. Some archaeologists see evidence for this encroachment, particularly on the territory of the Moabites, in the material culture of the region very late in Iron II. See Routledge, Moab in the Iron Age, pp. 46–7, and Dearman, ‘Moab and Ammon’, 103–4. 27 See I. Finkelstein, ‘Destructions: Megiddo as a case study’ in Exploring the Longue Durée: Essays in Honor of Lawrence E. Stager, ed. J. D. Schloen (Winona Lake, IN: Eisenbrauns, 2009), pp. 113–26, and Aren Maeir’s chapter in the same volume.

198

Genocide in Ancient Israelite and Early Jewish Sources

deporting thousands from the Levant to elsewhere in the Assyrian Empire apparently lessened overall population numbers, the tribute payments demanded by Assyria vitiated any positive effect this reduction might have had on the availability of land and other resources. The scale of violence experienced by the Israelites in both the Northern and Southern Kingdoms in 722–720 and 701 B C E was noteworthy, and this scale was matched by the Babylonian assaults against Judah in 597 and 586 B C E, which destroyed Jerusalem, forced thousands of Judeans to migrate to Babylon, and devastated the population of Judah through brute violence and starvation. Various scholars have written on the traumas suffered by the Israelites, the effects of which are clearly described in biblical sources.28 These traumas, which were social, cultural, psychological and not least of all material, with effects such as loss of life, destruction of agricultural lands and shortages of basic resources necessary for survival, kept extreme violence at the centre of the literature produced in these periods. While the Assyrian and Babylonian violence was not genocidal in character, its brutality was such that traumatised Israelite authors continued to produce texts featuring extreme or even genocidal violence while the Israelites were vassals of Mesopotamian empires. In fact, many scholars argue that Deuteronomy and Joshua were written during this time period. Although set in an idealised Mosaic past, what these most genocidal of biblical books were in actuality responding to were the interethnic tensions, the social pressures and the imperialist violence of the Iron II era.29 Complicating the portrait of the Israelite genocidal imaginary still further are the cultural features that very likely combined with the demographic and historical factors just described to produce not only genocidal longings but also actual interethnic violence in the Iron II era. Many biblical sources, Deuteronomy and Joshua especially, betray a focus on authoritarianism and conformity. These features reappeared in at least some genocidal contexts in modernity, with Rwanda being an important example.30 One very 28 See e.g. D. L. Smith-Christopher, A Biblical Theology of Exile (Minneapolis: Fortress Press, 2002); D. M. Carr, Holy Resilience: The Bible’s Traumatic Origins (New Haven: Yale University Press, 2014); and T. M. Lemos, ‘The apotheosis of rage: divine rage and the psychology of Israelite trauma’, Biblical Interpretation 23:1 (2015), 101–21. 29 As I argue at length in Lemos, ‘Dispossessing nations’. 30 For another, see D. J. Goldhagen, Hitler’s Willing Executioners: Ordinary Germans and the Holocaust (New York: Vintage, 1997), a work that has provoked controversy; and L. S. Newman, ‘What is a “social-psychological” account of perpetrator behavior?: The person versus the situation in Goldhagen’s Hitler’s Willing Executioners’ in Understanding Genocide: The Social Psychology of the Holocaust, ed. L. S. Newman and R. Eeber (Oxford University Press, 2002), pp. 43–67. On Rwanda: Mahmoud Mamdani, When Victims

199

t. m. lemos

much sees this emphasis in texts such as Deuteronomy 13:6–11 and 17:2–7, where disloyalty to Yahweh is punished by stoning, and Joshua 7, where failure to implement the rules of h¯erem to the required exacting standard led ˙ to the execution of not only the offender but also the offender’s entire household. The disposition of the books of Deuteronomy and Joshua in general is towards complete subordination to divine will and divine instructions, including divine instructions regarding genocidal violence. Another feature correlated with genocidal violence in both biblical texts and modern genocidal contexts such as Rwanda is a stark and highly dichotomised manner of framing ethnic difference. Rwandan genocidal propaganda constructed differences between Hutus and Tutsis in very exaggerated terms, despite the clear similarities between these groups. As Ben Kiernan writes: ‘Perhaps continuing to compensate for the lack of ethnocultural distinctions between Hutu and Tutsi, the propaganda of Hutu power focused on past injustices and, especially, fear of future contamination.’31 We find these very same strategies present in Israelite texts commanding genocide. As was the case with the Hutu and Tutsi, there seems to be an overcompensation, a frantic nature to the ethnic boundary-marking in both Israelite texts and the Mesha Inscription from Moab in which Levantine groups who were not very different from each other strove to maintain the decreasing amount of resources available to them. Texts recite past betrayals and past wrongs as justification for violence while also warning against how these groups would lead the Israelites astray in the future if they were not completely annihilated. Without the sharp differentiation between Israelite and other that we find in so many biblical sources, it is unlikely that genocidal commands and narratives could have taken shape or held force in the way they did.

Historical Context: Post-exilic Sources The patently genocidal narratives of Deuteronomy and Joshua presume a level of centralised leadership in their accounts. Joshua is depicted as a quasi-monarchic authoritarian leader who leads an army of Israelites become Killers: Colonialism, Nativism, and the Genocide in Rwanda (Princeton University Press, 2001), pp. 198–202; G. Prunier, The Rwanda Crisis: History of a Genocide (New York: Columbia University Press, 1995), pp. 141–2; C. C. Taylor, Sacrifice as Terror: The Rwandan Genocide of 1994 (Oxford: Berg, 1999), pp. 15–16, 25; R. Orth, ‘Rwanda’s Hutu extremist insurgency: an eyewitness perspective’ in Genocide in Cambodia and Rwanda: New Perspectives, ed. Susan E. Cook (New Brunswick: Transaction, 2006), p. 229. 31 B. Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007), p. 559.

200

Genocide in Ancient Israelite and Early Jewish Sources

against the cities and autochthonous peoples of Canaan. Biblical texts such as Amos 1:13 that refer to extreme interethnic violence between groups in the Levant were also written in an era when the Israelites had kings who commanded armies, and the Mesha Inscription from Moab, too, is the product of a monarchic state. These monarchies, while hardly of the scale of those of Egypt or Mesopotamia, nonetheless had the ability to marshall their own forces against surrounding groups in attempts, genocidal or otherwise, to protect or seize resources. However, in the sixth century B C E, the structure of Israelite society completely changed. The Northern Kingdom of Israel had already been eliminated by the Assyrians in 722–720 B C E, and Judah, the Southern Kingdom of Israel, met the same fate at the hands of the NeoBabylonian Empire in 586 B C E. The dynasty of David was no more, and the Judeans lost their sovereignty for hundreds of years. Thousands of Judeans were exiled to Babylon from 586 to 539 B C E. Judah became a province of Babylonia, then of Persia, and then of the Greek-speaking Ptolemaic and Seleucid states that ruled from Egypt and Syria, respectively, after the death of Alexander III of Macedon. After the Hasmonean revolt of 167–164 B C E, the Judeans did reinstate native rule for a hundred-year span of time, only to fall yet again under the oppression of foreigners – this time the Romans – in 63 B C E. This is not just an idle recitation of the facts of ancient history. Rather, this series of events raises the question of how biblical texts that command or describe genocide were received in the post-exilic period when the Israelite monarchy and Israelite armies no longer existed. The violence imposed upon the Israelites by the Assyrians and Babylonians was heinous, and biblical texts certainly respond to this fact. But what happened to Israelites’ genocidal imaginary in an era when carrying out genocidal longings was truly impossible? This question is especially worth considering in light of the increasingly authoritative status that texts dating to the monarchic period and Babylonian exile took on among Judeans living both within and outside of Judah. Before attempting to answer this question, it is worthwhile to examine the demographic and other major sociological features of Judah and the wider region of Syria–Palestine in the post-exilic era. In the Persian era, Judah was a province of the Persian Empire to which the exiled Judeans were allowed to return (even if not all chose to do so). Contrasting with the Assyrian and Babylonian periods, there is no extrabiblical evidence that the Judeans suffered mass violence under the Persians, though texts such as Esther seem to betray a fear that such violence could occur under this new empire just as 201

t. m. lemos

easily as it had previously. However, what our sources indicate did occur in this period were intense struggles among the Judeans themselves over how to define the boundaries of Judean ethnic identity, as well as a strong desire on the part of some Judeans to define ‘Israel’ in extremely narrow and rigid terms. And in the subsequent Hellenistic era, concerns over the boundaries of ethnic identity and questions of assimilation to Hellenistic culture become if anything even more prominent a theme in Judean literature.32 What also becomes prominent again in the Hellenistic era, as had been the case before the Babylonian exile, was relatively high population levels, urbanisation and social stratification in which divisions between rich and poor were very stark.33 After a major decrease in the number of inhabitants at the end of Iron II, in the Hellenistic period the population again grew quite urbanised, as it had been in the Middle Bronze, Late Bronze and Iron II eras, and in fact it even surpassed those eras in the Roman period, when the inhabitants of Palestine numbered somewhat less than 1 million and western Palestine became one of the most densely populated areas in the entire Roman Empire.34 In addition to these shifts, Judah becomes ever more factionalised, with boundary lines being strenuously and sometimes violently drawn not only between Judean and ethnic others but also between Judean and Judean, as the books of Maccabees, the works of Josephus and other texts make clear. This is the era in which a group of Judeans most scholars identify as the Essenes withdrew to the desert site of Qumran35 and wrote their War Scroll, envisioning a final battle in which the Sons of Light destroy sinful Israelites together with various foreign groups. It is also the era in which the Hasmoneans wrested sovereignty over Judah and its capital, Jerusalem, from the hands of their Greek-speaking, Judean-culturesuppressing overlords, the Seleucids (in 164 B C E). Because the Hasmoneans were descended neither from David nor from the high priestly line of Zadok, 32 For Persian-era struggles over ethnic boundaries, see Ezra–Nehemiah especially, and for the Hellenistic era, one can look to the books of Maccabees, Jubilees and other sources. 33 Lemos, Marriage Gifts, ch. 5 provides a summary of the evidence on these phenomena. 34 M. Broshi, Bread, Wine, Walls, and Scrolls (Sheffield Academic Press, 2001), pp. 87, 92–106. According to Broshi, the maximum carrying capacity of Palestine was reached in the Roman-Byzantine period, though late in this period (p. 92). As he notes, previous estimates have in fact estimated the population of Palestine in the Roman era to have been far higher than Broshi’s more conservative, and in my view more realistic, estimates. 35 This was likely sometime between 100 and 50 B C E. See J. Magness, The Archaeology of Qumran and the Dead Sea Scrolls (Grand Rapids, MI: Eerdmans, 2002), esp. p. 65.

202

Genocide in Ancient Israelite and Early Jewish Sources

their rule and oversight of the Jerusalem temple was not welcomed by all Judeans.36 Still, they asserted their presence in the region through acts of interethnic violence against neighbouring groups, seizing territory in all directions – Samaria, the Galilee, Iturea, Perea and Idumea, in other words, in the regions once known as the Northern Kingdom of Israel, Aram, Moab, Edom and territory alongside or partly including Ammon.37 Thus, they did as their ancestors in the pre-exilic monarchic era had done. The Hasmoneans even forced non-Judeans to take on Judean practices, much as the Seleucids had forcibly changed Judean practices just a few decades before.38 While land claims and a concern with territory are perhaps more prominent in sources predating the Babylonian exile, it hardly seems coincidental that interethnic violence among Levantine groups again markedly increased in the Hellenistic and Roman periods when population levels and social stratification also markedly increased. The ability of Judeans to respond to social unrest or resource challenges was limited, however, in the period of dominance by the Romans. I leave treatment of Roman violence against Judeans, which was severe in the first and second centuries C E, to a later chapter in this volume. What is worth noting here is that the social setting of the post-exilic period, especially in the Hellenistic and Roman eras, was such that some of the features that appear most connected with the genocidal imaginary of Israelites in the monarchic era now return in the post-exilic era. The genocidal and anthropocidal texts of the Hebrew Bible, by this time seen as authoritative scripture, provide inspiration for the writing of new texts promoting ethnocide, that is, the killing of specific ethnic groups, and featuring an anthropocidal timbre so intense that the flood narrative of Genesis appears almost benevolent by comparison.39 In this era of mostly foreign rule, however, Judeans, lacking human armies, were left to imagine angelic hosts who would carry out exterminationist violence for them. These visions of mass death were so 36 See the first century B C E Judean text Psalms of Solomon 8:11; 17:5–6, 21; H. Eshel, The Dead Sea Scrolls and the Hasmonean State (Grand Rapids, MI: Eerdmans, 2008); and V. Noam, Shifting Images of the Hasmoneans: Second Temple Legends and Their Reception in Josephus and Rabbinic Literature, trans. D. Ordan (Oxford University Press, 2018). 37 See Joseph. BJ 1.104–5, 156; 2.49; Joseph. AJ 13.250–8, 275–81, 393–4; 17.277; P. Schäfer, The History of the Jews in the Greco-Roman World, rev. ed. (London: Routledge, 2003), pp. 67– 9, 74–5; J. T. Fitzgerald, ‘Gadara: Philodemus’ native city’ in Philodemus and the New Testament World, ed. J. T. Fitzgerald, D. Obbink and G. S. Holland (Leiden: Brill, 2004), pp. 361–2. 38 John Hyrcanus forced Idumeans to submit to circumcision; see Joseph. AJ 13.254–8. 39 By ‘ethnocide’, I mean here the physical destruction of an ethnic group. I do not use it as a synonym for cultural genocide, as do some scholars, e.g. A. Jones, Genocide: A Comprehensive Introduction, 3rd ed. (London: Routledge, 2017), p. 35.

203

t. m. lemos

evocative that they played an important role in inspiring actual interethnic violence, with Judeans rising up not once or twice but three times against Roman rule (in 66–73 C E, 115 C E and 132–6 C E).40 Yet, far from fulfilling dreams of ethnocide, Jews themselves became the victims of ethnic violence on a tremendous scale.41

Relation to Modern Genocides and Modern Definitions of Genocide Due to the imagined and even fantastical nature of the mass violence described in some Israelite and early Jewish sources, one would be forgiven for wondering how this material even relates to modern cases of exterminatory violence or modern definitions of genocide. While understandable, this line of reasoning could lead one in the wrong direction when analysing biblical and post-biblical sources for mass violence. Previously, I have argued at length that the violence described in Joshua and commanded in Deuteronomy fits almost any definition of genocide, not just those used for this volume.42 Deuteronomy 7 and 20 command the Israelites to completely annihilate specific ethnic groups and Joshua describes the implementation of such commands. Targeting and annihilating specific ethnic entities is considered genocide according to either narrow or broad definitions of the term. However, few if any scholars today understand the narrative of Joshua to be historically accurate. The conquest of the land of Canaan it describes quite 40 Most scholars see apocalyptic texts and thinking as having played at least some part, if not a large part, in inspiring the Jewish revolts. See e.g. M. Hengel, The Zealots: Investigations into the Jewish Freedom Movement in the Period from Herod I to 70 A.D. (London: T&T Clark, 1989), and J. G. Gager, ‘Messiahs and their followers’ in Toward the Millennium: Messianic Expectations from the Bible to Waco, ed. P. Schäfer and M. R. Cohen (Leiden: Brill, 1998), pp. 37–46; but cf. T. Rajak, ‘Jewish millenarian expectations’ in The First Jewish Revolt: Archaeology, History and Ideology, ed. A. M. Berlin and J. A. Overman (London: Routledge, 2002), pp. 164–88, and M. Goodman, ‘Messianism and politics in the land of Israel, 66–135 CE’ in Redemption and Resistance: The Messianic Hopes of Jews and Christians in Antiquity, ed. M. Bockmuehl and J. C. Paget (London: T&T Clark, 2007), pp. 149–57, who are more sceptical of the role of apocalyptic thought in these revolts. A. J. Tomasino, ‘Oracles of insurrection: the prophetic catalyst of the Great Revolt’, Journal of Jewish Studies 59:1 (2008), 86–111, reaffirms the stance that apocalyptic thought did in part inspire these revolts: ‘Many factors contributed to the Jewish revolt against Rome. But a sense of God-ordained timing, disseminated by various prophetic figures, undoubtedly contributed to the idea that victory was certain’ (111). 41 In most languages, including Hebrew, there is no differentiation between the terms ‘Judean’ and ‘Jew’. Thus, when referring to the Roman era, I use ‘Judean’ and ‘Jew’ synonymously. 42 Lemos, ‘Dispossessing nations’.

204

Genocide in Ancient Israelite and Early Jewish Sources

simply never occurred. What, then, do these narratives have to do with modern genocides? In fact, examining biblical narratives of genocide is highly instructive. First, Deuteronomy and Joshua use language and genocidal concepts found also in the Mesha Inscription, which likely predates those biblical texts. The Mesha Inscription, unlike Deuteronomy and Joshua, is set not in some idealised past but rather in the ninth century B C E, exactly when it was written. Mesha claims to have slaughtered all of the Israelite inhabitants of ‘Ataroth and Nebo, in the latter case specifying that everyone from infants to pregnant women to adult males was massacred. While we do not have the evidence to ascertain whether or not Mesha succeeded in doing what he claimed, the violence he boasts of having committed is easily classifiable as genocide. The 1948 United Nations Convention on genocide states: ‘In the present Convention, genocide means any of the following acts committed with intent to destroy, in whole or in part, a national, ethnical, racial, or religious group’. The first act listed is killing members of the group, which is what Mesha claims to have done. The broad phenomenon of interethnic violence among Levantine groups that is attested by the Mesha Inscription is also referred to in many biblical sources, as discussed earlier in this chapter. Some of the actions described – for example, the cutting open of pregnant women – mirror not just the Mesha Inscription but also actions performed in modern genocides such as that in Rwanda. The Israelites even have a technical term in their language, h¯erem, used to refer to totalistic killing. Considering all this, ˙ it seems unwise to disregard biblical accounts of genocide in studies of genocide as a historical phenomenon. This is especially so because later on in history, for example in the settlement of New England by the Puritans and South Africa by the Dutch, the genocidal texts from the Bible were used to justify ethnic violence against the indigenous peoples whose lands Europeans wished to claim for themselves.43 If biblical texts have been used to legitimate extreme violence against ethnic others in recent centuries, it seems prudent to take the genocidal character of these texts very seriously and attempt to explicate why the Israelites imagined and described violence in the ways they did. Clearly, the anthropocidal narratives of Genesis and later apocalyptic literature have a different cast to them and are more difficult to relate to modern genocides such as the Holocaust or the Rwandan genocide. 43 See M. Prior, The Bible and Colonialism: A Moral Critique (Sheffield Academic Press, 1997), esp. pp. 81–4, 104, 263.

205

t. m. lemos

However, as I have attempted to demonstrate, many texts that refer to the primeval flood, a purely anthropocidal event not geared at specific ethnic groups, or to cosmic battles in which the great majority of the human race is destroyed, do, in fact, have ethnocidal elements that occur together with the anthropocidal ones. While the Judeans did not by themselves have the power to implement apocalyptic violence, the popularity of apocalyptic texts and thinking was part of what inspired Judeans to revolt against the Romans in the first and second centuries. Also, the interethnic and internal Judean violence of the earlier Hellenistic period was a major aspect of the social setting in which the earliest apocalyptic texts were written. To deny, then, a close connection between actual interethnic violence and the ethnocidalcum-anthropocidal violence of apocalyptic texts seems specious. If Israelite and early Jewish sources are any indication, genocidal imaginaries – whether taking anthropocidal or more limited ethnocidal forms – were not at all divorced from actual interethnic violence.

Conclusion In the ancient Levant, genocidal imaginings and impulses were closely connected with competition over land and other resources, demographic stress, and struggles for power both local and interregional in character. A multi-causal approach to genocide, in line with more recent research in Genocide Studies, encourages us to connect genocide with a complex interplay of factors, whether demographic, environmental, economic, social or cultural.44 This chapter has reviewed the most important evidence for genocide in ancient Israelite and early Jewish sources and attempted to demonstrate that genocidal texts and mass violence between ethnic groups in the ancient Levant related to both material and cultural features. Population increases at times led competition over resources to reach frenzied heights, while dichotomised processes of ethnic identity formation and social boundary-marking fostered ethnocidal impulses. Interregional imperialism, economic predation by regional elites and cultural emphases on conformity and authoritarianism also contributed to the prevalence of genocidal imaginaries in biblical and post-biblical sources. Genocidal texts reflecting regional competition and local interethnic violence took on a cosmic orientation in later eschatological literature, and both biblical conquest narratives and apocalyptic sources contributed to interethnic violence and 44 See my discussion in Lemos, ‘Dispossessing nations’, esp. pp. 30–1, 46.

206

Genocide in Ancient Israelite and Early Jewish Sources

genocide in later times, after Israelite texts had gained authoritative status among both Jews and Christians.

Bibliographic Note In the past many biblical scholars have been so hamstrung by theological concerns that they have refused to even acknowledge the extent of the violence in biblical texts, much less its genocidal dimensions. On occasion, scholars have pointed to the lack of historical evidence for military conquest of Canaan by the Israelites in order to sidestep analysis of the genocidal character of texts such as Deuteronomy and Joshua. In terms of theological interpretation, considering the status of the Bible in Christian and Jewish communities as authoritative religious scripture, it is not surprising that theological grappling with genocidal biblical texts has by necessity occurred. However, theological interpreters have typically been less interested in the historical realities of the ancient Levant and more in how contemporary communities should respond to, make use of or sideline biblical texts such as Joshua. Theological scholarship has too often been centred on apologetics, for example, attributing great villainy and sinfulness to the Canaanites to argue that the Canaanites deserved what they got, or used the general brutalities of ancient warfare to exculpate biblical texts and authors: see R. S. Hess and E. A. Martens, eds, War in the Bible and Terrorism in the TwentyFirst Century, Bulletin for Biblical Research Supplement 2 (Winona Lake, IN: Eisenbrauns, 2008); and C. S. Cowles, E. H. Merrill, D. L. Gard and T. Longman III, Show Them No Mercy: Four Views on God and Canaanite Genocide (Grand Rapids, MI: Zondervan, 2003). Among scholars whose interests have been historical rather than theological, research has often centred on narrow historical interpretation that neglects to take a broad view of Israelite political and social history and focuses instead on ‘religious’ aspects of biblical genocide. G. von Rad is interested in ‘holy war’ – see for instance his Holy War in Ancient Israel, translated and edited by M. J. Dawn (Grand Rapids, MI: Eerdmans, 1991; German original, Der Heilige Krieg im alten Israel, 1958) – while other scholars focus on putative ideological dimensions, such as a need to assert ‘order over chaos’ through genocidal violence and imagery (see P. D. Stern, The Biblical Herem: A Window on Israel’s Religious Experience (Atlanta: Scholars Press, 1991); C. L. Crouch, War and Ethics in the Ancient Near East: Military Violence in Light of Cosmology and History, BZAW 407 (Berlin: De Gruyter, 2009); and the critique in T. M. Lemos, ‘Order from chaos: comparing approaches to 207

t. m. lemos

violence in anthropology, Assyriology, and the study of the Hebrew Bible’, Currents in Biblical Research 18.2 (2020), 160–175)), or on scholarly disputes over the dating of relevant biblical passages. While S. Niditch in War in the Hebrew Bible: A Study in the Ethics of Violence (Oxford University Press, 1993) provides a useful assessment of h¯erem texts, she does not treat in any detail specific ˙ historical realities, whether those of Assyrian or Babylonian dominance or the demographic features of competition between Levantine groups. In many ways, the trends of scholarship on Israelite genocide mirror the trends of the field of Biblical Studies, where interest in texts over against historical contexts is in many cases an outgrowth of the religious backgrounds and interests of biblical scholars. Two exceptions are P. M. A. Pitkänen, Joshua, Apollos Old Testament Commentary 6 (Westmont, IL: InterVarsity Press, 2010) and ‘War in Deuteronomy’ in Oxford Handbook of Deuteronomy, ed. D. C. Benjamin (Oxford University Press, online 2020); and T. M. Lemos, ‘Dispossessing nations’ and ‘Neither mice nor men: dehumanization and mass killing in first-millennium Mesopotamian sources, He¯rem texts, and the War Scroll’ in With the Loyal You Show Yourself ˙Loyal: Essays on Relationships in the Hebrew Bible in Honor of Saul M. Olyan, ed. T. M. Lemos et al. (Atlanta: Society of Biblical Literature, 2020), pp. 249–65. Although Pitkänen is one of the few scholars who has engaged with interdisciplinary Genocide Studies in his treatments of biblical texts, he unfortunately argues for an extremely early date for the relevant texts that is not in line with the scholarly consensus, which limits the historical utility of his work. I have attempted in my research to shift the focus to a broader set of historical features – demographic changes, resource scarcities, identity construction – while incorporating frameworks from comparative Genocide Studies in a 2015 essay and in this current chapter. In general, however, one can only conclude that there is a good deal more work to be done on genocide and exterminationist violence in ancient Israel and the broader region of ancient West Asia.

208

7

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages t. m. lemos and seth richardson

Introduction To state that the empires of ancient Mesopotamia were brutal in their violence is no great historical insight. Their reputation is such that one finds the following statement in a book on the psychology of modern warfare: ‘Genocide is by no means new – the ancient Assyrians piled high the bones of their victims to make small mountains – but only in the nuclear age has genocide become a routine military contingency.’1 However, as historians of ancient West Asia as well as scholars of violence, we must interrogate further the question of whether or not the Assyrians or other Mesopotamian groups practised genocide on a regular basis, or at all. This chapter focuses on the three most relevant sets of sources from Mesopotamia: evidence from the twenty-third century B C E related to the violent reprisals of two kings of Akkad (Rimuš and Naram-Sin) against the ‘rebellious’ cities of Sumer and Elam; the Sumerian ‘city laments’ dating to a few centuries later; and first-millennium B C E Neo-Assyrian royal inscriptions – those very inscriptions that detail the Assyrians making small mountains of the corpses of their enemies. What this examination will demonstrate is that while mass violence is hardly rare in accounts of Mesopotamian history, finding evidence for genocide is not as easy. Whether or not the Bibliographic abbreviations used in this chapter include ETCSL (the Electronic Corpus of Sumerian Literature, http://etcsl.orinst.ox.ac.uk/); RIME 1 (D. Frayne, Presargonic Period (2700–2350 BC), Royal Inscriptions of Mesopotamia 1 (University of Toronto Press, 2008)); RIME 2 (D. Frayne, Sargonic and Gutian Periods (2234–2113 BC), Royal Inscriptions of Mesopotamia 2 (University of Toronto Press, 1993)); and RIME 3/2 (D. O. Edzard, Gudea and His Dynasty, Royal Inscriptions of Mesopotamia 3/2 (University of Toronto Press, 1997)). 1 R. W. Rieber and R. J. Kelly, ‘Substance and shadow: images of the enemy’ in The Psychology of War and Peace: The Image of the Enemy, ed. R. W. Rieber (New York: Springer Science, 1991), p. 4.

209

t. m. lemos and seth richardson

Mesopotamians practised genocide at all hinges on how one defines the term, and whether the targeted destruction of that most vaunted of Mesopotamian social configurations – the city – constitutes a type of genocide. In addition to an exploration of urbicide as genocide, the chapter also considers how Mesopotamian accounts of mass violence contributed to accounts and practices of violence, including genocidal violence, elsewhere in ancient West Asia in later periods.

Questions of Definition The United Nations Genocide Convention of 1948 defines genocide as any of the following acts committed with intent to destroy, in whole or in part, a national, ethnical, racial, or religious group (a) Killing members of the group; (b) Causing serious bodily or mental harm to members of the group; (c) Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part; (d) Imposing measures intended to prevent births within the group; (e) Forcibly transferring children of the group to another group.

If we take this definition as our starting point then we see that genocide need not be limited to ethnocide, which we define as the slaughter of an ethnic group, and the most prototypical or stereotypical form of genocide.2 In fact, more recent definitions of genocide have tended to be expansive, including more and more types of social groupings and more and more types of detrimental, life-destroying behaviours.3 The question of which social groupings fall under the rubric of genocide is not an idle one when dealing with the very large body of Mesopotamian evidence. 2 By ethnocide, we mean here the killing of an ethnic group. We do not use it as a synonym for cultural genocide, as some scholars do, e.g. A. Jones, Genocide: A Comprehensive Introduction, 3rd ed. (London: Routledge, 2017), p. 35. 3 Most major scholars of genocide in fact favour definitions that are broad – they favour inclusion of political and other collective entities and sexual minorities, as well as inclusion of actions that do not necessarily involve killing. On how to define genocide, see Jones, Genocide, pp. 22–39, who lists the definitions of over a dozen scholars; B. Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007), pp. 9–21; I. W. Charny, ‘The definition of genocide’ and H. Fein, ‘Defining genocide as a sociological concept’ in The Genocide Studies Reader, ed. S. Totten and P. R. Bartrop (New York: Routledge, 2009), pp. 36–40 and pp. 44–56 respectively; and J. Hughes, ‘Genocide’ in The Routledge Handbook of Ethnic Conflict, ed. K. Cordell and S. Wolff (New York: Routledge, 2011), pp. 122–39; among many others.

210

Map 7.1 Mesopotamian and other Near Eastern sites discussed. (Cartography by Steven Townshend)

t. m. lemos and seth richardson

Besides ‘genocide’, there are two other terms worthy of brief discussion before we proceed to the evidence. The first is ‘genocidal massacres’, a term first used by the noted genocide scholar Leo Kuper in the 1980s. This term has been adopted by scholars such as Ben Kiernan and refers to violence compris[ing] shorter, limited episodes of killing directed at a specific local or regional community, targeted because of its membership in a larger group but at the hands of perpetrators who may not (or cannot be proven to) harbour an intent to destroy the group ‘as such’ (which the [UN] Convention requires). Genocidal massacres often serve as object lessons for other members of the group. They may be state-organised, or more spontaneous, such as outbreaks of deadly communal violence.4

This term will figure in the analysis below, as will the term ‘urbicide’. While scholars tend to use ‘urbicide’ to refer to the destruction of cities as symbolic spaces, focusing less on the death of human inhabitants than on the violent dismantling of city architecture, cityscapes and the ‘built environment’,5 our interest in this chapter is actually the opposite. We will give consideration to the city as a social grouping of human beings and relatedly to whether the destruction of the people of a particular city should be considered a form of genocide akin to the destruction of an ethnic, racial or religious group.6

The Suppression of Revolt: Akkad, Twenty-Third Century B C E We turn now to the evidence from Mesopotamia, beginning with the first of what are arguably the three best-known episodes of mass violence in Mesopotamian history: the reprisals of the kings of Akkad, Rimuš (2279– 2270 B C E ) and Naram-Sin (2254–2218 B C E), against rebellious cities in Sumer and Elam. In fact we know little about why these rebellions took place, only that they were suppressed.7 Indeed, they are not explicitly called ‘rebellions’ in contemporary texts. We are left to infer ‘rebellion’ from the fact that these 4 B. Kiernan, ‘Is “genocide” an anachronistic concept for the study of early modern mass killing?’, History 99:336 (2014), 530–48, quoted at 543. 5 See e.g. M. Coward, Urbicide: The Politics of Urban Destruction (New York: Routledge, 2009). 6 On this issue see also H. van Wees, ‘Genocide in the ancient world’ in The Oxford Handbook of Genocide Studies, ed. D. Bloxham and A. D. Moses (Oxford University Press, 2010), pp. 239–58, esp. pp. 243–5, though his analysis of Mesopotamian and Israelite evidence is not always accurate or convincing. 7 See B. R. Foster, The Age of Agade: Inventing Empire in Ancient Mesopotamia (London: Routledge, 2016), esp. pp. 6–8, 44–6 and 323–4.

212

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages

cities were under Akkadian rule both before and after these events, and that some later historical traditions refer to them as ‘revolts’. This is logical enough, but virtually all of the information we have about these events is told to us by these kings in their royal inscriptions, and these texts do not distinguish the victims from the perpetrators by any terms other than local identity. Because of these conditions of the sources, then, we can have little hope of analysing the events themselves. Our attention is therefore more usefully directed to the intent the sources had in representing facts as they did, rather than their objective accuracy. This is, of course, only the usual caution to be exercised when dealing with royal inscriptions or propaganda of any kind. We must also recognise that it is not always clear from the texts whether the massacres are battlefield violence or reprisals against civilians. Typically, we do not consider violence against combatants to be genocidal.8 The expression of a clear intent to destroy city-states completely and permanently (i.e. urbicide) is the specific innovation of these texts among royal inscriptions. The texts describe not only the conquests of specific cities one by one, but the capture of their kings and the destruction of their city walls. Burial tumuli were erected in their vicinity, as if in mocking imitation of the destroyed city mounds.9 It is the extraordinarily high numbers of dead that the texts list that demands our attention in the present context. These mass killings were not only the first ever (recorded) of their kind, but they took place in a world where individual city-states were the largest political forms: to kill a city was to kill a society. The erasure of a city’s king, wall and citizenry was not merely the destruction of a city within a state – even as important a city, say, as a Stalingrad in the Soviet Union, or an Atlanta in the Confederacy. These urbicides were the eradication of entire sovereign states and their societies, of independent political, religious and literary cultures whose traditions are now lost to us forever. It is therefore the goals of the killings that seize our attention. The details of what the texts tell us have been known for decades and may be 8 Of course, this is precisely how many genocides are styled by their perpetrators: as military conflicts. Consider e.g. the official views of the United States, Turkey and Myanmar concerning the Native American, Armenian and Rohingya genocides as military operations in response to raids by ‘savages’, to ‘banditry’ or to ‘terrorism’. This may be the case here, as well, where our only historical sources are state documents. 9 See, S. Richardson, ‘Death and dismemberment in Mesopotamia: discorporation between the body and body politic’ in Performing Death: Social Analyses of Funerary Traditions in the Ancient Near East and Mediterranean, ed. N. Laneri (Chicago: Oriental Institute, 2007), pp. 189–208.

213

t. m. lemos and seth richardson

Table 7.1 Casualties of military campaigns of Rimuš and Naram-Sin CATEGORY

Rimuš Naram-Sin Total by category

1

CATEGORY

2

CATEGORY

3

‘struck down’

‘captive (men)’

‘expelled’ and ‘annihilated’

Total casualties

72,244 6,850 79,094

44,424 13,827 58,251

23,685 none? 23,685

140,353 20,677* 161,030

* Although the totals for Naram-Sin seem smaller than those for Rimuš, many of the relevant texts are lacunate; and we may note that in one text about the Great Revolt he gives an otherwise undescribed total figure for all enemy casualities of 137,400 men (RIME 2.1.4.2 vii 11).

quickly summarised. The revolts against Rimuš came at his accession in around 2279 B C E, following the death of Sargon, the founder of the empire; the revolts against Naram-Sin cannot be dated precisely within his reign, but came about fifty years later.10 Seven inscriptions – five of Rimuš11 and two of Naram-Sin12 – describe reprisals against more than a dozen cities and lands. The narratives use the strategy of providing specific numbers of rebels in three categories, namely those ‘struck down’, those ‘(made) captive’, and (for Rimuš only) those ‘expelled’ and ‘annihilated’ (Table 7.1). Even by modern standards, the claims made here are staggering. Did these kings really kill more than 100,000 people and imprison almost 60,000 more? Unfortunately, the answer to this question depends on translation issues that ultimately cannot be resolved. Two of the four terms by themselves, ‘captive’ and ‘expelled’, are relatively straightforward. All forms of the word ‘captive’ in category 2 have clear etymologies and parallels, and are written unambiguously (L U´ ×E´ Sˇ , 13 L U´ ×K A´ R , and Sˇ U . D U 8. A ) ; and the adjective 10 This is the view of Å. Westenholz, ‘The Old Akkadian period’ in Mesopotamien: AkkadeZeit und Ur III-Zeit, ed. P. Attinger et al., Orbis Biblicus et Orientalis 160/3 (Fribourg: Academic Press, 1999), pp. 15–117, esp. p. 42. 11 RIME 2 1.2.1 (4–13, 29–35, reprisals against Adab and Zabalam); .2 (4–13, 29–35, against Umma and K I . A N); .3 (4–13, 30–6, against Ur and Lagaš); .4 (8–15, 30–7; against Sumer; 44–55, against Kazallu); and .6 (9–23, against Zahara and Elam). ˘ vi 13–19, against Ur, Uruk, Nippur, 12 RIME 2 1.4.2 (iv 13–25; against Umma and Adab; Lagaš and others); and .6 (iii 1´–12´, against Kiš and other cities; iv 1´–24´, iv 36´–45´, two campaigns against Kiš). 13 L U´ ×E´ Sˇ is something like a ‘roped(-up) man’; L U´ ×K A´ R , prob. ‘captured’ but poss. ‘fled’; Sˇ U . D U 8. A, ‘held’.

214

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages

‘expelled’ in category 3 derives from a causitive form of the verb wasû (‘to go ˙ out’), literally, ‘(those) made to go out’. But the meanings of ‘struck down’ and ‘annihilated’ in categories 1 and 3, respectively, are not beyond dispute. To begin with, the basic form of the verb conventionally translated as ‘struck down’ (maqa¯tu) means ‘to fall or collapse’. Here appearing in causative form, it can be translated either as ‘to overpower, defeat’, or as ‘to strike down, to kill’, with rather different implications. Unfortunately, neither by context nor semantics can we determine certainly which idea is meant.14 And to the second word, the term ‘annihilated’ – ana g/k/qara¯šim išku¯n, literally ‘I assigned them to g/kara¯šu’ – is even more problematic because there is no consensus as to what g/kara¯šu means. We should read either something like gara¯šu, ‘destruction’,15 or kara¯šu, ‘a labour camp’.16 Thus, the ghastly sense of tens of thousands of people being ‘struck down’ and ‘annihilated’ together could mean instead merely those ‘defeated (in battle)’ and/or those sent to ‘labour camps’. The options form a bit of a maddening puzzle. On the one hand, it seems unlikely that both the terms maqa¯tu and g/kara¯šu indicated fatal punishments, since it makes no clear sense to consistently enumerate the killing of (presumably different) people by two different terms. On the other hand, it also seems unlikely that the inscriptions mean to distinguish between two different kinds of prisoners, either, that is as people ‘made captive’ and others ‘put in a labour camp’. Other possibilities exist as well, either that they were conscripted,17 or that they were banished and cut off as members of their political community, as a social rather than physical annihilation.18 Unfortunately, as Å. Westenholz puts it, there is simply ‘no proof either

14 A similar problem was identified long ago by H. Tadmor for the verb dâku, which can similarly mean ‘to kill’ or ‘to defeat’ (‘Historical implications of the correct rendering of Akkadian dâku’, Journal of Near Eastern Studies 17:2 (1958), 129–41); see discussion below. 15 We say ‘something like’ because although a root grš or qrš points to a sense of ‘slaughter’ or ‘obliteration’, the Assyrian Dictionary (Oriental Institute, University of Chicago, 1956–2010) (CAD) does not resolve the question. Its entry for gara¯šu (CAD G (1954), ‘to butcher’; cf. gara¯šu A (‘to copulate’) and B (‘to come/go’)) says ‘see qara¯šu’. But by the time CAD Q had been published (1982), qara¯šu had been softened to ‘to trim, carve (meat)’, with no mention of the Rimuš passages. Meanwhile, neither did its entry in volume K (1971) for kara¯šu, ‘labour camp’, include the Rimuš passages. 16 See the discussion by Westenholz, ‘Old Akkadian period’, 41 and n. 130. 17 In later times, a kara¯šu could also mean an army troop. 18 It is not impossible to understand qara¯šu as meaning men who were ‘trimmed away’ (following CAD) from their city. This would be consistent with the (exclusive) Old Akkadian usage of the verb with which it is here paired,wasû, ‘to cause to go out’, when ˙ it referred to someone permanently alienated from a household. By this understanding, qara¯šued people were those effectively ‘annihilated’ as sociopolitical beings.

215

t. m. lemos and seth richardson

way’ about these translation choices from etymological or contextual information.19 These uncertainties obviously affect whether or not we interpret these episodes as even urbicides, let alone genocides. In comparative historical terms, however, it does seem significant that other third-millennium Mesopotamian kings, either before, during or after the Akkadian Empire, did not list any casualty numbers in battle accounts. An Early Dynastic ‘prisoner-plaque’ of an unknown king provides the only precedent, listing around 36,000 prisoners taken in about twenty-five separate campaigns.20 And we now have a new Naram-Sin inscription describing his conquest of Armanum and Ebla – foreign wars, rather than rebellions – in which totals above 85,000 prisoners are given.21 But for the most part, other kings both before and after mostly did not describe the scale of their victories at all, only occasionally making vague claims that the ‘myriad corpses [of the enemies] reached the base of heaven’,22 were ‘annihilated in a swamp’,23 and so forth – but they never numbered them.24 Not even the other kings of the Akkadian dynasty25 used numbers in any of their many descriptions of battles, even as they described sweeping victories far and near. This is remarkable given both the frequency of battle inscriptions generally and the propensity of kings to otherwise count and measure all kinds of things in specific terms in royal inscriptions – land, barley, canals, bricks, prices, interest, burial mounds, loaves of bread, jugs of beer and so forth. Thus, contrary to what we might expect of deep antiquity, neither empires nor royal inscriptions normally mentioned large or specific numbers of enemy dead or captured; the appearance of numbers in these inscriptions, then – whatever their specific meaning – was an exceptional narrative strategy rather than a normal topos. The numbers stated by Rimuš and Naram-Sin, therefore, were deployed for very specific ideological purposes. With Westenholz, we see no need to doubt the numbers as a more or less 19 Westenholz, ‘Old Akkadian period’, 41 and n. 130. 20 See P. Steinkeller, ‘An archaic “prisoner plaque” from Kiš’, Revue d’assyriologie et d’archéologie orientale 107 (2013), 131–57; the prisoners are listed in groups as large as 6,300, but as small as fifty men. Steinkeller tentatively proposes to date the plaque to the Early Dynastic II period. 21 N. Alkhafaji and G. Marchesi, ‘Naram-Sin’s war against Armanum and Ebla in a newlydiscovered inscription from Tulul al-Baqarat’, Journal of Near Eastern Studies 79:1 (2020), 1–20; numbers of the slain appear to have been lost in breaks to the text preceding iii´ 1´. 22 RIME 1 9.3.1 vii 21–2. 23 RIME 3/2 1.2.35 iii´ 3–5. 24 Significantly, numbers are absent in the several battle accounts of the Early Dynastic Lagaš kingdom, and the Ur III accounts in RIME 3/2 1.2.35–7; 1.4.1, .3 and .5; and 1.5.2–.3. 25 I.e. Sargon, Maništušu and Šarkališarri.

216

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages

true description of what happened, at least as orders of magnitude.26 But our attention to the accuracy of the numbers should not distract us from thinking about those ideological purposes, whatever the intractable translation problems are. We shall offer two new observations about these numbers. First, despite our rather imprecise ideas about the size of armies in Early Bronze Age Mesopotamia, they generally did not number in the multiple tens of thousands. Sargon of Akkad famously bragged that ‘5,400 men ate before him daily’; Ur III units are known up to about 10,000 men.27 But even these numbers describe total forces, not casualties, for which one might expect much lower numbers. Thus, the numbers given by Rimuš and Naram-Sin are not only unusual on their face, but they could not have been meant or understood as the casualty figures from a mere battle; they were numbers openly meant to be read as episodes of slaughter. Second, though the Akkadian Empire has always enjoyed a reputation as a conquest state, its major legacy of governance was that it was also a registration state. A large quotient of administrative texts of the empire were lists of people assigned to reside and work in specific households and cities. Akkad was an organisation that was largely about tracking and assigning manpower to create a legible and manageable body of subjects. This function creeps into view in these battle accounts in that they reflect a sorting of people into useful categories, as much as they document a triumphal celebration of massacres. At the symbolic level, this does not solve our translation problems, because a generalised notion of state control over bodies must be expressed in all forms. At the practical level, one might posit that the empire had an interest in managing living, working and taxpaying bodies superior to its eagerness to gloat over dead and incapacitated ones. For these two reasons – both the comparatively small bodies of force available in the third millennium and the empire’s superior interest in manpower over body counts – it is more probable that these seven texts are best explained as the propaganda of a managerial state representing control over (rebellious) bodies through a military idiom, rather than actual 26 Westenholz, ‘Old Akkadian period’, 41–3. 27 S. Richardson, ‘Mesopotamia and the “new” military history’ in Recent Directions in the Military History of the Ancient World, ed. L. Brice and J. T. Roberts (Claremont, CA: Regina Books, 2011), pp. 11–51, esp. pp. 35–6. Later Middle Bronze Age coalition armies – that is, the combined armies of several states – could range up to 40,000–60,000 men. But these are all numbers for the largest known armies by period, not typical ones.

217

t. m. lemos and seth richardson

killing. It is therefore far more likely that what we are talking about (with reference to Table 7.1) are 79,094 men defeated in battle (many of whom may have been killed in the process) and another 23,685 banished as citizens – and not 102,779 men killed/executed. This reading cannot and does not dispute that these events produced mass death and destruction. Nor should we consider state control of bodies nonviolent just because it may not inflict physical pain or death. These texts more likely describe programmes of registration and resettlement than of corporal, carceral and capital punishments. But on the ideational level, the numbers were clearly meant to shock: they tell us (at the least) that something bad happened to about a third of the entire male population of lower Mesopotamia within fifty years’ time, in repeated and repeatable sanctions. If we accept the texts’ premise that the cities they described were indeed destroyed, these sanctions were primarily meant as object lessons for other cities that had not revolted, and whose male citizens still existed to be able to weigh the cost/benefit ratio of political compliance. In this context, the shock of urbicide rather than mere military defeat was precisely the operative threat: what was at stake was not only death or bodily punishment for individuals but also the destruction of kingship and cult, temple and city, urbs et cives. The novel scale of destruction lay at the heart of the message. Mutatis mutandis, however, even if this less bloody interpretation is correct, it is significant that later ages remembered the Akkadian repression in starkly literal terms, in stories whose factual dependability has often been called into question.28 The original inscriptions themselves remain known to us precisely because they were inscribed on magnificent display statues that remained on view for centuries and were copied as late as the first millennium B C E. Thus, and partly on the basis of such royal inscriptions (but also demonstrably from literary tales, some as old as the Akkadian state itself), a parallel body of historical-literary stories emerged, focused on the exploits of the Sargonic kings, including two about Naram-Sin, the Great Revolt and the Enemy Hordes. In these stories, the enemies unambiguously destroyed indeed number in the tens and hundreds of thousands, and are lent a sub- or

28 See M. Liverani, ‘Model and actualization: the kings of Akkad in the historical tradition’ in Akkad: The First World Empire: Structure, Ideology, Traditions, ed. M. Liverani (Padua: Sargon, 1993), pp. 41–68; S. Tinney, ‘A new look at Naram-Sin and the “Great Rebellion”’, Journal of Cuneiform Studies 47 (1995), 1–14; and Foster, Age of Agade, pp. 258–61.

218

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages

non-human aspect, as quasi-mythical beasts in behaviour, appearance and origin.29 Such monuments and stories had continuing referential force: the reconquest of the old Sumerian cities by the Babylonian king Samsuiluna five centuries after Naram-Sin, for instance, was recounted and perhaps enacted clearly based on these legends and inscriptions: Samsuiluna unambiguously claims to have ‘massacred’ (nêru) tens of thousands of men, including in some of the same cities (notably, Ur and Umma).30 And the fantastically racialised elements of the legends in turn folded into a broader literary stream which branded and dehumanised all manner of foreign enemies in later ages as the ‘Umman-manda’, a generic term for ‘enemy horde’ with the same sense as ‘barbarians’. The suppression of the revolts against Rimuš and Naram-Sin thus may not have been actual genocidal acts in fact. But their descriptions of shocking political violence were ideationally and historically impactful as full urbicides, meant to make a rupture with all past traditions and establish a kind of political year zero, worthy of re-enactment in later ages. The texts styled repression as a systematic programme meant to completely erase political communities through the mass destruction of bodies and bodies politic. In this, the Akkadian inscriptions contributed an influential myth of quasigenocide to later ages that sometimes took them at face value, and even valorised them as an apparent precedent to the dehumanisation and destruction of entire populations.

Urbicide: Sumerian City Laments, Twenty-First to Eighteenth Centuries B C E The events of the Akkadian period left an indelible imprint on Mesopotamian historical memory. This memory certainly lay in the background when, after the next large Mesopotamian state (that is, the Ur III kingdom), collapsed in 2004 B C E, a series of major Sumerian compositions emerged that were theologically explanatory of urbicide. Five ‘city laments’ are known, with a total of about 1,700 preserved lines of text.31 Four laments 29 See W. Sommerfeld, ‘Naram-Sin, die “Grosse Revolte” und MAR.TU-ki’ in Assyriologica et Semitica: Festschrift für Joachim Oelsner . . ., ed. J. Marzahn, H. Neumann and A. Fuchs, AOAT 252 (Münster: Ugarit-Verlag, 2000), pp. 419–36. 30 See W. G. Lambert and M. Weeden, ‘A statue inscription of Samsuiluna from the papers of W. G. Lambert’, Revue d’assyriologie et d’archéologie orientale 114 (2020), 15–62. 31 Quotations and lineation follow the online editions of ETCSL; the five laments are rubricised as follows: LUr = Lament for Ur (438 lines); LSUr = Sumer and Ur (519 lines);

219

t. m. lemos and seth richardson

are for Ur,32 Uruk,33 Eridu34 and Nippur;35 the fifth is for Sumer and Ur.36 The narratives begin with the disaffection of the tutelary deity from his or her city. The reasons for this alienation are never clearly stated, but from that point arise attacks on the city by foreign enemies, the onslaught of natural forces, famine, political collapse, and the breakdown of social order. These events explicitly verge on the extermination of all humanity. The rising action of the laments finally climaxes in the reconciliation of the god with city and temple, and the restoration of the divine order. They are, fundamentally, texts about the restoration of social order. A sixth and slightly earlier composition, The Cursing of Akkad,37 dating to the Third Dynasty of Ur itself, employed many of these same themes of citystate destruction as the laments, but to a substantially different purpose. The poem depicts the complete collapse of Akkad and straightforwardly celebrates its demise; by the end of the poem, the city lies damned and abandoned, with streets covered in grass and ghostly winds whistling through empty buildings. The dramatic hinge is the impiety of a king (Naram-Sin, in fact), who ignores the will of the gods that he not rebuild the temple of Enlil at Nippur. This city destruction is, therefore, a morality play about hubris and piety rather than a rejuvenation after cataclysm. Do these poems have any historical value? This question has produced scholarly debate for a century, which we will not rehearse here.38 Suffice it to say that the poems take at least superficial interest in historicising the action, and would certainly have been understood as historical dramas. The Lamentation for Sumer and Ur, for instance, is lightly historicised with plausible (if demonstrably inaccurate) details about attacking enemies and the fate of

32 33 34 35

36 37

38

LN = Nippur (323 lines); LU = Uruk (fragmentary, about 241 lines preserved); LE = Eridu (fragmentary, about 173 lines preserved). For edition and analysis see N. Samet, The Lamentation over the Destruction of Ur, Mesopotamian Civilizations 18 (Winona Lake, IN: Eisenbrauns, 2014). For edition and analysis see M. W. Green, ‘The Uruk Lament’, Journal of the American Oriental Society 104:2 (1984), 253–79. For edition and analysis see M. W. Green, ‘The Eridu Lament’, Journal of Cuneiform Studies 30:3 (1978), 127–69. For edition and analysis see S. Tinney, ‘The Nippur Lament: royal rhetoric and divine legitimation in the reign of Ishme-Dagan of Isin (1953–1935 B.C.)’ (PhD thesis, University of Pennsylvania, 1996). For edition and analysis see P. Michalowski, The Lamentation over the Destruction of Sumer and Ur, Mesopotamian Civilizations 1 (Winona Lake, IN: Eisenbrauns, 1989). Also following ETCSL, CA = The Cursing of Akkad (281 lines). For a full edition and analysis see J. Cooper, The Curse of Agade (Baltimore: Johns Hopkins University Press, 1983). See most recently N. Samet, ‘Reading literary texts historically: the Sumerian city laments as a test case’, Shnaton 24 (2016), 151–66 (in Hebrew).

220

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages

Ur’s king Ibbi-Sîn; the laments for Uruk and Nippur feature a cameo appearance by Išme-Dagan, a king of Isin, as the providential hero who restores city and temple; the Cursing of Akkad proposes to depict (but instead confuses) Akkad’s collapse as something occurring in the time of Naram-Sin, though we know this actually happened at least twenty-five years after his death. In most other terms, the texts’ fidelity to historical events is slight or irrelevant. But historical facts were hardly the point for the composers of these texts: their purpose was to assert ideal types of divine history. The laments present an agonistic theology of the state: its dependence on divine forces whose will is unknown; and the necessity of passing through catastrophe, where total ruin is a vividly and specifically imagined possibility. The renewal of the world is only possible for having survived its near annihilation. Here, destruction is the unavoidable cost of a greater renewal, as it is in other sacrifice narratives such as the Crucifixion and Resurrection. The texts evoke not only widespread fatal violence but apocalyptic levels of it – in the Lament of Ur, where it is clear that everyone is to be killed: The gods gave instructions that my city should be utterly destroyed . . . Its people littered its outskirts just as if they might have been broken potsherds . . . On its lofty city-gates where walks had been taken, corpses were piled. On its boulevards where festivals had been held, heads lay scattered. In all its streets where walks had been taken, corpses were piled. In its places where the dances of the Land had taken place, people were stacked in heaps. They made the blood of the Land flow down the wadis like copper or tin. Its corpses, like fat left in the sun, melted away of themselves. The heads of its men slain by the axe were not covered with a cloth. Like gazelles caught in traps, their mouths bit the dust. Men struck down by the spear were not bound with bandages. As if in the place where their mothers had laboured, they lay in their own blood.39

The cataclysm is ordered by the gods,40 but it is effected by an amorphous combination of natural forces on the one hand (characterised as the winds and waters of a ‘flood’ or ‘storm’) and hordes of faceless, animalistic foreigners on the other.41 The violence is utterly indiscriminate: strong and weak, good and bad, men and women, young and old all perish in its 39 LUr 161–8, 207–29. Similar depictions of total slaughter are found in all six compositions, e.g. LSUr 69–78, 225–34; LN 62–7, 95–102; LU D 10–20, E 51–65, 66–74, 89–99; LE 30–7; CA 201–44. 40 For example LE C 1–8: ‘(The god) Enlil, king of the lands, looked maliciously at Sumer. He demolished it.’ See similarly LN 95–102, LSUr 69–78, LU 136–43. 41 Variously Gutians, Subarians, Šimaškians and Elamites: see LSUr 27–37 and 69–78, LU E 89–99.

221

t. m. lemos and seth richardson

wake.42 The killing is not confined to battle combat, but includes grisly and torturous deaths – by burning, bleeding to death and the execution of the wounded.43 We see the abuse of human remains, with corpses piled up like heaps of grain, left floating in the river and ‘torn apart’; heads ‘sown like seeds’ and noses(!) heaped up.44 The deliberate nature of the brutality depicted in the poems comes closer to modern definitions of genocide than in the earlier Akkadian texts. It extends, for instance, to inflicting inimical ‘conditions of life’ on the group ‘calculated to bring about its physical destruction’, as the UN convention on genocide states. This comes first in the familiar form of the deprivation of subsistence, of food, water and shelter: ‘Those who went to fetch food, went away from the food and will not return; those who went to fetch water, went away from the water and will not return.’45 Homes are abandoned, settlements disperse and roads fade back into desert; refugees fill the plain like stampeding goats, chased like dogs.46 Everyday basics are in fact cruelly parodied, inasmuch as symbols of abundance – heaps of grain and streams of fresh water – are symbolically inverted as stacks of bodies and rivers of blood.47 The punishments are not limited to temporary or individual deprivations, but the wholesale collapse of Sumer’s productive landscape: fields and marshes dry up; the ground cracks ‘like a kiln’; rivers are filled with dust; orchards fill with brambles; the sheep and even the wild animals have fled meadow and steppe alike. The entirety of Sumer is made an uninhabitable hellhole. The poems repeatedly portray the infliction of ‘mental harm’ in evocations of affective and cognitive distress. Weeping suffuses the entire corpus – indeed, the entire (performed) genre is rubricised as ‘lament’.48 Words like ‘bitterness’, ‘despair’, ‘terror’ and ‘depression’ populate the refrains.49 The people cannot sleep for fear; their ‘minds are thrown into disorder’; ‘their brains are muddled’.50 The sense of entire urban populations en masse becomes ‘distorted’ and ‘deranged’.51 The texts are thus not content to report on these apocalyptic events as dry matters of fact; they depict the

LSUr 69–78, 207–29; LU E 51–65, 66–74; LN 62–7. 43 LUr 207–29. LUr 207–29; LU E 66–74; LN 62–7; LE 30–7; LSUr 79–103; CA 176–92. 45 LSUr 251–9. LN 42–9; LU E 66–74, 78–88; CA 149f.; LSUr 185–92. 47 LU D 10–20. From Sumerian er2, ‘weeping, mourning, tears; to weep’, forming the generic label er2 -šu3-a, ‘a lament’; also i-lu, ‘a sad song’, and a-nir, ‘a lament’. 49 From, respectively, Sumerian gig, ur5 uš2 (lit., ‘dead liver’), ni2, and šag4 sag3 (lit. ‘pounding heart’). 50 See LU E 32–40; LN 48; LU E 56. 51 See LU A 21–7. 42 44 46 48

222

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages

entire spectrum of emotional and mental distress one might expect of cataclysms. The compositions offer less clear evidence, however, for the disruption of a population’s reproduction by the ‘prevention of births’ or ‘transfer of children’, though they do contain some suggestive passages: LUr 218–40: The little ones lying in their mothers’ arms were carried off like fish by the waters. Among the nursemaids with their strong embrace, the embrace was pried open.52 CA 222–44: [The gods cursing the city:] May the cattle slaughterer slaughter his wife, may your sheep butcher butcher his child! . . . May your prostitute hang herself at the entrance to her brothel! May your pregnant priestesses and cult prostitutes abort their children!53 LN 201–10: [At the conclusion:] [The gods] have gathered back together the children whom they turned away from their mothers.

The image of children at risk is also carried over into another of the poems’ broad themes, the metaphorisation of the city as a stray and abandoned animal: Ur is depicted as a goat whose kid has perished and a ewe whose lamb has been ‘torn away’; Sumer as a land where ‘sheep will no longer multiply in the fold’.54 But it is clear that these set scenes are mostly symbolic extensions of the larger total destruction, not any specific programme to interrupt the biological or cultural reproduction of the target population. The focus is not so much on external controls to reproduction, but on the breakdown of the social fabric of the family, where father turns away from wife, mother turns away from child, families abandon one another and disperse. The total violence extends even beyond the grave, with the ghosts of the family dead chased away.55 These are themes common to all the compositions:56 it is central to the genre’s message that the breakdown of the divine order is reproduced not only at the level of the city, but extends downward to destroy even the most intimate family relations.57 52 From a later passage (230–40) in the same composition: ‘In the city the wife was abandoned, the child was abandoned . . .’ and similarly lines 402–10; lines 275–85, ‘My slave-girls and children have been carried off by boat’. 53 The translation of ‘pregnant’ and ‘abort’ is not at all certain; the literal meaning is only that these mothers would ‘return’ (gi4) their children. 54 LUr: 64–71; LSUr: 3–11. 55 LSUr 69–92. 56 With the possible exception of the fragmentary LE. 57 This theme finds its counterpoint in stressing the profound associality of the afflicting ‘storm’, which ‘knows no mother / father / wife / child / sister / brother / neighbor / female companion’ (LUr 230–40).

223

t. m. lemos and seth richardson

This alerts us to the fact that our conception of what the ‘total destruction of a people’ looks like may not match the emic conception of the same thing. The compositions also include motifs indispensable to an ancient Mesopotamian conception of a genocide not indicated in the UN Convention. Crucially, these motifs include, along with the symbolic homology of divine-royal-city-household collapse, the abandonment, destruction or defilement of the city’s temple; the departure of the god; divine wrath in the form of natural forces (wind, flood, flames, hailstones); and significant poetics of (eerie) silence and darkness. The extermination of a people would have been meaningless to Mesopotamians without these images and ideas. Yet even setting aside these culturally autonomous instances – even if what remains satisfies some necessary criteria for a ‘genocide’ – it is not clear that it is sufficient. Can genocide be committed by non-human agents (i.e. the gods)? That is, is anthropocide, the killing of the human race or the great majority of it, a kind of genocide? Can we call it a ‘genocide’ if the target group is everyone? Can we anyway count a fictional literary depiction? Again this becomes relevant at the conceptual rather than the documentary level, insofar as the laments (and the Cursing of Akkad) exercised influence over later compositions: total destruction became a subject worth elaborating at the compositional level. The destruction of humankind was a topic carried forwards from this point into the Sumerian Flood Story and Atrahasis (then folded into the ˘ Gilgameš epic), the story of Erra and Išum, and the Cuthaean Legend of NaramSin, later literature which was copied and read well into the first millennium. The termination of all humanity thus became an indispensable leitmotif for later remakings of the world. The theme was both a radical break from the worldcreating genesis stories of older Sumerian literature, which were set in a verdant Edenic world of talking animals, gods appointing the names and roles of all beings, and starkly different from one of the other major themes of later Akkadian literature, which focused on the individual alienated from society. These six compositions thus formed the models for later explanatory devices about the deportations and urbicides of Mesopotamia yet to come in its next, imperial phases, where themes of genocide acquired the referential truth of parable.

Genocide in Neo-Assyrian Sources, Tenth to Seventh Centuries B C E The third set of evidence from Mesopotamia that we will examine dates to the first half of the first millennium B C E. The Neo-Assyrian Empire is well known for its brutality, and the annals of Neo-Assyrian kings provide us with 224

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages

descriptions of assaults, at least some of them historically verifiable, against a whole host of enemies. This set of evidence will raise questions of definition similar to those raised by the earlier Mesopotamian evidence. While in some ways clearer than the earlier corpora of sources, still there is an ambiguity belied by the graphic nature of so much of the evidence. Many are familiar with the scenes of impalings and flayings and decapitations carved so painstakingly on Neo-Assyrian reliefs, and plentiful are the Neo-Assyrian royal inscriptions detailing with relish and brio the subjugation, dehumanisation and slaughter of this or that rebel king, this or that transgressive vassal city (Figure 7.1). In light of the bloodthirstiness of these records, it is logical to ask whether the Assyrian Empire that flourished in the first half of the first millennium B C E practised the most totalising form of violence – genocide – either on a regular basis or on

Figure 7.1 Assyrian relief from the palace of Sennacherib. (Photograph from Werner Forman / Getty Images)

225

t. m. lemos and seth richardson

a sporadic one, or ever. Despite the rational nature of the question, however, our analysis will conclude that it is difficult to arrive at a truly satisfying answer. As is well known, Assyrian royal inscriptions tend to be very repetitive in their descriptions of violence, with certain actions, metaphors and tropes being repeated again and again. In Middle and Neo-Assyrian sources, Assyrian kings describe themselves subjugating transgressive vassals by means of dehumanising violence dozens and dozens of times. Still, the phraseology of different kings – or at least the phraseology of their scribes – varies in the inscriptions they left to us. It is in these divergences where one finds evidence for practices that could perhaps be seen as genocidal, or as more exterminationist in intention than the violence Assyrian kings normally boast of committing. To put the matter more directly: although evidence for the Assyrians making use of genocide as a regular strategy of violence is in our view lacking, the language of certain kings in certain inscriptions could be interpreted as describing genocidal violence. A few representative examples will illustrate the point. The annals of the Neo-Assyrian king Sennacherib (704–681 B C E ) describe on several occasions the subjugation of the rebellious city of Hirimmu. Here are three of these ˘ accounts: I put to the sword the population of the city Hirimmu, a dangerous enemy ˘ to the kings, my ancestors, who since time immemorial had not submitted and I did not leave one alive. (Sennacherib 1, 58) I put to the sword the population of the city Hirimmu, a dangerous enemy, and I did not spare a single one. I hung their˘ corpses on poles and placed (them) around the city. (Sennacherib 4, 16) I put to the sword the inhabitants of the city Hirimmu, not sparing (any of) ˘ their offspring. (Sennacherib 42, 6)58

Despite a penchant for dehumanising violence, Sennacherib does not typically claim to have eliminated all of the population of a group. Yet his

58 It is worth noting that the differences in language that we address here are not (for the most part) due to translation issues, as was the case with the Akkadian material. All translations of Neo-Assyrian royal inscriptions quoted here are taken from the Open Richly Annotated Cuneiform Corpus, http://oracc.museum.upenn.edu/rinap/cor pus/. Most royal inscriptions from the Neo-Assyrian period can be found on this site. Translations of the annals of Sennacherib on the site are from A. K. Grayson and J. R. Novotny, The Royal Inscriptions of Sennacherib, King of Assyria (704–681 BC), Parts 1 and 2 (Winona Lake, IN: Eisenbrauns, 2012, 2014), abbreviated RINAP 3/1–2. Similar statements about Hirimmu can be found in: 2, 15; 3, 18; 15, ii 1b; 16, i 81; and elsewhere in ˘ Sennacherib’s inscriptions.

226

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages

treatment of Hirimmu is different and explicitly exterminationist – he intends ˘ to kill all the inhabitants of this city and claims to have done so. While this behaviour is an exception rather than the rule, it is not the only exception in Sennacherib’s inscriptions. The king also claims to have inflicted totalising, exterminationist violence against Babylon, a city he certainly hated no less than Hirimmu. His annals state: ˘ On my second campaign, I marched quickly to Babylon, which I planned to conquer, and (then) I blew like [the onset] of a storm and enveloped it like a (dense) fog. I besieged the city . . . Its people, young and old, I did not spare, and I filled the city squares with their corpses. I carried off alive to my land Šu¯zubu (Muše¯zib-Marduk), the king of Babylon, together with his family (and) his [. . .]s.59

While certain captives are kept alive, Sennacherib claims to have left no survivors from the people of Babylon, whether young or old. This merismus of young and old seems to imply eliminationist violence, as does the language of ‘not sparing’. The verb eze¯bu is used in Assyrian inscriptions to refer either to sparing or not sparing; it is not consistently used to refer to totalistic killing. But in this instance, its use is part of what implies eliminationist violence. Of course, we know that Sennacherib did not in fact destroy all the people in Babylon. The Babylonians, or at least a good number of them, in fact lived on to vex the Neo-Assyrian Empire another day. This fact could prompt us either to reinterpret Sennacherib’s language as not actually implying eliminationist violence, or to conclude that Sennacherib was, as so many ancient kings, a hyperbolic exaggerator, in this case claiming to have committed urbicide but not actually having done so. Of course, genocide can involve the intentional destruction of a group ‘in whole or in part’, and here the intent to destroy seems clear. Other statements by this king that could also possibly be taken as referring to genocidal violence are less clear. For example, Sennacherib states of Maniye, ruler of Ukku (modern-day Hakkâri in Turkey): ‘I destroyed, devastated, (and) bur[ned] with fire his cities, (and) made (them) like ruin hill(s) (created by) the Deluge.’60 If one were predisposed, one could see this as referring to eliminationist violence, but if so, then the inscriptions themselves seem more interested in discussing the destruction of the cities than they do the people in these cities. More importantly, elsewhere in his annals Sennacherib states that he carried off captives from these cities and then 59 Sennacherib 223, 43b.

60 Sennacherib 19, ii’ 12’.

227

t. m. lemos and seth richardson

razed them.61 Thus, his lack of reference to taking captives in this instance probably does not imply that the people along with the cities were eliminated. Also, Sennacherib uses similar language when speaking of various other cities while also mentioning taking both livestock and people as booty from those places. Sennacherib is not the only Assyrian king whose actions or alleged actions could be taken to refer to genocide. The inscriptions of Aššurnasirpal II state: I crossed over to Mount Kašiiari (and) approached the city Kinabu,62 the fortified city of Hula¯iia. With the mass of my troops (and) my fierce battle I besieged (and)˘ conquered the city. I felled with the sword 800 of their combat troops. I burnt 3,000 captives from them. I did not leave one of them alive as a hostage. I captured alive Hula¯iia their city ruler. I made a pile of ˘ boys (and) girls. I flayed Hula¯iia their their corpses. I burnt their adolescent ˘ city ruler (and) draped his skin over the wall of the city Damdammusa. 63 I razed, destroyed, (and) burnt the city.

And what had Hula¯iia done to merit such violence to himself and those under his power? The˘ inscriptions state that he, aided by some Assyrian men, ‘had rebelled; they had come to capture the city of Damdammusa, my royal city’.64 Complaints of rebellion are common enough in Assyrian inscriptions, as are descriptions of the violent subjugation of those who rebel. Nonetheless, this stands as an atypical case of an Assyrian ruler explicitly stating that no captives from the rebellious group were left alive. At other points in his inscriptions, as well, Aššurnasirpal II (883–859) states that he ‘razed, destroyed, burnt, (and) consumed’ particular cities.65 These other cases, however, do not explicitly state that no survivors were left alive. The phrasing ‘razing, destroying, burning’ does not necessarily or even usually feature exterminationist violence, but it is vague and could imply totalising killing. Another example of eliminationist killing in the corpus of Assyrian inscriptions is found in the annals of the tenth century B C E ruler Aššur-Dan II, where we read in a rather fragmentary section of text: With the support of Aššur, my lord, [I] mustered [. . . my chariots (and) troops]. I plundered their settlements from the city Ekal-pı¯-na¯ri [(. . .) to . . .] 61 See e.g. Sennacherib 17, iv 58. 62 This may be the city of Aktepe, in modern-day Turkey, about fifty miles from the Syrian border, or Taus¸antepe, according to E. Lipin´ski, Arameans, Their Ancient History, Culture, Religion, Orientalia Lovaniensia Analecta 100 (Leuven: Peeters, 2000), p. 148. 63 001 i 106–10. 64 001 i 101b. 65 The verbs appear in different combinations. See e.g. Aššurnasirpal II A.0.101 i 65, 110; A.0.101 ii, 44–7, 55–8, 116–17; A.0.101.17 i 76, iii 45–55; and elsewhere.

228

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages

(and) inflicted [upon them a major defeat]. Those that survived I slaughtered. [I carried off] their [. . . herds] (and) flocks without number. [I] burnt [(. . . and)] their [cities] (with) their inhabitants. I brought up from the Arameans [valuable booty].66

The text seems to describe totalising killing here, though its reconstructed nature makes that fact not entirely certain. In fact, there are many cases in Assyrian royal inscriptions where the killing described seems to be extensive, but there is no explicit claim to have destroyed everyone in a place, or the language is vague in some way. For example, the annals of Shalmaneser III state: ‘Moving on from the city Zaban I approached the city Mê-turnat. I laid siege to the city, captured (it), slaughtered its (people), and plundered it.’67 When Shalmaneser claims to have slaughtered this city’s people, does he mean every single person, or just many people? Either scenario could be argued to be genocidal violence depending on the parameters one sets for genocide, though the former case would be easier to justify categorising in such a fashion since the killing of many people in a particular place was in line with the general level of violence of ancient West Asian empires and rarely had an exterminationist purpose or result. It is useful to compare the phrasing and descriptions of mass killing just quoted with those from two highly relevant biblical texts. Deuteronomy 20:16–17 reads: ‘From the cities of these peoples that Yahweh your God is giving you as an inheritance, do not let anything that breathes remain alive. You shall surely annihilate them – the Hittites and the Amorites, the Canaanites and the Perizzites, the Hivites and the Jebusites – just as Yahweh your God has commanded you.’68 Deuteronomy 2:34 similarly states: ‘We captured all [of Sihon’s] cities at that time, and on each city we performed h¯erem against men, women, and children. We left no survivors.’ ˙ The term h¯erem is a technical one referring to totalistic killing, and both these ˙ texts feature it or verbs from the same root. Deuteronomy 20 also delineates specific ethnic groups that must be slaughtered. However, the point in quoting from these biblical texts is not to engage in a discussion of h¯erem ˙ 66 A.0.98.1, 7–15. Reconstructions also make it unclear whether or not Aššur-bel-kala (01, 1’) describes totalising killing. 67 A.0.102.5, iv 2–3. The uncertainty of what this means is an even deeper translation issue. Note the interpolation of the parenthetic ‘(people)’, and compare how CAD renders this and parallel Shalmaneser III phrases of GAZ.MEŠ-šu(-nu) a-duk, where GAZ.MEŠ with dâku is translated as the milder ‘to inflict losses/defeat’, not ‘slaughter’; see CAD D s.v. dı¯ktu. 68 All biblical translations are by T. M. Lemos.

229

t. m. lemos and seth richardson

but rather to emphasise how clear the texts from Deuteronomy are in describing totalistic killing – and Chapter 6 in this volume includes many more examples from the Israelites, as well as their Levanite neighbours the Moabites, of clear genocidal language. It is not just that the term h¯erem is ˙ used; no, the totality of the killing is made explicit in at least some cases with descriptions of the types of humans killed – men and women, children and adults, young and old – leaving no doubt as to the eliminationist aspect of the killing. In the biblical books of Deuteronomy and Joshua, there is a clear ethnocidal nature to the killing, with the autochthonous Canaanite inhabitants of the promised land being named and targeted for total destruction. This clarity highlights differences between biblical h¯erem texts and Assyrian ˙ royal inscriptions in terms of language, target groups and motivation for slaughter. No matter how extreme the violence inflicted upon victims, Assyrian inscriptions emphasise rebellion as the motive for repressive violence, not the ethnicity of the rebels. One could take the view that under the Genocide Convention, what matters is the ‘intent’, the deliberate nature of the act, rather than the motive, and that therefore the Assyrian cases could fit the contemporary legal definition of genocide. But they also remain distinct from the biblical cases in terms of the category of the victim group. In the Assyrian inscriptions, it is typically the rebellion of vassal kings or other leaders of foreign groups that is highlighted rather than foreign groups as a whole being described as inherently rebellious or prone to disobedience or evil. By contrast, Deuteronomy and Joshua portray particular groups in precisely the latter fashion – as inherently dangerous, culpable for past wrongs, and with the potential to lead Israelites to stray from Yahweh in the future. This inherent evil is what justifies totalising violence against them, violence that is explicitly exterminationist. Assyrian texts do not go in this direction. This is why one even has to ask the question whether or not they describe genocide, because they did not target ethnic groups as such and nor did they target as such other categories of groups listed in the UN Convention, such as racial or religious groups.69 Only rarely is totalising violence described clearly, and even then it is not, in our view, ethnocidal. How, then, should these cases of mass violence be classified? While they seem not to fall into the category of genocide as the term is defined by the UN Convention or in common parlance, we could perhaps see them as examples 69 While one could more easily argue that Mesopotamian states targeted ‘national’ groups, they rarely did so with the intention of eliminating them from the face of the earth, and the identity of the nation as such was not the focus but rather the actions of rebellion and insubordination that they had putatively committed.

230

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages

of ‘genocidal massacres’ – short bursts of extreme violence aimed at a regional community. The problem with this is in determining whether victims were ‘targeted because of [their] membership in a larger group’, which is the second part of the definition of ‘genocidal massacre’ quoted above from Kiernan. One could say that they were targeted for being part of a larger group of rebellious people or people led by a rebellious king, but here one treads on shaky ground. As with any technical term, if one continuously expands the parameters of genocide to include more and more cases of mass violence, then the term ceases to have explanatory value. The most compelling reason to regard the cases of totalising violence in Assyrian royal inscriptions as genocide is if we understand the city as a social group with a clear enough identity that targeting the population of a city could be characterised as genocide. In a recent essay, Jacob L. Wright has identified cases of urbicide in ancient Near Eastern texts, understanding urbicide as symbolic urban destruction aimed at the physical structures of cities.70 His analysis is persuasive. Certainly, we have quoted enough texts describing the razing, destroying and burning of cities to make clear that Mesopotamians targeted cities. This action held, it seems certain, a symbolic meaning for them, and the destruction of cities can be regarded, as Wright suggests, as a particular type of ritual violence. Our analysis above of the earlier corpora is in line with this conclusion. Of course, that the Mesopotamians attached great symbolic value to the city and to particular cities is well known. This attachment is present in third-millennium Sumerian texts and carries on for many hundreds of years in Akkadian ones. Thus, it seems plausible to say that the city was a social unit with enough importance and a clear enough identity in Assyrian culture that the destruction of the population of a city could be described as a type of genocide.71 If we posit this, we are on a stronger footing in claiming that the Assyrians occasionally practised genocide and were sometimes responsible for 70 J. L. Wright, ‘Urbicide: the ritualized killing of cities in the ancient Near East’ in Ritual Violence in the Hebrew Bible: New Perspectives, ed. S. M. Olyan (Oxford University Press, 2015), pp. 147–66. 71 On rare occasions, scholars have previously equated Mesopotamian urbicide with genocide. M. Freeman states of Sargon of Akkad: ‘The threat of perpetration of genocide, in the form of destruction of cities, constituted an economy of violence, not in the sense of minimising violence, but in that of maximising the efficiency of conquest’ (‘Genocide, civilization, and modernity’, British Journal of Sociology 46:2 (1995), 207–23, at 219). What Freeman cites is M. Mann, The Sources of Social Power: A History of Power from the Beginning to A.D. 1760 (Cambridge University Press, 1986), but Mann generalises without citing actual texts and does not actually apply the terms ‘genocide’ or ‘urbicide’ to Sargon or Mesopotamian history in the work cited.

231

t. m. lemos and seth richardson

genocidal massacres, particularly if ambiguous cases are not assumed to be innocuous and if we consider seriously the possibility of totalising violence being committed by a group that regularly and unselfconsciously claimed to dismember, impale, flay and display the mutilated bodies of victims. It would be difficult to assert that the Assyrians lacked the psychology, the appetite or the means to massacre the entire population of a city if they so chose. On the other hand, their extant texts do not always make this clear, and if they were obtrusive in describing draping city walls with flayed human skins and dying the countryside red with human blood as if it were sheep’s wool,72 then we might well argue that if they were in fact committing urbicide of the human and not architectural type then they would likely have made this clear for us, as well. In either case, Assyrian royal inscriptions, for all of their brutal, dehumanising violence, do not offer us the most straightforward instances of genocide found in ancient literature. Their victims the Israelites in fact offer us examples of a much more transparent genocidal imaginary.

Conclusion An examination of the three most relevant sets of Mesopotamian evidence has led us to a conclusion belied by the reputation of Mesopotamian empires: genocide was in no way a regular part of the practice of violence in this region and may not have occurred at all – if one limits genocide to ethnocide. If, however, one defines genocide such that it is inclusive of the city as a social grouping, then genocide was in fact part of the Mesopotamian imaginary of violence. Furthermore, if the many literary sources attesting to the destruction of cities are in any way instructive or reliable then urbicide was a part of the reality of violence as it was carried out by Mesopotamian states in different periods of the region’s history. The special status of the city in Mesopotamian culture and society from the earliest periods of written records encourages one to see urbicide as not just present in Mesopotamian sources but perhaps even as a quintessentially Mesopotamian sort of mass violence. After all, if city-states were quintessentially Mesopotamian, it does not seem unreasonable to regard the destruction of city-states as a particularly Mesopotamian form of genocide. It is worthwhile to note that, on the matter of genocide, the cultures of ancient West Asia diverged in their conceptions and practices of violence. In 72 See A. K. Grayson, Assyrian Rulers of the Early First Millennium BC (1114–859 BC) (University of Toronto Press, 1990), I I, pp. 194, 196(101.1); and ‘Kurkh Monolith’, translated by K. Lawson Younger Jr, in W. W. Hallo and K. Lawson Younger, eds., Context of Scripture (Leiden: Brill, 1997–2002), I I, 113A: 262.

232

Genocide in Ancient Mesopotamia during the Bronze and Iron Ages

contrast to Mesopotamian sources, the genocidal texts of Israelites and Moabites were ethnocidal in nature. While Deuteronomy 7 and 20 and the book of Joshua do contain urbicidal elements, they are fundamentally ethnocidal in their violence. Where anthropocide is concerned, however, there is more convergence. It appears that Mesopotamian literature was instrumental in spawning an anthropocidal imaginary in the Levant that would far outlive the Mesopotamian cultures of antiquity. Not only did Israelites adopt the anthropocidal flood narrative of the Mesopotamians but they also transferred the divinely orchestrated and totalistic killing of that narrative into the genre of apocalyptic literature, one in which eschatological violence on a cosmic scale is practically a sine qua non. Thus, even the more fantastical sorts of totalistic violence we find in Mesopotamian sources – the anthropocides carried out by Mesopotamian deities against all of humanity rather than against just one group of human transgressors – were pivotal in the development of genocidal imaginaries in other regions of ancient West Asia. In this, they thereby contributed in fundamental ways to the violent imaginaries of Jewish, Christian and Muslim groups both in past centuries and in contemporary times.

Bibliographic Note Genocide has not been directly addressed by scholars of ancient West Asia. In fact, this chapter is the first in-depth examination of whether Mesopotamian sources contain evidence of genocide. While one may find pieces of scholarship that refer to Mesopotamians carrying out genocide, or implying this, no real examination of the issue exists. For example, in ‘The ideology of the Assyrian Empire’ in Power and Propaganda: A Symposium on Ancient Empires, edited by M. T. Larsen (Copenhagen: Akademisk Forlag, 1979), pp. 297–318, M. Liverani implies that the Assyrians practised a kind of total violence (both cultural and physical) against foreigners, but he does not use the term ‘genocide’ or do an extensive analysis. B. S. Düring and T. D. Stek state in ‘Ancient empires on the ground: provincial and peripheral perspectives’ in The Archaeology of Imperial Landscapes: A Comparative Study of Empires in the Ancient Near East and Mediterranean World, edited by B. S. Düring and T. D. Stek (Cambridge University Press, 2018): ‘A second technique was to replace local populations (in part) with groups of productive colonisers, who owed their allegiance to the empire, through a combination of genocide, deportation, colonisation and the agricultural development of previously little cultivated landscapes’ (pp. 1–17, at p. 8). However, they say nothing 233

t. m. lemos and seth richardson

else about genocide. On occasion, overviews of genocide, such as Rieber and Kelly’s ‘Substance and shadow’ and Freeman’s ‘Genocide, civilization, and modernity’, state that genocide was practised by Mesopotamian empires, but typically do not support this conclusion with clear standards. One exception is van Wees, who in ‘Ancient world’ discusses genocide in Mesopotamia but is quite generalising and not quite on target in his analysis of ancient Near Eastern societies. Like van Wees, Wright discusses Mesopotamia in a recent (2015) chapter on urbicide (‘Urbicide: ritualized killing’) in the ancient Near East, though he does not deal with urbicide as a form of genocide in a concerted way. In contrast to scholars who assert or assume that genocide was practised by Mesopotamian groups, A. Fuchs writes that ‘at least racism, religious persecution, and genocide were phenomena unknown to the Assyrians’ (‘Assyria at war: strategy and conduct’ in Oxford Handbook of Cuneiform Culture, edited by K. Radner and E. Robson (Oxford University Press, 2011), p. 399). C. L. Crouch in War and Ethics in the Ancient Near East: Military Violence in Light of Cosmology and History (Berlin: De Gruyter, 2009), comparing war and ethics in ancient Near Eastern societies, unfortunately does not discuss genocide or urbicide, or describe h¯erem as genocide. The situation is similar ˙ with Z. Bahrani’s Rituals of War: The Body and Violence in Mesopotamia (New York: Zone Books, 2008). In light of the paucity of treatments by Assyriologists and other scholars of ancient west Asia, one can only conclude that the topic of genocide has been a lacuna of Assyriological scholarship, and one that the present chapter seeks in large part to fill.

234

8

Urbicide in the Ancient Greek World, 480–330 B C E paul cartledge

It is well known and widely appreciated that the ancient Greeks had some reputation for founding cities, hundreds of them. It has even been claimed that they invented the very idea of the city – or at least a significant version of it (the polis). What is less well known and certainly less widely appreciated is that they also were rather expert at destroying them, sometimes utterly. Pericles, a democratically elected leader of the city of the Athenians, referred loftily to the reputation of their usually hostile neighbours the Boeotian Greeks as mutually destructive executioners – as if the Athenians had not themselves more than dabbled in urbicide (the destruction of a city). This chapter ranges from the first recorded ancient Greek urbicide (effected on the island of Lesbos perhaps as early as the sixth century B C E ) to the – partial – destruction of the Persians’ ceremonial capital of Persepolis ordered (not by the Athenians, as it happens, but) by Alexander the Great in 330 B C E. Most of these urbicide cases are Greek on Greek, but some very significant exceptions, Persian or Carthaginian on Greek, are also considered. An important distinction must be drawn at the outset. For the ancient Greeks, the polis in its agent sense was its people, its fully embraced citizen people or politai. That was why it was always, for instance, ‘the Athenians’ who, acting in a state capacity, did this or that, never ‘Athens’ (which was a purely geographical descriptor). But ‘polis’ could also have a geographical sense, as a physical entity, sometimes but not often (Athens was one such example) one with a truly urban or ‘city’ core. ‘Urbicide’, therefore, in what follows could sometimes mean the physical destruction, even the permanent or temporary annihilation, of a polis, but more often it was the people, the citizen people, who were killed or otherwise removed from their native I am indebted to the editors for their most helpful comments and suggestions; and likewise to my colleague (and co-editor in another context) Greg Woolf, author of The Life and Death of Ancient Cities: A Natural History (Oxford University Press, 2020).

235

paul cartledge

scene. And under ‘citizen people’ one must comprehend also the female half of the adult population, together with the sub-adult children. Typically – if it’s legitimate to speak generically – under urbicide adult male citizens or most of them would be killed, their womenfolk and children sold – abroad – into slavery. An ancient Greek once famously wrote that there’s nothing new under the sun. I might beg to differ, slightly: in my Shorter Oxford English Dictionary (originally 1933; latest reprint in my possession 1972) there is no entry for ‘urbicide’. There is, however, one for the ‘bastard’ word ‘genocide’, which, like ‘homosexuality’, is formed of one Greek and one Roman parental root and was coined by Raphael Lemkin as recently as 1944 in order to come (literally) to terms with the Shoah or Holocaust.1 It now has all too common currency, and too often for the right reasons, but sometimes when used comparatively and cross-culturally it may be thought to stretch the bounds of strict accuracy.2 ‘Urbicide’ the word is therefore a palpable neologism;3 but urbicide the thing was something all ancient Greeks were all too familiar with, both as something they might themselves have actually experienced, or feared that they might, and as something central to their culture, indeed to the very plot of the earliest work of their – and indeed our, Western – literature. The fall and sack of Troy did not quite come within the very narrow temporal limits of the Iliad, but are mentioned retrospectively by Odysseus himself in Homer’s other monumental epic poem Odyssey. They are described in any detail only in later, derivative works. All the same, the Troy story lay at the heart of the ancient Greeks’ canon and served as a yardstick (the original meaning of Greek kanon, or measuring-rod) ever thereafter to gauge the extremity to which human brutality could be pushed (the callously brutal murders of a very young boy and of a very old man stick in the craw as well as the mind). The Homeric epics – the Iliad and the Odyssey – found their monumental form in the decades on either side of 700 B C E. A century or so later, a civilisation and culture quite close geographically to that of the Greeks but not yet in direct contact with it, experienced in real time and in real life a sack every bit as epoch-making as that of Troy in myth and legend. 1 Philippe Sands, East West Street: On the Origins of Genocide and Crimes against Humanity (London: Weidenfeld & Nicolson, 2016). 2 Ben Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007) (subtitle including ‘Sparta’). 3 It was first attributed to Michael Moorcock in his novella ‘Dead God’s homecoming’, Science Fantasy 20:59 (1963).

236

Urbicide in the Ancient Greek World, 480–330 B C E

Behold, I will command, saith the Lord, and cause them to return to this city; and they shall fight against it and take it, and burn it with fire: and I will make the cities of Judah a desolation without an inhabitant.

Thus spake the prophet Jeremiah (34 verse 22), and indeed it came to pass that the Babylonians wreaked havoc on the city of Jerusalem in the year we call 586 B C E. Not all that far up the coast from the Levant, on the island of Lesbos just off the Anatolian Turkish coast today, a far less momentous event of the same generic sort took place, probably in the same sixth century B C E: far less momentous on a world-historical scale, but nonetheless brutal and terrifying for those most directly affected.

Arisba, Lesbos, Sixth Century B C E Our source here is the Western world’s first historian, Herodotus, who came from the same general part of the eastern Greek world, but from further south, at Halicarnassus (modern Bodrum) on the mainland. He was born round about 480 B C E and was composing his Histories (literally, researches or enquiries) in roughly the third quarter of the fifth century B C E. We shall return shortly to his main subject, which also includes ours, but first we should try to take the measure of an earlier urbicide, almost casually mentioned, which occurred on Lesbos. The other two larger Greek islands lying to its south, Chios and Samos, each had just the one city or polis. By Greek standards this was parsimonious indeed: the ancient Greeks within the historical period we are now considering founded scores of cities, all round the Black Sea in the northeast, as far south as Cyrene in what is today eastern Libya, and as far west as southern France and eastern Spain. On Lesbos, however, they more approximated to the norm, and there were originally six cities on the island (though ‘city’ may give a misleading impression of size and scale). The inhabitants of Lesbos belonged to the ‘Aeolic’ branch of the Greek family, that is, they spoke a version of the Aeolic dialect, which they shared for example with the Thebans of the Greek mainland (to whom we shall return). Common speech was one of the chief ways in which all Greeks self-identified as Hellenes, and speakers of the same dialect or sub-dialect often felt a cultural as well as purely linguistic affinity. And yet, for whatever reason, the other five cities of Lesbos at some point ganged up on the sixth, Arisba, and, not to put too fine a point on it, they annihilated it.

237

Map 8.1 Ancient Greece and Sicily. (Cartography by the Ancient World Mapping Center, University of North Carolina)

Urbicide in the Ancient Greek World, 480–330 B C E

Herodotus (1.151) is actually even more brutally specific than that. The other five cities as he put it ‘andrapodised’ Arisba (literally enslaved it, such that those of the defeated who were spared were sold and treated as if they were mere tetrapoda, four-footed animals).4 It is well known that the ancient Greeks practised slavery. It is not perhaps so well known that, arguably,5 they invented probably the worst form of human servitude going, namely chattel slavery, under which human beings are reduced to things, mere chattels, to be bought and sold as commodities on the market. Consonant with that institution was the neuter noun lying behind Herodotus’ verb, namely andrapodon. Literally meaning a ‘(male) man-footed thing’, the word was calqued upon the Greek word for a beast or animal, tetrapodon, literally a ‘four-footed thing’. Andrapodon was only one of the dozen or so words the Greeks had to refer to a slave or unfree person, but it was peculiarly suited to and was the word of choice for people enslaved after a defeat in battle or, as in the case of Arisba, after the sack of a city. When the contemporary evidence becomes more plentiful, in the fifth century B C E, and to Herodotus is added the History of the Atheno-Peloponnesian War by Thucydides, it appears that the treatment of a defeated and sacked city was typically gendered. The men (male citizens) were normally killed, the citizen women and children were ‘andrapodised’, that is, sold into chattel servitude abroad – either to Greek owners in another city or to foreign, non-Greek purchasers. At the time when Arisba was annihilated, it is doubtful that the slave trade in the eastern Mediterranean was yet very developed, though we do hear already in the biblical Book of Ezekiel (chapter 27) that slaves were a regular item of commerce in those parts.6 But Herodotus’ language, if possibly anachronistic, does suggest some such division of spoils: the citizen men of Arisba were killed, their womenfolk and children sold into slavery elsewhere. In this way, but on a small scale, this early Greek case roughly prefigured the Roman destruction of Carthage in 146 B C E.7 4 On the terminology, see K. L. Gaca, ‘The Andrapodizing of war captives in Greek historical memory’, Transactions of the American Philological Association 140 (2010), 117–61. 5 M. I. Finley, Ancient Slavery and Modern Ideology, new ed., ed. B. D. Shaw (Princeton: Wiener, 1998). 6 D. M. Lewis, Greek Slave Systems in Their Eastern Mediterranean Context (Oxford University Press, 2018). 7 See N. Purcell, ‘On the sacking of Carthage and Corinth’ in Ethic and Rhetoric: Classical Essays for Donald Russell on His Seventy-fifth Birthday, ed. Doreen Innes, Harry Hine and Christopher Pelling (Oxford: Clarendon Press, 1995), pp. 133–48; and Chapter 10 in this Volume.

239

paul cartledge

Plataea, Thespiae and Athens 480 and 479 Herodotus’ overarching theme is relations between Greeks and non-Greeks, between about 550 and 479 B C E, and especially how and why they came to blows with each other during the two decades between 499 and 479. His huge work falls into two main halves: the first half sets the scene – what was the geopolitical situation in the eastern Mediterranean in c.550, and then in c.500; the second deals with the six great battles fought between (certain) Greeks and the forces of the mighty Achaemenid Persian empire: Lade (494), Marathon (490), Thermopylae and Salamis (both 480), and Plataea and Mycale (both 479).8 We shall confine ourselves in detail to the years 480 and 479 with special reference to the urbicides of those years carried out by the Persians. But it has to be said that the Persians had more than warmed up in advance by their sackings of Cypriot Paphos (498), Miletus (494) and Eretria (490, before the battle of Marathon). In spring 480 Great King Xerxes led a massive amphibious invasion of the Greek mainland and islands, setting out in person by land across the Hellespont (Dardanelles) strait dividing Asia from Europe. The Greeks exaggerated the size of his land and naval armaments, no doubt, but they did not exaggerate the threat he posed to their independence and indeed future existence. Most Greek cities in his line of fire therefore simply caved in, either positively embracing him or vainly seeking to stay or appear neutral. The Greek resistance such as it was began only south of Thessaly, and even then the most important city of Boeotia, Thebes, decided its future lay with Persia, not the paltry Greek coalition led by Sparta and Athens. Two other Boeotian cities, however, did decide to join the resistance: Thespiae and Plataea. But once Xerxes had forced the Thermopylae pass, overcoming the forces under Spartan king Leonidas, it was a simple matter for Xerxes’ troops to roll over, that is sack, Thespiae and Plataea. The latter destruction will have had the added savour of revenge, since Plataea had aided Athens’ victory at Marathon in 490. We are not told what form the sack took, that is, what other horrors were inflicted on those two cities beyond physical destruction. But in any case they were not the main object of Xerxes’ ire: that was Athens. Indeed, insofar as the Persians felt the need to advertise any sort of ‘just war’ motivation, it seems that they claimed they were mainly intent 8 William Shepherd, The Persian War in Herodotus and Other Ancient Voices (Oxford: Osprey, 2019).

240

Urbicide in the Ancient Greek World, 480–330 B C E

on exacting retribution on the Athenians for their sacrilege in 498, when they had not only torched a Persian provincial capital (Sardis in Lydia) but destroyed sacred property. That at any rate was a kind of motivation for war that all Greeks would have readily understood, since it certainly motivated them often and powerfully enough, and indeed they ‘had a word for it’, tisis, meaning ‘retribution’, often with a divine overtone. Xerxes therefore had no scruples about doing maximum damage to the city of Athens, which the Athenians had helpfully evacuated well in advance. Evidence of that damage is still visibly clear today, and it was no coincidence that the full force of Persian fury was concentrated on the sacred Acropolis rock, where the most important Athenian temples and shrines were located. Not long after, in that same summer of 480, the Athenians got their own back, up to a point, by being the leaders of the allied Greek navy that inflicted a swingeing defeat on the Persians off the islet of Salamis. Had the Greek loyalists lost, the game would have been up, and Persian occupation of all mainland Greece would have ensued sooner or later. But victory at Salamis by no means meant the opposite: total Persian withdrawal. So next year, at the beginning of the campaigning season in early summer 479, Athens – what was left of central Athens to destroy – was sacked again. Later that summer, conclusions were tried on land in a climactic, decisive battle fought not coincidentally near the site of sacked Plataea. The resistant loyalist Greeks, led by the Spartan masters of heavy infantry warfare, but significantly aided by Athenian and other troops, triumphed. Such were the destruction and devastation of Athens, however, that partly for lack of resources to do otherwise, partly to make a symbolic point, and partly to raise morale, the Athenians deliberately left much of central Athens and the Acropolis desolate for over thirty years, for a whole generation. When Herodotus uses the word barbaroi, ‘barbarians’, he usually means just ‘nonGreeks’, especially non-Greek speakers. But it was during those thirty years that the word acquired its negative, cultural overtones: barbarians were barbarous and barbaric, in a word, savage.9 It was therefore an irony, and not the least of several, that the war Thucydides described and analysed, a war chiefly of Greeks against other Greeks, could be made to look uncannily like a case of progressive or rather regressive barbarisation – of which urbicide was an unhappily prominent expression. 9 See e.g. E. Hall, Inventing the Barbarian: Greek Self-Definition through Tragedy (Oxford University Press, 1989).

241

paul cartledge

Aegina, and Thyrea 431 to 424 In the late 490s the island-state of Aegina (like Chios and Samos a single-city island) had sadly been inclined not to resist the post-Lade Persian demands for tokens of submission. The Aeginetans’ ‘treachery’ had not gone down well with their Athenian near neighbours. Indeed, such was the on–off antagonism between the two adjacent states that Pericles, always one for the snappy phrase, had derogated Aegina as the ‘stye in the eye of the Peiraieus’ (Athens’ chief port). It did not help matters that Aegina had fought gallantly on the same side as the Athenians at Salamis. For both before and more importantly after the Graeco-Persian Wars of 480–479 oligarchic Aegina was an ally of Sparta. From about 460 on Athens, while on the one hand proclaiming itself champion of a naval war against the Persian Empire, at the same time chose to prosecute another war – against major allies of Sparta. Quite a serious battle was fought, on land, at Tanagra, in Boeotia, in 458 or 457, which the Spartans just about won. In revenge the Athenians in 457 invaded and permanently occupied Aegina, partially replacing its indigenous population with its own settlers. Hardly surprisingly, the occupied and exiled Aeginetans were among the most vociferous of the Spartans’ allies in urging them in the later 430s to take on the pesky imperialist Athenians – in a war to end all wars. The Spartans in the end decided that their – that is, mainly their own, not their allies’ – cause was just, and that an aggressive war was indeed the choice way to end the Athenians’ hegemony of the Aegean and the consequential threat it posed to the stability of Sparta’s own alliance. War was duly declared by Sparta and embarked upon in the spring of 431. The Spartans were to be proved sorely mistaken, however, at least in the medium term, as the first phase of fighting lasted ten years, not the two or at most three that they had anticipated, and it ended in stalemate, with the strategic balance slightly tipped in the Athenians’, not their own, favour. Aegina remained a point of contention between the two protagonists. In the summer of 431, once war was well under way, the Athenians expelled all remaining Aeginetans, ‘men, women and children’, from their native island, and the Spartans had them resettled within their own civic territory in and near the town of Thyrea, in the region of Cynouria on the east coast of the Peloponnese.10 Seven years later, however, in 424 the Athenians exacted a further deadly revenge.11 They landed a large force by sea, captured, burned 10 Thuc. 2.27.

11 Thuc. 4.56–7.

242

Urbicide in the Ancient Greek World, 480–330 B C E

and plundered Thyrea, and took back to Athens those Aeginetans whom they had not already killed. Once back in Athens, they had all the Aeginetan captives summarily put to death. Thus they committed eventually the total – if temporary – urbicide of hated Aegina. Here was a clear case of what we would now call ‘genocide’.

Plataea 427 and 426 In 519 (probably), Boeotian Plataea had taken a surprising and definitive step: away from the proto-federal embrace of its fellow Boeotian cities and into the arms of – usually anti-Boeotian – Athens. Their famous joint victory over the Persians on the battlefield of Marathon in 490 cemented their special relationship, which was only strengthened by the events of the Persian Wars of 480–479 in which Plataea fought among the loyalist defenders under Sparta and Athens, whereas Boeotian Thebes opted to side with the Persian invaders and indeed provided them with their main base for the final, climactic battle of . . . Plataea in 479. We call it the ‘battle of Plataea’, but actually already by then the town of Plataea was in ruins, destroyed by the Persians in 480 as they emerged triumphantly from battle in the Thermopylae pass on their way to occupying and destroying much of central Athens. After 479 Sparta and Athens fell out in a big way, and as Thebes joined the Spartan alliance, so reconstructed and reoccupied Plataea’s alliance with Athens became all the more salient. When the great Atheno-Peloponnesian War broke out in 431, its very first act was an attack by Thebes on Plataea. In 429 that attack was renewed by the Spartans themselves, who indeed devoted three expensive and enervating campaigning seasons to reducing it eventually by siege (429–427). What followed was a parody of justice, a kangaroo court judgment, imposed by the Spartan victors at the urging of their Theban allies: only those – very few – captive Plataeans who were adjudged to have somehow benefitted the Spartan side were spared their lives, while the majority of the male citizen captives (some two hundred) were executed, the women and children having been already evacuated. Those Plataeans who escaped that ‘justice’ made their way naturally enough to Athens, where they were not only given sanctuary but also exceptionally granted a form of associate Athenian citizenship. That will only have enraged the Spartans and especially the Thebans even more, and, after allowing friendly foreign tenants to occupy and till Plataean farmland for about a year, in 426 the Spartans simply repeated what the 243

paul cartledge

Persians had done to Plataea in 480: they committed urbicide. Thucydides explains what he took to be the Spartans’ motivation as their perceived need of Theban military, strategic and political support against Athens.12 But this particular urbicide came at a huge ideological and symbolic cost, for Plataea – thanks to its role in 479 – was a sort of Panhellenic war memorial, a sacred location on which were buried Greeks from many cities who had died fighting for Greece and freedom, just exactly what the Spartans – mendaciously – claimed was motivating them in their current war on Athens.

Hysiae 417 and Mycalessus 413 Urbicide in classical Greece could take many forms, or have many gradations: from total destruction, even annihilation, through major destruction accompanied by mass slaughter and/or enslavement, sale and dispersal, to relatively minor physical damage to property and built structures but universal or near universal extinction of living creatures human and/or animal. Hysiae and Mycalessus in their different ways occupy the latter end of that spectrum. But even so I would not myself go so far as to invoke the term ‘genocide’, since such ad locum exterminations did not equate to or even imply a desire and intent to eliminate an entire genos (Greek word for ‘race’ or ‘kind’, or the entirety of a class of anything, for example, cattle). The Atheno-Peloponnesian War was a game of two halves: 431–421 and 413–404, each half being called by at least one separate name, but believed by the war’s chief historian, the Athenian Thucydides, to be in essence one 27year war (singular), the years between 421 and 413 being merely a period of phony peace regularly punctuated by outbreaks of significant armed hostilities. One such punctuation mark was the battle fought at Mantineia in the northeastern Peloponnese in 418, involving substantially the same antagonists as the ten years between 431 and 421. Alliances, however, had shifted significantly following the decision of the Athenians and the Spartans to make a separate treaty; separate, that is, from the principal diplomatic instrument of 421, the so-called Peace of Nicias. Mantineia, for example, had been an ally of Sparta until 420, when it went over to the Athenian side and fought against Sparta in 418. Sparta, however, won the battle of Mantineia and won it convincingly, but was unable immediately to expel the Athenians from its hated enemy Argos. Sparta under King Agis II in the winter of 418/417 therefore attacked Argos and did its city walls (something the Spartans 12 Thuc. 3.68; see further next example.

244

Urbicide in the Ancient Greek World, 480–330 B C E

themselves on principle did without) some serious damage, but it was what they did next that qualifies the episode for inclusion here. It might well be labelled ‘gendercide’. Not far from Argos lay the much smaller town of Hysiae, scene of Argos’ one and only major victory over the Spartans ever, and that had occurred in 669, two and a half centuries ago. But clearly the archived memory of that unique defeat still rankled, and the Spartans were anyway in a deeply unsettled state owing to a variety of major setbacks experienced in the first phase of the great war. Anyhow, not for the first time the Spartans chose to exercise extreme violence and brutality: they seized as hostages all the free males of Hysiae that they could lay their hands on – and then simply murdered them. Thucydides, who was not averse to describing atrocities in some detail, as we shall soon see, and indeed to dwelling on and bringing out the brutality involved, here13 merely allots to the Hysiae massacre-cum-atrocity a oneliner: the Spartans ‘capturing the Argive town Hysiae and killing all the freemen that fell into their hands’. So understated indeed is this formulation that its true significance was often until quite recently overlooked.14 So keen was Thucydides to play up, to highlight and spotlight, the brutality of the Athenians – a trait shared, probably for different reasons, by one of his immediate successor historians, Xenophon – that it is that, rather than the Spartans’ almost casual and gratuitous brutality at Hysiae, which most readers take away with them in their mental picture of the war’s ghastly evolution. Unless it is the image conveyed by Thucydides’ account15 of a massacre that took place further north, at Mycalessus, in Boeotia, in 413. The Athenians were again responsible but indirectly this time, unlike in the next case study that I’ll be considering. The year 413 was when the war started again in earnest, by land and by sea, and this time it was more the Athenians than the Spartans who had started it. In this latter phase it became more and more common for both sides to hire mercenaries. This had been the case from the start in naval warfare, since not even Athens, with its by Greek standards huge population, ever had the citizen capacity to man all the trireme warships it wanted to put on the sea and so had to rely on Greek allied sailors to make up the numbers for pay. But in the particular case under review in 413 it was a bunch of non-Greek land-soldier mercenaries that the Athenians had hired, from Thrace (roughly, modern Bulgaria). 13 Thuc. 5.16. 14 See the brilliant short work, Kenneth Dover, Thucydides, New Surveys in the Classics (Oxford University Press, 1973). 15 Thuc. 8.21.

245

paul cartledge

However, having hired them, the Athenians found that they had no work to give them, so they sent the Thracians back home, by sea, under an Athenian commander. He stopped off en route on the eastern coast of Boeotia and was then guilty of a massive dereliction of duty, since he simply let the Thracians (Thucydides says he ‘led them against Mycalessus’) do what they willed. And what they willed, apart from securing food and booty, was to commit indiscriminate slaughter – of every living breathing creature, man or beast, in the town of Mycalessus. Thucydides wrote: The Thracians bursting into Mycalessus sacked the houses and temples, and butchered the inhabitants, sparing neither youth nor age, but killing all they fell in with, one after the other, children and women, and even beasts of burden, and whatever other living creatures they saw; the Thracian race, like the bloodiest of the barbarians, being even more so when it has nothing to fear. Everywhere confusion reigned and death in all its shapes; and in particular they attacked a boys’ school, the largest that there was in the place, into which the children had just gone, and massacred them all. In short, the disaster falling upon the whole town was unsurpassed in magnitude, and unapproached by any in suddenness and in horror.

As it happened, Thucydides was a man with Thracian connections, indeed Thracian family (his very father’s name was Thracian) and Thracian property holdings, so he felt confident enough to generalise about the Thracian mentality and spirit. When their blood is up (he actually says ‘boldness’, tharsos), they are the most savage of all humankind. Thucydides had a taste for superlatives, and he is surely guilty here of at least some Hellenic ethnocentrism, but presumably he did not unduly exaggerate the nature and extent of the Thracian mercenaries’ indulged bloodlust.

Melos 416/415 What of his own Athenians? Were they qualitatively different, or significantly more civilised? One’s answer depends on how far one allows power politics, Realpolitik, to serve as a justification and mitigation, and on what moral standards one applies in a situation that by ancient Greek standards had degenerated into one of no-holds-barred, total war. That at any rate was what Thucydides perceived the situation to be by 416 B C E, when the Athenians – technically at peace with the Spartans – chose to single out for exemplary punishment a small island city in the southwest Cyclades group that had tangled with Athens ten years before, that advertised its sympathy or even kinship with the Spartans and that had resolutely refused and was still

246

Urbicide in the Ancient Greek World, 480–330 B C E

refusing to accept Athens’ offers or demands to include Melos (the island in question, source of the Venus de Milo in the Louvre) in its alliance. The reason Melos has such a high profile on the urbicide agenda is because Thucydides16 chose to make a song and dance about it, as he conspicuously had not done in the case of Hysiae the previous year. He had, however, made what one might call a ‘dry run’, when it came to his describing and analysing the Athenians’ attitude to and treatment of the city of Mytilene on Lesbos in 427. Unlike Melos, Mytilene was an ally of Athens, and a very important one indeed too. Its revolt therefore caused a major crisis for the Athenian alliance as a whole and in particular for Athens’ management of it. Hence Thucydides’ decision to write up his own full-scale version of the debates that actually took place inside and outside the Assembly of Athens. One of the options of exemplary punishment considered, once the revolt had been quelled, was indeed urbicide – or rather gendercide: killing all adult male citizens whether or not they had (actively or passively) revolted. In the event the – relatively – mild measure of executing a thousand or so Mytilenaean male citizens was adopted. By comparison, Melos in itself was small fry, an annoyance rather than a threat to Athens, though Thucydides puffed up its significance into being a symbol of Athens’ imperial hubris that inevitably produced – or at least made much more likely – a forthcoming sooner or later nemesis. Melos had perhaps as many or rather as few as five hundred adult male citizens, making a total citizen population of some 2,000. As such, it was within the modal range of a citizen population for the totality of circa a thousand Greek poleis in the entirety of Hellas. But compared to Athens’ 25,000 or so adult male citizens, this was nugatory. More to the point were two facts about Melos: it was governed by a small oligarchy, the antithesis of Athens’ own democracy, and nor was it simply Sparta-leaning but it claimed actually to be originally a Spartan foundation. The Athenians ‘negotiated’ with the Melian authorities, after a fashion; crucially, the oligarchs would not allow the Athenian delegates to address the Melian citizenry in open assembly. Whence Thucydides’ wholly invented ‘Melian Dialogue’, an exercise in imperial logic – and potential folly. Naturally enough, given that they quite foolishly anticipated receiving serious military support from Sparta, the Melians refused to accede to any of the Athenians’ demands, whereupon the town was besieged, taken and occupied. The male citizens whom the Athenians captured were summarily killed. Their wives, daughters, other female relatives and their children of 16 Thuc. 5.84–116.

247

paul cartledge

both sexes were sold into slavery abroad. Melos as was ceased to exist as such. Instead, the Athenians sent out some of their own citizens – either as ‘colonists’, that is, they gave up their Athenian citizenship to become New Melians, or as cleruchs, that is, they occupied and farmed land on Melos but remained citizens of Athens. A sad tale, for sure, but not as ghastly as Thucydides has made it, and in particular not that much ghastlier than the Spartans’ treatment of Hysiae, despite what Thucydides’ earlier one-liner would suggest.

Selinus and Himera 409 The Sicilian Greeks always stood somewhat apart from the history of those Aegean Greek cities which had originally founded them, beginning with Syracuse founded from Corinth in (traditional date) 733 B C E. Geopolitically, this was because besides themselves the island was occupied from the eighth century on by two main other groups of population, or ethnic groups: the ‘natives’, whom the Greeks partially assimilated and incorporated, partially conquered and enslaved, and in part reached an accommodation with; and the inhabitants of the far northwest of the island, whose founders had come from what the Greeks called Phoenicia (roughly modern Lebanon). The latter group therefore saw themselves as, and were seen as, having shared ancestry with the city of Carthage in north Africa (founded from Phoenician Tyre), and it was the ambitions of fifth-century Carthage both to exploit the island’s commercial opportunities and to extend its power and reach within Sicily that brought it into military conflict with individual Greek cities and with Greek power blocs at either end of the island. Greek Selinus (Selinous in Greek) in southwest Sicily was founded in the late seventh century from mainland Megara, probably with help from another of its Sicilian foundations, Megara Hyblaea. It figures prominently in Thucydides in connection with the Athenians’ ill-fated Sicilian expedition, since its conflict with neighbouring non-Greek Segesta had brought in Athens on Segesta’s side, whereupon Selinus had allied with Syracuse, the main target of Athens between 415 and 413. Our main source for what happened next is the native Sicilian Greek compiler Diodorus, who in his Library of History from the first century B C E describes how a further territorial dispute between Segesta and Selinus in 410 led to the intervention of Carthage and ultimately to Carthage’s destruction of Selinus, when houses were burned or torn down.17 Greek 17 Diod. Sic. 13.43, 13.55–7.

248

Urbicide in the Ancient Greek World, 480–330 B C E

historians are never at their most reliable when dispensing round figures, but for the sake of completeness Diodorus alleges that no fewer than 16,000 Selinuntines (women and children as well as adult men) were killed, while as many as 2,600 managed to escape east to the Greek city of Acragas (modern Agrigento). Greek Himera was an internal Sicilian foundation, founded from Zancle (Messana) on the far northeast of the island (which was itself of ultimately Euboean in origin). Himera lay closest of all to the northwest ‘Phoenician’ sphere and had played a famous role in 480 B C E. The battle of Himera, in which the Greeks defeated a Carthaginian invasion possibly co-ordinated with the vast Persian invasion of Aegean Greece, was said honorifically to have been fought on the very same day as the battle of Salamis. Seventy-one years later the tables were turned. Again, our source is Diodorus, and again he not only records destruction but also offers figures. Modern scholars on a variety of bases have calculated or guesstimated that the fifth-century population of Himera was normally of the order of 20,000 or rather more. Diodorus18 reports that about 3,000 andres (adult males) were captured and ‘butchered’ during the Carthaginian destruction and pillage of 409. The women and children, presumably the survivors, were therefore enslaved and distributed among the victors. The site was at least temporarily abandoned.19

Mantinea 385 Another important ally of Sparta, also a member of its ‘Peloponnesian League’ alliance, was Mantinea, in northern Arcadia, one of the two most important Arcadian cities (the other being Tegea further south). It owed its importance to geography, both being uncomfortably close to always antiSpartan Argos and possessing the sort of level plain on which major pitched battles could be – and were (418, 362) – fought. Mantinea caused further discomfort to always pro-oligarchic Sparta by flirting too often with and sometimes actively embracing democracy, and/or by getting into bed with Sparta’s democratic enemies (Athens, Argos) and sometimes even fighting on their side against Sparta – an act of open rebellion. Such was the case during the so-called Corinthian War (395–386) waged by a coalition headed by Thebes and Athens that Sparta ultimately won with Persian aid – a repeat in a minor key of the Atheno-Peloponnesian War result. 18 Diod. Sic. 13.62.4.

19 Diod. Sic. 11.49.4; Strabo, Geography 6.2.6.

249

paul cartledge

Sparta led by its fiercely anti-Persian and anti-democratic king Agesilaus II (r. c.400–360) singled out democratic Mantinea as the first target of its revenge and retribution. A siege was begun, a river was diverted, and the Mantineans were quickly compelled to surrender. Sparta’s terms were extremely harsh: in essence, nothing short of the termination of the existing polis of Mantinea and its replacement by one of a different kind. The process imposed was termed diokismos, literally a ‘housing apart’, the exact antithesis of a normal process of polis formation called sunoikismos, or ‘housing together’. In Mantinea’s case, that meant a forcible subdivision of the formerly politically unified, centralised and fortified city into its original four or five constituent ‘villages’ (komai). As it happened – or rather, it was no coincidence that – Sparta town was likewise a loose agglomeration of five villages lacking any surrounding city walls. Our main source, Athenian exile Xenophon,20 was himself very pro-Spartan (indeed, he was a personal client of Agesilaus) and very anti-democratic, so he both entirely endorsed Agesilaus/Sparta’s aims and empathised with the (antidemocratic) ‘men of property’ among the Mantineans, who allegedly rejoiced at being once again able to live without fear in their villages on their large landed estates without having to bother their heads with the Mantinean masses and their pesky demagogue leaders. Unlike Diodorus, however, Xenophon was far less interested in figures and numbers than he was in the distribution of political and social power. So we do not get an idea of the human toll in this particular case of urbicide.

Orchomenus 364 Xenophon’s Hellenic History covers, sketchily, the period 411 to 362. But despite its title, it is not so much a Hellenic as a Peloponnesian history, told within an overwhelmingly Spartan and personal perspective. So far as the destruction of Orchomenus in 364 is concerned, it does not help that Xenophon also shared Agesilaus’ violent hostility to Thebes, for it was the Thebans who effected the destruction, partly because Orchomenus in 395 had chosen to take the side of its enemy Sparta in the ‘Corinthian War’ (see above). Partly – but only partly: actually, the rivalry tending to open hostility between the two most important cities of the region of Boeotia was far deeper and far more long-standing even than that.

20 Xen. Hell. 5.2.

250

Urbicide in the Ancient Greek World, 480–330 B C E

The Boeotians were an ethnos (a people) as well as a koinon (federal state). But that did not mean they necessarily always all got on with each other as one big happy family. Far from it. Relations between Thebes and Orchomenus were but the limit case of chronically bad relations. Orchomenus lay in northern Boeotia, on the far side of Lake Copais from Thebes in the southeast. A moderately oligarchic federal regime had been in force throughout most of Boeotia since 447, and the Boeotians had taken the Spartans’ side, most effectively, throughout the Atheno-Peloponnesian War. The immediate aftermath of that war, however, had seen Thebes distance itself from Sparta to the point of actually rebelling and going to war with its former alliance leader in 395. Orchomenus had gone in the exact opposite direction. As with Sparta and Mantinea, so with Thebes and Orchomenus, retribution was on the cards. Orchomenus’ misfortune was that, after Thebes’ famous defeat of the Spartans at Leuctra in 371, its principal enemy no longer exercised hegemony merely in its own Boeotian backyard but throughout much of mainland Greece. Eliminating Orchomenus was therefore a bagatelle, only a matter of time; especially as Thebes had a track record of at least partial urbicide, in the case of another Boeotian city, Thespiae in 424, and more extreme urbicide in the case of Plataea in 373. In 370 Orchomenus had been forced back into a now (since 378) no longer moderately oligarchic but moderately democratic Boeotian federation after a quarter of a century’s independence.21 In 364 an abortive independence-fuelled coup was crushed as the Thebans conquered and destroyed the city and exposed it to an andrapodismos in which all the men captured were killed and the women and children sold into slavery.22 The city of Orchomenus was not, however, razed. As with Aegina and Melos in the fifth century (above), its site was given over to outside settlers, in this case naturally, Thebans.

Olynthus 348 Rather different is my next example, where political elimination did equal physical annihilation. Like the Boeotians (above), the Chalcidians of Thrace were both an ethnos and a koinon. Spectacular development including impressive domestic construction in the later fifth century had made of Olynthus a major player in northern Greek geopolitics and, in particular, an annoyance, even a threat, to its neighbouring kingdom of Macedon, first properly unified 21 Diod. Sic. 15.57.1.

22 Diod. Sic. 15.79.3–6.

251

paul cartledge

under Archelaus (413–400). Sparta had taken a brief interest in northern Greek affairs in the 420s and again in the 380s, when it still had pretensions to being a regional superpower – before the catastrophic defeat at Leuctra in 371. But the long siege of Olynthus between 381 and 379, undertaken largely because Sparta was not keen on federations, had not only cost the life of a Spartan king but also unwittingly paved the way for the even bigger catastrophe that overtook Olynthus in 348. In 359 the fortunes of the kingdom of Macedon were transformed again, when the younger brother of a king killed in battle against a neighbouring non-Greek people ascended the Temenid throne. That younger brother was Philip son of Amyntas III, who reigned as Philip II from 359 to 336. One contemporary called Philip the greatest man Europe had ever produced; that is as may be, but certainly he transformed first northern Greek and then all mainland Greek political affairs. After achieving the mastery of Thessaly and extending his reach as far east as the Black Sea, he was poised to descend upon the Chalcidians led by Olynthus. The golden-mouthed patriot and democrat Demosthenes delivered his famous three empathetic ‘Olynthiac Orations’ before the Athenian Assembly in 349 – to no avail. What aid the Athenians sent was too little and too late. In 348 Philip marched into Thracian Chalcidice and, as an object lesson for anyone else who might think of opposing him, razed Olynthus together with its port of Mecyberna to the ground. Olynthus had experienced andrapodismos once before, in 479, at the hands of the Persian invaders.23 But so successful, so total, was the annihilation of 34824 that the site was never reoccupied. So it remained there in its pristine, destroyed state ready for the spades and trowels directed in the 1930s by American archaeologists, who thereby excavated by far the best preserved classical Greek habitation site on record. So extensive and rich are the deposits that excavation has been renewed, successfully, within the second decade of the twenty-first century.

Thebes 335 Philip II it was too who was indirectly responsible for my next urbicide. One of his much later historians, the Sicilian Greek Diodorus, writing his Library of History in the first century B C E on a universalist, comparativist basis, interestingly discusses both Olynthus (above) and my next example, Thebes, along with some others as examples of using fear or terror to maintain hegemony.25 23 Herod. 8.127.

24 Diod. Sic. 16.53.3.

25 Diod. Sic. 32.2.

252

Urbicide in the Ancient Greek World, 480–330 B C E

Thebes and Athens were near neighbours, but until the late fifth century far more often enemies than friends, let alone allies. The rise of Philip of Macedon, however, threw Thebes and Athens together in 340, renewing their collaborations of 404, 395 and 378. Demosthenes was instrumental in bringing about the rapprochement, which was eased by the fact that in 340 both cities were democracies. Philip, however, was merely biding his time, waiting for a good excuse or opportune moment to complete the conquest of mainland Greece that he had begun in 352 (above). In 339 he finally decided to advance south with his highly proficient, professional Macedonian army through the pass of Thermopylae and, after settling some lesser business in central Greece, came to decisive blows with Athens, Thebes and their allies on the battlefield of Boeotian Chaeronea in summer 338. The result was a complete victory for Philip, ably assisted by his eighteenyear-old son and heir Alexander. Thebes and Athens were at his mercy. Anticipating the Romans, he practised a version of divide and rule. Athens he treated quite lightly – no regime change, no garrison, no indemnity – because he felt he needed Athens’ fleet for his next major enterprise, conquest of at least some part of the Persian Empire. Thebes, however, experienced his iron fist with all its force: democracy was replaced by a pro-Macedonian oligarchy, and to ensure subservience a Macedonian garrison was imposed. So far, so good, but then – in 336, on the point of joining the advance force he had already sent over to northwest Asia Minor – Philip was assassinated. Alexander, who may or may not have had something to do with his father’s death, certainly acted ruthlessly to ensure his succession and thereby inherit his Persian project. Before setting off for Persia himself, however, he felt it necessary to secure his rear: to make as sure as he could that affairs in mainland Greece and indeed as far north as the Danube were sufficiently stabilised to make it unlikely that unrest back home would necessitate his return from Asia. While he was on campaign in the far north, a rumour reached central Greece that he, like his uncle, had been killed, and that was the signal for the Thebans to raise the flag of revolt. Big mistake. Not only was Alexander still very much alive but within two weeks he had covered some 500 kilometres and was outside the walls of Thebes with his army. With his father’s example of Olynthus to guide him, and with the prospect of his leading the attack on the Persian Empire so imminent, Alexander had Thebes utterly destroyed as a red light warning to others – with just a couple of exceptions. There was a great deal of built environment in Thebes to destroy. A great wall made of mudbrick superimposed on a stone base and covered with tiles 253

paul cartledge

surrounded the central city, of which the Cadmeia acropolis was the dominating feature. Most Thebans lived on or immediately around the Cadmeia. There too were located a dozen religious shrines, together with the principal official government offices, such as that of the Boeotarchs (chief executive officers and generals) and the council chamber. In the lower city there were another dozen shrines, and there also lay the Agora (town square) flanked by colonnades and a theatre. All that went up in smoke in an orgy of destruction, or rather almost all. Of the two exceptions one was generic, the other very specific. Alexander was anyway very religious, so it was second nature to him not to destroy religious shrines and structures. But he had a further reason for sparing them: one of his alleged motives or reasons for attacking Persia was to effect retribution and restitution for the sacrilegious damage done by the Persians to Athenian and other Greek sacred sites in 480 and 479, during wars in which Thebes had been on the wrong, Persian side. The specific case of exemption was the house on the Cadmeia in which the great Theban praise-poet Pindar (c.518–448) had lived. Again, there was an anti-Persian point to be made: Pindar was the Panhellenic – all-Greek – poet par excellence, since the victors he praised were victors in the four great Panhellenic religious festival games. But not only the city’s buildings were destroyed. More than 6,000 men were reportedly eliminated, a sixth of the city’s population. The remaining huge number of 30,000 men, women and children was subjected to andrapodismos. The site was more or less completely abandoned for some twenty years, until in a different era another Macedonian warlord sanctioned its resurrection.

Persepolis 330 At last, in 334, Alexander was ready, able and willing to cross over from Europe via the Hellespont/Dardanelles strait into Asia. By a unique combination of strategic brilliance, bravery, cunning and sheer good fortune (his life was once saved by a whisker), Alexander led his mainly Macedonian troops to a string of victories on land in Asia Minor (334, 333), the Levant (332) and finally – after a major Egyptian detour (332) – in what is today northern Iraq (330). The decisive battle of Gaugamela was fought on what we call 1 October 330. Persian emperor Darius III survived the battle but lost his life soon after, murdered by a rival would-be king of kings. The way was opened for Alexander to the Persian imperial heartland in southern Iran where lay the palaces of Pasargadae, Susa and (what the Greeks called) Persepolis. 254

Urbicide in the Ancient Greek World, 480–330 B C E

The Pasargadae palace was the creation of the empire’s founder Cyrus II (r. c.550–529). Those of Susa and Persepolis were owed to the empire’s second founder, Darius I (r. 522–486). There was a certain division of function: if Susa was the chief administrative capital, Persepolis was the ceremonial capital. Its vast palace – enhanced not least by Darius’ son and successor Xerxes – was adorned by an extraordinary, enormous, multi-columned reception hall (Apadana), the way to which was up through an elaborately decorated monumental stairway. Persepolis too was the location not only of massive fortifications but also of one of the biggest of the empire’s treasuries, stuffed full of the tribute garnered from the subjects of a hypertrophied empire (above). It was inevitable therefore that Alexander would make a big song and dance of his conquest, occupation and triumphal entry into Persepolis. What followed, however, remains controversial for religious and cultural reasons to this very day. To put it succinctly, the Persepolis palace complex suffered severe damage: the historical questions are – just how extensive was the damage? was it deliberate? and, if so, why? Emotion tends to cloud the issue and the answers. My own are as follows. The destruction was very severe and extensive but it was far from total. It was indeed quite deliberate and, more to the same point, rather carefully targeted. Burning of Persepolis was the carefully orchestrated and advertised revenge and retribution for the sacking by the Persians of the Athenian acropolis in 480 and 479. Chief author of that earlier sacking had been Xerxes son of Darius I, so correspondingly and proportionately the chief object of Alexander’s destruction were the parts of the Persepolis palace that could be associated specifically with Xerxes. A sad end to a sorry tale – but from the mythical Troy to the all too real Persepolis, Greek urbicide was never pretty.

Summary To conclude, M. H. Hansen and T. H. Nielsen summarise ‘the four different phenomena’ involved in ancient Greek urbicide. These ‘often occurred together: the destruction of a polis was almost always accompanied by the annihilation or enslavement or expulsion of its population or with its fragmentation into village communities’. True, the impression of total destruction or expulsion conveyed by the sometimes too sensationalist sources can be belied by both material and written evidence of continuity or at least speedy revival and recovery. So far as the extant evidence goes, ‘annihilation of a polis seems to have been achieved in about a score of the 112 individually 255

paul cartledge

attested poleis’.26 Several of these cases of urbicide also approached the level of what we would now call genocide. There is no gainsaying that the ancient Greeks were notably unable to translate Hellas, a cultural-geographical term, into any voluntary, lasting political union: urbicide was both a cause and symptom of that chronic inability or failure.

Bibliographic Note James Crawford’s Fallen Glory: The Lives and Deaths of the World’s Greatest Lost Buildings (London: Old Street Publishing, 2016) covers monumental destructions sometimes involving urbicide from biblical Babel to the death of doomed internet GeoCities, but alas from our standpoint it skips from Egyptian Amarna (‘Died 1331 BC’) to the Jerusalem Temple (‘Died AD 70’). E. M. Harris and colleagues’ proceedings of a conference held at the American School of Classical Studies, Athens, in 2019 (www.academia.edu/ 38801696/Destructions_Survival_and_Recovery_in_Ancient_Greece_Ameri can_School_of_Classical_studies_at_Athens) addressed ‘Destructions, survival and recovery in ancient Greece’. Paul’s ‘Urbs capta: sketch of a literary motif’ (Phoenix 36 (1982), 144–55) deals with literary tropes such as the Fall of Troy. But Hansen and Nielsen’s Inventory of Archaic and Classical Poleis is the indispensable historical tool for addressing any question regarding the thousand or so poleis (cities, citystates, citizen-states) of Hellas (the ancient Greek world) between, in round figures, 600 and 300 B C E, including therefore urbicide. Ancient sources are always minutely referenced, and it has an invaluable introductory section, ‘Destruction and disappearance of poleis’ (pp. 120–3, with an appendix on the principal meaning of the Greek phrase ten polis anastaton poiein/genesthai, which they identify as being to expel a population or depopulate a city, rather than physically ruin or destroy it). For recent consideration of the topic of city destruction in Greece more generally, see S. Fachard and E. M. Harris, eds., The Destruction of Cities in the Greek World (Cambridge University Press, 2021).

26 M. H. Hansen and T. H. Nielsen, eds., An Inventory of Archaic and Classical Poleis (Oxford University Press, 2004), pp. 120, 122.

256

9

Violence, Emotions and Justice in the Hellenistic Period michael champion

While we know of no genocidal episodes in the hundred years after Alexander’s death, in the period from around 223 B C E onwards, instances of genocidal violence in the form of the complete destruction of cities, massacres which did not distinguish between combatants and non-combatants, and mass enslavements in the course of military campaigns return to the historical record. This chapter explores key episodes of extreme violence between Greek-speaking groups, focusing on the events of the Cleomenic and Social wars in the period 223–217 B C E.1 Case studies include the destruction of Mantinea (222 B C E ) (woven into Polybius’ account of the fall of Megalopolis in 223 B C E), Lyttus (221/220 B C E) and the related destruction, in the late second century, of Drerus, Cynaetha in Arcadia (220 B C E), as well as the fate of Dium, Dodona and Thermum (219–217 B C E ) (see Map 9.1). Across these different episodes, competing discourses about justice, including different emotions related to justice, conspire to both moderate and intensify violence throughout this period. The effects of contests between different traditions of justice and associated emotional regimes of anger, pity and shame is a significant element necessary for understanding episodes of extreme violence in the period. These traditions of justice in turn sit within a range of other cultural and political commitments, including the development and maintenance of civic identity, religious norms and considerations of economic and political power. The chapter therefore begins with discussion of the contexts of Hellenistic violence to identify conditions for extreme violence and concludes by weighing these different features of the phenomenon of extreme violence in the Hellenistic period. 1 For genocidal violence associated with Rome’s destruction of Carthage and Corinth in 146 B C E, see B. Kiernan, ‘The first genocide: Carthage, 146 BC’, Diogenes 51:4 (2004), 27–39; and Chapter 10 of this volume.

257

michael champion

Dium Thermum Aetolia Thebes Cynaetha

Dodona Attica Athens

Arcadia Mantinea Megalopolis LLaconia

Sparta

Crete

20 km 20 mi

Lyttus

Drerus

Map 9.1 Sites of city destruction in Hellenistic Greece, 223–217 B C E. (Cartography by the Ancient World Mapping Center, University of North Carolina)

Terms and Tensions I employ Helen Fein’s sociological definition of genocide as a ‘sustained purposeful action by a perpetrator to physically destroy a collectivity directly or indirectly, through interdiction of the biological and social reproduction of group members, sustained regardless of the surrender or lack of threat offered by the victim’.2 This definition provides some analytic clarity since it casts several episodes of Hellenistic violence as genocidal while excluding some nevertheless horrendous events.3 A relevant political, social and sub-ethnic grouping in the period is the polis – an urban centre and its surrounding districts – so violence can be counted as genocidal should it directly or indirectly aim at the destruction 2 H. Fein, ‘Defining genocide as a sociological concept’, Current Sociology 38:1 (1990), 8–31, at 24. Cf. Convention on the Prevention and Punishment of the Crime of Genocide, Paris, 9 December 1948, United Nations, Treaty Series, vol. 78, no. 1021; www.hrweb.org/legal/ genocide.html. 3 The following analysis applies Fein’s definition and resulting ‘paradigm’ to the Hellenistic context. Compare Fein, ‘Defining genocide’, 25–30.

258

Violence, Emotions and Justice in the Hellenistic Period

of a polis.4 In Hellenistic military campaigns, there were clear command structures overseeing organised campaigns, and occasions where military leaders authorised city destruction following the surrender of the city or its elimination as a military threat. Such devastation cannot be deemed accidental, since it differs from indiscriminate looting by out-of-control soldiers. While it is not always possible to distinguish between intentional and unintentional city destruction, killing inhabitants, selling them into slavery or razing buildings and arable land were effective and planned means of eliminating or grossly interfering with the reproduction of the group, with long-term and sometimes irremediable consequences for the victims’ group. Political structures and accounts of justice were deployed to render such actions ethical and legally sanctioned in the eyes of the perpetrators, an element in the purposiveness of such action. Such ideologies, alongside factors shaped in part by actions and intentions of the perpetrators of extreme violence and often present in cases of genocide (for example identity contests, ethnic stigmatisation and marginalisation, political fragmentation and transformation, and features of expansionist ambition or loss of power and prestige) could contribute to such city destruction. In identifying cities as the most relevant collectivity in the case of Hellenistic genocide, we should note that there is no recorded instance of citizens being systematically hunted down and killed if they were not present in the city at the time of its destruction. This does not mean that such violence is not genocidal (since the Genocide Convention definition includes the partial destruction of a group). But the fact that extreme violence could be directed at a collectivity but was not sustained beyond the time taken to destroy the city, its inhabitants and its surrounds may make the contemporary legal category of the crime against humanity of ‘extermination’ – mass killing or massacre of civilians and the imposition of intolerable conditions through systematic attacks – more applicable. At least for some purposes, it may usefully distinguish Hellenistic episodes of genocidal violence from some more recent cases by foregrounding differences in the duration of violence and the different role place rather than civic or ethnic identity 4 For the polis as a relevant collectivity for the study of genocide in the period, see H. van Wees, ‘Genocide in archaic and classical Greece’ in Our Ancient Wars: Rethinking War through the Classics, ed. V. Caston and S.-M. Weineck (Ann Arbor: University of Michigan Press, 2016), pp. 19–37; and H. van Wees, ‘Genocide in the ancient world’ in The Oxford Handbook of Genocide Studies, ed. D. Bloxham and A. D. Moses (Oxford University Press, 2010), pp. 241–58.

259

michael champion

could play as a motivator of violence (although such identities, as I will argue, remain important for explaining genocidal violence in the Hellenistic period).5 Key to this distinction is that the violence I analyse always occurred in the context of war. The ancient world did not conceptualise or enact the possibility that a collectivity could be destroyed or terrorised outside the context of war. That depravity is restricted to a later time period. Following Plato, it was the common view that war (at least between Greeks) should not aim at total destruction, but war did result in such violence.6 There were instances of city destruction when the military threat posed by that city had been eliminated or was at a particularly low ebb and extreme violence could be deployed to secure power once empire had been won.7 Diodorus of Sicily wrote that empire is ‘gained by intelligence and courage, consolidated by equity and philanthropy, and secured by fear and terror’.8 Such acts of terror could include political repression and harshly applied justice in peacetime, although it is clear that both Diodorus and Polybius have war in mind when they refer to fear inspired by major powers. Moreover, Polybius codes violence over a ruler’s own populace as tyrannical: outside the context of war, rulers should gain the willing consent of citizens through philanthropy rather than acts of terror more suited to tyrants.9 War was thus the context for genocidal violence in Hellenistic Greece.

Violence in the Hellenistic Period The Hellenistic world was shaped and ravaged by war. Writing towards the end of the Hellenistic period, the historian Philippus grieves that ‘all conceivable calamities and continuous mutual slaughter have occurred in 5 For ‘extermination’, see the 1998 Rome Statute of the International Criminal Court, Article 7, section 1(c), 2(b). Genocide and extermination may be compared with other ‘war crimes’, understood as defined by the 1998 Rome Statute, Article 8. See www.icccpi.int/resource-library/documents/rs-eng.pdf. 6 Pl. Resp. 469b–471c. Text: Plato, Republic, ed. J. Burnet (1900; Oxford: Clarendon Press, 1967). Cf. Arist. Pol. 7.7, 1328a7–8, on the difference between Greek and non-Greek violence. Text: Aristotelis politica, ed. W. D. Ross (1957; Oxford: Clarendon Press, 1964). On the right limits to warfare, see Polyb. Hist. 5.11.5. Text: Polybii historiae, ed. T. Büttner-Wobst, 4 vols. (1893–1905; Leipzig: Teubner, 1962–7). 7 For example Aegina: Thuc. 2.27.1–2, 4.57.1–4. Text: Thucydidis Historiae, ed. H. S. Jones and J. E. Powell, 2 vols., 2nd rev. ed. (Oxford: Clarendon Press, 1942). 8 Diod. Sic. 32.2.1–4. Text: Diodorus Siculus, Bibliotheca historica, ed. and trans. F. R. Walton, vols. 11–12 (Cambridge, MA: Harvard University Press, 1957, 1967). L. I. Hau, ‘Diodoros of Sicily (32.2 and 4) and Polybios’, Classica et Mediaevalia 57 (2006), 67–102. 9 Polyb. 5.11.5.

260

Violence, Emotions and Justice in the Hellenistic Period

our time throughout Asia and Europe and the peoples of Africa and the cities of islanders’.10 These calamities were from the first the defining feature of the historical narrative. Polybius, the major historian of the three hundred years after the death of Alexander the Great from the Wars of the Successors to the battle of Actium, offers the first ‘entangled’ global history, where major conflicts between Rome and Carthage, Achaeans and Macedonians and Aetolians, and the kings of Egypt and Syria intersect and become bound up in each other, spilling over into further devastating confrontations leading to the rise of Rome.11 This ‘ubiquitous’ warfare marked all aspects of life, as Angleos Chaniotis has demonstrated.12 The Hellenistic physical environment was shaped by the presence of trophies, civic inscriptions and other monuments of war memorialisation, which dotted the landscape and dominated civic architecture. War was inscribed in civic and religious ceremonials, its scars were on the bodies of individuals and the city-scapes of poleis, it had caused the devastation of the countryside. It shaped civic education, and was represented in coinage, public art, historiography and drama.13 Significant geopolitical changes radically affected the magnitude, political framing, technology and economics of Hellenistic warfare.14 Hellenistic kingdoms sustained their power on the back of mercenaries and the increasing professionalisation of war, seen also in the rise of technical manuals of warfare and developments in siege engines and other military technology. Warfare thus unsettled the political power of the polis, which found itself needing to innovate to negotiate a new place in the world order. If war had been central to democratic power and civic ideology, it now became 10 FGrH 95 F1 = IG iv 12 687. FGrH: F. Jacoby (ed.), Fragmente der griechischen Historiker (Berlin: Weidmann, 1923– ). IG: Inscriptiones Graecae (Berlin: Preussichche / BerlinBrandenbürgische Akademie der Wissenschaften, 1873– ). Late Hellenistic dating: A. Chaniotis, Historie und Historiker in den griechischen Inschriften. Epigraphische Beiträge zur griechischen Historiographie (Stuttgart: Steiner, 1988), pp. 314–17; P. Goukowsky, ‘Philippe de Pergame et l’histoire des guerres civiles’, Hellènika symmikta 2 (1995), 39. 11 Polybius, The Histories, trans. W. R. Paton, revised F. W. Walbank and C. Habicht, Loeb Classical Library, 6 vols. (Cambridge, MA: Harvard University Press, 2010–12); F. W. Walbank, A Historical Commentary on Polybius, 3 vols. (Oxford: Clarendon Press, 1957–79); B. McGing, Polybius’ Histories (Oxford University Press, 2010). 12 A. Chaniotis, War in the Hellenistic World: A Social and Cultural History (Malden, MA and Oxford: Blackwell, 2005), pp. 1–17. 13 Ibid., pp. 1–5; M. Champion and L. O’Sullivan, eds., Cultural Perceptions of Violence in the Hellenistic World (London and New York: Routledge, 2017). 14 See further M. Champion and L. O’Sullivan, ‘“War is the father and king of all”: discourses, experiences, and theories of Hellenistic violence’ in Cultural Perceptions of Violence, ed. Champion and O’Sullivan, pp. 2–6.

261

michael champion

associated with the institution of kingship.15 The polis found itself now part of new political conglomerations within the Hellenistic kingdoms, and had to negotiate its own identity with and within these new power blocks, sometimes through military means.16 The increasing size and religious and ethnic heterogeneity of Hellenistic cities also led to conflict and to new forms of moderating it, just as war against a range of ethnic others – Carthaginians, Romans, Gauls – also helped to shape and define Greek identity.17 Manuals of advice for kings sought to school the new monarchs in gentle justice towards their cities.18 Inscriptions record increasing diplomatic activities between cities to secure lasting friendship between different groups. Grants of asylia, which recognised a city, sanctuary and its surrounds as ‘sacred and inviolate’, increased, partly to prevent devastation in war. Violence could also be moderated through attempts to define a Greek identity that held good beyond the boundary of an individual polis, for example in common civic architecture, Panhellenic festivals and the increased standardisation of a shared language (koine) across formerly linguistically distinct regions. If claims to homonoia or fellow feeling between cities could – then as now – signal chauvinistic claims to dominance by the more powerful cities, they could also join cities in a common purpose.19 Together, this amounts to a geopolitical context from Africa to Rome where civic identity was in flux and under threat, where diverse cultures were militarised across significant parts of their life, where civic tensions associated with ethnic and religious heterogeneity, power instabilities and social inequalities were always close to the surface, and where a range of different political groupings had access to advanced military technology and professional forces capable of enormous destruction. The conditions for genocide, or at least for devastation on a grand scale, were present. 15 Suda s.v. βασιλεία. See Suidae Lexicon, Lexicographi Graeci 1.1–1.4, ed. A. Adler, 4 vols. (Leipzig: Teubner, 1928–35). 16 J. Ma, ‘Fighting poleis of the Hellenistic world’ in War and Violence in Ancient Greece, ed. H. van Wees (London: Duckworth and Classical Press of Wales, 2000), pp. 337–76. 17 On the large topic of identity in the period, see J. Hall, Ethnic Identity in Greek Antiquity (Cambridge University Press, 1997), and, for an overview focused on the Hellenistic period, S. Burstein, ‘Greek identity in the Hellenistic period’ in Hellenisms: Culture, Identity and Ethnicity from Antiquity to Modernity, ed. K. Zacharia (Farnham: Ashgate, 2008), pp. 59–78. See also note 40. 18 M. Haake, ‘Writing down the king: the communicative function of treatises On Kingship in the Hellenistic period’ in The Splendors and Miseries of Ruling Alone: Encounters with Monarchy from Archaic Greece to the Hellenistic Mediterranean, ed. N. Luraghi, Studies in Ancient Monarchies 1 (Stuttgart: Franz Steiner, 2013), pp. 165–206. 19 For example Cytenium supported by fellow Dorians from Lydia: SEG 38.1476. SEG: Supplementum Epigraphicum Graecum, ed. A. Chaniotis et al. (Leiden: Brill, 1923– ).

262

Violence, Emotions and Justice in the Hellenistic Period

But genocidal violence must be distinguished from other ‘criminal’ atrocities, especially given that modern legal restraints on actions in war do not map directly to Greek laws or conventions.20 The notion of ‘war crime’ – actions in war that were thought of as against the law – existed in the period. For example, Polybius describes the ‘lawless’ actions of a group of Campanian mercenaries, who enter Messene professing false friendship, but act falsely, take the city, kill some inhabitants, exile others, capture the women and children, and plunder the city.21 Further, we know of a limited number of treaties restricting acts of hostility.22 But the Hellenistic ‘laws’ or ‘rules’ of war differ in important ways from modern legal frameworks. Partly, the notion of law differs: there is a stronger emphasis on violation of custom and cultural norms than on codified laws, although some wartime examples of treaty violation were considered criminal; partly, the actions taken to be ‘criminal’ may differ, with a strong emphasis on protecting religious sanctuaries, lands, festivals and the people involved in them,23 but less concern (as a matter of cultural norm) for restrictions on actions in wartime based on humanitarian considerations.24 Further, as I will argue, cultural norms and emotions associated with justice could license extreme violence. The horrifying fate of the city of Abydus when it was besieged by Philip V (200 B C E ) usefully distinguishes wartime atrocities from genocide itself while pointing to terrifying consequences of normalised extreme violence.25 Faced with a demand of unconditional surrender by Philip, the Abydeans resolved to liberate the slaves to fight alongside the free men, and place women in the temple of Artemis, the children with their nurses in the gymnasium, and all the wealth of the city in two ships. Then they swore to fight to the death and designated fifty men to burn the ships and kill all the women and children, so that they would not fall into the hands of the enemy, should the inner wall fall to Philip’s forces.26 Fierce fighting ensued, but despite the extraordinary resistance of the Abydeans, it became clear that Philip’s army would prevail. The old men who had sworn to kill the women 20 For a detailed overview see C. Phillipson, The International Law and Custom of Ancient Greece and Rome, 2 vols. (London: Macmillan, 1911). 21 Polyb. 1.7.1–4. 22 For example the oath of the Amphictyonic Council: Aeschines, Orations, 2.115. Text: V. Martin and G. de Budé, Eschine: Discours, vol. I (1927; Paris: Les Belles Lettres, 1962). 23 For example Polyb. 5.10–11. 24 For a recent discussion see A. Lanni, ‘The laws of war in ancient Greece’, Law and History Review 26:3 (2008), 469–89. 25 Polyb. 16.31–4; Livy, Ab urbe condita 31.18. Text: Titi Livii, Ab Urbe Condita, ed. A. H. McDonald, Oxford Classical Texts (Oxford University Press, 1965). 26 Polyb. 16.31–2.

263

michael champion

and children now resolved instead to plea to Philip for mercy. Polybius characterises this as a cowardly act which ‘cast down the honourable and admirable decision’ that had conferred glory on the city.27 Philip takes the city but is appalled by what he finds: ‘citizens, children, women slaughtered, throats cut, the city burning down, hanging, throwing themselves into wells or off high roofs’.28 He decrees three days’ grace for those ‘who wished to hang or slaughter themselves’.29 Whereupon the ‘Abydenians . . . judging that they had become traitors to those who had fought and died for their city, could not bear to live on any terms; each family therefore rushed headlong to their death’.30 The extermination of the city is devastating, but cannot count as genocide. Yet within the history of genocide, the siege of Abydus provides evidence that the terror instilled by city destructions in the period was so great that some were prepared to swear and carry out murder-suicide pacts of the most violent kind in order to escape the evils of submitting to the conqueror. Their furious anger (ὁρμή) leads to, and is an index of, the imagined justice of these terrifying actions, which Polybius describes in religious terms.31 While this is the most horrifying example, it is not the only one. Polybius records that other cities had made similar pacts, even if the ultimate outcome in their case was not so brutal.32 Genocidal violence begat other atrocities in antiquity.

Destruction of Lyttus and Drerus The example of Lyttus should make us alert to key difficulties in identifying genocidal violence from our sources. Polybius records that forces from Cnossus and Gortyn decided to put an end to Lyttus as an example to all other towns in Crete. Lyttus of all the Cretan towns had resisted them most strongly, and Cnossus and Gortyn were, we learn, ‘urgently trying to drive them completely from their homes as an example and terror to the rest of Crete’.33 While the armed forces of Lyttus were campaigning elsewhere, Cnossus and Gortyn took their chance, killed non-fighting men and enslaved all the women and children, removing them to Cnossus. The town was torched and razed. When the soldiers return, they find what Polybius refers to as ‘irretrievable disaster’.34 We seem to have a clear example of genocidal city destruction, with the (albeit partial) killing of members of the city, serious bodily and psychological harm, and mass enslavement, including 27 Polyb. 16.33.4. 28 Polyb. 16.34.9–10. 29 Polyb. 16.34.10. 30 Polyb. 16.34.11–12; Livy 31.18.7. 31 For example Polyb. 16.33.4, 34.9. 32 For example Polyb. 16.31. 33 Polyb. 4.53.3. 34 Polyb. 4.53.3.

264

Violence, Emotions and Justice in the Hellenistic Period

the transfer of women and children to Cnossus, severely threatening the existence of the polis of Lyttus as a cultural group. Yet the epigraphic record shows that Lyttus was in fact rebuilt. Its refoundation was celebrated alongside the commemoration of the (later) destruction of Drerus, which does seem to have been completely obliterated by the forces of its sworn enemy Lyttus at the end of the third or beginning of the second century B C E.35 Drerian men of military age had long pledged perpetual hatred against Lyttus, swearing to fight to the death against their enemies.36 Their allegiance with Cnossus would have magnified Lyttus’ enmity towards them following the events of 221/220 B C E. As Drerus discovered to its ruin, Lyttus seems to have learned from bitter experience what it takes to destroy a city definitively. Archaeological finds suggest that public buildings were destroyed in part by being torn down and elements of the buildings that carried inscriptions thrown into cisterns in what seems likely to have been a deliberate damnatio memoriae.37 Florence Gaignerot-Driessen has written movingly of the ritual killing of the city, as, in her reading, the archaeological record suggests that Drerus was rendered incapable of functioning in the world of the living through its destruction. The city became a tomb.38 Drerus, which does not enter the literary historical record at all, is a better example of genocide than its destroyer Lyttus, whose own genocidal destruction is lamented by the most famous historian of the period. Militarised cultures, together with long-standing and institutionalised enmity, resulted in genocide.

Destruction of Mantinea and Megalopolis, 223/222 B C E Mantinea is probably most well known to historians for two famous battles (418 and 362 B C E ). 39 What is left out of this account of its significance for Greek history is the fate it suffered under the Peace of Antalcidas (387 B C E), when the Spartans first dispersed the city into its surrounding villages and 35 See F. Gaignerot-Driessen, ‘The “killing” of a city: a destruction by enforced abandonment’ in Destruction: Archaeological, Philological and Historical Perspectives, ed. J. Driessen (Louvain: Presses universitaires de Louvain, 2013), pp. 288, 292; Chaniotis, War in the Hellenistic World, pp. 9–11, 46–7. 36 IC I, ix, 1 in Inscriptiones Creticae, ed. M. Guarducci, 4 vols. (Rome: Libreria dello Stato, 1935–50). 37 Gaignerot-Driessen, ‘“Killing” of a city’, 290, with relevant literature. 38 Ibid., 293–5. 39 418 B C E: Thuc. 5.66.1–5.74.3. 362 B C E: Diod. Sic. 15.84–7; Xen. Hell. 7.4.39–5.27. Text: Xenophon, Hellenica: Xenophontis Opera Omnia, ed. E. C. Marchant, vol. I (1900; Oxford: Clarendon Press, 1968).

265

michael champion

then, upon meeting resistance, razed it to the ground (385 B C E).40 The Arcadians restored the city after the battle of Leuctra in 371 B C E, and it would enjoy another 150 years of existence before falling in brutal fashion to the Macedonian king Antigonus III Doson in 222 B C E. Polybius’ account of the ultimate destruction of Mantinea by Antigonus is woven into an argument about the excesses of earlier historians like Phylarchus, who allegedly does not adequately provide an account of the causes of historical events, instead only offering an overly emotional or sensationalist account of slaughter.41 It is simultaneously interwoven with his account of the destruction of Megalopolis.42 The two accounts of the sack of Mantinea could not be more different. In Polybius’ own narrative, it is disposed of in a subordinate clause: ‘after the Macedonians quickly terrified the city and took it into their power, he broke camp and set out for Heraea and Telphusa’.43 What matters is the movement of kings, not the obliteration of a city. It is left to Phylarchus – whose narrative, in Polybius’ view, is marked as unwarranted sensationalising – to fill out the details of what being ‘terrified into surrender’ means. His emotional and gendered account focalises ‘women embracing, tearing their hair, and exposing their breasts’.44 We see and hear the ‘tears and lamentations of men and women driven into slavery pell-mell with their children and aged parents’.45 This is an instance of mass enslavement of non-combatants which destroys the collectivity – something the interpreter must work hard to extract from Polybius’ brief account. His brevity is partly explained by his view that selling people – combatants and non-combatants, women and children – into slavery after defeat is normal and licit under the ‘laws of war’.46 Yet it is clear that this ‘normal’ event in the context of ancient warfare entirely destroyed the city’s culture. Antigonus renamed the city Antigonea, and handed it over to Aratus, who founded a new Achaean colony there.47 40 Xen. Hell. 5.1.35–5.2.7; Ephorus FGrH 70 F 79; Diod. Sic. 15.5.4, 15.12.2. P. Funke, ‘Between Mantinea and Leuctra: the political world of the Peloponnese in a time of upheaval’ in The Politics of Ethnicity and the Crisis of the Peloponnesian League, ed. P. Funke and N. Luraghi, Hellenic Studies 32 (Washington, DC: Center for Hellenic Studies, 2009). 41 Polyb. 2.54, 56–8. See Walbank, Historical Commentary, and J. Marincola, ‘Polybius, Phylarchus and tragic history: a reconsideration’ in Polybius and his World: Essays in Memory of F. W. Walbank, ed. B. Gibson and T. Harrison (Oxford University Press, 2013), pp. 73–90. 42 Polyb. 2.55, 61. 43 Polyb. 2.54.12. 44 Polyb. 2.56.7. 45 Polyb. 2.56.7. 46 Polyb. 2.58.9–13. 47 Polyb. 2.56.6–7; Plut. Vit. Arat. 45. Text: Plutarch, Aratus: Plutarchi vitae parallelae, ed. K. Ziegler, vol. 3.1, 2nd ed. (Leipzig: Teubner, 1971). Pausanias’ confused narrative is evidence for the eradication of local Mantinaean memory. See further M. Pretzler,

266

Violence, Emotions and Justice in the Hellenistic Period

Mantinea was then written out of the historical record from 222 B C E up until Hadrian restored the ancient name of the city.48 In Polybius’ framing, we see how genocide can be enabled through legal frameworks. Such a legal and rhetorical strategy helps to render the exceptional, murderous and violent nature of city destructions normal and unproblematic. Phylarchus’ account, of course, also acts as an emotional intensifier which fills out Polybius’ sparser narrative. The narrativisation of violence seems at times in Polybius’ narrative to overflow and break the boundaries of the social and legal norms by which he explicitly evaluates events. The piteous ruin of Mantinea both expands his heavily compressed statement about its ‘surrender’ and also contextualises Cleomenes’ sack of Megalopolis. While Polybius’ account of this city destruction is more expansive, it is still deliberately curtailed. He briefly informs his readers that Cleomenes completely possessed Megalopolis, enslaved or exiled the inhabitants, and ‘destroyed the city so vindictively and with such enmity that there was no hope that it might ever be rebuilt as a city again’.49 The Megalopolitans lost ‘their territory, tombs of the ancestors, and their temples, homes, and property . . . everything, in fact, which people value most’.50 It suffered this fate because it had threatened to undermine Cleomenes’ power and status by spurning his earlier calls for aid.51 A slight to Cleomenes’ power through the exercise of independence is reason to incur extreme wrath. The narrative of the destruction of Mantinea and Megalopolis illumines the conceptual structures which could be used to critique or normalise such events and set them within conventional ethical frameworks. The destruction of Mantinea could be justified by the treachery and impiety of the Mantineans. The Mantineans were guilty, in Polybius’ account, of the worst form of treachery, betraying their Achaean League allies who had come to their assistance to the Spartans, who then kill them in cold blood in contraventions of the ‘laws of war’ (οἱ τοῦ πολέμου νόμοι).52 This is an action that rightly justifies extreme anger (ἡ ὀργή).53 In fact, the extreme anger of the Achaeans is an index of the justice of their genocidal actions.54 Similarly, the drowning at sea of Aristomachus by Antigonus following events at

48 49 53 54

‘Pausanias at Mantinea: invention and manipulation in local history’, Cambridge Classical Journal 51 (2005), 21–34. IG v, 2, 263 (contemporary with the sack) and IG v, 2, 264–6, 307 (Flavian), cited in Pretzler, ‘Pausanias at Mantinea’, 24. Polyb. 2.55.6–7. 50 Polyb. 2.61.10. 51 Polyb. 2.55.8. 52 Polyb. 2.58.15. Polyb. 2.58.15. See further D. Konstan, ‘Anger, hatred, and genocide in ancient Greece’, Common Knowledge 13:1 (2007), 170–87.

267

michael champion

Mantinea was just in that it was merciful: in Polybius’ view, though less intuitively, he was a man so vile, criminal, that it would have been just had he been tortured in every city in the Peloponnese. He is tarred with the brush of tyranny, which Polybius claims includes ‘every injustice and transgression of law’.55 It is therefore just to feel anger and dispense the greatest punishment rather than feeling sympathy; acting with vengeance would, given the magnitude of Aristomachus’ crimes, have been reasonable and just.56 Yet the Achaeans consistently acted towards him justly, in Polybius’ view, by showing not vengeance by ‘gentleness and goodness’.57 His ultimate death was not as vengeful as it might have been had justice been conceptualised merely as retributive. Here Polybius rescues the Achaeans from the charge of injustice levelled by Phlyarchus by drawing on traditions of justice as equity, which is always more lenient and is characterised by the emotions of pity and mercifulness rather than anger.58 This yields two accounts of justice that may either complement each other or pull in different directions in the context of extreme violence. The pathos one may feel for Mantinea as framed by Phylarchus’ narrative is moderated by the insistence that the most extreme forms of violence may be reasonable and just, and even punishment resulting in death could be counted just and merciful. The Achaean anger was exceptional, in this account, because the injustice it rights is so great. Anger maps to righteousness and may lead to extreme violence understood as just. On the other account, understanding justice as equity allows extreme violence to be coded as just and even appropriately gentle.

Destruction of Cynaetha, 220 B C E Violent othering through strategies of dehumanisation is often a feature of genocides. The destruction of Cynaetha is congruent with this pattern. Its destruction was partly associated with long-standing views held by other Greeks about its allegedly barbarous nature. Polybius describes the 55 Polyb. 2.59.6. 56 Polyb. 2.59.4; 2.60.1–2. 57 Polyb. 2.60.3. 58 Arist. Eth. Nic. 1137b34–1138a3, 1143a19–35. Text: Aristotle, Aristotelis Ethica Nicomachea, ed. I. Bywater (1894; Oxford: Clarendon Press, 1962). Arist. Rh. 1374b2–10. Text: Aristotle, Aristotelis ars Rhetorica, ed. W. D. Ross (1959; Oxford: Clarendon Press, 1964). Cf. Hdt. 3.53. Text: Herodotus, Herodoti Historiae, ed. N. G. Wilson, 2 vols. (Oxford University Press, 2015). Gorgias, Epitaphios DK 82B6. Text: Gorgias, Fragmenta, ed. H. Diels and W. Kranz, in Die Fragmente der Vorsokratiker, vol. I I, 6th ed. (Berlin: Weidmann, 1952). See M. Nussbaum, ‘Equity and mercy’, Philosophy and Public Affairs 22:2 (1993), 83–125.

268

Violence, Emotions and Justice in the Hellenistic Period

Cynaethans as barbarians. He explains their uncouth nature in an extended discourse on the improving power of education. The Cynaethans are Arcadians, and Arcadians in general have the worst natural characteristics of all the Greeks. For Polybius, this is because the geography and climate of the region generates their primary nature and militates against the development of culture. The rest of Arcadia has successfully overcome their first nature by putting on a second, cultured nature through education, especially through music.59 They ‘humanise and make their merciless soul more gentle through the exercise of such habituation’.60 Again, Polybius draws on understandings of justice as gentle equity. The Arcadians are not originally capable of such justice given that they are not capable of mercy; education civilises by making them just – gentle, human, merciful. But the Cynaethans, alone among all the Arcadians, neglected such musical education, when they needed it more than others cities in the region, since their climate and geography is the worst of all Arcadia. Their resulting unimproved primary nature therefore predominates, explaining their bloody mutual animosity, unremitting civic strife and ‘atrocious crimes’.61 Cynaetha is riven by violently warring factions, its civic history one of banishments, forcible confiscation and redistribution of property and land, and massacres. Unlike civilised Greeks, the Cynaethans are harsh, unremitting, unjust, violent and hateful. Antagonism based on this essentialising account of the barbarous nature of the Cynaethans is a significant catalyst for the extreme violence directed against them, which results in the torture and extermination of citizens and the plunder and ultimate physical destruction of the city.62 Polybius characterises this violence as ‘supreme justice’ that rightly punishes Cynaethan treachery.63 The force of explanations from nature can be seen in Polybius’ account of the reception of Cynaethan ambassadors after the sack of their 59 Polyb. 4.20. First and second nature: Arist. Eth. Eud. 1220a–b. Text: Aristotle, Aristotelis Ethica Eudemia, ed. F. Susemihl (1884; Amsterdam: Hakkert, 1967). On Polybius’ Greek– barbarian distinction and Roman political commentary, see C. Champion, ‘Polybian barbarology, flute-playing in Arcadia, and fisticuffs at Rome’ in Polybius and His Legacy, ed. N. Miltsios and M. Tamiolaki (Berlin: De Gruyter, 2018), pp. 35–42, and C. Champion, Cultural Politics in Polybius’s Histories (Berkeley: University of California Press, 2004). 60 Polyb. 4.21.4. 61 Polyb. 4.21. 62 Polyb. 4.19. Narrative of the city destruction: Polyb. 4.16.11–21.9. If Strabo is to be believed, the city did not recover, but Pausanias lists statuary (including of Hadrian) in the agora and altars, along with a sanctuary of Dionysus: Strabo, Geography 8.8.2. Text: Strabo, Strabonis Geographica, ed. A. Meineke, 3 vols. (Leipzig: Teubner, 1877). Paus. 8.19.1–3. Text: Pausanias, Description of Greece, trans. W. H. S. Jones (Cambridge, MA: Harvard University Press, 1918–55). 63 Polyb. 4.19.13.

269

michael champion

city. The barbarous nature of the Cynaethans is such that no other Greek city will come to their aid or help the survivors in the aftermath of the destruction.64 Nor were other Greeks prepared to stand in the way even of the ‘naturally violent’ Aetolians in this case.65 Here we see natural characteristics of the Aetolians appealed to as an explanation for their actions too. For Greek thinkers, appeals to nature provided a crucial explanatory framework which enabled different groups to be stigmatised in ways that provided necessary conditions for extreme violence. The destruction of Cynaetha was at least rendered more likely by the general Greek view that the Cynaethans are beyond the bounds of cultured human society. Similar dynamics can be seen in Polybius’ account of Prusias’ massacre of the Gauls in Asia.66 Soldiers are put to the sword, women and children are ‘nearly all’ slaughtered. But the atrocity is positively narrated: Prusias’ actions are described as a deliverance from the fearful and dangerous licentious violence of barbarians (βάρβαροι). In the case of Cynaetha, cultural difference does not completely erase ethical problems with the Aetolian attack; the Achaeans nominate the destruction of Cynaetha as one of the causes of war the Aetolians have allegedly provided to justify the Social War.67 But passionate animosity towards the group because of their allegedly natural characteristics weakened obstacles to extreme violence.

Destruction of Dium, Dodona and Thermum, 291–217 B C E Pathological violence is often linked to acts of sacrilege. This may be seen in Polybius’ accounts of the destruction of Dium and Dodona.68 These Aetolian atrocities extinguished the possibility of cultural life in the two cities. In the case of Dium, the Aetolians ravaged the countryside, caused the inhabitants of the city to flee, and then razed the city walls, houses and gymnasium and destroyed the colonnades around the sacred temple enclosure, along with all the votives that had been placed in it.69 Furthermore, all the statues of the kings and rulers of the city were destroyed. We recall the case of Drerus and Lyttus, where statuary was thrown into cisterns in an attempt to obliterate objects of cultural memory and civic and religious significance.70 At Dodona, while the city was able to recover over time, we learn that the general 64 Polyb. 4.21. 65 Polyb. 4.16. 66 Polyb. 5.111.2–7. 67 Polyb. 4.25. 68 Polyb. 4.62, 4.67. 69 Polyb. 4.62.1–3. 70 Chaniotis, War in the Hellenistic World, pp. 231, 239–41 for Lyttus and ‘places of memory’ more generally.

270

Violence, Emotions and Justice in the Hellenistic Period

Dorimachus wasted the land around Epirus.71 His pathological violence was revealed by the fact that he was motivated in this act of devastation not by the prospect of booty but in order to harm the Epirotes. It was also, and most prominently revealed, in his acts of sacrilege.72 As in the case of Mantinea, where the impiety of the Mantineans was taken as an index of their depravity, acts of sacrilege set the Aetolians outside the bounds of acceptable human behaviour. Polybius records that when Dorimachus reached the city, ‘he burnt down the colonnades, destroyed the votive offerings, and even destroyed the sacred building . . . the Aetolians did not uphold the norms (ὅρον) of peace and war, but in both circumstances assaulted the common customs and laws of humanity’.73 Since Greek religion stood at the heart of a city’s communal and cultural life, the destruction of religious objects and precincts is an effective means of erasing civic identity. The twin atrocities of Dium and Dodona frame Polybius’ account of Philip V’s brutal destruction of Thermum. As in these earlier cases, which are explicitly presented as providing a cause for the extreme violence of Philip’s actions, Philip plunders the surrounding districts, ravages the surrounding countryside, steals everything portable he can carry and burns the rest.74 This, Polybius tells us, was all ‘done rightly and justly in accordance with the laws of war’.75 But then, ‘remembering the Aetolian atrocities’, ‘they burnt down the colonnades, destroyed the rest of the votive offerings . . . and not only did they utterly ruin the roofs with fire, but they also destroyed them right to their foundations. They overturned the statues . . . and destroyed most of them, except for those which had been inscribed with the names or images of gods’.76 Polybius notes that Philip believed himself to be acting justly by retributively punishing the Aetolians for their earlier acts of impiety.77 Sparing the statues of gods may, in this view, have distinguished Philip’s actions from the destruction of Dium or Dodona. But Polybius disagrees. Philip I of Macedon showed equity, moderation and philanthropy and the associated emotions of ‘mildness’ which avoided ‘vindictive passion’.78 By contrast, Philip V, he claims, was motivated by quite different, and more vengeful, concepts of justice that overflowed into sacrilegious violence of which even the harsh Alexander the Great was not guilty.79 Indeed, Polybius argues that Philip V would have been seen by his victims as fully justified in exacting vengeful justice for their crimes at Dium and Dodona but would have won their allegiance by acting still more justly had 71 Polyb. 4.67.1. 72 Polyb. 4.67.1–5. 73 Polyb. 4.67.3–4. 74 Polyb. 5.8.7–9. 75 Polyb. 5.9.1. 76 Polyb. 5.9.2–3. 77 Polyb. 5.9.6. 78 Polyb. 5.10.1–5. 79 Polyb. 5.10.6–11.

271

michael champion

he instead demonstrated equity through the gentle generosity characteristic of the ‘great-souled’, most virtuous, ruler.80 Thus, in the case of Philip V’s destruction of Thermum, Polybius’ account implicitly pits the two concepts of justice we saw operating in earlier case studies against each other. Philip’s own view – that the destruction of Thermum is just retributive punishment for Aetolian sacrilege – is met with counter-examples of justice as equitable moderation and mildness; his overpowering just anger is contrasted with the gentleness of equity. Further, Polybius’ condemnation of Philip V’s destruction of temples is evidence for the capacity of religion to moderate violence in the Hellenistic context. Across these three episodes, the obliteration of religious sites is viewed as evidence of unjustifiable violence. Philip’s assault on the temple precinct at Thermum is justified as a punishment of more extreme sacrilege, and he steps back from complete destruction of religious statuary. This speaks to the centrality of religious practice and its close ties to conceptions of both civic and Panhellenic identity. It also suggests that just as mechanisms of othering functioned in the Cynaethan case to facilitate violence, ideologies of unity and shared culture between different Greek groups, including through shared religious practices, could moderate violence. Nevertheless, such norms were violated under stress of war, and erasing a city’s religious institutions is a key feature of the most extreme violence associated with city destructions.

Justifying Genocide These case studies of city destructions point to several intersecting justifications for Hellenistic genocidal violence. There are strong economic factors which justify extreme violence. Mass enslavement and the prospect of plunder are clear motivations which may justify extreme violence on pragmatic or realist grounds. Such considerations may mean that the collectivity is destroyed without mass killing or the complete destruction of property, since there is an economic motivation to keep people alive in order to profit by enslaving them and to restrict destruction to maximise booty.81 Nevertheless, enslavement and impoverishment, including through the destruction of arable land needed to sustain civic life, could result in the annihilation of a polis. Similar pragmatism could be used to justify violence in order to establish power. The destruction of Drerus, Megalopolis and 80 Polyb. 5.11.8.

81 Cf. van Wees, ‘Archaic and classical Greece’, 28–9.

272

Violence, Emotions and Justice in the Hellenistic Period

Thermum are all narrated in part as arising from a desire to establish power and terrify other cities into submission. Similarly, threats to status can become catalysts for extreme violence, amplified by the centrality of honour and shame in Hellenistic culture, and their relationship to norms of masculinity and the claims of civic identity.82 This points to the ways in which cultural norms and prejudices can be mobilised to justify genocidal violence. In the case of Cynaetha, we saw that violence could be justified in part through dehumanising the enemy. The evidence does not support the claim that Greeks set out to kill other Greeks based solely on cultural prejudice, but it is naive to think that dehumanising discourses are unrelated to extreme violence. Further, while Hellenistic conceptions of ethnicity do not seem to have been primary motivators of violence, ethnic bonds between different cities were sufficiently strong to motivate action, seal military alliances and lead to conflict, including extreme violence, between groups, as conflict between Aetolians and Achaeans suggests.83 In the Achaean characterisation of the natural viciousness of the Aetolians recorded in Polybius’ narrative, we saw evidence of similar essentialising of enemies that promoted violence. Conversely, larger ethnic groupings could join diverse cities to one another in a common purpose and minimise violence. Religion could similarly function both to justify and to moderate violence. It could moderate extreme violence in asylia decrees and through other conventions precluding destruction of religious precincts, and it could work against practices of othering by foregrounding shared practices and commitments between different Greek groups. But where such norms were contravened, aggressors could cite religious transgressions as grounds for brutal violence. Recall, for example, Philip V’s destruction of Thermum, or Cleomenes at Cynaetha, where violation of oaths had religious force. This is an instance of the main justification of extreme violence: violence justified as legitimate punishment.84 Polybius explicitly argues that ‘good men do not wage war against those who have done wrong to destroy and exterminate them but to correct and change what has been done wrong’, 82 Cf. ibid., 31, 33–5; van Wees, ‘Ancient world’, 241–58. 83 Achaea and Aetolia as ethnic groups: F. W. Walbank, Polybius, Rome and the Hellenistic World: Essays and Reflections (Cambridge University Press, 2002), p. 147. Greek ethnicity: C. Morgan, ‘Ethnicity and early Greek states: historical and material perspectives’, Proceedings of the Cambridge Philological Society 37 (1991), 131–63; Hall, Ethnic Identity in Greek Antiquity; J. Hall, Hellenicity: Between Ethnicity and Culture (University of Chicago Press, 2002). 84 Konstan, ‘Anger, hatred, and genocide’, 187; van Wees, ‘Archaic and classical Greece’, 34.

273

michael champion

distinguishing between the guilty and the innocent.85 Anything else is characteristic of ‘tyranny’ and a ‘raging spirit’.86 The destruction of Cynaetha punishes extreme duplicity and treachery, Thermum’s fate is sealed by sacrilege at Dium and Dodona, Mantinean treachery is reasonably published by the city’s obliteration. This takes us into central areas of Greek ethics and institutions. Genocidal violence can be justified through legal norms, both those explicitly related to international relations and warfare and those more commonly instantiated in the courtroom in the context of civil or criminal cases. As David Konstan has argued, in the legal context, anger can be stimulated by a perceived injustice and jurors in court cases are regularly encouraged to feel anger towards perpetrators of crimes.87 An effect of this view of anger is that it becomes a proxy for judgements about just action.88 Genocidal violence, therefore, may be justified as legitimate by coding it as just. Across multiple cases of city destructions, we have seen that two accounts of justice, with different emotional contours, are in play. The punishment paradigm makes justice as retribution a significant factor in the justification of genocide. This is associated with the emotions of anger which help in turn to magnify violence. But another tradition of thinking about justice may also, equally insidiously, justify extreme violence, this time by coding it as reasonable and gentle. In this view, genocidal violence is civilised by associating it with leniency, reason and pity. Some actions, on this account, are so far beyond the pale that even someone dispensing justice equitably – with reason, gentleness and leniency – will level a devastating punishment. While justice as equity and its associated emotional norms could be activated to critique the brutal anger of retributive justice, it too could be marshalled to justify extreme violence. The Hellenistic period saw significant episodes of extended and extreme violence in the context of war that did not distinguish between combatants and non-combatants and resulted in the destruction of multiple civic collectivities, including through mass slaughter, enslavement and the obliteration of civic, agricultural, religious and political infrastructure. Such extreme violence meets the sociological definition of genocide with which I began, so it is plausible to speak of multiple episodes of genocide in Hellenistic Greece. But I conclude by registering hesitations over this characterisation. Partly, my hesitation is generated by the near impossibility of conceptualising the 85 Polyb. 5.11.5. 86 Polyb. 5.11.4,6. 87 See e.g. Arist. Eth. nic. 1135b25–29. See also Konstan, ‘Anger, hatred, and genocide’, 181, 187. 88 Konstan, ‘Anger, hatred, and genocide’, 187.

274

Violence, Emotions and Justice in the Hellenistic Period

enormity of the phenomenon. Partly, I worry that there is something ethically significant about restricting the category of genocide, since anything less could relativise the absolute horror of genocides like the Holocaust, for all the political utility of extending the category to assist in the early mobilisation of forces to reduce or prevent new genocidal destruction.89 But there is also a danger that too strict a demarcation of genocide can allow us to play down the seriousness of the sorts of violence this chapter has explored. In Classical Studies, blindness to the devastation of ancient violence has been a besetting sin of the profession, which has been readier to gloss over such violence in admiration for the more palatable elements of classical culture, or the technological and political advances generated by warfare. In that context, testing the applicability of the category of genocide to ancient violence has its own an ethical impetus. With the rise of the glorification of so-called ‘western’ civilisation within far-right movements today, this is a particularly pressing concern.

Bibliographic Note Sources The best ancient historical account of the period is Polybius, The Histories, trans. Paton, rev. Walbank and Habicht, in the Loeb Classical Library. Polybius was himself a war victim, who lived much of his life as a (rather atypical) hostage sent to Rome to guarantee Greek submission. He travelled on military campaigns with the Roman general Scipio and was an eyewitness to the devastating genocidal ruin of Carthage in 146 B C E , for which see Kiernan, ‘First genocide’. While his Histories are not completely extant, his is the best continuous narrative of the period. Walbank’s Historical Commentary on Polybius provides much illuminating context for the narrative. Diodorus Siculus and Livy also provide illuminating historical narratives, while Plutarch’s Lives also provide important material. There is much evidence about the devastation of warfare and the ideological conditions for Hellenistic violence in contemporary epigraphic evidence. The Supplementum Epigraphicum Graecum (SEG) collects these inscriptions and provides useful commentary and literature. Fragmentary historians of the period can be consulted in Jacoby’s Fragmente der griechischen Historiker (FGrH). A useful 89 Samantha Power (The Education of an Idealist (London: William Collins, 2019), p. 127) cites a US Office of the Secretary of Defense discussion paper warning against defining the Rwandan atrocities as genocide, because a ‘Genocide finding could commit [the US government] to actually “do something”’. Making such sophistry more difficult seems warranted.

275

michael champion

collection of sources for the period in general is Austin, Hellenistic World from Alexander to the Roman Conquest. Modern Scholarship For orientation to the political, economic, social, intellectual religious and ethnic changes associated with the transition from the classical world to the Hellenistic kingdoms, see G. Shipley, The Greek World after Alexander, 323–30 B.C. (London: Routledge, 2000) and F. W. Walbank et al., The Cambridge Ancient History, vol. V I I, part I: The Hellenistic World, 2nd ed. (Cambridge University Press, 1984). For a comprehensive and illuminating account of Hellenistic warfare in cultural, ideological, technical, political and economic perspective, see Chaniotis, War in the Hellenistic World. Chaniotis’ study advances understanding of the complex of factors that motivated extreme violence in the period and explores in detail the social, political, economic and cultural effects of warfare. P. Ducrey’s Le traitement des prisonniers de guerre en Grèce ancienne: des origines à la conquête romaine (Paris: De Boccard, 1968) and Warfare in Ancient Greece, trans. J. Lloyd (New York: Schocken Books, 1986) provide a comprehensive account of military campaigns and mass enslavements across the period, collating evidence on the degree of destruction, loss of life and forced displacement caused by war. Excellent overviews of warfare in the period are W. K. Pritchett’s five-volume The Greek State at War (Berkeley: University of California Press, 1974–91), and P. Sabin, H. van Wees and M. Whitby, The Cambridge History of Greek and Roman Warfare (Cambridge University Press, 2007). On the military role of the polis in the Hellenistic period, see Ma, ‘Fighting poleis of the Hellenistic world’. There is much excellent scholarship on conceptions of Greek identity and ethnicity in the period: see for example Morgan, ‘Ethnicity and early Greek states: historical and material perspectives’ and Hall’s Ethnic Identity in Greek Antiquity and his Hellenicity: Between Ethnicity and Culture. Such scholarship is relevant to the question of the degree to which notions of ethnicity, religion and identity motivated or moderated extreme violence. Similarly, studies of connections between education and military ideology are significant for a picture of connections between identity construction and the conditions for extreme violence in periods of social transition. Many studies of extreme violence in Greek antiquity tend to focus on earlier periods, although they do include the study of central aspects of Hellenistic violence (archaic to classical Athens, Sparta and the campaigns of Alexander the Great). Important here are A. B. Bosworth, Conquest and Empire: The Reign of Alexander the Great (Cambridge University Press, 1988); K. Raaflaub, ed., War and Peace in the

276

Violence, Emotions and Justice in the Hellenistic Period

Ancient World (Oxford: Blackwell, 2007); and van Wees’ edited volume War and Violence in Ancient Greece. There is also extensive literature on connections between democracy and war, including the influential work of J. Ober, The Athenian Revolution: Essays on Ancient Greek Democracy and Political Theory (Princeton University Press, 1996). Ober also provides a useful study of norms surrounding rules of war. Since earlier military events and ideology continued to shape Hellenistic norms, understanding Hellenistic violence means setting it in this longer context, while studies of connections between democracy and violence are relevant by way of contrast with the political conditions of the Hellenistic kingdoms. Particular studies of the category of genocide include van Wees, ‘Genocide in the ancient world’ and ‘Genocide in archaic and classical Greece’. Van Wees sees shame playing a strong motivating role, while also pointing to elements of realist political theory, economic considerations (which might tend towards mass enslavement rather than mass killing of the vanquished) and the gendered nature of violence. Konstan, ‘Anger, hatred, and genocide in ancient Greece’ asks whether hatred of a group is a motivator for genocide and concludes that hatred is not needed when anger will suffice, drawing attention, as this chapter has, to the role anger plays in Greek ideas of justice.

277

10

A Tale of Three Cities The Roman Destruction of Carthage, Corinth and Numantia tristan s. taylor

As Corinth followed Carthage, so Numantia followed Corinth, nor then was there anywhere in the world preserved from Roman arms. Florus 1.33

The destruction of a city in the ancient Mediterranean was a potent statement of power in a ‘well-known symbolic language’.1 To Greco-Roman writers, there was a clear connection between the Roman destruction of the citystates of Carthage in north Africa and Corinth in Greece in 146 B C E and the elimination in 133 B C E of the Spanish Celtiberian stronghold Numantia. These three acts of what we now would call urbicide – ‘the obliteration of urban living-space as a means of destroying the viability of an urban civilisation and eroding its collective values’2 – were fundamental in ancient thinking on the symbolic establishment of the Roman Empire in the second century B C E: acts of terror that symbolised and secured the empire’s power.3 Rome’s destruction of these three civic communities has attracted attention in debates on the application of the concept of genocide to antiquity; this is the case particularly with the destruction of Carthage, which has been given the label ‘genocide’ by some,4 including Raphael Lemkin, who coined the term and had intended to write on Carthage in his unfinished History of 1 N. Purcell, ‘On the sacking of Carthage and Corinth’ in Ethics and Rhetoric: Classical Essays for Donald Russell on His Seventy-fifth Birthday, ed. D. Innes, H. Hine and C. Pelling (Oxford: Clarendon Press, 1995), pp. 131–48, at p. 131. 2 A. Jones, Genocide: A Comprehensive Introduction, 3rd ed. (London: Routledge, 2017), p. 37. 3 Diod. Sic. 32.2–4. 4 For example B. Kiernan, ‘The first genocide: Carthage, 146 BC’, Diogenes 203:3 (2004), 27–39 and Blood and Soil: A World History of Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007), pp. 49–58; Jones, Genocide, p. 7; against: S. T. Katz, The Holocaust in Historical Context, vol. I: The Holocaust and Mass-Death before the Modern Age (Oxford University Press, 1994), p. 92 n. 107, and N. Barrandon, Les massacres de la république romaine (Paris: Fayard, 2018), pp. 323–9. See also F. R. Chalk and K. Jonassohn, The History and Sociology of Genocide: Analyses and Case Studies (New Haven: Yale University Press, 1990), pp. 75–93.

278

Map 10.1 The Mediterranean of the Roman Republic, showing relative positions of Carthage, Corinth and Rome. (Cartography by the Ancient World Mapping Center, University of North Carolina)

tristan s. taylor

World Genocide.5 Controversy continues, however, as to whether genocide is an appropriate concept for these events, or for the premodern period in general.6 Some believe it is anachronistic to apply the concept to antiquity due to the specific, negative connotation it has in modernity,7 when the value system of the Roman world, which universally accepted slavery,8 differed significantly from the modern. Nonetheless, even in antiquity, mass violence could both be praised and critiqued.9 Beyond questions of morality, definitional issues also arise in applying the concept of genocide. Lemkin’s original broad conception of genocide as a systematic attack against a group’s foundations, including its cultural foundations, would seem to embrace actions such as the destruction of a city, and the enslavement and dispersal of any surviving population.10 However, the 1948 Genocide Convention, which focuses on a group’s physical destruction,11 defines genocide more narrowly as certain acts conducted with an intent to destroy ‘in whole or in part . . . a national, ethnical, racial or religious group, as such’.12 Dissatisfaction with this definition, in particular with its exclusion of various social groups, has led to a plethora of competing definitions as well as to the proposal of other ‘-cides’ – including ‘urbicide’.13 5 R. Lemkin, Lemkin on Genocide, ed. S. L. Jacobs (Lanham, MD: Lexington Books, 2012), pp. 17. 6 Against its use, see Barrandon, Massacres, pp. 329–37; for its application, see H. van Wees, ‘Genocide in the ancient world’ in The Oxford Handbook of Genocide Studies, ed. D. Bloxham and A. D. Moses (Oxford University Press, 2010), pp. 239–58; see also T. S. Taylor, ‘Introduction’ to A Cultural History of Genocide in the Ancient World, ed. T. S. Taylor (London: Bloomsbury, 2021), pp. 1–29, and F. Quesada-Sanz, ‘Genocide and mass murder in Second Iron Age Europe’ in The Routledge History of Genocide, ed. C. Carmichael and R. C. Maguire (London: Routledge, 2015), pp. 9–22. On anachronism, see B. Kiernan, ‘Is “genocide” an anachronistic concept for the study of early modern mass killing?’, History 99:336 (2014), 530–48. 7 N. Roymans, ‘A Roman massacre in the far north: Caesar’s annihilation of the Tencteri and Usipetes in the Dutch river area’ in Conflict Archaeology: Materialities of Collective Violence from Prehistory to Late Antiquity, ed. M. Fernández-Götz and N. Roymans (London and New York: Routledge, 2018), pp. 167–81, at p. 179. 8 Justinian, Institutes, 1.3.2 describes slavery as part of the ius gentium, the laws applying to all peoples. 9 D. J. Colwill, ‘Responses’ in Cultural History of Genocide, ed. Taylor, pp. 141–2, 145–50; Taylor, ‘Introduction’, 25–7. See R. Bauman, Human Rights in Ancient Rome, (London: Routledge, 2000) for Roman views of ‘human rights’. 10 R. Lemkin, Axis Rule in Occupied Europe: Laws of Occupation, Analysis of Government, Proposal for Redress (Washington, DC: Carnegie Endowment for International Peace, 1944), pp. 81–90. 11 Kiernan, Blood and Soil, pp. 12–13. 12 Convention on the Prevention and Punishment of the Crime of Genocide, Paris, 9 December 1948, United Nations, Treaty Series, vol. 78, no. 1021, 277–322, art. 2, www .hrweb.org/legal/genocide.html. 13 Jones, Genocide, pp. 22–37.

280

A Tale of Three Cities: Roman Destruction

These definitions either widen or contract the Convention understanding of the perpetrators or the victims of genocide, the actions that may constitute genocide, the motivation for the violence and the scale of the intended destruction.14 Many of these definitions, and popular understandings of genocide, are implicitly shaped by the paradigmatic impact, or ‘conceptual constraint’, of the Holocaust.15 Following the approach of The Cambridge World History of Genocide, which takes the Convention definition as its starting point, and taking into account the need for the consistent, transhistorical application of concepts to facilitate comparative research,16 this chapter will consider whether these urbicides involved an intent to destroy these three civic communities ‘as such’.

Sources First, we need to briefly consider our historical sources for this period. Here it is important to bear in mind two general limitations of Greco-Roman historiography, which had rhetorical purposes removed from modern approaches. First, to flesh out their narratives, ancient historians favoured the use of stock rhetorical topoi, accepted understandings of what ‘usually’ happened in events, such as the sack of a city.17 Second, Greco-Roman historians liberally included speeches in their narrative that often reflect what they believed ‘ought’ to have been said rather than what was actually spoken. Despite such limitations, we remain largely dependent on our flawed literary sources, even though all three sites have been the subject of archaeological study; such excavations can shed light on the nature of the devastation that occurred, but cannot tell us of the Roman motives and intentions.18 In addition to these limitations, we have only fragmentary contemporary witnesses for Roman actions, and almost nothing of contemporary Roman views of imperialism.19 Our most important contemporary witness is the Greek historian Polybius, who was present at the fall of Carthage and wrote during the mid-second century B C E. A friend of Scipio Aemilianus, the destroyer 14 Ibid., p. 28; S. Straus, ‘Contested meanings and conflicting imperatives: a conceptual analysis of genocide’, Journal of Genocide Research 3:3 (2001), 349–75, at 361. 15 D. Moshman, ‘Conceptual constraints on thinking about genocide’, Journal of Genocide Research 3:3 (2001), 431–50. 16 Kiernan, Blood and Soil, p. 12. 17 Quintilian, Institutio oratoria (Institutes of Oratory) (Quint. Inst.), 8.3.67–70. 18 On sources: Taylor, ‘Introduction’, 5–11; van Wees, ‘Ancient world’, 242–3. On archaeology: N. Roymans and M. Fernández-Götz, ‘The archaeology of warfare and mass violence in ancient Europe’ in Conflict Archaeology, ed. Fernández-Götz and Roymans, pp. 1–10. 19 S. Mattern, Rome and the Enemy (Berkeley: University of California Press, 1999), p. 212.

281

tristan s. taylor

of Carthage and Numantia, he also helped Rome organise Greece after Corinth’s destruction. Polybius wrote a history to explain to fellow Greeks how Rome had achieved world domination. He originally ended his history in 167 B C E with the defeat of Macedon under Perseus, but then continued to c.146 and also wrote a (now lost) history of the Numantine War.20 His work is only fragmentary for the years in question, and we thus depend on later Greek writers who draw upon his work and other lost sources. The most important of these is Appian, a historian writing in the second century C E, who provides our most detailed accounts of Carthage and Numantia. In addition, Polybius is a key source for the fragmentary historian Diodorus Siculus in the first century B C E, and the geographer Pausanias in the second century C E, who describes the destruction of Corinth. The well-read Greek polymath Plutarch, writing in the late first/ early second century C E, also provides useful information in some of his biographies of statesmen from the period based on now lost sources.21 Aside from fragments from the second century B C E, all the relevant Latin authors are from the first century B C E or later. The Roman historian Livy, writing in the first century B C E, drew upon Polybius for his patriotic account of Roman history, but for this period his work survives only in summaries (Periochae). Livy’s history possesses a strong moral purpose, and juxtaposes Carthage against Rome as its moral antitype.22 Also important is the late Republican philosopher, statesman and orator Cicero, who reflects on these three urbicides and Roman imperialism in a variety of contexts. It is always uncertain whether these later authors reflect earlier values. However, the views espoused in Cicero’s era remain relatively constant for centuries thereafter, suggesting some stability in core components of Roman imperial ideology.23

Roman Imperialism in the Second Century B C E Given the link between Roman imperialism and the destruction of Carthage, Corinth and Numantia, it is important to say something briefly on the nature of the Roman Empire in the second century B C E,24 acknowledging both that 20 Polyb. 3.1–4. 21 The bibliography on individual sources is large. For an introduction, see the author entries in the Oxford Classical Dictionary, ed. S. Hornblower, A. Spawforth and E. Eidinow, 4th ed. (Oxford University Press, 2012). 22 R. Miles, Carthage Must Be Destroyed: The Rise and Fall of an Ancient Civilization (London: Allen Lane, 2010), pp. 359–61. 23 Mattern, Rome and the Enemy, p. 216. 24 A good introduction with translated sources is A. Erskine, Roman Imperialism (Edinburgh University Press, 2010), esp. pp. 2–87.

282

A Tale of Three Cities: Roman Destruction

the concept of ‘Roman imperialism’ is controversial25 and that we lack contemporary Roman sources from the mid-second century that tell us of Roman imperialism through Roman eyes.26 Polybius believed that Rome had achieved world domination with the defeat of King Perseus and the dismantling of the Macedonian kingdom in 167 B C E.27 Polybius was not here principally concerned with direct territorial rule – although Rome did exercise such control, as in Sicily and northern Italy – but rather with those peoples over whom Rome might command authority. Such authority was embraced by the polyvalent term imperium.28 Such authority rested in part on respect for Rome’s status and the pretext, at least, for these urbicides was disrespect of this status. Let us now turn to the distinctive motivations for these three acts of destruction, before considering how they related to more general aspects of Roman imperialism.

Carthage: Predetermined Destruction In 149 B C E Rome commenced a siege of its long-time nemesis Carthage, ostensibly on the pretext that Carthage had violated the treaty that had ended the Second Punic War by responding to years of incursions by its Numidian neighbour with a war, undertaken without Rome’s permission.29 The Roman siege then lasted three years, with little progress made until Scipio Aemilianus took command.30 When the Romans finally breached the city’s walls in 146 B C E, an intense six-day carnage followed as they fought house-tohouse, setting fire to buildings and demolishing them, with many Carthaginians – including the old, women and children – perishing inside.31 Eventually, Scipio yielded to the supplications of Carthaginian survivors, and all those who came as suppliants, whom Appian numbers at 50,000, were enslaved rather than killed. The city was then plundered by the soldiers, and the spoils (including the survivors, who were all taken as captives) sold.32 Estimations of Carthage’s prewar population vary from 200,000 to 400,000.33 25 For example ibid., pp. 32–49. 26 Mattern, Rome and the Enemy, p. 212. 27 Polyb. 3.1. 28 Erskine, Roman Imperialism, pp. 6, 39; R. M. Kallet-Marx, Hegemony to Empire: The Development of Roman Imperium in the East from 148–62 B.C. (Berkeley: University of California Press, 1996), pp. 22–9; E. Gruen, The Hellenistic World and the Coming of Rome, 2 vols. (Berkeley: University of California Press, 1984), p. 279. 29 Miles, Carthage, pp. 337–8. 30 For a summary of the siege, see ibid., pp. 341–7; J. Levithan, Roman Siege Warfare (Ann Arbor: University of Michigan Press, 2013), pp. 112–13. 31 Appian, Punica (also known as Punic Wars or Libyke) (App. Pun.) 128–9. 32 App. Pun. 130–3. 33 Kiernan, Blood and Soil, p. 53.

283

tristan s. taylor

Even using the most conservative estimation of the population, this suggests that well over 100,000 perished through the combined effects of the war, siege and sack of the city. The Senate sent a commission of ten to arrange African affairs and they declared that what remained of Carthage was to be razed. Religious rituals were also invoked, with the Carthaginian gods summoned from the city prior to its destruction by the process of euocatio,34 and subsequently the land was in some form consecrated so that none were allowed to live there.35 Five of Carthage’s allied towns were also destroyed, with their lands divided among those neighbours who had supported Rome.36 Carthage stands apart from both Corinth and Numantia in that, while justified on the pretext of a Carthaginian treaty breach, it appears that at least a significant faction at Rome had decided on Carthage’s destruction before this rationale existed. Thus, Polybius states that the Romans had decided to destroy Carthage long before they did, but were waiting for a pretext acceptable to others.37 Appian places this decision in the 150s when Cato the Elder – a veteran of the Second Punic War and an implacable opponent of Carthage – visited Carthage on an embassy associated with a Numidian seizure of Carthaginian territory in 152 B C E.38 Cato was reportedly alarmed at Carthage’s prosperity after its humbling at the end of the Second Punic War, finding it ‘teeming with vigorous fighting men, overflowing with enormous wealth, filled with arms of every sort and with military supplies, and not a little puffed up by all this’.39 Cato then reported Carthage’s growing power to the Senate and (in)famously ended every subsequent speech he made by saying that Carthage must be destroyed – apocryphally rendered today as ‘Carthago delenda est’.40 Another senator, Scipio 34 Macrobius, Saturnalia (Macrob. Sat.), 3.9.7–13. 35 Cicero, De lege agricultura (On the Agrarian Law) (Cic. Leg. agr.) 1.5. On Carthage’s consecration and the apocryphal nature of the legend that the land was sown with salt, see R. Ridley, ‘To be taken with a pinch of salt: the destruction of Carthage’, Classical Philology 81 (1986), 140–6, with S. T. Stevens, ‘A legend of the destruction of Carthage’, Classical Philology 83 (1988), 39–41. 36 App. Pun. 135. 37 Polyb. 36.2.1; App. Pun. 74; Polybius’ sources for this have been doubted: P. J. Burton, Friendship and Empire: Roman Diplomacy and Empire in the Middle Republic (353–146 BC) (Cambridge University Press, 2011), pp. 313–14. However, Polybius was at Carthage’s destruction and as Scipio’s friend had access to some senators’ thoughts: W. V. Harris, ‘Roman expansion in the west’ in The Cambridge Ancient History, vol. V I I I : Rome and the Mediterranean to 133 BC,ed. A. E. Astin et al., 2nd ed. (Cambridge University Press, 1989), pp. 107–62, at p. 148. 38 App. Pun. 69. 39 Plutarch, Cato Major (Cato the Elder) (Plut. Vit. Cat. Mai.) 26.2 (trans. Perrin); see also App. Pun. 69; Livy, Periochae (Per.) 47. 40 The Greek writers Appian (Pun. 69) and Plutarch (Cat. Mai. 27.1) report that Cato’s wish was that Carthage ‘not exist’ (Καρχηδόνα μὴ εἶναι), rather than that it ‘must be destroyed’.

284

A Tale of Three Cities: Roman Destruction

Nasica, opposed Cato with an insistence on Carthage’s survival, reportedly on two bases: first, that ongoing fear of Rome’s rival was necessary to encourage domestic harmony and good governance of the allies, lest they revolt;41 and second, that there was no proper pretext for war.42 Once Carthage’s treaty violation provided a pretext, the diplomatic exchanges that followed as Carthage tried to avert war showed the Senate resolved at the very least on war, if not on Carthage’s destruction. Although Carthage executed the general who commanded the battle that violated the treaty, Rome deemed this insufficient. Instead, Appian reports that the Roman Senate declared war, and secretly instructed the consuls not to desist until Carthage was razed.43 Carthage responded by making a formal submission or deditio in fidem (surrender into good faith) to Rome. Such a surrender was a key component of Roman imperialism whereby a people surrendered themselves into the good faith or fides of the Romans, which symbolised Roman superiority and would usually result in lenient treatment if enacted prior to conflict.44 However, Rome continued its war preparations, demanded hostages and stated that further instructions would be given by the Roman army, which was already heading to Africa.45 The Romans in Africa then demanded that Carthage surrender its arms. This was followed by Rome’s ultimatum: that the Carthaginians abandon their city and move around sixteen kilometres inland. The Carthaginians refused, and the Third Punic War began.46 While it has been questioned whether this ultimatum was intended to drive Carthage to war,47 its full existential import must be appreciated. It struck at the heart of Carthage’s existence as a maritime state that relied upon its sea connections for its wealth and power. It effectively amounted to a ‘death sentence’ for Carthage as, in addition, ‘there was no precedent for a state’s surviving the eradication of what constituted it on the sacred plane’, such as temples and graves, despite Roman assurances that these would be spared.48 Scholars debate why Rome persisted in this unusual determination for conflict even after Carthage had made a deditio and complied with Rome’s 41 Plut. Cat. Mai. 27.2; Diod. Sic. 34/35.33. 42 Livy, Per. 48. 43 App. Pun. 74–5. 44 Mattern, Rome and the Enemy, pp. 217–18; F. Marco Simón, ‘Insurgency or state terrorism: the Hispanic wars in the second century BCE’ in Brill’s Companion to Insurgency and Terrorism in the Ancient Mediterranean, ed. T. Howe and L. L. Brice (Leiden: Brill, 2015), pp. 217–47, at pp. 224–5; Burton, Friendship, pp. 114–21. 45 App. Pun. 76; Harris, ‘Roman expansion’, 152. 46 App. Pun. 80–1. 47 For example A. E. Astin, Scipio Aemilianus (Oxford: Clarendon Press, 1967), p. 274; cf. Harris, ‘Roman expansion’, 152n257. 48 S. Lancel, Carthage: A History, trans. A. Nevill (Oxford University Press, 1995), pp. 158–9; Miles, Carthage, p. 340; App. Pun. 89.

285

tristan s. taylor

demands.49 Various factors have been proposed, including the Roman elite’s desire for glory and the general desire for the plunder from sacking a rich city like Carthage.50 While they all may have played some role, the ancient sources emphasise one factor: fear.51 Here, the views of Cato, regarded as the author of Carthage’s demise, are particularly important.52 In calling for Carthage’s destruction, Cato emphasised several themes: its resurgent resources, power and confidence, and its preparedness for war against Rome: ‘For the Carthaginians are already enemies to us. For whoever prepares everything against me so that he might bring war against me at whatever time he wishes, this person is already my enemy, even if he does not yet act with arms.’53 These words must be read in the context of Roman stereotypes of maritime powers corrupted by the sea’s influence: they possessed a grasping nature that would cause them to strive to rival Rome.54 Thus, Carthage, with its population, harbours and walls, was seen as jutting out and threatening Rome’s most productive islands.55 This anxiety was reflected in Rome’s demand that Carthage move sixteen kilometres inland – the distance Plato claims was necessary to avoid the sea’s corrupting impact.56 In addition, Cato’s theatrical gesture of letting a fresh fig drop from his toga in the Senate and claiming that it was from Carthage, a place only three days’ sail from Rome, emphasised Carthage’s proximity to Rome.57 Finally, in a hostile speech, fragmentarily preserved from his successful efforts to rebuff last-ditch Carthaginian peace overtures, Cato tapped into the ‘standard’ hostile Roman stereotypes of Carthaginians.58 He emphasised memories of the past devastation they had wrought on Italy in the Second Punic War, and their reputation for perfidiousness (Punica fides, ‘Punic faith’, had become an ironic byword at Rome for untrustworthiness) and cruelty:59 ‘Who are they who have often broken treaties? The Carthaginians. Who are they who have waged war with severest cruelty? The Carthaginians. Who are they who have marred the face of Italy? 49 Harris, ‘Roman expansion’, 160; Kiernan, ‘First genocide’, 28. 50 For example Astin, Scipio, p. 52, and Cato the Censor (Oxford: Clarendon Press, 1978), p. 283; Harris, ‘Roman expansion’, 152–7; U. Vogel-Weidemann, ‘Carthago delenda est: aitia and prophasis’, Acta Classica 32 (1989), 79–95; Burton, Friendship, pp. 311–22. 51 Polyb. 36.9.3–4; Cicero, Pro Murena (On Behalf of Murena) (Cic. Mur.) 58; Cicero, De republica (The Republic) (Cic. Rep.) 1.71; Velleius Paterculus (Vell. Pat.) 2.4.5; Astin, Scipio, p. 52. 52 Plut. Cat. Mai. 36. 53 Oratorum Romanorum Fragmenta3 fr. 195; Miles, Carthage, p. 339. 54 Cic. Rep. 2.7–8. 55 Cic. Leg. agr. 2.87. 56 Plato, Leges (Laws) (Pl. Leg.) 704d; Harris, ‘Roman expansion’, 156. 57 Plut. Cat. Mai. 26.1. Actually six days: Kiernan, Blood and Soil, p. 53. 58 B. Isaac, The Invention of Racism in Classical Antiquity (Princeton University Press, 2004), pp. 327–30. 59 See also Livy’s description of Hannibal as of ‘inhuman cruelty and a faith worse than Punic’ (Per. 21.4.9).

286

A Tale of Three Cities: Roman Destruction

The Carthaginians. Who are they who now ask for pardon? The Carthaginians. See then how appropriate it is for them to gain their request.’60 Cato’s logic is clear, as is his emphasis on the Carthaginian threat: the Carthaginians are near; the Carthaginians are preparing for war; the Carthaginians are untrustworthy and cruel; the Carthaginians cannot be pardoned: Carthage must be destroyed. Not all shared this perspective – Scipio Nasica and others resisted, and some were driven by motives such as glory or greed – but this intersection of anxiety and prejudice provides a framework for understanding the unusual Roman persistence in pursuing Carthage’s destruction. In addition to the external threat to Roman liberty of a renascent Carthage, later sources also suggest Cato was concerned about an internal crisis:61 anxieties about the corrupting impact of empire on Roman character from the influx of wealth and increased contact with external, particularly Greek, culture.62 Consequently, Cato believed that Rome needed to eliminate external threats so that it could focus inwardly on internal challenges. Thus, Plutarch claims that while Scipio Nasica wanted a powerful Carthage as a counterweight to Rome, It was precisely this that seemed [to Cato] to be a cause for alarm, that a city that had always been great [Carthage] . . . was threatening the Roman people when they were to a great extent inebriated and staggering with their power. Such external threats to their authority ought to be done away with altogether, he thought, that they might be free to devise a cure for their domestic failings.63

We cannot know whether Plutarch had access to an account that directly provided Cato’s thoughts, or whether this instead represents Plutarch’s reasonable supposition from Cato’s well-known anxiety to preserve traditional Roman mores. Nonetheless, such Roman anxieties are recorded for this period in general, and for Cato in particular.

Corinth: Retribution In contrast to the premeditated nature of Carthage’s destruction, the razing of Corinth appears as reactive retribution for an ally’s failure to respect Roman authority.64 While a pretender, Andriscus, assumed the Macedonian throne in 60 Cicero, Rhetorica ad herennium (To Gaius Herrenius on the Theory of Public Speaking) 4.14.20 (trans. Caplan). 61 App. Pun. 69; Kiernan, ‘First genocide’, 32–3. 62 See A. W. Lintott, ‘Imperial expansion and moral decline in the Roman republic’, Historia 21 (1972), 626–38. 63 Plut. Cat. Mai. 27.3 (trans. Perrin, adapted). 64 For a succinct narrative, see P. S. Derow, ‘Rome, the fall of Macedon and the sack of Corinth’ in Cambridge Ancient History, V I I I, pp. 290–323, at pp. 319–23.

287

tristan s. taylor

c.149 B C E,65 a dispute erupted in Rome’s ally the Achaean League, the meeting place for which was Corinth,66 over Sparta’s desire to withdraw from the League. After some delay due to Andriscus’ revolt, a Roman embassy arrived in Corinth in 147 to resolve the dispute. It demanded that several places, including Lacedaemon, Corinth and Argos, be detached from the League. This threatened the League’s continuity, and provoked a violent response in Corinth towards both Spartans and the lodgings of the Roman ambassadors where some Spartans were housed. This mistreatment of the ambassadors, when the hospitable treatment of ambassadors was regarded as sacrosanct in antiquity,67 was used as a Roman pretext for the subsequent war.68 Despite another Roman embassy, the League voted at Corinth for war nominally against Sparta, though in reality against Rome.69 Metellus, who had suppressed Andriscus’ revolt and was eager for further glory, attempted to end the matter before the consul Mummius’ arrival, and he won some initial victories over the Achaeans. Mummius then arrived and defeated what remained of the Achaeans at the battle of the Isthmus of Corinth. Mummius’ forces pursued the Achaeans to Corinth, the gates of which were found open after the remainder of the Achaean army and most of the inhabitants had fled. After some initial hesitation from fear of an ambush, Mummius’ soldiers stormed the city, killing the men, enslaving the women and children, and plundering and burning the city.70 Subsequently, Mummius attempted to enslave those Corinthians who had escaped the sack.71 How many were killed in the sack or subsequently enslaved is unknown, both because our sources tell us that many of the inhabitants had fled, and because the pre-sack population of Corinth is very uncertain. For the earlier classical period, when the city was more prosperous,72 an average of the various modern estimates for the urban population based on reported army figures ranges from as little as around 26,000 to as many as approximately 54,000.73 As at Carthage, an euocatio had been performed to call out the city’s gods, and a subsequent senatorial delegation

65 On this revolt, see Kallet-Marx, Hegemony, pp. 31–6. 66 E. R. Gebhard and M. W. Dickie, ‘The view from the Isthmus, ca. 200 to 44 B.C.’ in Corinth: Results of the Excavations Conducted by the American School of Classical Studies at Athens, vol. X X: Corinth, the Century 1896–1996, ed. C. K. Williams II and N. Bookidis (American School of Classical Studies at Athens, 2003), pp. 261–78, at p. 262. 67 Van Wees, ‘Ancient world’, 255. 68 Pausanias (Paus.) 7.14.1–3; Livy, Per. 51–252; Cicero, Pro lege Manilia (On the Manilian Law) (Cic. Leg. Man.) 11. 69 Polyb. 38.12–13; Derow, ‘Rome’, 322–3. 70 Paus. 7.16.5. 71 Zonaras (Zonar.) 9.31. 72 Gebhard and Dickie, ‘View from the isthmus’, 262. 73 R. Willet, ‘Whirlwind of numbers – demographic experiments for Roman Corinth’, Ancient Society 42 (2012), 127–58, at 147.

288

A Tale of Three Cities: Roman Destruction

determined that the city should be razed, and the territory was given to neighbouring peoples.74 It is tempting to see parallels between Corinth and Carthage, both destroyed in 146.75 Like Carthage, Corinth was a city of great antiquity and reputation, and a maritime city – thus also liable to the sea’s corrupting influence.76 Indeed, Cicero lays emphasis on Corinth’s position as a motivation for its destruction:77 on the isthmus, it held the ‘key’ to northern and southern Greece by land, and, located on both the Corinthian and Saronic gulfs, it united two seas.78 It has even been argued that the symmetry between Carthage and Corinth, emphasised by the similar rituals used (e.g. euocatio), marks these two acts of destruction as twin intentional statements of Roman power in both the eastern and western Mediterranean.79 However, since the Romans embarked on their conquest of Carthage years before they knew of any conflict at Corinth, such symbolism would have been opportunistic following the taking of Corinth, rather than an intentional plan from the beginning.

Numantia: Shame and Humiliation Polybius termed Rome’s conflict with the Numantines in Spain in the second century B C E the ‘fiery war’ for its violent and unceasing nature (πύρινος πόλεμος).80 The eventual destruction of Numantia – a stronghold of the Arevaci, a Celtiberian tribe – in 133 B C E evinces a different retributive motivation for the Romans – that of atoning for shame and humiliation, a motivation that ‘features prominently in the most extreme manifestations of aggression: murder, war and genocide’ (Figures 10.1, 10.2).81 Despite their relatively small population, the Numantines both resisted Roman authority and inflicted a series of humiliating defeats upon Rome’s most senior magistrates – the consuls. Sporadic conflict with the Celtiberians had existed after Rome first created two regular provinciae in Spain – areas of command (Hispania Ulterior and Hispania Citerior, further and nearer

74 Macrob. Sat. 3.9.7–13; Zonar. 9.31; Florus (Flor.) 1.32.6; Kallet-Marx, Hegemony, pp. 85–6. Livy, Per. 52: this was done by senatorial decree. See also Vell. Pat. 1.13.1. 75 See Purcell, ‘Carthage and Corinth’. 76 Cic. Rep. 2.7–8. 77 For example De Officiis (Of Duties) (Off.) 1.35. 78 Cic. Leg. agr. 2.87. 79 Purcell, ‘Carthage and Corinth’, 139 and 141. 80 Polyb. 35.1; Diod. Sic. 31.40. Narrative in Harris, ‘Roman expansion’, 118–42; see also Marco Simón, ‘Insurgency’ for Roman actions in Spain. 81 Jones, Genocide, p. 540.

289

tristan s. taylor

Figure 10.1 Ruins of Numantia. (Photo: Ben Kiernan)

Figure 10.2 Reconstruction of the ruins of Numantia. Museo Numantino, Ministerio de Cultura. (Photo: Ben Kiernan)

290

A Tale of Three Cities: Roman Destruction

Spain) – during the Second Punic War. That conflict ended with a treaty in 179 B C E.82 In the 150s, however, two interconnected wars erupted involving the Lusitanians, in roughly modern-day Portugal, and the Celtiberians in central Spain (see Map 10.2). After some humiliating defeats for the Romans,83 the consul M. Claudius Marcellus secured the deditio of the Celtiberians in 151.84 However, the Celtiberian Arevaci, Belli and Titthi were persuaded by a subsequent successful revolt in Lusitania to again defect from Roman authority in c.144–143 B C E.85 While the other two Celtiberian groups surrendered in 142 B C E, the Arevaci balked at the last minute at surrendering their arms.86 Then, if not before, the Senate decided it would only accept the Numantines’ unconditional deditio to end the war, and it rejected several peace efforts as a series of Roman commanders failed to defeat Numantia.87 First, Pompeius, suffering under the harsh winter, failed to take Numantia and proposed a peace favourable to the Numantines, which was to be covered under the pretence of a deditio. That was rejected by the Senate in 139.88 The Numantines then defeated Pompeius’ successor, M. Popillius Laenas, and the next magistrate – C. Hostilius Mancinus – was ambushed by the Numantines and compelled to surrender on equal terms under oath in 137.89 The Senate also rejected Mancinus’ humiliating capitulation, and he was handed over bound and naked to the Numantines to appease the gods’ anger at his broken oath.90 After so many conspicuous failures to defeat Numantia, the war had become a source of public shame (publicus pudor). Consequently, despite a law against re-election, Scipio Aemilianus – Carthage’s destroyer – was reelected as consul in 134 to overcome Numantia.91 Scipio raised additional volunteers for the army, restored discipline in Spain and cut off potential Numantine supplies through a scorched earth policy against the neighbouring Vaccaei. He refused battle and besieged Numantia, aiming to starve its population into submission.92 Eventually, after resorting to cannibalism, the Numantines offered their deditio – a number committing suicide beforehand to preserve their liberty.93 After selecting fifty of the survivors to march in his Roman triumph, Scipio sold the remainder of the city’s population into slavery.94

82 84 87 89 92 94

Appian, Hispania (Iberike or Spanish Wars) (Hisp.) 43–4. 83 App. Hisp. 45–8, 56. App. Hisp. 49–50. 85 App. Hisp. 66. 86 Diod. Sic. 33.16. See Astin, Scipio, pp. 147–8, and Richardson, Hispaniae, pp. 140–9. 88 App. Hisp. 76–9. App. Hisp. 80; Livy, Per. 55. 90 App. Hisp. 83; Livy, Per. 56. 91 Livy, Per. 56. App. Hisp. 85–90; Levithan, Siege Warfare, pp. 113–15. 93 App. Hisp. 96–7. App. Hisp. 98.

291

S

E

AT L A N T I C OCEAN

W

N

Durius

Gades

Tagus

Baetis

Carteia

Italica

Anas

Alcántara

B

H RT

ER

A SP

IN

N

R

ER

SP

Saguntum

EA

M A R E

Carthago Nova

Numantia

I N T E R N U M

FU

Corduba

PENINSULA

IBERIAN

D

A

IN

0

0

50

75

E

100

150

Emporiae

Balearic Islands

Tarraco

P Y R E N E Gracchurris E S A N S

Ebr o

C E L T I B E R I

C

150 Miles

225 Km

Massilia

Map 10.2 The Iberian peninsula in the second century B C E. (Cartography by the Ancient World Mapping Center, University of North Carolina)

Ancient World Mapping Center 2011

4

3

2

1

A

LU

IA

AN

SI T

Rhodan us

A Tale of Three Cities: Roman Destruction

It is not known how many perished in the siege or were subsequently enslaved, although at the start of the war – prior to deaths in various battles – Appian gives a warrior population of about 8,000 and Velleius Paterculus claims never more than 10,000 were armed, though Livy, Florus and Orosius give a figure of only 4,000 soldiers. The latter, smaller figure would suggest a total Numantine population of approximately 16,000.95 However, given the small size of the town itself – it was about 8.36 hectares, of which 5.1 hectares were inhabited; or between 500 and 600 households – an urban population of 1,500 to 2,500 seems more likely, although it is probable that more fled to the stronghold from the surrounding countryside during times of conflict.96 Allowing for an influx of rural refugees, these figures suggest several thousand perished in the siege or were enslaved. Scipio did not wait for senatorial authorisation, but razed the stronghold, dividing its territory among its neighbours.97 In so doing, Velleius says that Scipio freed Rome from ‘humiliation’.98 A committee of ten senators then arrived to reorganise the region.99

Motivations and Justifications: Empire, Symbolic Violence and Existential Anxiety In discussing how empire was obtained and preserved in the Greco-Roman world, Diodorus Siculus observes that ‘Those whose object is to gain dominion over others use courage and intelligence to get it, moderation and consideration for others to extend it widely, and paralysing terror to secure it against attack.’100 He then observes that the Romans were renowned for clemency in creating their empire, ‘[b]ut once they held sway over virtually the whole inhabited world, they confirmed their power by terrorism and by the destruction of the most eminent cities. Corinth they razed to the ground, the Macedonians (Perseus for example) they rooted out, they razed Carthage and the Celtiberian city of Numantia, and there were many whom they cowed by terror.’101 While Diodorus exaggerates (such Roman tactics predate this period),102 he highlights well 95 App. Hisp. 97; Vell. Pat. 2.1.4; Livy, Per. 54; Flor. 1.34; Orosius, Historia contra paganos (History against the Pagans) (Oros.) 4.7.3; A. Jimeno, R. Liceras, S. Quintero and A. Chaín, ‘Unraveling Numantia: Celtiberian and Roman settlement (Soria, north-central Spain’, in Archaeology in the River Duero Valley, ed. J. C. Sastre Blanco and Ó. Rodríguez-Monterrubio (Newcastle upon Tyne: Cambridge Scholars, 2018), pp. 199–220, at p. 207. 96 Jimeno et al., ‘Numantia’, 209. 97 App. Hisp. 98. 98 Vell. Pat. 2.4.3. 99 App. Hisp. 99; Flor. 1.34. 100 Didod. Sic. 32.2 (trans. Walton). 101 Diod. Sic. 32.4 (trans. Walton). 102 Kallet-Marx, Hegemony, p. 88.

293

tristan s. taylor

the connection between the use of symbolic violence – what van Wees has termed ‘conspicuous destruction’ – and empire in the Greco-Roman world.103 Polybius similarly observes that it was the Roman custom (ethos) to massacre not only people but even animals they encountered while they sacked a city, in order to inspire ‘terror’.104 This terror was ‘exemplary’ – either to cow others to surrender and thus avoid future conflict, or to punish those who had disobeyed or rebelled against Rome, and so deter others from doing likewise.105 This use of exemplary collective violence reflected views that vengeance was a just cause for war,106 and that unavenged disrespect of Rome’s authority posed a threat to Rome’s dominance, which depended on Rome’s reputation (existimatio) to deploy force in response to disobedience.107 Although writing later, Cicero expresses this view succinctly when he describes his own destruction of a small mountain town for siding with Rome’s rival, the Parthians: ‘I came to a decision that it was a matter of upholding the reputation (existimatio) of our imperium to crush their impertinence, so that the spirit of others who might be hostile to our imperium might be more easily broken’.108 Consequently, there was a fear that should one people get away with failing to respect Roman authority by not yielding to Rome’s demands, by revolting, or by attacking Rome’s allies despite Rome’s warnings, others would soon follow, bringing about the downfall of Rome’s imperium. This sentiment is expressed in Cicero’s assessment that, in destroying Carthage and Numantia, Scipio Aemilianus eliminated dread threats (terrores) both to Roman imperium and the city.109 The casting of these cities as terrores of Roman authority sits uneasily with a rational assessment of the relative position of these peoples.110 However, the successful resistance of a small community such as Numantia was perhaps an even more serious threat to 103 Van Wees, ‘Ancient world’, 250. 104 Polyb. 10.15.4–5. For discussion of this passage’s idealised nature, and the Roman practice of sacking cities, see A. Ziolkowski, ‘Urbs direpta, or how the Romans sacked cities’ in War and Society in the Roman World, ed. J. Rich and G. Shipley (London: Routledge, 1993), pp. 69–91, and Kallet-Marx, Hegemony, pp. 84–5. 105 For example Justin 34.2.6 on Corinth. See also Taylor, ‘Introduction’, 11–13; van Wees, ‘Ancient world’, 256–7. 106 Cic. Rep. 3.35. 107 On these ideas, see Mattern, Rome and the Enemy, esp. pp. 212–22, on the republican period. 108 Cicero, Epistulae ad familiares (Letters to His Friends) (Cic. Fam.) 15.4.10, quoted in van Wees, ‘Ancient world’, 251 (adapted). 109 Cic. Mur. 58; Cic. Rep. 1.71; Vell. Pat. 2.4.5. 110 For example Harris, ‘Roman expansion’, 151–3. On Carthage, though, cf. Burton, Friendship, pp. 315–16; J. S. Richardson, Hispaniae (Cambridge University Press, 1986), p. 152.

294

A Tale of Three Cities: Roman Destruction

Rome’s reputation and authority than that of a serious imperial rival: it could inspire further revolt, just as the Numantines themselves had been inspired by the Lusitanians.111 Importantly, though, mass violence does not necessarily flow from rational calculation, even when, as in the Roman world, such violence is regularly used ‘pragmatically’. Here we might consider the implications of what Scott Straus has termed the ‘domination–vulnerability paradox’: those that perpetrate mass violence possess overwhelming force yet are also motivated by a feeling of vulnerability that becomes focalised on a particular people and can only be assuaged by the removal of that threat.112 Here, the impact on Roman thought of ongoing turmoil in the mid-second century B C E, including Andriscus’ revolt and instability in Spain, cannot be discounted. While such a paradigm sits uneasily with displays of Roman arrogance, like its insistence on the breaking up of the Achaean League, haughty ultimata to Carthage and insistence on a Numantine deditio, it does offer a paradigm for considering Roman fears of a renascent Carthage. As much as humiliation by a small people like Numantia required avenging, there was also symbolic potency to be gained by destroying a formidable rival, and thus the impact of the destruction of Carthage and Corinth was enhanced by their scale and antiquity.113 Carthage had been Rome’s keenest rival in the western Mediterranean, and Corinth had symbolic importance: it was the location where, following the Roman defeat of the Macedonian king Philip V in 196 B C E, the Roman commander Flamininus had ostentatiously decreed the Greeks to be free.114 Its destruction was therefore a punctuation mark of Greece’s submission to Rome, and considered so by Polybius.115 In addition, its location on the isthmus between northern Greece and the Peloponnese made it well placed as a visible reminder of Rome’s destructive power.116 111 Van Wees, ‘Ancient world’, 253. 112 S. Straus, Making and Unmaking Nations: War, Leadership and Genocide (Ithaca, NY: Cornell University Press, 2015), pp. 53–6; S. Kelley, ‘Perpetrators’ in Cultural History of Genocide in the Ancient World, ed. Taylor, pp. 72–101, at pp. 74–6. For similar themes see M. Levene, Genocide in the Age of the Nation State, vol. I: The Meaning of Genocide (London: I. B. Tauris, 2005), pp. 129–43. 113 Purcell, ‘Carthage and Corinth’, 133–4. 114 Polyb. 18.44–6; Livy, Per. 32.10–33.15. 115 Polyb. 38.1. 116 Kallet-Marx, Hegemony, p. 88. E.g. Diod. Sic. 32.27 (modelled on Polybius, but emphasising the visual impact of Corinth); Cicero, Tusculanae disputationes (Tusculan Disputations) (Cic. Tusc.) 3.54; Servius Sulpicius in Cic. Fam. 4.5.4.

295

tristan s. taylor

Prejudice and Justification of Violence As noted above, Roman anxieties about maritime cities and stereotypes of Carthaginian perfidy and cruelty played some role in motivating and justifying mass violence against Carthage, and possibly Corinth. Does the connection between violence and prejudice extend further?117 Examinations of ancient mass violence have generally argued for an absence of racial motivation, instead insisting that people were targeted not for ‘who they were’ but for ‘what they did’.118 Further, the absence of a general ethical objection to mass violence is argued to mean that there was no need to ‘dehumanise’ victims.119 This is consistent with the absence of a campaign to destroy ‘in whole’ the wider ethnic or national groups that Carthage, Corinth and Numantia represented. That said, the sources do show traces of ethnic prejudice playing some role in the destruction of Carthage and Numantia. As we have seen, Cato emphasised the Carthaginians’ cruelty and perfidy.120 Although writing later, Cicero joins Carthaginian stereotypes with the Iberians’ barbarous reputation – placed by Greco-Roman sources at the fringes of civilisation, such that the early imperial geographer Strabo describes the Celtiberians as the most ‘brute-like’ (θηριωδέστατοι) people.121 In his philosophical work De officiis (On Duties), Cicero outlines appropriate behaviour in warfare, distinguishing between using discussion and force to resolve disputes: the former being suitable for ‘people’ (homines), the latter for ‘beasts’ (beluarae). He then observes that one should spare those who have not been ‘cruel’ (crudeles) or ‘inhuman’ or ‘savage’ (immanis). That is, those who behave like beasts (beluarae/immanis) are not suited for diplomacy, but may be destroyed. Cicero explicitly includes Carthage and Numantia within this category. That is, they behaved in an inhuman, savage way and were thus correctly destroyed.122 In contrast, Cicero wishes that (civilised) Greek Corinth had not been destroyed, and speculates that its destruction, rather than based on its inhabitants’ behaviour, was conducted because of pragmatic concerns about its location.123 Polybius expresses similar thoughts and states that Greeks, 117 See also Taylor, ‘Introduction’, 18–19; Isaac, Invention of Racism, pp. 215–24. 118 For example D. Konstan, ‘Motivation and justification’ in Cultural History of Genocide in the Ancient World, ed. Taylor, pp. 51–71, at p. 71. 119 So Isaac, Invention of Racism, p. 222. But see on ancient Israel and Mesopotamia, T. M. Lemos, Violence and Personhood in Ancient Israel and Comparative Contexts (Oxford University Press, 2017). 120 Isaac, Invention of Racism, pp. 327–30. 121 Strabo 3.2.15. See also Diod. Sic. 33/34.4.2. 122 J. Barlow, ‘Scipio Aemilianus and Greek ethics’, Classical Quarterly 68 (2018), 123. 123 Cic. Off. 1.35.

296

A Tale of Three Cities: Roman Destruction

when fighting among themselves, ought not exterminate their enemies, but correct them.124 However, when discussing the destruction of the Aegosage, a Gallic tribe, by a Hellenistic king, Polybius’ concern about extermination disappears, and he praises the action as an exemplary lesson to barbarians.125 While it is not the inherent characteristics of Carthaginians or Numantines, but rather their cruel or inhuman behaviour, that is used to justify violence against them here – not ‘who they were’ but ‘what they did’ – this should not obscure the dynamics at work. The Romans frequently identified and punished those individually responsible for wrongdoing. Indeed, while besieging Numantia, Scipio had the hands amputated of only those residents of a neighbouring town who actively advocated aiding the Numantines, rather than punishing the whole people (although he threatened to do so).126 However, in relation to Carthage and Corinth, and to Numantia itself, individual responsibility was effaced, and the majority of the group’s members were targeted not for ‘what they did’, but for ‘who they were’, based on the collective’s ascribed guilt.127

Empire, Land, Agriculture and Mass Violence Land scarcity and associated expansionist aspirations, combined with the idealisation of agriculture, have historically been a powerful combination of factors driving mass violence.128 Did these factors play a role in these three urbicides? The destruction of Carthage and Corinth can be connected with Roman prejudices about the corrupting effects of trade and sea travel, in contrast with the moral benefits of agriculture. Here we might note that the staunchest advocate of Carthage’s destruction, Cato, opens his manual on farming (De agricultura) by contrasting the trader with the idealised citizen farmer, stating that ‘it is from the farming class that the bravest men and the sturdiest soldiers come, their calling is most highly respected, their livelihood is most assured and is looked on with the least hostility, and those who are engaged in that pursuit are least inclined to be disaffected’.129 In addition to this ideology in Roman thought, we can identify contemporary Roman perceptions that a land shortage existed, in particular for those who served in the Roman army. This had inspired first Caius Laelius around 140 B C E, then in 133 Tiberius Gracchus – who had served in turn under both 124 127 128 129

Polyb. 5.11.5. 125 Polyb. 5.111.6. 126 App. Hisp. 94. Taylor, ‘Introduction’, 29. Kiernan, Blood and Soil, pp. 29–33. De agricultura, preface (trans. Hooper and Ash), quoted in Kiernan, ‘First genocide’, 29–30.

297

tristan s. taylor

Scipio Aemilianus at Carthage and Mancinus at Numantia – and then his younger brother Caius Gracchus in 122–121, to attempt the redistribution to Roman citizens of public land acquired by conquest (ager publicus) in Italy.130 However, it was only with Caius that a proposal arose to settle the territory taken from Carthage and Corinth.131 It was unsuccessful. These contemporary concerns do not appear to have driven the destruction of Carthage, Corinth and Numantia, as, rather than resettled by Romans, the land of Carthage was consecrated, and that of Corinth and Numantia redistributed to neighbours. This contrasted with Roman actions in northern Italy in the first half of the second century B C E, where programmes of massacre, enslavement and deportation had been combined with the settlement of Roman colonies or colonia on seized territories until around 170 132 B C E.

Greed and Glory Although acquisition of territory for settlement may not have been a priority, Romans knew well the material benefits that flowed from the conquest of urban communities, including plunder, indemnities, land and taxation or tribute. Indeed, it has been argued that the possibility of plunder from wealthy Carthage explains the relative ease of recruitment for the Third Punic War, in contrast to the wars in Spain.133 Mass enslavement of the surviving inhabitants, as happened in all three cases here, frequently accompanied a city’s sacking. However, the desire to profit from plunder did not require that the city, or its population, be destroyed; indeed, the possibility of profiting from enslavement could militate against full-scale massacre.134 Beyond the material benefits that could flow from conquest, Romans attached significant ideological and political importance to military glory. This is most clear for the elite, who were in intense competition to enhance their reputation and that of their families through two interrelated achievements: public office and success in war. As each Roman magistracy only lasted one year, elites were incentivised to engage in military conflict in this 130 For ancient perceptions of the land crisis, see Appian, Bella civilia (Civil War) (App. B Civ.) 1.7–10; Plutarch, Tiberius Gracchus, 8–9. 131 Plutarch, Caius Gracchus, 10–11. 132 See e.g. S. P. Oakley, ‘The Roman conquest of Italy’ and J. Rich, ‘Fear, greed and glory: the causes of Roman war-making in the middle Republic’ in War and Society in the Roman World, ed. Rich and Shipley, pp. 9–37 and pp. 38–68, at p. 49 respectively; cf. Harris, ‘Roman expansion’, 106–18 (Cisalpine Gaul) with 138–9 (Spain). 133 Harris, ‘Roman expansion’, 156. 134 Van Wees, ‘Ancient world’, 252; Taylor, ‘Introduction’, 24; Konstan, ‘Motivation and justification’, 55.

298

A Tale of Three Cities: Roman Destruction

Figure 10.3 Inscription erected by Mummius dedicating the temple to Hercules Victor. Inscriptiones Latinae Selectae 20. (Center for Epigraphical and Palaeographical Studies, The Ohio State University)

Transcription and translation of Figure 10.3 L(ucius) Mummi(us) L(uci) F(ilius) co(n) Lucius Mummius, Son of Lucius, Consul, s(ul). Duct(u) | auspicio imperioque | under whose leadership and auspices eius Achaia capt(a). Corint(h)o | deleto Achaea was captured. When Corinth Romam redi[e]it | triumphans. Ob had been destroyed, he returned to hasce | res bene gestas quod | in bello Rome in triumph. Because these things voverat | hanc aedem et signu(m) | had been well accomplished, the vicHerculis Victoris | imperator dedicat. torious commander (imperator) dedicated this temple and statue of Hercules Victor, which he had vowed in war.

period to find an opportunity for military success.135 The destruction of a city was a particularly potent symbol for Roman audiences, and Mummius and Scipio were each awarded one of Rome’s highest honours for a victorious 135 Erskine, Roman Imperialism, pp. 39–42. W. V. Harris, War and Imperialism in Republican Rome 327–70 BC, rev. ed. (Oxford: Clarendon Press, 1985) is a key proponent of this factor.

299

tristan s. taylor

commander – a ‘triumph’ or victory parade through the city – for the destruction of Carthage, Corinth and Numantia.136 Similarly, Mummius boasted both of the capture of Achaea (Achaia Capta) and his destruction of Corinth (Corinto deleto) when he dedicated a temple to Hercules Victor after his return (Figure 10.3).137 For the elite, military success and the profit from plundering a city were intimately connected – part of the wealth obtained would be displayed in the triumph, and would help fund the leaders’ future political careers. Roman sources readily imagine the quest for wealth and glory as also fuelling violence. Thus Appian speculates that perhaps one reason Scipio destroyed Carthage and Numantia was his belief that great calamities were the foundation of glory.138 Wealth and glory are also how Appian explains massacres perpetrated in Spain in the 150s and 140s by the governors L. Licinius Lucullus and Ser. Sulpicius Galba, and the unauthorised Spanish campaign of the governor M. Aemilius Lepidus Porcina.139 According to Appian, Lucullus, spurred by his desire for fame and money, attacked the Vaccaei without provocation or senatorial authorisation, when peace had been made with the Arevaci by his predecessor that otherwise denied him the opportunity for conquest, and also massacred the Vaccaean Caucaei despite their deditio.140 Similarly, Galba divided a group of Lusitanians who were seeking peace into three groups, marched them separately into the countryside and systematically massacred each group.141 During another hiatus of fighting with the Numantines, Porcina again attacked the Vaccaei – according to Appian, in a quest for glory and a triumph rather than for Rome’s advantage – on the spurious pretext that they had aided the Numantines.142 While the conquests of Scipio and Mummius resulted in a triumph, the actions of these commanders drew criticism. Thus, Appian condemns Lucullus as bringing infamy upon the Roman name and Scipio restored the survivors of the Caucaei to their homes on the basis of Lucullus’ violated oath.143 Appian also criticises Galba as having avenged treachery with treachery. An unsuccessful attempt was

136 On celebration, see Colwill, ‘Responses’, 142–4. On the triumph, see e.g. M. Beard, The Roman Triumph (Cambridge, MA: Harvard University Press, 2009) and F. J. Vervaet, The Roman Republican Triumph: Beyond the Spectacle (Rome: Edizioni Quasar, 2014). 137 Inscriptiones latinae selectae 20. 138 App. Hisp. 98. 139 App. Hisp. 52, 80–3, 89. 140 App. Hisp. 52; Marco Simón, ‘Insurgency’, 232–3. 141 App. Hisp. 60; Marco Simón, ‘Insurgency’, 233–4. 142 App. Hisp. 80. 143 App. Hisp. 52, 89.

300

A Tale of Three Cities: Roman Destruction

made to prosecute Galba for violating his oath. Finally, the Senate tried to stop Lepidus’ campaign and fined him on his return.144

Reactions and Ethical Constraints on Mass Violence The contrast between the contemporary celebrations of the urbicides perpetrated by Mummius and Scipio and the critiques of Lucullus and Galba shows that mass violence could be celebrated in the Roman world, but only in the correct ethical context. Indeed, the Romans appear to have been intensely concerned with justifying their actions.145 This is clear in relation to Carthage, where Polybius stresses how the Roman desire to destroy Carthage was tempered by the wish for a just pretext.146 The justification most often used in Roman sources is ‘defence’ of themselves or their allies, which has led some to argue that Rome followed a policy of ‘defensive imperialism’, only engaging in conflict for defence, and thus they were almost ‘accidental’ imperialists.147 This perspective must be tempered by the aforementioned importance Romans attached to military glory and an acknowledgement that Rome opportunistically forged alliances with peoples already under threat, which thus made war, if not inevitable then highly likely.148 People were not always convinced by Rome’s pretexts. This is clear in relation to Carthage. Polybius surveys four broad Greek viewpoints on Carthage’s destruction. First, that Rome had acted appropriately as Carthage was a perpetual threat to Roman domination.149 Second, that the Romans had departed from the principles by which they had gained their empire and destroyed Carthage out of a lust for power despite Carthage’s yielding to all Rome’s demands.150 Third, Rome had acted with impiety and treachery towards the Carthaginians.151 And finally, that the Romans had acted appropriately due to Carthage’s disobedience of Rome’s authority.152 The themes here are clear, as is the overlap with patterns in Roman thought – the destruction of a city is acceptable where there is a threat, or disobedience of authority, but not where there is treachery or a people has surrendered and obeyed all terms. The destruction of Carthage was later regretted by some Romans – such as the historian Sallust writing in the first century B C E – not for the destruction 144 App. Hisp. 60, 80–3. 145 For example Burton, Friendship, pp. 314–15. 146 Polyb. 36.2.1. 147 Summarised in Erskine, Roman Imperialism, p. 36. E.g. Cic. Rep. 3.35, ‘our people has power over the whole world through defending our allies’. 148 See P. Brunt, ‘Laus imperii’ in Roman Imperial Themes (Oxford: Clarendon Press, 1990), pp. 288–323, at p. 307. 149 Polyb. 36.9.3–4. 150 Polyb. 36.9.5–8. 151 Polyb. 36.9.9–12. 152 Polyb. 36.9.13–17.

301

tristan s. taylor

itself, but as marking a negative turning point in terms of Roman morality: the removal of Rome’s rival ushered in moral decline.153 Scipio’s destruction of Numantia also appears to have drawn criticism for its severity.154 The greatest Roman regret, however, was Corinth’s destruction, with the sight of the city’s ruins evoking strong reactions.155 Thus, Cicero says the Romans did wrong (peccatur) in destroying Corinth under the ‘appearance of expedience’ (utilitatis specie).156 Writing later, Florus terms it an ‘unworthy crime’ (facinus indignum).157 These assessments reflect ideas that the destruction of a leading Hellenic city, and the ‘uncomprehending barbarism’ of Mummius’ troops in looting the artwork, were acts of cultural vandalism.158 We can recall here Cicero’s contrast between the justified elimination of the savage (immanis) Carthaginians and Numantines, and the destruction of Corinth discussed above.159

Religious Context The Roman concern with the preservation of a reputation for good faith or fides and with perceptions that their actions were justified reflect the Roman belief that their success stemmed from, in part, divine support arising from Roman piety and virtue.160 This sentiment is reflected in Mummius’ dedication of a temple to Hercules Victor for his victory (see Figure 10.3). Following a typical Roman practice, he had vowed on campaign to build the temple if successful – the success of the campaign then reflected the god’s favour. A similar thought pattern underlay the use of religious ritual in the destruction of a city. First, by the ritual of euocatio, the gods were summoned from the city. The city then underwent ritual deuotio, whereby the commanding Roman general called upon the infernal gods to ‘consider those cities and fields and the people’s lives and lifetimes cursed and execrated according to those laws under which enemies have at any time been cursed’.161 These rituals occurred at Carthage and Corinth, and perhaps Numantia.162 The 153 Sallust, Bellum Catilinae (Catilinarian Conspiracy) 10; Sallust, Bellum Iugurthinum (Jugurthine War), 41; Sallust, Historiae 1.10 and 1.12. 154 Astin, Scipio, p. 154. 155 On the extent of the destruction, Gebhard and Dickie, ‘View from the isthmus’, 263–4. 156 Cic. Off. 3.46. 157 Flor. 1.32.1. 158 Colwill, ‘Reponses’, 142; Gebhard and Dickie, ‘View from the isthmus’, 263; e.g. Strabo 8.6.23 referring to Polybius as eyewitness. 159 Cic. Off. 1.35; Astin, Scipio, pp. 153–4. 160 See further Brunt, ‘Laus imperii’, 294–7. 161 Macrob. Sat. 3.9.7–13. See on vows euocatio and deuotio, M. Dillon, ‘Evocatio: taking gods away from enemy states and peoples’ in Religion and Classical Warfare: The Roman Republic, ed. M. Dillon and C. Matthew (Barnsley, UK: Pen & Sword, 2020), pp. 53–103. 162 Macrobius refers to ‘towns of our Spanish enemies’ (oppidaque hostium . . . Hispanorum) (Sat. 3.9.13).

302

A Tale of Three Cities: Roman Destruction

destruction of these cities following such rituals gave the appearance of divine sanction for Rome’s actions.163 Despite this conviction of divine support for Rome’s authority, and occasional hostility to foreign religions – in particular the (in)famous suppression of the imported Bacchic cult in 186 B C E in Italy – Rome in this period did not try to eliminate those who worshipped gods other than the traditional Roman gods.164

Genocide? While all three acts can readily be identified as ‘urbicide’, were they ‘genocide’, that is, intentional actions to destroy, in whole or part, a racial, ethnical, national or religious group ‘as such’? An initial question is the nature of the victim group. In particular, as all three cities were representatives of larger ethnic groups – Punic for the Carthaginians; Greeks, Dorians or Achaeans for the Corinthians; and Arevaci or Celtiberians for the Numantines165 – does the fact that other members of these groups and the cultures as a whole were not targeted166 mean that these events were not genocide? The only further actions taken were against those who had supported the main protagonist, such as the destruction of the cities of Carthage’s Punic allies, and the demolition of the walls of those cities that had supported the Achaean League’s revolt.167 However, the city-state itself was a highly developed form of state identity in the ancient Mediterranean,168 and performed similar social, economic and political roles to a modern ‘nation-state’ – it was analogous to a ‘national group’.169 Certainly, the Romans conceived of the Carthaginians and Corinthians as distinct groups, thus Cato’s emphasis on Carthaginian perfidy and cruelty and Mummius’ attempt to enslave Corinthians, rather than others.170 Different considerations perhaps apply to Numantia, which is often spoken of as a city or fortified town of the larger Arevaci group, rather than as a separate city-state,171 reflecting the fact that the Celtiberians were a less 163 Miles, Carthage, p. 351. 164 The bibliography on the suppression of the Bacchic cult is vast; see e.g. R. A. Bauman, ‘The suppression of the Bacchanals: five questions’, Historia: Zeitschrift für alte Geschichte 39 (1990), 334–48. 165 Van Wees, ‘Ancient world’, 244; on Numantia, see Strabo 3.4.13. 166 Barrandon, Massacres, p. 328. 167 App. Pun. 135; Paus. 7.16.7. We know little of the remainder’s fate: Gebhard and Dickie, ‘View from the isthmus’, 264. 168 Purcell, ‘Carthage and Corinth’, 133. 169 Van Wees, ‘Ancient world’, 243; Taylor, ‘Introduction’, pp. 15–16. 170 Rhet. ad her. 4.14.20; Zonar. 9.31. 171 For example App. Hisp. 45–6; Strabo 3.4.13.

303

tristan s. taylor

urbanised people than the Greeks or the Carthaginians.172 Yet the matter is complex: Appian speaks of war against the ‘Numantines’, and Scipio received the unofficial honorific agnomen ‘Numantinus’ for his conquest,173 suggesting the Romans also conceived of them as a distinct group. Even if the Arevaci fall short of a distinctive ‘national’ group, the Convention definition of genocide does not require the intentional destruction of an entire group, but only a significant ‘part’ of that group.174 In this regard, Numantia was the ‘most notable’ (ὀνομαστοτάτη) Arevacan town,175 whose destruction could readily be regarded as destruction of a significant part of the Arevaci. Alternatively, their targeting could be regarded as a ‘genocidal massacre’:176 destruction of members of the Arevaci – or Celtiberians – to deter others from revolt. A further objection is that even if these civic communities were distinct groups, the ultimate targets here were not the people – the Carthaginians, Numantines or Corinthians – but rather the state:177 that is, not a ‘physical’ or ‘biological genocide’, but a civic death. This reflects the notion that a citystate might be regarded as its people rather than its physical fabric. Appian uses this concept in his speech for the consul Censorinus encouraging the Carthaginians to abandon their city: ‘we consider you to be Carthage, not the ground’.178 It is also reflected in the Roman practice of referring to peoples, ‘Carthaginians’, rather than places, ‘Carthage’, as subject to Roman imperium.179 While this conception may facilitate the survival of a group that has been physically displaced, it problematises separating an attack on a ‘state’ from an attack on its constituent population. In these three examples, the destruction of the physical city was accompanied by the massacre, enslavement and dispersal of the group, both eliminating the population and the physical fabric foundational to the group’s cultural practices.180 In this regard, the fact that the Romans enslaved the population of these cities that survived the siege and sack, rather than killed them, raises the question of whether the intention was to ‘destroy’ the group ‘as such’.181 Although enslavement itself was not necessarily a path to swift death in the Greco-Roman world (not all were sent to live short lives in the mines),182 the 172 Richardson, Hispaniae, p. 16. 173 App. Hisp. 76; 98. 174 See W. Schabas, ‘Law and Genocide’ in Genocide Studies, ed. Bloxham and Moses, pp. 123–41, at pp. 135–7. 175 Strabo 3.4.13. 176 Kiernan, Blood and Soil, pp. 13–14. 177 Barrandon, Massacres, p. 324. 178 App. Pun. 89. 179 Erskine, Roman Imperialism, p. 7. 180 Kiernan, ‘First genocide’, 28; Quesada-Sanz, ‘Genocide’, 12. 181 Jones, Genocide, p. 28. 182 Levene, Meaning of Genocide, pp. 146–7.

304

A Tale of Three Cities: Roman Destruction

impact of mass enslavement and dispersal of the population was as consequential as massacre for a group’s elimination. Most would be dispersed to separate unknown fates in unknown destinations, unable to continue the group of which they were a part through both the prevention of endogamous biological reproduction and their inability to perform the various cultural acts constitutive of the group’s collective identity.183 In considering the Convention definition of genocide, mass enslavement could thus be considered as an act that imposed conditions of life calculated to bring an end to the group.184 In relation to children, enslavement also involves their forcible transfer between groups, and indeed as the entire surviving group was dispersed, endogamous biological reproduction was prevented, acts also listed in the Convention definition.185 It cannot be doubted that the Romans were aware that mass enslavement would have such consequences. Thus, Polybius claims that the Carthaginians were ‘totally destroyed’ (ἄρδην ἀφανισθέντες), and that none were left to mourn the city’s passing (in contrast to what happened in Greece).186 There is an element of hyperbole here for Polybius to emphasise his point (that Greece suffered worse than Carthage). Certainly, a Carthaginian diaspora must have survived for the exiled Carthaginian philosopher Clitomachus to write a consolation to them.187 Yet, it is also true that Polybius envisioned that this description of the Carthaginians’ fate could be received as appropriate, even though he and his readership knew that the surviving Carthaginians had been enslaved, not exterminated. Importantly, these acts of mass enslavement and the intentions they evince must be understood in the context of the wider Roman actions towards each community. While violent sacking of cities was a common accompaniment to siege warfare, as is the deliberate infliction of starvation during a prolonged siege, the conscious determination that that community would never rise again was less common.188 In at least the case of Carthage and Corinth, their gods were summoned from the city, and in all three cases here a conscious decision was made that the civic community would end. In the 183 Van Wees, ‘Ancient world’, 245; Taylor, ‘Introduction’, 16–17. 184 H. van Wees, ‘Genocide in archaic and classical Greece’ in Our Ancient Wars: Rethinking War through the Classics, ed. V. Caston and S. Weineck (Ann Arbor: University of Michigan Press, 2016), pp. 19–37, at p. 21. 185 Colwill, ‘Responses’, 138. 186 Polyb. 38.1.6–9. 187 See Cic. Tusc. 3.54; Colwill, ‘Responses’, 137–8. Cicero also claims he saw Corinthians in his youth at Corinth – although whether these were descendants of the polis’ inhabitants is uncertain: Cic. Tusc. 3.53 with Gebhard and Dickie, ‘View from the isthmus’, 263–4. 188 Purcell, ‘Carthage and Corinth’, 141.

305

tristan s. taylor

case of Carthage, this was further accompanied by the consecration of the ground, along with the use of the plough to ‘undo’ the city’s foundation.189 Thus, the massacre in the sack of Carthage and Corinth, and death through depravation deliberately inflicted in the prolonged sieges of Carthage and Numantia, in combination with the mass enslavement and dispersal of the surviving populations and destruction of the urban fabric of each community, are representative of an intent to destroy the Carthaginians, Corinthians and Numantines ‘as such’.190 A final question is that of the relative roles of motivation and intention.191 As is clear from the foregoing, these three urbicides were driven by a cluster of motivations including fear, and retributive ‘conspicuous destruction’ to ensure the maintenance of the Roman imperium, and to avenge humiliation. As much as Roman stereotypes might have fuelled and justified violence, they did not represent an ideology that dictated the extermination of all maritime powers, Greeks, Punic peoples or Celtiberians. For some scholars, when compared to modern events such as the Holocaust, such ‘pragmatic’ rather than ‘ideological’ destruction lies beyond the realm of ‘genocide’.192 However, any requirement of an ideological motivation is outside the Convention genocide definition, which requires only an intent to destroy a group ‘as such’, whatever the motive.193 Here it is clear that the Romans ‘intended’ to permanently destroy these civic communities as communities: following significant loss of life through a prolonged siege (Carthage and Numantia) or massacre accompanying the cities’ sack (Carthage and Corinth), the surviving populations were enslaved, and significant parts of the urban fabric were consciously razed. These actions prevented both the respective population’s biological continuity and the performance of the social practices foundational to the community. It is only for Carthage that we have evidence of prolonged agitation for the city’s destruction, with Corinth’s elimination seeming perhaps opportunistic. Numantia’s demise, however, seems to have been predetermined, at least from the point when Scipio arrived. Scipio’s refusal of battle and initiation of siege works suggests 189 On the plough, see the third century C E jurist Modestinus: Digest 7.4.21. See also Cic. Leg. agr. 1.5, Ridley ‘Pinch of salt’, 140–6 with Stevens, ‘Carthage’, 39–41. 190 Kiernan, ‘First genocide’, 28; Quesada-Sanz, ‘Genocide’, 12. 191 See Taylor, ‘Introduction’, 4–5. 192 For example Barrandon, Massacres, pp. 332, 336 on Caesar. Purcell sees an ‘ideological’ motivation in Rome’s opposition to the ‘democratic’ leaders of Roman resistance at Carthage and Corinth (‘Carthage and Corinth’, 136–7). However, Rome favoured whoever supported them, without any particular political orientation: Kallet-Marx, Hegemony, p. 76. 193 Kiernan, Blood and Soil, pp. 18–20.

306

A Tale of Three Cities: Roman Destruction

not only an anxiety about fighting the Numantines but also that the intention was to redress past shame at defeat by the Numantines through their destruction.194 Thus, while we cannot discount the psychological impact a long siege can have on the besiegers,195 often leading to a violent release, the destruction of these communities was not the haphazard product of an army running amok,196 lacking the specific intent necessary for genocide.197 Even in Carthage, where Appian claims the carnage that resulted was the unintentional product of the passion of war, this particular ‘release’ involved six days and nights of continuous violence, with the soldiers rotated so that they would not be worn out by the toil of the slaughter and the horrid sights they were seeing.198 The commanders were thus aware of the slaughter occurring, planned for it, and certainly did nothing to stop it, however difficult that might have been.199 Finally, the cluster of motivations involved has numerous touch-points with factors that have driven modern genocides, including:200 pragmatic, retributive violence associated with imperialism; existential anxiety – internal and external – about threats posed to Rome’s imperium by a particular group, such as Carthage’s renascent power, in a context of Rome’s perceived internal crisis;201 and memories of past wrongs, such as Carthage’s ravaging of Italy, or humiliation, as at the hands of Numantia.202

Conclusion Rome intentionally destroyed the communities of Carthage, Corinth and Numantia, killing many and dissolving their surviving populations through their enslavement and demolishing the physical fabric of the community – enacting their symbolic as well as practical undoing – and preventing the continuation of the cultural practices foundational to the communities’ identities. Such action was not unprecedented, but rather part of a spectrum of violence deployed by the Romans reflecting the perceived importance of exemplary ‘conspicuous destruction’ to maintain their power. As the Romans focused on the obedience of peoples rather than on territory, it is not surprising that Rome would ignore or efface individual responsibility, Levithan, Siege Warfare, pp. 114–15. 195 See ibid., pp. 205–27. Contra Barrandon, Massacres, p. 328. 197 Van Wees, ‘Ancient world’, 243. App. Pun. 130. 199 Ziolkowski, ‘Urbs direpta’, 86–7. For example the inclusion of ‘retributive’ genocide and genocide ‘removing a threat’ in typologies by Fein and Chalk and Jonassohn: Taylor, ‘Introduction’, 4–5 and 28–9. 201 For example Levene’s genocide definition: Meaning of Genocide, p. 35. 202 For example Jones, Genocide, pp. 339–41.

194 196 198 200

307

tristan s. taylor

and instead collectively target the group involved. Such violence did not occur in an ethical vacuum, but rather within a paradigm that could celebrate retributive mass destruction, where it was perceived as justified. While contemporary or later questions were raised in relation to the justification of all three cases, it is Carthage that stands out. While both Corinth and Numantia were destroyed as retribution for revolt, resistance and humiliation, a determination had arisen at Rome among some to destroy Carthage because of its perceived threat, disproportionate perhaps to reality, with a wrong subsequently seized upon as justification. This paranoid, existential anxiety is analogous to motivations found in modern instances of mass violence. Thus, while the ideas here are removed from modern, essentialist views of race, the example of Carthage is perhaps closest to the ideological complex of paranoid fears attached to a particular group so common in modern genocides: if the determination to destroy the Carthaginians had arisen before they had wronged Rome, it would seem that they were destroyed much more for ‘who they were’ rather than ‘what they had done’.

Bibliographic Note on Sources in English The bibliography on all topics considered here is large; the examples provided here are good places to start. On Roman imperialism, see in general Erskine, Roman Imperialism, and on its manifestations in the west, Harris, ‘Roman expansion’; and in the east, Kallet-Marx, Hegemony. For Spain in particular, see Richardson, Hispaniae, and Simón, ‘Insurgency’. For connections between the destruction of Carthage and Corinth, see Purcell, ‘Carthage and Corinth’. For the destruction of Carthage as genocide, see Kiernan, ‘First genocide’ and Blood and Soil. For a contrasting view see Barrandon, Massacres. On genocide in antiquity see van Wees, ‘Ancient world’ and Quesada-Sanz, ‘Genocide’; see also the chapters in Taylor, ed., Cultural History of Genocide in the Ancient World.

308

11

Caesar’s Gallic Genocide A Case Study in Ancient Mass Violence tristan s. taylor

From the years 58 to 50 B C E a campaign of conquest in the region north of the Alps known to the Romans as Gaul was led by the ambitious Roman statesman Gaius Julius Caesar (Figure 11.1). The cost in human lives of this campaign was prodigious, with the biographer Plutarch recording in the first century C E that 1 million people were killed and 1 million enslaved – miraculously balanced numbers.1 Such extraordinary figures have unsurprisingly attracted attention in discussions of the application of the concept of ‘genocide’ to the ancient world. Indeed, Raphael Lemkin, the Polish jurist who coined the term ‘genocide’, had himself intended to write a chapter on ‘Genocide in Gaul’ in his own unfinished History of World Genocide.2 While some have been willing to use the label ‘genocide’ for Caesar’s actions,3 the matter remains controversial, both in terms of the specific case of Caesar, and the general question of whether ‘genocide’ is an appropriate concept to apply to the premodern era.4 As Nico Roymans summarises, ‘the concept (of genocide) has a specific, highly negative 1 Plutarch, Caesar (Plut. Vit. Caes.) 15. 2 R. Lemkin, Lemkin on Genocide, ed. S. L. Jacobs (Lanham, MD: Lexington Books, 2012), pp. 17–18. 3 For example A. Riggsby, Caesar in Gaul and Rome: War in Words (Austin: University of Texas Press, 2006), p. 178. See also B. Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007), pp. 58– 9; N. Roymans, ‘A Roman massacre in the far north: Caesar’s annihilation of the Tencteri and Usipetes in the Dutch river area’ in Conflict Archaeology: Materialities of Collective Violence from Pre-History to Late Antiquity, ed. M. Fernández-Götz and N. Roymans (London and New York: Routledge, 2018), pp. 167–81, at p. 180. 4 Against its use, see N. Barrandon, Les massacres de la république romaine (Paris: Fayard, 2018), pp. 329–37; for its application, see T. S. Taylor, ‘Introduction’ to A Cultural History of Genocide in the Ancient World, ed. T. S. Taylor (London: Bloomsbury, 2021), pp. 1–30. H. van Wees, ‘Genocide in the ancient world’ in The Oxford Handbook of Genocide Studies, ed. D. Bloxham and A. D. Moses (Oxford University Press, 2010), pp. 239–58; on the question of anachronism, see B. Kiernan, ‘Is “genocide” an anachronistic concept for the study of early modern mass killing?’, History 99:336 (2014), 530–48.

309

tristan s. taylor

Figure 11.1 Bust of Julius Caesar. Posthumous, but thought to be first century B C E. Museo Pio-Clementino, Musei Vaticani. (Photograph from Getty Images)

meaning in our modern society, which is why some scholars have argued that it is anachronistic when applied to mass killings in pre-modern periods’.5 Although, as we shall see, ancient mass violence did not occur in a moral vacuum, nonetheless it is clear that the intentional destruction of a people could, in certain circumstances, be regarded as not only morally acceptable but even praiseworthy. Indeed, the question of the application of the concept of genocide to antiquity in general, and to Caesar’s conquest of Gaul in particular, is not only one of morality but also one of definition. The 1948 Genocide Convention defines genocide as various acts carried out with an intent to bring about the destruction ‘in whole or in part, of a national, ethnical, racial or religious group, as such’. This definition only partially reflects Lemkin’s broader conception of a systematic attack against a group’s foundations, including its cultural bases, in addition to mass killing. Dissatisfaction with the Convention definition, including with its 5 Roymans, ‘Roman massacre far north’, 179.

310

Caesar’s Gallic Genocide: A Case Study

exclusion of various social groups, has led to a plethora of alternative scholarly definitions of genocides as well as the invention of other ‘-cides’.6 These alternative definitions generally take the Convention as their starting point, and either expand or contract the categories of perpetrators, the acts involved (narrow definitions focus only on mass killing), the groups targeted and the motivations involved.7 Implicit in many of these definitions, and in popular understandings of genocide, is the paradigmatic impact, or ‘conceptual constraint’, of the Holocaust.8 The impact of this is reflected in rejections of the application of genocide to antiquity, including to Caesar’s conquests, on the basis that, for example, ‘genocide’ is killing conducted to ‘purify’ a community, or to implement an ‘ideology’, neither of which are applicable to Caesar’s campaigns.9 Such an understanding of genocide imports the notion that mass killing must be conducted because of a particular motive, an idea absent from the genocide Convention, where what matters is an intention to destroy a group ‘as such’.10 In keeping with the approach of The Cambridge World History of Genocide, which takes the Convention definition as a starting point, and with the need for consistent application of concepts across time in order to facilitate comparative research,11 we will consider here whether any of Caesar’s actions involved an intent to destroy, as such, the groups he massacred or enslaved in his conquest. In addition, bearing in mind the various ways in which the very concept of genocide is contested, we will also consider how Caesar’s actions might map on to alternative definitions of genocide in terms of the perpetrators, actions, targets and motivations involved. We will further consider whether, rather than ‘genocide’, some of the actions involved might be considered ‘genocidal massacres’, as defined by the pioneering scholar of Genocide Studies Leo Kuper: shorter, more limited episodes of mass killing directed towards a specific local or regional community on account of their membership of a larger group, often to serve as an object lesson to that group.12

6 A. Jones, Genocide: A Comprehensive Introduction, 3rd ed. (London: Routledge, 2017), pp. 22–37. 7 Jones, Genocide, p. 28. 8 D. Moshman, ‘Conceptual constraints on thinking about genocide’, Journal of Genocide Research 3:3 (2001), 431–50. 9 Barrandon, Massacres, pp. 322, 336. 10 Kiernan, Blood and Soil, p. 18. 11 Ibid., p. 12. 12 L. Kuper, Genocide: Its Political Use in the Twentieth Century (New Haven: Yale University Press, 1981), pp. 59 ff.; Kiernan, Blood and Soil, pp. 13–14.

311

tristan s. taylor

A First-Person View Caesar’s conquest of Gaul makes for a particularly interesting case study of mass violence in antiquity, not least because we possess one of the few eyewitness accounts of such a campaign in Caesar’s own commentaries, the Bellum Gallicum (The Gallic War). As such, we have a rare window into the mind of a perpetrator of mass violence that can perhaps enable us to answer the question of his intentions – a key plank of the modern legal definition, and implicit in most scholarly definitions, yet in most cases notoriously difficult to identify. When and how Caesar describes mass violence can also tell us much about contemporaneous Roman attitudes towards it. Caesar was an astute Roman politician, and certainly one of the most ambitious. Therefore, we can assume that where Caesar describes mass violence, it is in a context that he considered would be regarded as politically positive. As Andrew Riggsby states of the Bellum Gallicum, ‘nearly all scholars . . . would agree that the work was designed to make its author look good to his audience’.13 This of course complicates the extent to which Caesar’s Commentaries can be used to recover the ‘truth’ about events in the Gallic conquest, as opposed to an ancient form of ‘spin’. A further complication in our recovering what actually happened is Caesar’s ‘bare’ style that was already noted in antiquity.14 For example, Cicero praises the Commentarii as nudi (bare), devoid of unnecessary ornamentation.15 We might therefore at times expect brief notices of events rather than detailed elaboration, with the audience expected to fill out the details. Thus, as the first-century C E teacher of rhetoric Quintilian states, the mere mention that a town has been captured would connote for the audience scenes of killing, enslavement, destruction and pillage.16 Therefore, a brief notice 13 Riggsby, Caesar in Gaul and Rome, p. 191. On the potential audience, see T. P. Wiseman, ‘The publication of de Bello Gallico’ in Julius Caesar as Artful Reporter: The War Commentaries as Political Instruments, ed. K. Welch and A. Powell (London: Duckworth, 1998), pp. 1–9, at pp. 2–4. On Caesar’s positive portrayal of his actions, see the other chapters in Caesar as Artful Reporter, ed. Welch and Powell, and also J. H. Collins, ‘Caesar as a political propagandist’, Aufstieg und Niedergang der römischen Welt 1:1 (1972), 922–66, and M. Rambaud, L’art de la déformation historique dans les Commentaires de César (Paris: Les Belles Lettres, 1966). 14 See van Wees, ‘Ancient world’, 240, on the problem of brevity in general in ancient sources. 15 Cicero, Brutus (Cic. Brut.) 262. At one stage this was thought of as a generic feature of commentarii: they were to be the bare material for future histories, but see now Riggsby, Caesar in Gaul and Rome, pp. 145–50 and the references therein; contra Wiseman, ‘Publication of de Bello Gallico’, 6. 16 Quintilian, Institutio ortatoria (Institutes of Oratory) (Quint. Inst.) 8.3.67–9.

312

Caesar’s Gallic Genocide: A Case Study

that Caesar had taken a Gallic town or oppidum might be expected to indicate to the Roman reader instances of mass violence such as those adumbrated by Quintilian. Indeed, Caesar himself places in the mouth of the Gallic leader Vercingetorix the statement that the usual fate following the sack of a city was slavery for the women and children, and death for the men.17 Such a ‘gendercide’, the killing of all persons of one gender, could in the appropriate context ensure the destruction of a group as a group, which is the essence of genocide.18 The frequency of this topos cannot be denied, nor can it be denied that many suffered this fate following the sack of cities in the ancient world.19 However, if it were the usual consequence, or necessary fate, then it is surprising that the evidence, such as it is, suggests that the slave population of Rome contained significantly more men than women.20 If the usual fate of conquest was slavery for women and children, and death for the men, one might expect the gender imbalance to be less. This suggests that either episodes of massacre involving men following the sacking of a city, the mass enslavement of women and children – or both – were less common than the topos of the captured city suggests. A further evidentiary challenge presented here is that a historical fact (that a city was taken) is vulnerable to literary elaboration in line with features of the classical topos. As a consequence, we should not confuse more elaborate detail with the ‘truth’. As Quintilian makes clear, when elaborating on events classical writers could include not only what actually happened but also fictitious events of the kind that may also be expected to occur.21 Neither political spin, nor brevity, nor the possibility of fictional embellishment, however, renders the study of Caesar’s text in relation to mass violence without merit. An examination of how mass violence is represented can tell us much about attitudes towards it in a particular culture. Similarly, 17 Caesar, Gallic War (Caes. BGall.) 7.14. 18 Van Wees, ‘Ancient world’, 244–5. In the modern context, the massacre of a significant proportion of the male population of Srebrenica by Bosnian Serb forces in the war in Bosnia that followed the dissolution of Yugoslavia was considered genocide by the International Criminal Tribunal for the Former Yugoslavia: Kristic ́ (Appeals Chamber – 19 April 2004) ICTY-89–33-A [28]. 19 See further G. M. Paul, ‘Urbs capta: sketch of an ancient literary motif’, Phoenix 36 (1982), 144–55. 20 W. V. Harris, ‘Demography, geography and the sources of Roman slaves’, Journal of Roman Studies 89 (1999), 62–75, at 65. But cf. W. Scheidel, ‘The Roman slave trade’ in The Cambridge World History of Slavery, ed. K. Bradley and P. Cartledge (Cambridge University Press, 2011), I, pp. 287–310, at p. 307. 21 Quint. Inst. 8.3.70: ‘et licebit etiam falso adfingere quidquid fieri solet’ (‘and it will be permitted to append also what is false, but might be customary to occur’).

313

tristan s. taylor

Tribes and cities in Gaul at the time of Julius Caesar

Tencteri

Britain Morini

En

s gli

h h C

a

e nn

l

Venelli

Usipetes Menapii Atrebates Ubii Nervii Aduatuci Suebi Belgae Eburones

Bellovaci Remi Suessiones Paris

Treveri

Parisii

Osimii Veneti

Lingones Alesia

Nammetes

Carnutes Serones

Aedui

Pictones

Bituriges

Sequani Vesontio

Bribcte

Helvetii

Avaricum

Santones

Gergovia

Averni

Allobroges

N W S

Vocontii

Uxellodunum

E

Aquitania

Tolosa Massalia

Tectosages

Mediterranean Sea 0 0

100 50

100 km 100 mi

Map 11.1 Peoples and main settlements of Gaul at the time of Julius Caesar. (© Citypeek/ Wikimedia commons)

where Caesar chooses to use elaboration, rather than brevity, one can assume that this has been done for reasons of emphasis that can be revealing.

Mass Killings Let us now turn to Caesar’s narration of his conquest of Gaul in the fifties B C E (Map 11.1). Caesar describes four instances of violence that exceeded the infliction of military defeat on his opponents and instead involves largescale killing of non-combatants: the execution of the senate of the Veneti, the massacre of the Usipetes and Tencteri, the intentional destruction of the Eburones and the massacre at Avaricum. It is possible, if not probable, that

314

Caesar’s Gallic Genocide: A Case Study

more violence occurred than is narrated. Thus Caesar reports of the Helvetii that, following their conflict with him, only 110,000 remained from their initial population of 368,000, despite his offering no explicit narrative of any massacre of non-combatants.22 Leaving aside for the moment the execution of the senate of the Veneti, discussed below in relation to its accompanying mass enslavement, the first massacre we meet in the Bellum Gallicum is of the migrating Usipetes and Tencteri in 55 B C E. These two German tribes had left their homeland under pressure from the powerful Germanic tribe the Suebi, and crossed the Rhine into Gaul, where they had seized some villages and the supplies of the Gallic Menapi. As Caesar presents it, he was concerned about the disruption caused by this dislocation, and decided to make war upon them and to reject a request from the two tribes either to be granted lands or to hold those they had seized from the Menapi.23 In Caesar’s version, after some inconclusive negotiation, the German cavalry attacked the Roman cavalry, killing seventy-four of them, supposedly in a time of truce. The next day, Caesar apprehended the German delegation, comprising their leaders, and then launched an attack on the unsuspecting German camp.24 After some fighting, the two tribes retreated and were pursued by Caesar’s men to a river. There, those who did not jump into the river were massacred: Our soldiers, aroused (incitati) by the treachery (perfidia) of the previous day, burst into the camp. In there, those who could take up arms quickly resisted us for a while . . . But the remaining throng of children and women (for they had left their home and crossed the Rhine with all their belongings) began to flee in all directions. Caesar sent the cavalry to pursue them. The Germans, when they heard the noise behind them and saw their own being killed, after throwing down their arms and leaving behind their standards, burst out from the camp and, when they had reached the confluence of the Moselle and Rhine, with hope of escape abandoned and many of them already killed, the remainder threw themselves headlong into the river and there, worn down by fear, fatigue and the force of the river, perished.25

The survivors escaped to the territory of the Sugambri and the two tribes appear later in Roman history.26 Caesar gives the numbers present in the tribes before the massacre as over 400,000, and Plutarch claims that number perished.27 It 22 Caes. BGall. 1.29.1–3. 23 Caes. BGall. 4.1–8. 24 Caes. BGall. 4.9–13. 25 Caes. BGall. 4.14–15. 26 Caes. BGall. 4.16. Later survival, e.g. Tac. Ann. 1.51 in 14 C E: Roymans, ‘Roman massacre far north’, 169n9. 27 M. M. Sage, Roman Conquests: Gaul (Barnsley, UK: Pen & Sword, 2011), p. 75.

315

tristan s. taylor

seems almost certain that this number, the largest given by Caesar anywhere in the Bellum Gallicum, is inflated significantly, with a more likely figure – based on the size of the military forces that Caesar reports – being a population of 140,000.28 Not all perished, but probably most did. Caesar stresses two points in his narrative here as motivating factors. The first we might term an imperial concern: the threat to stability in Gaul posed by the invading Germans and their impact on the ‘fickle’ Gauls.29 A powerful show of force was thought necessary to deter other Germans from crossing the Rhine and causing similar instability.30 The second point is the emphasis on the treachery (perfidia) of the Germans, both for Caesar, who alleges the Germans started the fighting after seeking peace ‘through deceit and tricks’ (per dolum atque insidias),31 and in particular, Caesar claims that his soldiers, as they invaded the camp harbouring the women and children, had been ‘inflamed’ (incitati) by this treachery of the Germans.32 Of the massacres that Caesar reports, it is for this one alone that we have some information of the Roman reaction. On the one hand, a supplicatio (thanksgiving) was voted for him at Rome. On the other, Caesar’s bête noire, Cato the younger, demanded that Caesar be handed over to the Germans to expatiate the blood guilt that Cato claimed stemmed from Caesar’s breaches of good faith (fides).33 These breaches possibly entailed allegations of mistreating the ambassadors of the Germans and/or attacking during a truce, or negotiations for such a truce.34 All this is notable for two reasons. First, that such mass killing could justify public celebration in Rome. Second, that when mass killing was objected to, it was not on the basis of the killing per se, but in light of the surrounding circumstances: that the Romans had attacked in breach of good faith, the very claim used by Caesar about the Germans to justify his attack on them.35 We have here a Roman dispute about the legality of how the fighting began, not how it ended. Should we see this massacre as a genocide or as a genocidal massacre? Certainly, Caesar expressed no intention to destroy these peoples – although, as we shall see, he was not reluctant to express such an objective. However, it is clear he must have known that his assault would result in the near 28 30 33 34 35

Roymans, ‘Roman massacre far north’, 169. 29 For example Caes. BGall. 4.5–6. Sage, Roman Conquests: Gaul, p. 57. 31 Caes. BGall. 4.13. 32 Caes. BGall. 4.14. See Plut. Caes. 22.3; Plut. Cat. Min. 51.1. K. Morrell, ‘Cato, Caesar and the Germani’, Antichthon 49 (2015), 73–93, at 76. See A. Powell, ‘Julius Caesar and the presentation of massacre’ in Julius Caesar as Artful Reporter, ed. Welch and Powell, pp. 111–37, who argues that Caesar is unusually defensive in his narrative here, perhaps in response to Cato: 127; see also Morrell, ‘Cato, Caesar and the Germani’, 83–90.

316

Caesar’s Gallic Genocide: A Case Study

annihilation of the people, attacking directly the poorly prepared camp, in full awareness that these peoples had brought their women and children with them. Caesar presents the anger of his men as explanatory of their violence, but does not indicate that he wishes to check it. When the women and children flee, he does not allow escape but dispatches the cavalry in pursuit. This is directly followed by the statement that the Germans ‘saw their own killed’ (suos interfici uiderent),36 which suggests the massacre of those fleeing, not their capture or enslavement. Although Caesar elsewhere records enslavement during his campaign, he does not do so here. Archaeological evidence tentatively connected with this massacre also includes women and children as victims.37 It could be argued that Caesar’s demand that the Sugambri surrender those of the Usipetes and Tencteri cavalry who had escaped to them indicates a desire to ensure as complete a destruction as possible of these people,38 although he did release those leaders whom he had taken prisoner.39 The next massacre that Caesar discusses is perhaps the most interesting from the perspective of genocide. In 54 B C E, Caesar suffered one of his most significant defeats in the Gallic campaign when a legion under the legates Sabinus and Cotta was ambushed and nearly wiped out by the tribe on the western side of the Rhine, the Germanic Eburones led by Ambiorix.40 The blow to Roman prestige created by this defeat enabled Ambiorix to persuade others to join his revolt.41 In 53 B C E, Caesar decided to punish the tribe responsible by rousing neighbouring Germanic tribes to attack the Eburones: If he [Caesar] wished to finish the business and destroy the stock of the criminal men (stirpemque hominum sceleratorum), he must send many bands and divide the soldiers . . . Caesar sent men to the neighbouring tribes and invited them all to him with the hope of booty to pillage the Eburones, so that he might endanger the lives of Gauls rather than legionaries in the forest, and at the same time, by surrounding them with a great multitude, the stock and name of the tribe might be destroyed (stirps ac nomen civitatis) for so great a crime (facinus) [supporting the rebel Ambiorix].42

This passage is important for at least three reasons. First, it is clear that Caesar felt that he could explicitly state an intention to annihilate a people, just as in relation to the massacre of the Usipetes and Tencteri he felt he could describe 36 Caes. BGall. 4.15. 37 Roymans, ‘Roman massacre far north’, 179. 38 Ibid., p. 180. 39 Caes. BGall. 4.15–16. 40 Sage, Roman Conquests: Gaul, p. 88 lists this defeat and Gergovia in 52 as Caesar’s two most serious defeats. 41 Ibid., p. 89. 42 Caes. BGall. 6.34 (emphasis added).

317

tristan s. taylor

a massacre, showing clearly that he envisaged that a description of such actions could receive a favourable reception in Rome. Second, Caesar’s stated goal has much in common with what we might term ‘genocide’: his intention is to destroy the very ‘stock and name’ of the Eburones, to destroy them as a biological group. Third, this intention is couched in the language of justification – the Eburones have committed a ‘crime’ (facinus) and they are ‘criminal men’ (homines scelerati). It was thus collective punishment and, no doubt, intended to deter others in Gaul from revolting against the Romans.43 A parallel here exists with his massacre of the Usipetes and Tencteri, where Caesar places emphasis in his narrative on the ‘treachery’ (perfidia, dolus, insidia) of those two Germanic peoples. Despite Caesar’s rhetoric, the stirps ac nomen (‘stock and name’) of the Eburones were not destroyed. Hirtius, who wrote an eighth book to complete the narrative of Caesar’s campaigns, reports that Caesar himself subsequently led a campaign against the people of Ambiorix (i.e. the Eburones), employing violence against the citizens: he killed and captured them and lay waste to their homeland in order to make Ambiorix hated by his own people, ‘if fortune left any remaining’ (si quos fortuna reliquos fecisset).44 The archaeological evidence suggests that fortune in fact left few, if any, and the territory they previously inhabited suffered a severe population drop at that time.45 Although mentioned by the Augustan geographer Strabo, writing decades after Caesar’s conquest, perhaps based on old information, they disappear from history thereafter.46 The third massacre occurred when the Romans took the town of Avaricum, an oppidum or fortified town of the tribe of the Bituriges in 52 B C E , within the context of the famous Gallic revolt against Rome led by Vercingetorix. This revolt itself erupted with what we might term a genocidal massacre, when the Carnutes, a Gallic tribe, murdered all the Roman citizens in the town of Cenabum within their territory.47 That echoed the outbreak of some other revolts against Roman rule, in particular the massacre of Romans, Italians and freed slaves of Italian birth in Asia Minor that had accompanied the outbreak of the first Mithridatic War in 88 B C E.48 Inspired by the massacre at Cenabum, the Gallic noble Vercingetorix of the 43 As Sage writes, ‘an object lesson for all of the Gauls’: Roman Conquests: Gaul, p. 103. 44 Caes. BGall. 8.24–5. 45 N. Roymans and M. Fernández-Götz, ‘Caesar in Gaul: new perspectives on the archaeology of mass violence’ in TRAC 2014: Proceedings of the Twenty-fourth Annual Theoretical Roman Archaeology Conference, ed. T. Brindle et al. (Oxford: Oxbow Books, 2015), pp. 70–80, at pp. 77–8. 46 Strabo 4.3.5. 47 Caes. BGall. 7.3. 48 Appian, Mithridatic War (App. Mith.) 22.

318

Caesar’s Gallic Genocide: A Case Study

Averni raised a revolt, attracting a number of tribes to his side and threatening some others who remained loyal to Rome. As Caesar marched to the aid of one such tribe, the Boii, he first attacked and then swiftly took Vellaunodunum, home of the Senones, from whom he seized hostages before moving swiftly to Cenabum, site of the initial massacre. Here Caesar appears to have captured much of the population as they attempted to escape (see below), and plundered and burnt the town.49 Caesar then marched on the Bituriges, who had thrown their lot in with Vercingetorix. Caesar first took the town of Noviodunum, where like Vellaunodunum he took hostages after routing some of Vercingetorix’s cavalry that had come to their aid and had momentarily forestalled the town’s surrender.50 After these reverses, Vercingetorix had initiated a ‘scorched earth’ policy, burning supplies and even towns that could not resist Roman assaults, to deprive the Romans of resources.51 The town of Avaricum was spared from this policy following the entreaties of the Bituriges that the city was defensible and was also both one of the most beautiful cities of Gaul and the ‘bulwark and ornament of their state’ (praesidio et ornamento sit ciuitati).52 Caesar laid siege to this town next, suffering no small hardship as a consequence of want of supplies and harassment by Vercingetorix’s forces.53 After a siege lasting nearly four weeks, Caesar’s soldiers stormed the town, showing no mercy to those they encountered: Casting away their arms, they (the Gauls) made in a continuous rush for the most distant parts of the town; some as they pressed on each other in a narrow passage of the gates, were slain there by the troops, some as they came out from the gates were killed by the cavalry – nor was there anyone who showed enthusiasm for plunder. Thus aroused (incitati) by the slaughter of Cenabum and the labour of the siege works, they [the Romans] spared neither those worn down by age, women nor infants. Finally, from the whole number, which was around forty thousand, there were scarcely eight hundred who, roused by the first shout, had burst out of the town and safely came to Vercingetorix.54

Again, there is an emphasis on retribution – the Romans are angered by the earlier massacre of Roman traders at Cenabum (although Cenabum itself was spared this action, instead undergoing mass enslavement; see below). Caesar had earlier emphasised this point as a motivation for his troops in their maintaining the siege (the desire to avenge the death of Roman citizens on 49 Caes. BGall. 7.11. 50 Caes. BGall. 7.13. 51 Caes. BGall. 7.14–15. 53 Caes. BGall. 7.16–17. 54 Caes. BGall. 7.28.

319

52 Caes. BGall. 7.15.

tristan s. taylor

account of the ‘treachery of the Gauls’: Gallorum perfidia).55 While Caesar emphasises the will of his soldiers here, as with the massacre of the Usipetes and Tencteri – where similar language is used – the soldiers are ‘aroused’ (incitati) in that instance too by anger at treachery – he could not have been unaware at the likely consequences when the town was taken, and he shows no interest in restraining his men. While this was a mass killing on a large scale, can we see it as targeting a ‘group’ to destroy it ‘as such’ so as to be considered a genocide? In the cases of the massacres of the Usipetes and Tencteri, and the Eburones, Caesar’s targeting of the whole group is clear. This is implicit in relation to the Usipetes and Tencteri, where Caesar is aware that the whole people have crossed into Gaul, or explicit in relation to the Eburones, where Caesar expresses a desire to destroy the ‘stock and name’ of the people (ciuitas). In regard to Avaricum, the identity of the target is more complex. While one might think of the target as the inhabitants of Avaricum, this does not appear to be the focus of Caesar, or how he conceptualises the focus of his men. Unlike city-states in other parts of the Mediterranean, where an individual town can be seen as somewhat analogous to modern concepts of a ‘nation’,56 Caesar’s ethnographic description of the Gauls represents them as larger ethnic groupings or ciuitates – Averni, Aedui, Bituriges and so on, groups that we would call ‘tribes’ – rather than the inhabitants of a particular town, as emphasised in his famous opening description of Gaul and its inhabitants by nation or ciuitas.57 In a similar vein, Caesar emphasises Avaricum as one town of the larger group the Bituriges, although an important town as the bulwark and ornament of their state, as noted above.58 As such, the ‘conspicuous destruction’ of the town, the best defended of these people, would represent an object lesson to intimidate the Bituriges, to lead them to abandon their support for Vercingetorix: indeed, the Gallic leader is portrayed as expecting a mutiny to follow this loss.59 Here we might note Caesar’s progress, indeed escalation, as he moves from one people to another – targeting first Vellaunodunum of the Senones, then Cenabum of the Carnutes, then Noviodunum and Avaricum of the Bituriges, in each case taking the town and then taking hostages (Vallaunodunum and Noviodunum), enslaving (Cenabum) or massacring (Avaricum) the populace. Thus, there is a deliberate pattern of targeting strategic locations of ciuitates to intimidate those peoples. In this context, the identities of the inhabitants of Avaricum as 55 Caes. BGall. 7.17. 58 Caes. BGall. 7.15.

56 Van Wees, ‘Ancient world’, 243. 59 Caes. BGall. 7.28.

320

57 Caes. BGall. 1.1.

Caesar’s Gallic Genocide: A Case Study

Bituriges, and that town as the bulwark (praesidium) and ornament (ornamentum) of the people, are both significant:60 this was a targeted massacre of members of a group because of their membership of that group to change the behaviour of that group, though unaccompanied by an intention to destroy the Bituriges as a whole. Such a mass killing is arguably embraced by Kuper’s concept of a ‘genocidal massacre’ – a targeted killing of members of a group that lacks the intent to destroy the wider group ‘as such’.61 The perception Caesar attributes to his soldiers is different – as we have seen, they are portrayed as driven by a desire to avenge the massacre of Romans at Cenabum through the treachery of the Gauls as a people (perfidia Gallorum) – a decision that had apparently been made before the town was captured as it is given as their motivation for maintaining the siege.62 From this perspective, the identities of the Gallic peoples as different groups or nations (ciuitates) are effaced and they are all grouped together, and all to be held violently accountable for the massacre at Cenabum, even though this had not been perpetrated by the Bituriges but rather by the Carnutes and the inhabitants of Cenabum itself had already been enslaved. That is to say, the people of Avaricum were massacred, at least in part, because of their identity in the soldiers’ minds as Gauls and the collective guilt of that people, although this is not coupled with a desire to go further and destroy the wider group. From this perspective also, this seems a clear case of Kuper’s concept of a genocidal massacre.63 Importantly, all these cases of explicitly described massacres of civilians are united as examples of what Hans van Wees has termed ‘conspicuous destruction’,64 couched in the language of exemplary, collective punishment, part and parcel of a wider pattern of collective violence he has identified in the ancient world.

Mass Enslavement One striking difference between the Roman world and modernity is the presence of slavery as a universally accepted cultural institution.65 As noted above, it was considered a common fate for the defeated for the women and children to be enslaved following the massacre of the men, and it was also possible for the men to be enslaved rather than massacred. Once enslaved, 60 Caes. BGall. 7.15. 61 Kuper, Genocide, pp. 59ff. 62 Caes. BGall. 7.17. 63 Kuper, Genocide, pp. 59ff. 64 Van Wees, ‘Ancient world’, 257. 65 See the sixth century C E Roman legal textbook, the Institutes, 1.2: slavery is part of the ius gentium, rules common to all civilised peoples.

321

tristan s. taylor

a variety of fates might await, all fatal to the ongoing existence of the group as a group: those enslaved might be sold on the spot, or sold later, or dispersed among the soldiers as booty – a commodity that they themselves might sell or keep. While it is conceivable that a portion could be bought, or ransomed, within their home territory, most would be dispersed to unknown fates, in unknown destinations, unable to maintain the existence of the group of which they were a part through both the prevention of biological reproduction and their inability to perform the various collective cultural acts constitutive of the group’s collective identity. That is, enslavement could have the consequence of genocide.66 However, the absence of immediate physical destruction of the members of the group places mass enslavement outside the bounds of ‘hard’ genocide definitions that limit genocide to killing.67 Looking at the Genocide Convention, one might consider mass enslavement an act imposing conditions of life calculated to bring an end to the group ‘as such’, if one can identify the requisite intent to do so.68 In some respects, enslaving and dispersing a population from a territory could also be seen as a type of ‘ethnic cleansing’, a concept that is frequently seen as related to genocide, and part of the same ‘spectrum of violence’ directed towards groups as groups.69 In Caesar’s narrative, when we turn from instances of mass killing to mass enslavement, we find that such events occur in similar circumstances, couched in the rhetoric of justified retribution. As with mass killing, we should not assume we have a complete catalogue of events, as reporting the enslavement of captives seems to have rarely excited historians.70 As we have seen, Plutarch records a million captured in Caesar’s campaigns,71 but in Caesar’s narrative descriptions instances of mass enslavement are relatively rare. Despite that, there is clear evidence that Caesar’s campaigns did lead to a substantial flow of slaves into the Roman world. This can be traced through the ebb and flow of Roman coins found in hoards in Romania. In short, as Michael Crawford has argued, Republican denarii can be found in significant numbers in Romanian coin hoards from the sixties, 66 Van Wees, ‘Ancient world’, 245. 67 Jones, Genocide, p. 28. 68 H. van Wees, ‘Genocide in archaic and classical Greece’ in Our Ancient Wars: Rethinking War through the Classics, ed. V. Caston and S. Weineck (Ann Arbor: University of Michigan Press, 2016), pp. 19–37, at p. 21. 69 See e.g. R. Gellately and B. Kiernan, ‘The study of mass murder and genocide’ in The Specter of Genocide, ed. R. Gellately and B. Kiernan (Cambridge University Press, 2003), pp. 19–20; B. Lieberman, ‘“Ethnic cleansing” versus genocide’ in Oxford Handbook of Genocide Studies, ed. Bloxham and Moses, pp. 42–60. 70 K. Bradley, ‘On captives under the principate’, Phoenix 58 (2004), 298–318, at 302. 71 Plut. Vit. Caes. 15.

322

Caesar’s Gallic Genocide: A Case Study

dropping off in numbers during the fifties during Caesar’s campaigns and increasing again in volume in the forties, after Caesar’s Gallic wars. This suggests that the Gallic campaign produced a significant flow of slaves, reducing the need to import them from Romania, but that purchases resumed once the Gallic flow ceased.72 What does Caesar say of mass enslavement? The first case described is that of the Aduatuci in 57 B C E. This again must be placed in its historical context: the wider uprising of the Belgae against Rome. Caesar follows a pattern in suppressing the opposition of the Belgae: either with or without combat, he brings about the surrender of the various tribes of the Belgae. He then takes hostages, but does not appear to engage in further enslavement. This pattern is followed with the Suessones, Bellovaci and Nervi.73 It is only with the last tribe that Caesar reaches, the Aduatuci, that the pattern changes. Here the Aduatuci appear to surrender their arms, but have secretly held some back. They then make a sally in the night to attack the Romans. In response, Caesar assaults and takes the town, and then sells – on the spot – 53,000 Aduatuci as slaves.74 The number is clearly sizeable, the largest figure Caesar explicitly gives. Importantly, in Caesar’s narrative all the Aduatuci have gathered in this one stronghold after abandoning their other towns and strongholds.75 Thus – as portrayed – this was the mass enslavement of the Aduatuci as such, not just a portion of them, and the intention therefore could be cast as bringing about the dissolution of the Aduatuci as a group. However, in fact this was more a rhetorical display of Caesar’s success than a real dissolution of this people, as the Aduatuci appear later in Caesar’s narrative as a not insignificant force.76 Thus, in this series of events, Caesar’s usual pattern is to secure the deditio, or surrender, of his opponents. Mass enslavement is used in response to the Aduatuci’s betrayal of their surrender. It is a collective punishment, also intended as a deterrent to others – that is, to set an example. The second instance of mass enslavement is in response to the rebellion in 56 B C E of the Veneti, who – in the hope of regaining their own hostages – had detained Roman ambassadors who had come seeking corn. Caesar attributes to the Veneti an awareness of one of the cardinal principles of international ‘law’ in the ancient world: the inviolability of ambassadors.77 Caesar again lays emphasis on the breach of a previous surrender, and also concern that if 72 M. Crawford, ‘Republican Denarii in Romania: the suppression of piracy and the slave trade’, Journal of Roman Studies 67 (1977), 117–24, at 121–2. 73 Caes. BGall. 2.12 (Suessones); 2.14–15 (Bellovaci); 2.28 (Nervi). 74 Caes. BGall. 2.33. 75 Caes. BGall. 2.29. 76 For example Caes. BGall. 5.39. 77 Caes. BGall. 3.9.

323

tristan s. taylor

this particular revolt were not dealt with, others would follow suit.78 After defeating the Veneti in a sea battle, Caesar writes that all the fighting men, and even old men, had gathered in the area and had been captured. Caesar has those of the Veneti senate executed, numbering some 600, and the remainder sold off as slaves. The execution of the 600 senators is the remaining instance of mass killing of civilians described by Caesar, who here is more explicit in stating his motive than he is with his enslavement of the Aduatuci – Caesar acted ‘so that the right of ambassadors might be more readily observed in the future’ (quo diligentius in reliquum tempus a barbaris ius legatorum conseruaretur),79 that is, a collective, exemplary punishment. Riggsby has described these actions as ‘verging on the genocidal’.80 Certainly, as Caesar portrays it, all the men are captured, and enslaved, which would bring about the end of the Veneti as a distinct group, both because of the importance of men to group identity in the patriarchal societies of antiquity and in terms of the impossibility of further biological reproduction of the group.81 Nonetheless, here again, as with the Aduatuci, Caesar’s efforts were less complete than he implies, as the Veneti are also later described as contributing troops to subsequent battles in Caesar’s Gallic conquest.82 Further mass enslavement appears to have resulted from the looting of Cenabum by the Roman army. Here again, Caesar contextualises his actions in the form of justified retribution. As noted, Cenabum was the location of the first main action of the revolt to be led by Vercingetorix: the slaughter of the Roman traders in the town.83 Caesar tells us that he was able to capture nearly everyone in the town, both those men who were trying to escape by night and all inside. Caesar then says succinctly that his troops plundered and burned the town, and the booty went to the soldiers (oppidum diripit atque incendit, praedam militibus donat) – no mention is specifically made here of the fate of all the people who were captured; presumably they were part of the booty.84 Although this action is not explicitly described as retributive, given that this town’s massacre of its Romans is later adduced as justifying the Romans’ actions at Avaricum,85 it is easy to characterise this enslavement as an act of retribution from the surrounding narrative context. However, unlike with the 78 Caes. BGall. 3.10. 79 Caes. BGall. 3.16. 80 Sage, Roman Conquests: Gaul, p. 68 describes this terror as a ‘necessary instrument of policy’. Riggsby, Caesar in Gaul and in Rome, p. 178; B. Levick, ‘The Veneti revisited: CE Stevens and the tradition on Caesar the propagandist’ in Julius Caesar as Artful Reporter, ed. Welch and Powell, pp. 61–83, at p. 70, describes the reported sale of the tribe as ‘implausible’. 81 Van Wees, ‘Archaic and classical Greece’, 25. 82 Caes. BGall. 7.34. 83 Caes. BGall. 7.3. 84 Caes. BGall. 7.33. 85 Caes. BGall. 7.16.

324

Caesar’s Gallic Genocide: A Case Study

Aduatuci and Veneti, Caesar’s narrative does not imply that all or even most of the people who lived at Cenabum – the Carnutes – were present. Further, as with the other examples of mass enslavement, Caesar’s actions here were perhaps less comprehensive than he describes: a year later Cenabum still possessed both sufficient infrastructure and victuals to support two legions.86 The last reported incident follows the capture of Alesia at the conclusion of Vercingetorix’s unsuccessful revolt. After Vercingetorix’s surrender, Caesar reports that he distributed as plunder one of the captives from Alesia to each of his men, with the exception of those previously captured from the Aedui and Averni.87 It is unclear how many people were involved here. Caesar states 80,000 men were in Alesia and that he restored some 20,000 to the Averni and Aedui, which means around 60,000 must have either escaped or been killed or enslaved.88 As this force represented a collective body of different tribes, the targeting of a particular people, or peoples, ‘as such’ is difficult to identify. Although Caesar is not explicit here, it is perhaps not surprising that grave consequences should follow the most serious threat to the Roman conquest. Thus, this episode can, like the others, be categorised as an example of retribution and exemplarity, and performed for the same reason: to deter further revolt.

A Spectrum of Violence The above discussion suggests that mass enslavement and mass killing directed at groups are related phenomena: both form part of the ‘spectrum of violence’ utilised by the Romans in the exercise of their imperialism and both are forms of retribution and exemplary deterrence. Such actions appear in Caesar’s narrative as relatively exceptional events.89 The surrounding context and justificatory language used in all the cases of massacres and instances of mass enslavement are identical: collective punishment.90 We can find contemporary confirmation of this ideology of collective retribution and deterrence in Roman imperialism in the writings of Cicero, who speaks of vengeance as justification for a ‘just war’ . Cicero uses a similar language of deterrence through violence in discussing his motivation for his own campaigns during his time as governor in Cilicia.91 86 Caes. BGall. 8.6. 87 Caes. BGall. 7.89. 88 Caes. BGall. 7.71 and 7.90. 89 See also Lieberman, ‘“Ethnic cleansing” versus genocide’, 45, on ethnic cleansing and genocide as part of the same spectrum of violence. 90 See also Riggsby, Caesar in Gaul and Rome, pp. 188–9: Caesar is in these instances responding to bad faith. 91 See the fragment of Cicero’s De republica quoted in Isidorus, Eytmologiae, 17.1.3. On Cicero in Cilicia, see Epistulae ad familiares (Letters to His Friends) 15.4.10, and van

325

tristan s. taylor

It is unclear whether any firm patterns emerge as to when massacre, rather than enslavement, was utilised. Hans van Wees has argued that the prospect of profiting from conquest commercially through mass enslavement always militated against massacres.92 In three of the cases where Caesar reports mass enslavement – the Aduatuci, Veneti and Cenabum – these peoples appear later in Caesar’s narrative (in the case of Cenabum, just one year later), and as peoples of some power and numbers not long after their supposed mass enslavement. This suggests that Caesar exaggerated his success in taking captives and/or that a significant number of those enslaved escaped or were actually ransomed on-site by relatives rather than sold abroad. Thus even though mass enslavement could, theoretically, be a means of bringing about the end of a group as such, it does not appear to have been utilised that way in this Gallic campaign. Similarly, descriptions of mass killing also seem to have been much less thorough than described by Caesar, with all the groups he mentions as being victims of such actions appearing in some numbers later in his narrative. This suggests that the message sent by such acts of mass violence was more important to Caesar than the actual outcome: it was not the actual enslavement or destruction of a people, whether in its entirety or a significant part, that Caesar necessarily desired, but the impression that he was capable of it. It was this that would achieve Caesar’s stated or implicit goals of retribution and deterrence. We may thus distinguish, in this regard, Caesar’s actions from other, more recent, genocidal actions where the elimination of the group qua group was seen as a central goal in and of itself, such as the Nazi’s ‘Final Solution’ in World War Two.93 In a legal context, however, the motivation to punish and deter must be distinguished from the intention to destroy a group whatever the motive, in this case in order to achieve punishment or deterrence.94 In this regard, Caesar’s intention to destroy the stirps ac nomen of the Eburones was undeniably genocidal within the Convention’s definition. Wees, ‘Ancient world’, 250–2. On vengeance: Riggsby, Caesar in Gaul and Rome, pp. 159, 171 and ch. 6 in general. 92 Van Wees, ‘Ancient world’, 252. 93 The bibliography on the Holocaust is immense, and it is the topic of Lemkin’s own work Axis Rule in Occupied Europe: Laws of Occupation, Analysis of Government, Proposal for Redress (Washington, DC: Carnegie Endowment for International Peace, 1944), in which he introduces the term ‘genocide’. For an introduction see C. R. Browning, ‘The Nazi empire’ in Oxford Handbook of Genocide Studies, ed. Bloxham and Moses, pp. 407–25. 94 On motivation versus intention, see ICTR in Kayishema and Obed Ruzindana (Appeal Chamber – 1 June 2001) ICTR-95–1-A [161] and ICTY in Jelisiç (Appeal Chamber – 5 July 2001) IT-95–10-A [49].

326

Caesar’s Gallic Genocide: A Case Study

Prejudice? Prejudice is frequently seen as a central force in driving mass violence, particularly genocide.95 Caesar certainly draws at times on ideas common in Rome of the Gauls as a divided, fickle and bellicose people.96 While antiGallic prejudice, including the ability to characterise them as enemies or hostes,97 certainly existed in Rome, Caesar does not draw upon this in a systematic way to justify the destruction of the Gauls in general or particular Gallic peoples; rather, mass killing and mass enslavement in Caesar are a response to particular actions.98 Rather than marked for destruction, it has been argued that the Gauls are described as people capable of appropriation and subjection to empire, as opposed to the Germani.99 In this regard, it may be of some significance that two of the three instances of mass killing are directed towards Germani. Throughout, Caesar goes to some lengths to establish the ‘otherness’ of these Germanic peoples. One brief example will suffice: There is no private or separate use of land among them, nor do they remain for longer than a year in one place for cultivating land. They do not use much corn, but mostly they live off milk and cattle and are greatly involved in hunting . . . from childhood they have no sense of duty and discipline, and do nothing at all against their will.100

Notice here the key role played by negatives – the Germani are described as much by who they are not as who they are – and of course, the things that they are not, are things that Romans are. This is a powerful way of reinforcing ‘otherness’. Furthermore, as Riggsby argues, the additional reference to Germans as practising banditry (latrocinium) operates to indicate the separation of the Germani from human civilisation: by declaring the Germani effectively bandits or latrones, they are neither robbers subject to the legal process (such as for theft or furtum) nor a foreign power against whom war might be declared. This characterisation by negation is utilised much less in 95 See Kiernan, Blood and Soil, pp. 21–2. 96 See J. Gardner, ‘The “Gallic menace” in Caesar’s propaganda’, Greece & Rome 30 (1983), 181–9; A. N. Sherwin-White, Racial Prejudice in Imperial Rome (Cambridge University Press, 1967), pp. 21–33. 97 For example Cic. Font. 27–32. 98 B. Isaac, Invention of Racism in Classical Antiquity (Princeton University Press, 2004), p. 250, on the general principle that the Romans lacked the need for a proto-racist justification for mass killing. 99 Riggsby, Caesar in Gaul and Rome, p. 82; C. Ruggini, ‘Intolerance: equal and less equal in the Roman world’, Classical Philology 82 (1987), 187–205, at 192–3. 100 Caes. BGall. 4.1.

327

tristan s. taylor

relation to the Gauls.101 Riggsby further argues that the perception of the Germani as peoples who would do ‘nothing at all against their own will’ made them undesirable candidates for enslavement.102 It is also possible that the fact that the Germans were perceived as nomadic and lacking fixed assets made it more likely that the Romans would direct violence against the population itself, as there were no other assets they might attack.103 However, it is important to note that while this ‘otherness’ may indeed have facilitated the violence directed towards the Germani, it nonetheless did not dictate a programme of their total destruction: Caesar was quite happy to have other Germani attack the Eburones.

Roman Imperialism The use of the spectrum of mass violence only in limited circumstances can also be connected with the nature of Roman imperialism – in particular, the manner in which Rome governed its empire and what she desired from it. Such a topic is beyond the scope of this chapter, but it suffices to note that while the extension of Roman imperium was frequently accomplished through extensive violence, Rome by and large depended upon local assistance in the governance of her empire, particularly co-operation from local elites, in order to achieve her aims of extracting revenue and manpower, and maintaining order. The complete, or near complete, destruction or deportation of local peoples was not compatible with such a system of imperial governance.104

Conclusion In conclusion, it appears that, in the context of Caesar’s expansionist imperialism at least, mass enslavement and mass killing directed against groups were part of a spectrum of mass violence used in the context of exemplary collective punishment. Such ‘conspicuous destruction’ was intended both to punish a betrayal of the Romans and at the same time to deter others from similar action. In some instances, particularly the 101 Riggsby, Caesar in Gaul and Rome, pp. 61, 122; Caes. BGall. 6.23. 102 Caes. BGall. 4.1; Riggsby, Caesar in Gaul and Rome, pp. 69–70. 103 E. Luttwak, The Grand Strategy of the Roman Empire for the First Century A.D. to the Third (Baltimore: Johns Hopkins University Press, 1976), pp. 45–6. 104 See e.g. N. Morely, The Roman Empire: Roots of Imperialism (London: Pluto Press, 2010), pp. 43–50; P. Brunt, ‘Laus imperii’ in Roman Imperial Themes (Oxford: Clarendon Press, 1990), pp. 288–323, at pp. 317–18.

328

Caesar’s Gallic Genocide: A Case Study

Eburones, this mass killing had all the elements that enable us to use the modern term ‘genocide’. Nonetheless, from Caesar’s writings it appears that such mass violence was considered as part of a particular spectrum of actions available in the exercise of Roman imperialism, rather than a distinct phenomenon. All such actions, however, are conspicuous by their effacement of the individual and the direction of violence at a group qua group. Although the actions of Caesar attracted criticism, for example by Cato, Caesar’s comfortable portrayal of instances of mass violence and the positive reception of such acts at Rome, such as the proposed thanksgiving for the massacre of the Usipetes and Tencteri, indicate that Caesar’s conception of the appropriate use of mass violence was not at odds with that of many of his fellow Romans. The fact that Caesar frequently seems to have overstated the extent of destruction or enslavement strongly suggests that his goals were the perception of the potential for mass violence at the hands of the Romans and the consequent deterrent effect, rather than actual elimination of various peoples. Mass violence was thus seen as an appropriate response to the specific actions of smaller groups of peoples – the Eburones, the Tencteri and Usipetes – rather than to the perceived characteristics of, or underlying threats posed by, broader groups of peoples, the Gauls or Germani. Such a conclusion is in harmony with the general nature of Roman imperialism, which sought to extend Roman imperium through intense violence, but then to govern through local elites rather than to supplant or eliminate the existing population.

Bibliographic Note The bibliography on Caesar, his conquest of Gaul and his narrative is immense. For an excellent overview of Caesar’s narrative and campaign, see Riggsby, Caesar in Gaul and Rome. For a discussion of Caesar’s narrative technique, see the various essays gathered in Welch and Powell, eds., Caesar as Artful Reporter. For a discussion of the contribution archaeology can make to the vexed question of Caesar’s use of mass violence in Gaul, see Roymans and Fernández-Götz, ‘Caesar in Gaul’, and Roymans, ‘Roman massacre far north’. For more general discussions of genocide in antiquity, see Taylor, ed., Cultural History of Genocide in the Ancient World; Barrandon, Massacres; Isaac, Invention of Racism, chapter 3; and van Wees, ‘Genocide in the ancient world’.

329

12

Genocidal Perspectives in the Roman Empire’s Approach towards the Jews gil gambash

The relationship between the Roman Empire and the Jewish population that inhabited parts of it has been studied in depth, possibly even beyond proportion. The wide body of research produced by modern scholars of the topic has tended to emphasise – again, perhaps disproportionally – the violent aspects of the relationship. The history of the Jews during the centuries and millennia that followed the Jewish War, as Josephus named it, naturally dictated the multiple biases that afflict the historiography of the topic, and must be admitted in the background of any relevant investigation, particularly one that aims to appraise potential genocidal aspects in the relationship. Thus, this chapter focuses on the extended conflict between the Roman administration and the Jewish population of the Roman Empire during the first and second centuries C E, while taking into consideration the available primary sources as well as dominant modern theories. Both are analysed here through the scope of genocide, in an attempt to determine whether there ever existed a Roman intent to destroy the Jewish group or parts of it. While the examination portrays a provincial group that may have practised more insubordination than others, the forms of violence and destruction exercised against this group occurred in the context of war, and are relevant for discussing the Jews as much as they are for studying the wars of the Gauls, Britons or Dacians. Whole communities, mostly in the form of cities or towns, were certainly targeted and destroyed, but only because of the opposition they presented to Roman rule, not for their being Jewish.

Genocide in Antiquity The topic of genocide in antiquity has been discussed by modern scholarship on several occasions. The sources are often vague and open to interpretation, even when claiming to represent deliberate actions aimed at the destruction

330

The Roman Empire and the Jews: Genocidal Perspectives

of whole groups. In theory, the idea of genocide was pronounced in writing already in the Homeric epos, as demonstrated by the words of Agamemnon to Menelaos with regard to the Trojans: ‘Not even one of them must escape sheer destruction at our hands.’1 Still, among modern historians of the ancient world, it may be tentatively agreed that such reports often fall short of representing accurately the reality they claim to mirror. The discourse of violence usually left much room for exaggeration, whether by the leaders and generals taking credit for their deeds or by the historians and poets singing their praises or simply aggrandising the story. Numerous groups, tribes, peoples and particularly cities allegedly reached the point of ultimate destruction, only to rise again and return to prosper shortly thereafter.2 The reasons for destruction on a grand scale may have varied but were often connected with different modes of political tension, suspicion or fear, ultimately leading to a war of conquest or to an outbreak and subjugation of revolt. Hypotheses have been formulated regarding the employment of deliberate annihilation as a grand strategic policy, with scholars referring, for example, to ancient empires’ ‘institutional genocide’, which would have been ‘routinized . . . and motivated by the desire to create terror, to display one’s power, and to remove the possibility of future retaliation’.3 It is thus barely surprising that a recent dissertation goes as far as to suggest that ‘Roman magistrates and the Senate became increasingly aware of their ability to fashion the geospatial and political landscape of their world. Building roads and colonies was one way of furthering this revisionism and augmenting their imperial hegemony; destroying communities was another.’4 Given the equivocal evidence available to us in most cases, it is necessary to steer away from such crude generalisations, just as it is essential to remain cautious about theories of ‘grand strategy’ in the ancient world more broadly.5 Close familiarity with specific cases is essential in approaching the 1 2

3

4

5

Hom. Il. 6.54–65. See for example the survey of archaeological and literary sources for the famous destruction of Corinth and Carthage in 146 B C E offered by S. A. James, ‘The last of the Corinthians? Society and settlement from 146 to 44 BCE’ in Corinth in Contrast, ed. S. J. Friesen, S. A. James and D. N. Schowalter (Leiden: Brill, 2013), pp. 21–7. R. Smith, ‘State power and genocidal intent: on the uses of genocide in the twentieth century’ in Studies in Comparative Genocide, ed. L. Chorbajian and G. Shirinian (Berlin: Springer, 2016), pp. 3–14. D. J. Colwill, ‘Genocide and Rome, 343–146 B C E: state expansion and the social dynamics of annihilation’ (PhD thesis, Cardiff University, 2018), p. 241. E. N. Luttwak, The Grand Strategy of the Roman Empire from the First Century A.D. to the Third (Baltimore: Johns Hopkins University Press, 1976).

331

gil gambash

Map 12.1 The Roman Near East in the early second century C E (Cartography by Gil Gambash)

topic of genocide in antiquity with accuracy, and a close and critical reading of the sources is a sine qua non. Also required is a standard employment of definitions, particularly where a comparative approach is adopted. Following the 1948 UN Convention on the Prevention and Punishment of the Crime of Genocide, the term ‘genocide’ will be considered here as: any of the following acts committed with intent to destroy, in whole or in part, a national, ethnical, racial, or religious group, as such: killing members of the group; causing serious bodily or mental harm to members of the group; deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part; imposing measures intended to prevent births within the group; forcibly transferring children of the group to another group.

Also relevant will be the further expansion offered by the legal definition of the crime against humanity of ‘extermination’, included in the 1998 Rome Statute of the International Criminal Court. This crime is legally described as any kind of conduct that ‘constituted, or took place as part of, a mass killing of members of a civilian population’ and was ‘committed as part of a widespread or systematic attack directed against a civilian population’. Using the category of extermination will allow us to include mass violence against a broader set of social groups, such as political groups or socio-economic classes.

332

The Roman Empire and the Jews: Genocidal Perspectives

Extermination covers most of those cases of genocide that are advanced by genocide scholars yet not covered by the UN Genocide Convention, such as those involving cities, towns and even rural communities. Thus, this chapter examines Roman approaches towards the Jews against the background of crimes against groups, collectivities or communities that are encompassed by the terms ‘genocide’ and ‘extermination’. Further elaboration of terminology that I find relevant comes from sociological approaches to the topic, such as Helen Fein’s, generally recognising four categories of genocide: developmental, despotic, ideological, and retributive.6 Attention to the perceived levels of threat or challenge presented by the victim group are also relevant here. Thus, Fein defines genocide as ‘sustained purposeful action by a perpetrator to physically destroy a collectivity directly or indirectly, through interdiction of the biological and social reproduction of group members, sustained regardless of the surrender or lack of threat offered by the victim’.7 And Hans van Wees refers directly to genocide in antiquity as ‘an act of conspicuous destruction, a display of force, designed to assert the power and status of the perpetrator in the face of a perceived challenge’.8

Provincial Routine Taking all of these definitions into consideration, an evaluation of the conflict between the Jews and the Roman administration must include the background for its outbreak, starting at least as early as the time of the annexation of Judea by the Roman Empire in 6 C E. The six decades that followed this moment, up until the mid-sixties, present a functioning system not unlike the one experienced by the other provinces of the empire. Earlier scholarly treatments saw Judea’s procurators as either generally incompetent or hostile to the Jews, and such hostility, if true, would have been relevant for our discussion of eventual violence turned against the Jews.9 But it is widely 6

7 8

9

H. Fein, Genocide: A Sociological Perspective (London: Sage, 1993). Essentially, developmental genocide is employed against a population which is seen to impede the exploitation of natural resources; despotic genocide aims to eliminate threats to political power; ideological genocide is motivated by a system of belief; and retributive genocide is performed in retaliation to a perceived threat on a privileged position. Ibid., p. 24. H. van Wees, ‘Genocide in the ancient world’, in The Oxford Handbook of Genocide Studies, ed. D. Bloxham and A. D. Moses (Oxford University Press, 2010), pp. 239–57. For example E. Schürer, The History of the Jewish People in the Age of Jesus Christ (London: Bloomsbury, 1973), I , p. 455: ‘It might be thought, from the record of the Roman procurators to whom, from [the year 44], public affairs in Palestine were entrusted, that they all, as if by secret arrangement, systematically and deliberately set out to drive the people to revolt.’

333

gil gambash

acknowledged today that such a narrative is widely off the mark, and based more on Josephus’ formulaic discourse than on the events he depicts. Above all, it is heavily influenced by the faults and failures of the one procurator during whose tenure a real crisis actually started – Gessius Florus. However, according to the routine portrayed in our sources, it is clear that most Roman officials came to understand the needs and limits of the Jewish population, and, with a few exceptions, strove to comply with them. The universal and inevitable hardships that accompanied Roman imperialism aside, the Roman administration was aware of the specific concerns of the Jews, and flexible enough to supply solutions in most cases. Breaches in this routine should not be mistaken for common practice, but rather approached as the exceptions that prove the rule. Thus, respecting Jewish religious beliefs, Roman officials regularly refrained from presenting image-bearing standards in Jerusalem and were also careful not to produce image-bearing coins in the mint at Caesarea. Due to an understanding of their particular restrictions, Jews were exempt from appearing before a Roman magistrate on the Sabbath.10 Such examples of routine consideration represent the reality of foreign rule in Judea that is much more accurate than those few occasions when a Roman procurator or emperor caused provocation. The insignificant military presence in the region, and the low rank of the officials sent to govern it during this period, indicate no significant tensions. Occasional riots and disturbances, on the other hand, would have been a regular part of the reality of imperialism, whether for universal reasons, such as the pressure of taxation or military drafts, or for more specific causes particular to each province and its particular nature.11 Dwelling on the period preceding 66 C E is important even if Judea can be seen here as a typical province of the empire, neither more problematic nor more violent or tense. All provinces negotiated with Rome on a regular basis, though not all provinces had Josephus to record the spirit of negotiations, if not their verbatim narrative. All provinces experienced moments of tension, disappointment and disagreement, some even open revolt – but peaceful routine always predominated, as it did in pre-66 Judea. This should also be the point of reference for the question asked in this chapter. Certainly, war, once it started, had the power to change realities – both physical landscapes and the nature of relationships. It was perhaps less common for war – even when 10

11

M. Goodman, Rome and Jerusalem: The Clash of Ancient Civilizations (London: Penguin, 2007), p. 417. Pekáry lists about a hundred events of unrest across the empire in the period from Actium to Commodus; see T. Pekáry, ‘Seditio: Unruhen und Revolten im Römischen Reich von Augustus bis Commodus’, Ancient Society 18 (1987), 133–50.

334

The Roman Empire and the Jews: Genocidal Perspectives

long and difficult – to change deeply engrained cultural structures, such as the conceptual place provincial populations played in the routine of empire, to which the genocidal perspectives discussed here are related. An important question that should be asked, therefore, is what routine was like, and to what extent the mechanism of Roman imperialism aspired to return to it when tension arose, and then when war erupted. Pontius Pilate’s term as Judea’s governor between 26 and 36 C E demonstrates this shift between routine and tension. The two incidents that occurred during this decade do not constitute a case for regarding it as ‘stormy’, or Pilate himself as incompetent.12 It is probably the evaluation by Tacitus of Tiberius’ reign as quiet that captures the spirit of the period much more accurately.13 The Jews never refrained from complaining to higherranking Roman magistrates when they were dissatisfied with their direct rulers, and these complaints often proved to be effective, as in the banishment of King Archelaus in 6 C E and the willful provincialisation of the Herodian kingdom. The fact that Pilate remained in Judea for so long is thus convincing evidence that both the Roman government and the Jewish population in the region were for the most part satisfied with his governance. Pilate’s term as governor was in fact terminated as a result of the complaints of another provincial population, the Samaritans, who felt that the procurator had used unnecessary force against them during a gathering – apparently religious in nature – on Mount Gerizim.14 Their specific pleading to Vitellius, the governor of Syria, that they did not gather in that location for the sake of revolt, may hint at the reason for Pilate’s suspicion. As a result of this provincial appeal to a higher authority, an emissary was sent from Syria to pacify tensions, and Pilate himself was ordered back to Rome, to explain matters to the emperor, and never returned.15 The process that brought about the interruption of his term is telling of an administration generally mindful of local grievances, surfaced openly and effectively by Jews and other groups.

12

13 14

15

J. J. Price, Jerusalem under Siege: The Collapse of the Jewish State 66–70 CE (Leiden: Brill, 1992), p. 6. Tac. Hist. 5.9.2. On the Samaritans, see J. D. Purvis, ‘The Samaritans’ in The Cambridge History of Judaism, ed. W. D. Davies and L. Finkelstein (Cambridge University Press, 1989), pp. 591–613. For Josephus’ treatment of the Samaritans, see R. J. Coggins, ‘The Samaritans in Josephus’ in Josephus, Judaism, and Christianity, ed. L. H. Feldman and G. Hata (Leiden: Brill, 1987), pp. 257–73. Joseph. AJ 18.85–9; J-P. Lémonon, Pilate et le gouvernement de la Judée: textes et monuments (Paris: Libraire Lecoffre, 1981), p. 238.

335

gil gambash

The governor of Syria’s reported visit to Jerusalem upon Pilate’s removal would have made clear to the Jews themselves the consideration and even sympathy of the administration: Vitellius, on reaching Judea, went up to Jerusalem where the Jews were celebrating their traditional feast called the Passover. Having been received in magnificent fashion, Vitellius remitted to the inhabitants of the city all taxes on the sale of agricultural produce and agreed that the vestments of the high priest and all his ornaments should be kept in the temple in custody of the priests, as had been their privilege before.16

The policy was persistent. When marching some time later against Aretas IV of Nabatea, Vitellius is said to have diverted the course of the army in accordance with the request of the Jews, in order to avoid once again the presentation of image-bearing standards in Judea.17 Loyal to the general Roman tolerance of other religions, he is also reported to have offered sacrifice to the god of the Jews.18 Vitellius was hardly exceptional in his approach towards the Jews – and, one might perhaps assume, towards his other subjects as well. His successor had to face a most difficult choice between disobeying the direct command of the emperor Gaius and provoking the Jews in Judea and the diaspora alike. Publius Petronius, who is said by Philo to have studied Jewish philosophy in preparation for ruling a region containing large Jewish communities, took the side of the Jews, possibly at the risk of his own ruin.19 Gaius’ orders arrived in 39–40, in direct response to Jewish aggression against non-Jewish communities in Jamnia. Petronius was instructed to place a statue of the emperor at the Temple in Jerusalem. He soon realised that the Jews were set to revolt across the region if Gaius’ instructions were implemented. Refusing the request of the Jews to send an embassy of their own to Rome, he took it upon himself to write to the emperor and try to persuade him to annul the decree. His care at this stage for the cultivation of the land is of particular interest, since the Jews were neglecting their agricultural duties 16 18 19

Joseph. AJ 18.90. 17 Joseph. AJ 18.121. News of the death of Tiberius reached him in Jerusalem at that time (ibid.). Philo Leg. 245: ‘he still had himself some sparks of the Jewish philosophy and piety, since he had long ago learnt something of it by reason of his eagerness for learning, and had studied it still more ever since he had become governor of the countries in which there are vast numbers of Jews scattered over every city of Asia and Syria’. Smallwood relates this to the famous reference to Genesis 1 in De sublimitate (9.9), indicating familiarity with Hebrew literature among educated Greeks and Romans; see E. M. Smallwood, Philonis Alexandrini Legatio ad Gaium (Leiden: Brill, 1961), pp. 279–80.

336

The Roman Empire and the Jews: Genocidal Perspectives

as a result of the tension.20 Beyond his wish to avoid a revolt, sources mention his fear of a resulting famine as a crucial incentive to solve the problem promptly, demonstrating the wide array of considerations regularly involved in the administration’s pursuit of keeping the provinces peaceful, even when an emperor’s instruction had to be challenged.21 The tensions that emerged during the short reign of Gaius extended also to Alexandria, where non-Jewish groups prevailed upon the governor to curtail the privileges of the Jewish population.22 Such events motivated Claudius to act shortly after his accession in a decisive way in order to steer the relationship with the entire Jewish population of the empire back to its familiar calm routine.23 Claudius’ first decision was to enlarge the territory of King Agrippa, consisting of Gaulanitis, Auranitis, Batanaea and Trachonitis, by adding to it Judea – an organised and stable province of the Roman provincial system for thirty-five years by that time. This step should not be understood as a personal token of friendship between emperor and king – Agrippa had also been very close to Gaius – but as a diplomatic move, devised to pacify further the Southern Levant after the turmoil caused by Gaius’ threats, and calculated to rebuild the trust between the Roman government and the local population.24 Rome’s considerations in determining the status of provinces and client kingdoms usually emanated from a motivation to serve the empire’s best interests.25 Just as the decision to annex a kingdom could stem from reasons of security and stability, so the reverse process of returning a province to the status of a client kingdom could probably derive from a similar motivation. Quite likely, Claudius’ new organisation of the region was meant to last, and only Agrippa’s sudden death in 44 C E, and the young age of his son at the time, necessitated the return of Roman officials and the reannexation of Judea. By all accounts, Agrippa’s brief reign was a stable and 20

21

22 23

24

25

Philo (Leg. 249) talks about harvest time (May–June), whereas Josephus (BJ 2.200, AJ 18.272) refers to the time of sowing (October). Cf. the aftermath of the Boudican revolt (Tac. Ann. 14.38): ‘Nothing however distressed the enemy so much as famine, for they had been careless about sowing corn, people of every age having gone to the war, while they reckoned our supplies as their own.’ Philo In Flacc. 16–96. For the stasis in Alexandria between the Jews and Greeks, see B. Ritters, Judeans in the Greek Cities of the Roman Empire: Rights, Citizenship, and Civil Disorder (Leiden: Brill, 2015), pp. 132–83. Tacitus (Ann. 12.54) is also aware of the Jews’ restlessness at the time: ‘When they had heard of the assassination of Gaius, there was no hearty submission, as a fear still lingered that any of the emperors might impose the same orders.’ See also E. M. Smallwood, The Jews under Roman Rule from Pompey to Diocletian: A Study in Political Relations (Leiden: Brill, 1981), p. 193. D. Braund, Rome and the Friendly King: The Character of Client Kingship (London: Croom Helm, 1984), p. 189.

337

gil gambash

prosperous period for the Judean population, in accordance with the emperor’s probable aims. Claudius’ handling of the conflict in Alexandria clearly supports such a view and demonstrates how the emperor’s policy regarding the Jews was persistent and universal. Tension between various groups in the city was perennial and notorious, and at this point it was the insecurity of the Alexandrian Jewish community that encouraged Claudius to issue an edict reaffirming the citizenship rights of the Alexandrian Jews.26 But the emperor did not only aim to solve the specific problem of that Jewish community alone. Recognising that, because of the restrictions of their religion, the position of the Jews was fragile across the empire and even in Judea – as Gaius’ attack most recently demonstrated – he declared his predecessor’s madness in his Alexandrian edict, and then issued a second edict, which reaffirmed the rights and freedom of cult of the Jews throughout the empire.27 The policy stipulated in the decrees had an occasion to be implemented soon thereafter, in 41–42, when enforced by the governor of Syria in response to the infringement of the Jewish law by the Greek community at Dora. Petronius’ letter to the city magistrates referred directly to Claudius’ edicts, and included the administration’s strong wish to keep the peace with the Jewish community: ‘Both I myself, and King Agrippa, for whom I have the highest honor, have nothing more under our care, than that the nation of the Jews will have no occasion given them of getting together, under the pretense of avenging themselves, and become tumultuous.’28

Provincial Conflict It was not so much the outbreak of the Jewish revolt in 66 as the subsequent defeat of the governor of Syria, Cestius Gallus, that constituted a significant turning point in the Roman approach towards rebelling Jews in Judea. While standard provincial revolts usually instigated focused efforts by small, locally stationed troops, the disastrous retreat of Gallus from Jerusalem persuaded the Roman administration to approach the subjugation of the revolt from that point onwards as a full-scale foreign campaign, to be supervised by a 26

27 28

Joseph. AJ 19.280–5. For a full bibliography and a discussion of the edict, see M. Pucci Ben Zeev, Jewish Rights in the Roman World (Tübingen: Mohr Siebeck, 1998), doc. 28. Joseph. AJ 19.287–91. See Pucci Ben Zeev, Jewish Rights, doc. 29. Joseph. AJ 19.303–11. For a full bibliography and a discussion of the letter, see Pucci Ben Zeev, Jewish Rights, doc. 30.

338

The Roman Empire and the Jews: Genocidal Perspectives

general with outstanding authority.29 This basic fact is important in understanding key aspects of the events that followed, not least the employment of violence by Roman officials and troops. The task force assigned to the commander of the operation, Vespasian, amounted to 60,000 troops – twice as large as that under Cestius Gallus just a year earlier. It assembled in Ptolemais and was ready for action early in 67. The core of the force comprised three legions, only one of which, the X Fretensis, was taken from Syria’s regular garrison of four legions.30 The legionary forces were supplemented by twenty-three auxiliary cohortes and six alae of cavalry. Four client kings – Antiochus IV of Commagene, Agrippa II of Chalcis, Sohaemus of Emesa, and Malchus II of Nabatea – sent an additional 18,000 troops.31 The fierceness of the revolt and its initial success created a reality whereby the Southern Levant shook off all tokens of Roman control and had to be reconquered. Vespasian’s army was thus similar in scale to forces put together for the purpose of foreign campaigns. It was larger than the army that had invaded Britain in 43 C E.32 The history of conflict that follows is not limited to the event of this first revolt and its subjugation; it spans more than seven decades and includes also the Diaspora and Bar Kochba revolts. It is this history that accounts for the inclusion of the Roman–Jewish relationship in a volume on genocide. It must be read, however, not only against the wider background of peaceful routine presented above but also with careful attention to Roman policies and the way they corresponded with that provincial routine, particularly at the end of each eruption of active resistance. To be sure, the progress of Vespasian’s army followed the familiar patterns of conquest campaigns. Josephus’ account of Roman actions in the years 67– 70 abounds with reports of atrocities against the Jewish population, and Vespasian’s treatment of Gadara is a case in point: So Vespasian marched to the city of Gadara, and took it upon the first onset, because he found it destitute of any considerable number of men grown up and fit for war. He came then into it, and slew all the youth, the Romans 29

30

31

32

G. Gambash, ‘Foreign enemies of the empire: the Great Jewish Revolt and the Roman perception of the Jews’, Scripta Classica Israelica 32 (2013), 173–94. The other two legions were the XV Apollinaris, brought by Titus from Alexandria, and the V Macedonica, which participated in the Armenian campaign in 61/2. F. Millar, The Roman Near East, 31 BC – AD 337 (Cambridge, MA: Harvard University Press, 1993), p. 72. For the comparison, see F. Millar, ‘Last year in Jerusalem: monuments of the Jewish war in Rome’ in Flavius Josephus and Flavian Rome, ed. J. Edmondson, S. Mason and J. B. Rives (Oxford University Press, 2005), pp. 101–28.

339

gil gambash

having no mercy on any age whatsoever; and this was done out of the hatred they bore the nation, and because of the iniquity they had been guilty of in the affair of Cestius. He also set fire not only to the city itself, but to all the villas and small cities that were round about it; some of them were quite destitute of inhabitants, and out of some of them he carried the inhabitants as slaves into captivity.33

Next to such retributive measures, violence erupted also in numerous mixed cities, where tensions that had existed between Jewish and non-Jewish populations were now exacerbated by the revolt and the loss of trust in the Jews, also on the part of the administration itself.34 In Skythopolis, non-Jewish inhabitants slaughtered some 13,000 and plundered their property; Ascalon, Ptolemais, Tyre and other Syrian cities saw between 2,000 and 2,500 Jewish citizens massacred in each.35 In Alexandria, 50,000 Jews were killed in the clashes that erupted with the Greek community of the city, and the consequent involvement of the military. The orchestrator of the subjugation operation was no other than Tiberius Alexander, the prefect of Egypt and himself a Jew, who would later serve as Titus’ second-in-command during the siege of Jerusalem.36 The abundance of such reports may appear striking, but only when compared to other provincial rebellions.37 Numerous wars of conquest and expansion, from Asia Minor and North Africa to Gaul, Spain and Britain, share in such descriptions. One distinct feature, common to such campaigns, is the focus of the campaign on particular towns and cities as centres of resistance. Regardless of how popular resistance movements could become, usually not all settlements actively resisted or even nominally joined the opposition to Roman rule. Even during war, Roman armies knew to

33

34

35 37

Joseph. BJ 3.132–4. Another example may be found in Joseph. BJ 6.417–18: ‘So this Fronto slew all those that had been seditious and robbers, who were impeached one by another; but of the young men he chose out the tallest and most beautiful, and reserved them for the triumph; and as for the rest of the multitude that were above seventeen years old, he put them into bonds, and sent them to the Egyptian mines. Titus also sent a great number into the provinces, as a present to them, that they might be destroyed upon their theatres, by the sword and by the wild beasts; but those that were under seventeen years of age were sold for slaves.’ E. Gruen, The Construct of Identity in Hellenistic Judaism (Berlin: De Gruyter, 2016), pp. 313–32. Joseph. BJ 2.466–8; 477–8. 36 Joseph. BJ 2.487–98. Joseph. BJ 3.62–3; 3.304–5; 3.338–9. The available accounts of the Pannonian–Dalmatian revolt hardly give a notion of Roman action that is not taken purely out of military considerations; neither Cassius Dio nor Velleius give more than a little testimony to the employment of brutal measures against the local population and the rebels themselves. The same applies to the war in Africa against Tacfarinas.

340

The Roman Empire and the Jews: Genocidal Perspectives

differentiate closely between supportive and opposing elements, which fact allows us to follow their employment of violence in a more focused way. Villages, towns and cities that joined the Jewish revolt and opposed the Roman subjugation effort were obviously targeted, besieged or attacked, and ultimately conquered. Van Wees maintains that, despite the great damage caused as a result of such measures, genocide should not be considered if the population had not been eliminated; he also argues that the devastation caused by soldiers not following the direct orders of their commanders – for example during the process of breaking into a long-besieged city – may not be treated in such terms.38 Titus himself is stated at one point to have declared that he had had no intention for the Temple in Jerusalem to be burnt down, after his soldiers had done just that.39 However, whole settlements, like Jerusalem itself, were intentionally razed to the ground, with their population killed or sold into slavery, and the question regarding the relevance of genocide in such cases is valid. It should be acknowledged, firstly, that much like in most other cases in antiquity, it was not the gens or ethnos that were targeted for destruction during the first Jewish revolt. Not far from Gadara stood the Jewish town of Sepphoris, which refused to join the Jewish revolt and instead opened its gates to Vespasian’s army voluntarily, winning the favour of the Roman administration for the future to come.40 Therefore, the term ‘urbicide’ may be more relevant for the purpose of our discussion here.41 Settlements in Judea and the Galilee were for the most part not organised politically as citystates, but nonetheless contained distinct features of local identity and selfrepresentation, even while associating themselves more broadly with the Jewish people. The planned destruction of such towns as Gadara and Jotapata, while it was not conducted as part of a grand plan to destroy the Jewish population as a whole, did result in the annihilation of whole organic communities, and may be included in the discussion evoked by this volume, even if these communities did not define themselves first and foremost as ethnic, national, racial or religious groups. Despite this, one fact may be stated with confidence: all cases of provincial rebellions and wars of conquest saw the gradual emergence/resumption of peaceful routine under the empire, even if the pace of the process varied.42 Recent numismatic evidence, dated to the very first moments after the fall of Jerusalem, shows that this was also the initial intention of the Roman 38 41 42

Van Wees, ‘Ancient world’, 245–50. 39 Joseph. BJ 6.254. 40 Joseph. BJ 2.18–20. Van Wees, Genocide. G. Gambash, Rome and Provincial Resistance (New York: Routledge, 2015).

341

gil gambash

administration for Judea. The aureus at issue depicts on its reverse a palm tree, and a woman is standing next to it with her legs crossed, her arms bent, and her head resting on her right hand. The legend on the reverse reads I V D A E A – R E C E P T A (Judea returned), and, together with the iconography, it is unique and has never been found before on coins or elsewhere in the official Roman propaganda advertising the conclusion of the first Jewish revolt.43 The particular iconography of Vespasian’s portrait on the obverse, and the titulature given by the legend – I M P C A E S A R V E S P A S I A N V S A V G – allow numismatists to date the coin soon after September 70 C E. Unlike capta-type coins, which convey the idea of conquest by force of new territory and the status of a province recently added by force to the empire, the term recepta was employed in the imperial-provincial context in the aftermath of the return of an erring province to the embrace of the empire. The issue of Iudaea recepta coinage would have been in line with Rome’s normal practice, which was to return as quickly and as efficiently as possible to the antequam situation. Such an effort could have included the avoidance of retributive measures and humiliating commemoration projects. Where the threat of local resistance was thought to have subsided, trust in local loyalty could be established, potentially followed by demilitarisation. Rebelling provinces could even obtain a prompt resumption of their former status.44 This is precisely what took place once the Batavian revolt, led by Civilis, had been put down by Petilius Cerialis in the year 71 C E.45 The Batavians are known to have been exempt from taxes since the time of their incorporation into the empire, and their revolt, serious as it was, did not lead to the annulment of this significant privilege: They [the Batavians] still retain this honor, together with a memorial of their ancient alliance; for they are neither insulted by taxes, nor oppressed by tax farmers. Exempt from fiscal burdens and extraordinary contributions, and kept apart for military use alone, they are reserved, like a magazine of arms, for the purposes of war.46

As Tacitus’ report shows, once the revolt was subjugated, rather than engaging in punitive measures, the Roman administration restored the beneficial status of the Batavians, and trust in them was quick to resume. Their cavalry and troops continued to support the military efforts of the 43

44

45

G. Gambash, H. Gitler and H. Cotton, ‘Iudaea recepta’, Israel Numismatic Research 8 (2013), 89–104. G. Gambash, ‘To rule a ferocious province: Roman policy and the Boudican revolt’, Britannia 43 (2012), 1–15. Tac. Germ. 29. 46 Tac. Germ. 29.

342

The Roman Empire and the Jews: Genocidal Perspectives

empire in Britain and along the Danube for decades and centuries. In commemoration of the subjugation of the revolt, Vespasian advertised through coinage only the return of the military standards that had been lost in the conflict. The caption merely read S I G N I S R E C E P T I S (Military standards returned). Other forces, however, came to play a significant part in the aftermath of the first Jewish revolt. They redefined the Roman approach towards the Jews soon after the issuance of the recepta coin, making it, and the entire policy it aimed to represent, all but redundant. Among these forces, most significant were, firstly, the need of the new imperial dynasty for symbolic and material gains; and, secondly, the tangible threat still perceived on the Jewish front. Regardless of the reality of the war, Vespasian, and his sons after him, launching as they did a new dynasty of Roman emperors, were intent on monumentalising the campaign’s importance, and therefore emphasised and probably exaggerated also the brutality shown towards the Jews – much in the way that theatrical re-enactments of the conquest of Britain had been celebrated in Claudian Rome, or the images of Dacian slaves would later flood the city after Trajan’s conquest of the region. On this account, the entire Flavian commemoration operation, including Josephus’ report, should be suspected of exaggeration. Beyond symbolic acts, practical measures were also enacted. The Temple in Jerusalem, destroyed by the Romans, was not to be rebuilt; and a new tax came into existence – the fiscus Judaicus – exacted from all Jews across the empire. The tax came to replace the traditional temple donation of two drachms, which had been sent annually by every Jew to the Temple in Jerusalem, and which the Romans thought had helped in financing the revolt.47 It is best understood in the context of war indemnities and of the Flavian administration’s need for resources. Additionally, Judea now had a legion permanently stationed in it, changing its status into a propraetorian province. These measures – punitive and practical alike – were harsh and even unique in the Roman world. But they do not fall under any of the categories examined here in relation to genocide and extermination, as they do not appear to inflict on the entire group of the Jews conditions of life calculated to bring about its physical destruction. The recollection of Gallus’ colossal defeat in 66 would still have been vivid. And persistent rebel groups in Herodium, Machaerus and Masada would keep the fire of resistance going 47

Joseph. BJ 6.335.

343

gil gambash

until 73/4, depriving the Roman government of all motivation to resume trust in the local population.48 Evidence of unrest within Jewish communities is seen for example in messianic hopes nurtured within some Diaspora communities, possibly inspired by Sicarian elements that appeared in Egypt and Libya after the revolt.49 The presence of a legion in Jerusalem and the ban on rebuilding the temple demonstrate just how the image of the Jews remained threatening even after their defeat. On the other hand, the writings of contemporary Roman authors – such as Seneca, Tacitus and Juvenal – do not represent universal Roman apprehension of Jews.50 Of course, it would have been mostly operational circles of officials and military men who would have perceived any threat.51 Beyond the new reality in the province itself, it was mostly the Jewish tax, implemented across the empire, which would have symbolised the shift in Jewish status after the revolt. And as soon as the Flavian dynasty had been removed from power, a first attempt was made to cancel that tax. On Nerva’s ascent to power in 96 C E, after the assassination of Domitian and the termination of the Flavian dynasty, his coins began to advertise the removal of an ‘abuse’ committed by the institution of the Jewish tax – the fiscus Judaicus: F I S C I J U D A I C I C A L U M N I A S U B L A T A (The abuse of the Jewish tax abolished).52 It is debatable what the exact nature of the abuse was, and who suffered from it under Domitian, but the new measure would have been devised on behalf of the Jews, whether in abolishing the Jewish tax completely or in putting an end to its presumed imposition by Domitian even on Jews who had abandoned their ethnic 48

49

50

51

52

Schürer, History of the Jews, pp. 511–13; H. M. Cotton and J. J. Price, ‘Who captured Masada in 66 CE and who lived there until the fortress fell?’, Zion 55 (1990), 449–54; J. Roth, ‘The length of the siege of Masada’, Scripta Classica Israelica 14 (1995), 87–110. For the historiography of Masada, see also N. Ben-Yehuda, The Masada Myth: Collective Memory and Mythmaking in Israel (Madison: University of Wisconsin Press, 1996); and N. Ben-Yehuda, Sacrificing Truth: Archaeology and the Myth of Masada (New York: Humanity Books, 2002). M. Pucci Ben Zeev, Diaspora Judaism in Turmoil, 116–117 CE: Ancient Sources and Modern Insights (Leuven: Peeters, 2005), pp. 128–42; S. Applebaum, Jews and Greeks in Ancient Cyrene (Leiden: Brill, 1979), pp. 242–60. See E. Gruen, ‘Roman perspectives on the Jews in the age of the Great Revolt’ in The First Jewish Revolt, ed. A. M. Berlin and J. A. Overman (London and New York: Routledge, 2002), pp. 27–42. See also E. Gruen, Diaspora: Jews amidst Greeks and Romans (Cambridge, MA: Harvard University Press, 2002). Woolf also emphasises elite perspectives on provincial revolts more broadly (G. Woolf, ‘Provincial revolts in the early Roman Empire’ in The Jewish Revolt Against Rome, ed. M. Popović (Leiden: Brill, 2011), pp. 27–44). H. Mattingly, Coins of the Roman Empire in the British Museum, vol. I I I: Nerva to Hadrian (London: British Museum Publishing, 1976), pl. 15 no. 88, pl. 17 no. 98, pl. 19 nos. 105–6.

344

The Roman Empire and the Jews: Genocidal Perspectives

identity.53 A wish to distance himself from his predecessors may have driven Nerva to adopt a less hostile approach towards the Jews.54 However, as with the Flavians, here also imperial policy may not be analysed separately from political reality. Observing the practical aspects of the new step, Nerva may be seen here to be acting to resume the routine status of the Jews as a standard provincial population of the empire, now thought to have paid the symbolic debt of the revolt, and gradually regaining the trust of the Roman administration. Observed from the broader viewpoint of Rome’s relationship with the provinces, one must consider the possibility that Nerva’s act took into consideration the frustration of the Jews with their current shaky position within the provincial system. The Jewish tax could not have been popular, and the fall from power of the dynasty that had destroyed Jerusalem would have created anticipation for improvement. Claudius had introduced exactly such optimistic prospects for improvement five decades earlier, after Gaius’ oppressive measures; and Vespasian briefly adopted such an approach with his Iudaea recepta coinage. Nerva was publicising exactly such a turn in Roman policy, one that was bound to arrive at some point, just as it did for every other provincial population in the history of the empire. Evidence from Egypt shows that the Jewish tax was still exacted there in the early second century, under Nerva’s successor, Trajan. This has been ascribed to the new emperor’s strong affiliation through his father to the Flavian legacy, though Egypt’s case may have been outstanding and does not necessarily mean that Nerva’s measures were universally repealed.55 Still, the Temple in Jerusalem was not allowed to be rebuilt, and the legion stationed there since the subjugation of the great revolt remained a strong and vivid presence in the routine of the province. This may have been, at least in part, the background for the second insurrection of the Jews, which took place primarily outside of Judea, and is therefore known as the Diaspora revolt (115–117 C E).

53

54

55

On the abolition of the tax, see M. Hadas-Lebel, ‘La fiscalité romaine dans la littérature rabbinique jusqu’à la fin du III siècle’, Revue des Études Juives 143 (1984), 5–29; M. Goodman, ‘The Fiscus iudaicus and gentile attitudes to Judaism in Flavian Rome’ in Flavius Josephus, ed. Edmondson, Mason, and Rives, pp. 167–80. On apostate Jews, see M. Goodman, ‘Nerva, the Fiscus judaicus, and Jewish identity’, Journal of Roman Studies 79 (1989), 40–4. M. Goodman, ‘Trajan and the origins of Roman hostility to the Jews’, Past & Present 182:1 (2004), 3–29. Ibid.

345

gil gambash

Regardless of whether the oppressive measures of the Romans against the Jews had persisted because of symbolic aspects of dynastic legacy, tangible apprehension of revolt, or, most likely, a combination of the two, by the beginning of the second century the situation of the Jews grew anomalous within the Roman provincial sphere. It is difficult to find other groups that did not manage to reach a state of peaceful routine within the provincial system, and to be considered eventually a provincia fidelis (loyal province) by the government. It is just as challenging to find examples of provincial populations that rebelled more than once. While much of the Diaspora revolt occurred outside of Judea, its Jewish focus defines it as connected at the core to the broader Roman–Jewish conflict. Resistance emerged simultaneously in the summer of the year 116 in Cyrenaica, Egypt, Cyprus and Mesopotamia.56 From the very beginning, Trajan clearly treated the movement seriously, assigning leading generals to deal with it. In Mesopotamia, the mission was entrusted to Lusius Quietus, who had already distinguished himself in commanding operations on the Parthian front.57 The Jews here joined a larger opposition movement to Roman conquest, but Quietus managed to recapture within a year the cities of Nisibis, Seleucia, on the Tigris, and Edessa, the capital of Osrhoene. He was then appointed as consul and sent to govern the province of Judea itself, indicating Trajan’s apprehension of the rise of opposition there, as well.58 With this appointment, Judea’s status now changed to a pro-consular province, and a second legion was added to the one that had been permanently stationed there since 70 C E.59 In the eastern Mediterranean, Trajan likewise invested significant efforts to quell the revolt. He abandoned the siege of Hatra and sent to the region two more of his best generals, Q. Marcius Turbo to Egypt and perhaps also to Libya, and Caios Valerius to Cyprus.60 According to papyrological evidence, two legions fought the Jews in Egypt under the command of Turbo, one of which – the Legio XXII Deiotariana – had been stationed in Egypt since the time of Augustus; the other was the Legio III Cyrenaica, sent back from the Parthian front.61 Great sacrifices were made in Trajan’s eastern campaign in 56 57 59

60 61

For the chronology of the events, see Pucci Ben Zeev, Diaspora Judaism, pp. 143–56. Prosopographia Imperii Romani2 L 439. 58 Cass. Dio 68.32.5. Pucci Ben Zeev, Diaspora Judaism, pp. 253–5; H. M. Cotton and W. Eck, ‘Governors and their personnel on Latin inscriptions from Caesarea Maritima’, Proceedings of the Israel Academy 7:7 (2001), 219–23. Pucci Ben Zeev, Diaspora Judaism, pp. 144–6. Corpus Papyrorum Judaicarum II 438. Another, less likely possibility, is that it was Legio XV Apollinaris; see Pucci Ben Zeev, Diaspora Judaism, pp. 178–86.

346

The Roman Empire and the Jews: Genocidal Perspectives

order to pacify the Diaspora revolt, with significant troops and leading generals allocated to the more urgent mission, which was obviously considered a great threat to the empire’s stability, both because of its scope and due to the fact that it was a recurring Jewish revolt. If these circumstances also entailed fierce subjugation, it should, once again, be viewed through the prism of war rather than that of genocide. Violence demonstrated during the campaign and the reprisals that followed in its aftermath would have been congruous with the generally suspicious approach towards the Jews prior to the revolt. As always, extreme views of hostility towards the Jews should be approached with caution. Dio, for example, goes as far as to report about Cyprus that ‘no Jew can set foot on that island, and even if one of them is driven upon the shores by a storm, he is put to death’.62 Appian writes that Trajan ‘exterminated’ the Jewish race in Egypt, and Arrian writes about the emperor’s determination to destroy the nation entirely.63 Much like the descriptions of alleged destruction of various other cities and peoples with which this chapter opened, such statements represented views and sentiments at the end of a long and difficult campaign more than they did reality itself.64 Papyrological evidence from Egypt presents a more balanced picture. It reveals the confiscation of Jewish property as ordained by the issue of an imperial decree, which condemned the Jews for having rebelled against Rome – a measure reminiscent of the fiscus Judaicus of earlier decades.65 In Judea as in the diaspora, Jewish communities continued to prosper without facing measures that would threaten their existence, either in purely Jewish settlements, as was the common case in the Galilee, for example, or in mixed towns and cities, in which they comprised a minority. A telling case may be found in Dora, which had seen in 41–42 C E the violation of the status quo between the Greek and Jewish communities and the quick implementation of the fresh pro-Jewish Claudian decrees by Petronius, the governor of Syria. There is evidence of the dedication of a statue by the entire community of the city to its patron, the outgoing governor of Judea and incoming governor of Syria, Gargilius Antiquus, dating to the middle of the 120s C E, shortly before the Bar Kochba revolt.66 62 64 65

66

Cass. Dio 68.32.3. 63 App. B. Civ. 2.90; Arr. Parth. fr. 85. Gruen, Construct of Identity. For example P.Oxy. 500. See J. Mélèze Modrzejewski, ‘Ioudaioi apheremenoi: la fin de la communauté juive d’ Égypte (115–17 de n.è)’ in Symposion 1985: Vorträge zur griechischen und hellenistischen Rechtsgeschichte, ed. G. Thuer (Cologne: Böhlau, 1985), pp. 337–61; Pucci Ben Zeev, Diaspora Judaism, pp. 186–90, 264. G. Gambash and A. Yasur-Landau, ‘Governor of Judea and Syria: a new dedication from Dor to Gargilius Antiquus’, Zeitschrift für Papyrologie und Epigraphik 205 (2018), 158–64.

347

gil gambash

The final event of the prolonged Roman–Jewish conflict was the Bar Kochba revolt, which, to a large extent, should be approached here similarly to the Diaspora revolt, despite the fact that it took place primarily in Judea. The starting points of both outbreaks of resistance are similar, combining a suspicious and untrusting administration and a frustrated and humiliated provincial population. Like his predecessor Trajan, Hadrian appointed his best generals to subjugate this revolt. His apprehensive approach may be seen also in the large number of Roman troops allocated to the campaign.67 These measures arguably say as much about the charged history of the relationship as they do about the actual scope of the opposition. At this point, the prevailing perception of the Jews as threatening and perennially resistant would have dictated the nature of the Roman response as much as the size of the force recruited by Bar Kochba. As with the Diaspora revolt, here too we have reports of widespread warfare and destruction, ascribed to both sides. Eusebius focuses, unsurprisingly, on the Jewish agency: ‘The Jews, who took up arms, devastated Palestine during the period in which the governor of the province was Tineus Rufus, to whom Hadrian sent an army in order to crush the rebels.’68 Another Christian writer, Sulpicius Severus, extends the scope of the revolt to Syria: ‘Then under Hadrian the Jews attempted to rebel, and endeavoured to plunder both Syria and Palestine.’69 The most extreme – and problematic – description comes from an eleventhcentury abridgement of Cassius Dio: ‘Fifty of their most important outposts and nine hundred and eighty-five of their most famous villages were razed to the ground; five hundred and eighty thousand men were slain in the various raids and battles, and the number of those that perished by famine, disease and fire was past finding out.’70 Dio adds that also many Romans perished in the war, and that the emperor refrained from opening his address to the Senate with the customary statement: ‘I and the army are in good health.’ As with other ancient reports containing such numbers, here also most historians dismiss the description, either as a gross exaggeration – possibly introduced by the editor of the epitome, Xiphilinus – or as an attempt of Dio himself to justify the losses suffered by the Romans.71 Scholars who tend to accept the 67

68 70 71

W. Eck, ‘The Bar Kokhba revolt: the Roman point of view’, Journal of Roman Studies 89 (1999), 76–89. Euseb. Chron. s.v. Hadrian 16. 69 S. Severus, Hist. Sacr. 2.31. Cass. Dio, Epitome 69.14. For example P. Schäfer, Der Bar-Kochba Aufstand (Tübingen: Mohr Siebeck, 1981); M. Mor, The Second Jewish Revolt (Leiden: Brill, 2016).

348

The Roman Empire and the Jews: Genocidal Perspectives

source, at least in spirit, see in it first and foremost testimony to the serious military challenge faced here by Rome.72 The fact that the fighting took place, as before, within the settlements themselves channels our discussion once again to the sphere of siege warfare and ‘urbicide’. The aftermaths of the Diaspora and Bar Kochba revolts were similar, demonstrating a modest scale of commemoration, far less grand and universal than that of the first revolt. There is, for example, no evidence of celebration of the victories of Quietus in Mesopotamia, though only a few months earlier the Senate had conferred on the emperor the title of Parthicus for his conquest of Ctesiphon.73 Instead, only private funerary monuments are erected referring to the campaign, suggesting that the revolt and its subjugation had not been publicised across the empire.74 Private monuments commemorating the subjugation of the Bar Kochba revolt may have been dedicated in Rome, and three generals are known to have received the ornamenta triumphalia.75 A triumphal arch was dedicated to Hadrian in Tel Shalem (12 kilometres south of Scythopolis) in honour of his victory, yet such a monument would not have been known outside the rebelling province itself. Even coinage, Hadrian’s most popular tool of propaganda, was not employed for the commemoration of the victory, leaving the empire largely in ignorance of its occurrence. Despite popular portrayals of the whole conflict as ruinous to the Jews, and the Bar Kochba revolt as the war that nearly put an end to Jewish life, Jewish communities continued to flourish, in Palestine as in the diaspora, while shifting the focus of their religious life from Jerusalem to other centres, such as Yavneh and Tiberias, and while producing, not long after the conclusion of the conflict, such prospering cities as Sepphoris, and such intellectual achievements as the Mishnah.

Conclusion This volume as a whole, looking as it does at a range of cases, may join the scholarly effort to generalise on those few enemies of empire with whom no room for reconciliation was left, according to Roman imperial logic. In such a consideration, the threatening image of those enemies should receive much 72 73 74 75

Eck, Bar Kochba. Cass. Dio 68.28.2. See Pucci Ben Zeev, Diaspora Judaism, pp. 212–17. For example L’Année epigraphique 1929, 167. W. Eck, ‘Kaiserliche Imperatorenakklamation und Ornamenta Triumphalia’, Zeitschrift für Papyrologie und Epigraphik 124 (1999), 223–7.

349

gil gambash

more attention than their actual potential to damage the empire. In the particular case of Rome and the Jews, all evidence points to the Roman perception of the Jews as a significant threat throughout the decades of the conflict. The evidence also demonstrates persistent Roman motivation to return to a state of peaceful routine with the Jews, according to provincial standard. In his important article published in 2004 in Past & Present, Martin Goodman refers to the foundation of Aelia Capitolina by Hadrian as a deliberate move against Jewish aspirations, aiming, uniquely for Hadrian, ‘not to flatter but to suppress the natives’. Goodman goes as far as to employ – very consciously – the emotional term ‘final solution’. He does so, though, only as an answer to a Jewish problem that he recognises as incessant rebelliousness. So: ‘Hadrian’s final solution to Jewish rebelliousness’.76 Employing such terminology is useful for the discussion developed here, since it encourages us to consider Roman– Jewish relations through Holocaust discourse. Rome no doubt faced a ‘Jewish problem’, but it was much more a practical problem than an ideological one – namely, incessant rebelliousness in Judea and the diaspora; and Rome no doubt searched for a ‘final solution’ for this problem – namely the pacification of the Jews and the restoration of their status to a standard provincial population. This, then, constituted the Roman goal, or in legal terms, motive. On the path from the destruction of the Temple in Jerusalem to the emergence of Yehuda Hanassi and the thriving Beit Shearim, mass and indiscriminate but intentional killing of Jews did occur, though likely on a scale more moderate than reported by the written sources. It happened in the context of half a century of long and aggressive conflict, mostly during three high points where actual war and fighting erupted. Of the various categories encompassed under ‘genocide’ in modern research, ‘ideological genocide’ appears irrelevant to our case, whereas ‘retributive genocide’ appears closest to describing the situation accurately. When employed, it was usually more in the form of ‘urbicide’ or even ‘oppidocide’ (coined here to denote the destruction of settlements smaller than cities) – each settlement being approached differently, depending on the levels of resistance it had presented. In any event, genocide was not directed against the Jewish people as such, and violence disappeared from view altogether with the disappearance of resistance and war, allowing a return to free and undisturbed routine – different from the one practised before the wars, but recognisably Jewish nonetheless. 76

Goodman, ‘Trajan’, 27.

350

The Roman Empire and the Jews: Genocidal Perspectives

Bibliographic Note Roman violence directed against Jewish groups occurred almost exclusively as part of military campaigns aimed at subjugating repeated revolts. Background reading, which may prove relevant also for other chapters in this volume, should thus include publications on the measures, tactics and conduct of the Roman Empire and Roman legions in the course of war, and the role ascribed to violence, as examined for example by The Cambridge History of Greek and Roman Warfare, ed. P. Sabin, H. van Wees and M. Whitby (Cambridge University Press, 2007). Also relevant are particular forms of violence practised outside of the immediate sphere of the battlefield, which may be observed as potentially leading to genocide: see for example Elisabeth Vikman’s ‘Ancient origins: sexual violence in warfare’, Anthropology and Medicine 12:1 (2005), 21–31. A specific form of warfare that presents itself as relevant for numerous events during the Roman–Jewish conflict is siege warfare, as discussed for example by Paul B. Kern in Ancient Siege Warfare (Bloomington: Indiana University Press, 1999). More specific treatment of genocide in classical antiquity has been attempted in the recent past, mostly as part of a broader discussion on genocide, where ancient history may serve as an early documented starting point for a chronologically oriented survey, or as a basis for comparison with modern representations of the phenomenon. Thus, The Oxford Handbook of Genocide Studies, edited by Donald Bloxham and A. Dirk Moses (Oxford University Press, 2010), includes under a section dedicated to ‘Premodern and early modern genocide’ a single chapter (pp. 239–57) on ‘Genocide in the ancient world’, authored by Hans van Wees. The section then proceeds to present medieval Europe and colonial America in four additional contributions. Van Wees provides a wide scope for antiquity, and employs the Roman Empire as a prominent example, but the case of the Jews as a distinct group is not included, and may mostly be considered in relation to his preferred model of punitive measures and revenge as the main cause for Roman violence of annihilation. On the other hand, his approach to urban and even rural communities as groups of distinct character and particular identity makes his use of the concept of ‘urbicide’ highly relevant for the reality of the Roman–Jewish conflict. When discussing genocide in the ancient world next to modern events, the risk of anachronism naturally intensifies. Seemingly unaware of current views regarding the ‘grand strategy’ of the Roman Empire as mostly non-existent, Smith ascribes destructive Roman violence to a well-devised plan, based on an imperial ideology of power, in ‘State power and genocidal intent’. Shawn

351

gil gambash

Kelley also examines ideological perspectives in the context of the Bible’s endorsement of genocide in his book on Genocide, the Bible, and Biblical Scholarship (Leiden: Brill, 2016). To be sure, when ideology is imagined in the context of the Roman–Jewish conflict, wider room emerges for the anachronistic employment of terms such as antisemitism, ideas of a civilisational clash, and notions of actual genocide – none of which is supported by the sources.

352

13

Religious Violence in the Later Roman Empire The Tetrarchic Persecutions, 302–313 C E carl r. rice

[S]oldiers surrounded a whole town in Phrygia that to a man was full of Christians, put it to the flame, and burned them along with their children and women as they cried out to the God who is over all. In fact, the whole populace living in the city, the logiste¯s himself and the strategoi and those in power and the whole populace confessed themselves Christians, refusing to be swayed by those who ordered them to worship idols. Eusebius, Historia ecclesiastica1

With these words, Eusebius of Caesarea described a massacre in central Anatolia in his account of the Tetrarchic persecution of Christians, which had begun in Spring 303 with a series of increasingly severe edicts. The assault in Phrygia was straightforward, yet brutal: on their leaders’ orders, Roman soldiers mercilessly targeted Christians simply for their refusal to worship the gods of Rome. Despite the violence, Christians steadfastly displayed their devotion and piety to their own God. The passage’s brevity, moreover, wherein Eusebius fails to identify the town, to name any of the victims or even to clarify the chronology, reveals just how enigmatic and elusive this episode really was, despite its long-lasting impact on Christianity. Scholars have dedicated volumes to understanding what has been called the Great Persecution.2 Yet, to my knowledge, no scholar has considered these events in the context of Genocide Studies. In this chapter I aim to situate the Tetrarchic persecutions of Christians within a longer, global history of genocide. I argue that although these persecutions may not satisfy modern juristic definitions of the term, which overwhelmingly emphasise the 1 Euseb. Hist. eccl. 8.11.1; Eusebius of Caesarea, The History of the Church: A New Translation, trans. J. Schott (Berkeley: University of California Press, 2019). The logiste¯s and the strategoi were municipal magistrates. 2 I prefer ‘Tetrarchic persecutions’ rather than the traditional ‘Great Persecution’.

353

Map 13.1 Selected sites in the late Roman Empire. (Cartography by Tom Elliott, with modifications by the Author)

Religious Violence in the Later Roman Empire

physical consequences of genocidal behaviour, the emperors nonetheless aimed to eliminate alternative systems of knowledge and to extirpate particular religio-cultural populations in the early fourth century C E, acts which share many features with other genocidal events in world history.3 Moreover, they employed tactics easily identifiable as ‘cultural genocide’ to do so. Based on this analysis, I suggest scholars consider more prominently means of sociocultural extermination alongside physical ones, an approach which serves well the types of sources that survive from premodernity.

The Sources While nothing like the detail-rich archives of a modern regime exists for the ancient world, in the case of the Tetrarchic persecutions, a great deal of information survives. Indeed, the early fourth century is an unusually welldocumented period of ancient history. Like any body of texts, however, these sources come with their own perspectives and biases, which require careful criticism and engagement to arrive at a cogent understanding of the period and its actors. As with any historical analysis, we must acknowledge the limitations of our sources and draw any potential conclusions with caution. This is especially true of the literary accounts of Lactantius and Eusebius, two prominent Christian writers who wrote in the immediate aftermath of the Tetrarchic persecutions as members of the victim group. Each was writing his work with his own aims and interests and in the context of a new Christian historiography that featured the relationship between Christianity and the Roman Empire and that saw the triumph of Christianity fulfilled by the inauguration of Constantine I’s reign.4 3 In recent decades many scholars have followed the work of J. Z. Smith (e.g. ‘Religion, religions, religious’ in Critical Terms for Religious Studies, ed. M. Taylor (University of Chicago Press, 1998), pp. 269–84) and have cautioned against looking for ‘religion’ in non-Western societies. The literature on this question is extensive. For a good overview of religion in the ancient Mediterranean, see R. S. Falcasantos, Constantinople: Ritual, Violence, and Memory in the Making of a Christian Imperial Capital (Berkeley: University of California Press, 2020), ch. 1. Of particular note are B. Nongbri, Before Religion: A History of a Modern Concept (New Haven: Yale University Press, 2013), and C. A. Barton and D. Boyarin, Imagine No Religion: How Modern Abstractions Hide Ancient Realities (New York: Fordham University Press, 2016). They are right to challenge modern scholars’ assumptions at the outset of our analyses. However, for convenience, I use the noun ‘religion’ and the adjective ‘religious’ to describe a pervasive element of private and public life comprised of a variety of social obligations and the networks they produce. 4 B. Colot, ‘Historiographie chrétienne et romanesque: le De mortibus persecutorum de Lactance (250–325 ap. J.C.)’, Vigiliae Christianae 59:2 (2005), 140–50, in fact, reads these two authors together and highlights the similarities of their projects.

355

carl r. rice

Lactantius was a Latin rhetorician who became a teacher in Nicomedia (modern-day Izmit, Turkey), an eastern imperial capital. In the years after the persecutions, Lactantius composed an important tract titled On the Deaths of the Persecutors, a text which T. D. Barnes has called an ‘anti-hagiography’ and ‘political satire’.5 In his treatise, which was an unapologetic polemic in response to the events of his own day, Lactantius quickly surveys the most notable Roman ‘persecutors’ – the emperors Nero, Domitian and Decius – and the grisly ends they met at God’s hands, before proceeding to a lengthy account of the Tetrarchic persecutions. Lactantius was closely allied with Constantine I after his rise to power in the persecutions’ aftermath, and, indeed, helped shape the increasingly Christian empire.6 Yet, because of his teaching post at Diocletian’s court at the outset of the persecutions, Lactantius had seen at first hand the early events of the Christian persecutions unfold. In other words, Lactantius provides a valuable report, but one steeped in a pro-Christian perspective. Eusebius of Caesarea likewise offers a firsthand perspective of the persecutions, though from the point-of-view of a provincial subject. As a cleric at Caesarea during the persecution, Eusebius witnessed its effects on Palestine’s Christian populations. Eusebius’ first relevant text, the Ecclesiastical History (also known as the History of the Church), presents a historical account of Christianity’s development from the time of Jesus to his own day, and includes a detailed treatment of the Tetrarchic persecutions in its eighth book. His second, the Martyrs of Palestine, offers a detailed, though literary, description of the individuals who died throughout Palestine during the Tetrarchic persecutions.7 In both texts (and, indeed, the rest of his oeuvre), Eusebius presents a theologically inflected view of historical development that promotes a narrative of Christian triumph.8 5 T. D. Barnes, Early Christian Hagiography and Roman History (Tübingen: Mohr Siebeck, 2010), pp. 115–16. 6 For a thorough treatment of Lactantius’ relationship with Constantine and his subsequent influence on the Christian Roman Empire, see E. D. Digeser, The Making of a Christian Empire: Lactantius and Rome (Ithaca, NY: Cornell University Press, 2000). Colot, ‘Historiographie chrétienne’, 135–51 has also argued that, although Lactantius’ account contains some fictionalised elements, we can still consider many parts authentic. 7 Eusebius, Bishop of Caesarea, The Ecclesiastical History and the Martyrs of Palestine, trans. H. J. Lawlor and J. E. L. Oulton (London: SPCK, 1927), I , pp. 327–400; hereafter Euseb. MP. 8 On Eusebius’ writings as historical and literary texts, see T. D. Barnes, Constantine and Eusebius (Cambridge, MA: Harvard University Press, 1981); A. Cameron, ‘Eusebius of Caesarea and the rethinking of history’ in Tria Corda: Scritti in onore di Arnaldo Momigliano, ed. E. Gabba (Como: Edizioni New Press, 1983), pp. 71–88; C. Rapp,

356

Religious Violence in the Later Roman Empire

In addition to Eusebius’ and Lactantius’ works, a genre of texts known as the martyr acts contain short narratives about Christians who faced persecution and died for their faith.9 Highly encomiastic in nature, they emphasise their subjects’ spiritual and religious fortitude. The textual histories of these accounts – when, by whom and for what purpose they were written – is often difficult to ascertain, leading historians to cast a sceptical eye on the literary and fantastical elements of such texts.10 Some scholars have, however, begun to turn increasingly to these texts for their perspectives – even highly stylised and symbolic ones – on early Christian communities. If we evaluate these texts individually and read them in relation to other documents, the martyr acts can offer a great deal of useful information to historians of early Christian persecutions, often supplementing other sources.11 A variety of documentary sources, particularly legal proclamations, edicts and constitutions, survive, though in a haphazard and scattered fashion. Some appear in the literary sources: both Lactantius and Eusebius include quotations of official documents (for example, both quote an edict by Galerius that ended the persecutions) that are sometimes verifiable by other means. Some texts also survive in special compilations, though these are usually the product of editorial selection and curated with a particular purpose in mind.12 Other documents, which allow a view beyond literary records, survive more organically. For instance, some papyri have emerged from Egypt that allow us to ascertain day-to-day experiences and not just the normative descriptions that abound in literature or law. In addition, some epigraphic material recording ‘Imperial ideology in the making: Eusebius of Caesarea on Constantine as bishop’, Journal of Theological Studies 49:2 (1998), 685–95; Colot, ‘Historiographie chrétienne’, 135– 51; E. D. Digeser, ‘Lactantius, Eusebius, and Arnobius: evidence for the causes of the Great Persecution’, Studia Patristica 39 (2006), 33–46; D. Singh, ‘Eusebius as political theologian: the legend continues’, Harvard Theological Review 108:1 (2015), 129–54; and, most recently, J. Corke-Webster, Eusebius and Empire: Constructing Church and Rome in the Ecclesiastical History (Cambridge University Press, 2019). The essays in A. P. Johnson and J. Schott, eds., Eusebius of Caesarea: Tradition and Innovations (Washington, DC: Center for Hellenic Studies, 2013), are also quite useful. 9 H. Musurillo, ed. and trans., Acts of the Christian Martyrs (Oxford: Clarendon Press, 1972). 10 C. R. Moss, Ancient Christian Martyrdom: Diverse Practices, Theologies, and Traditions (New Haven: Yale University Press, 2012), for example, cautions for close attention to the rhetorical and generic aims of these martyr acts. 11 G. W. Bowersock, Martyrdom and Rome (Cambridge University Press, 1995), pp. 23–40. For a more sceptical view, see É. Rebillard, ed., Greek and Latin Narratives about the Ancient Martyrs (Oxford University Press, 2017), pp. 15–17. 12 R. M. Frakes, Compiling the Collatio Legum Mosaicarum et Romanarum in Late Antiquity (Oxford University Press, 2011); hereafter Leg. Coll.

357

carl r. rice

legal texts has also survived. Though the famous ‘epigraphic habit’ was on the decline by the early fourth century, some individuals and municipalities inscribed important documents on stone for public display well into late antiquity.13 While the survival, discovery and publication of such texts varies by happenstance, several documents which confirm or complete aspects of the literary record have emerged, like Maximin’s response to anti-Christian municipal petitions.14 Although these sources are fragmentary and unevenly distributed, they provide an important counter-perspective to the literary accounts. In sum, this rich and varied body of evidence, while riddled with biases and lacunae, nonetheless offers a relatively clear view of state actions toward Christianity and Manichaeism between 302 and 313.

The World of Diocletian and the Tetrarchy When Diocletian rose to power in 284, he came to rule over an empire that had been rattled by economic, political and social strife for half a century.15 Early in his reign, he appointed his friend Maximian co-emperor; they were to share imperial rule, each as an Augustus, a title born by emperors of Rome since the empire’s inception. In 293, Diocletian and Maximian elevated two others, Galerius and Constantius, appointing them to the more junior rank of Caesar. The four emperors – two senior Augusti and two junior Caesares – each presided over a particular territory but ruled collectively as a single imperial college now known as the Tetrarchy, the ‘rule of four’ (Figure 13.1). As part of their shared rulership, the Tetrarchs developed an imperial ideology underpinned by traditional modes of Roman religion.16 Each Augustus adopted a divine honorific: Diocletian added the title Jovius to his name, after Jupiter (Jove in some Latin forms), while Maximian took on the 13 R. MacMullen, ‘The epigraphic habit in the Roman Empire’, American Journal of Philology 103:3 (1982), 233–346. 14 S. Mitchell, ‘Maximinus and the Christians in A.D. 312: a new Latin inscription’, Journal of Roman Studies 78 (1988), 105–24. 15 The best overviews of Diocletian and the Tetrarchy’s reign are S. Williams, Diocletian and the Roman Recovery (London: Batsford, 1985) and R. Rees, Diocletian and the Tetrarchy (Edinburgh University Press, 2004). J. Harries, Imperial Rome AD 284 to 363: The New Empire (Edinburgh University Press, 2012) and D. S. Potter, The Roman Empire at Bay, AD 180–395, 2nd ed. (London: Routledge, 2014) present good textbook overviews of the period. 16 Williams, Roman Recovery, pp. 153–62; P. S. Davies, ‘The origin and purpose of the persecution of AD 303’, Journal of Theological Studies 40 (1989), 92ff.; W. Portmann, ‘Zu den Motiven der Diokletianischen Christenverfolgung’, Historia: Zeitschrift für Alte Geschichte 39:2 (1990), 221–2; W. Kuhoff, Diokletian und die Epoche der Tetrarchie: Das römische Reich zwischen Krisenbewältigung und Neaufbau (284–313 n. Chr.) (Frankfurt am Main: Peter Lang, 2001), pp. 246ff.; B. Leadbetter, Galerius and the Will of Diocletian (London: Routledge, 2009), p. 120.

358

Religious Violence in the Later Roman Empire

Figure 13.1 Porphyry sculpture portrait of the Four Tetrarchs, c.300 C E. Originally in Constantinople, now at St Mark’s Basilica in Venice. (Photograph from Education Images / Getty Images)

name Herculius, after Hercules. The Tetrarchs also extensively built and restored traditional sacred sites at Rome and Carthage, among other significant cities in the empire.17 In some cases, like Diocletian’s capital at Nicomedia, the emperors visibly integrated religious and imperial ideologies in public building projects and art.18 17 For example Aur. Vict. Caes. 39.45; Lactant. De mort. pers. 7.8. 18 T. S¸ are Ağ türk, ‘A new Tetrarchic relief from Nicomedia: embracing emperors’, American Journal of Archaeology 122:3 (2018), 411–26 offers only one example. Numismatic evidence is also an important source for Tetrarchic ideologies, but I will not discuss the coins here. Rees, Diocletian and the Tetrarchy offers more on both art and coins.

359

carl r. rice

In Rome, religion and empire had long been closely intertwined. For example, one of the prominent roles Octavian Augustus – the first to successfully claim imperial power at Rome – assumed was that of pontifex maximus, the chief priest of Rome and a title held by nearly all subsequent Roman emperors. By the Tetrarchic period, this Roman political theology had also become diffuse and visible throughout society. For example, a papyrus from Egypt records an oath sworn by the emperors’ ‘heavenly fortune’.19 In the preamble to the 301 Edict on Maximum Prices, the emperors confirm this sentiment, claiming that they had attained ‘the most profound peace’ through their alliance with the Roman gods. The ‘benign favor of the gods’ had enabled the emperors’ successes as rulers.20 Such rhetoric suggests a profound conceptual link between proper religion and the Tetrarchy’s imperial rule. This profound engagement with traditional Roman religion met in the early fourth century with an increasingly visible Christian movement. Christianity had flourished in the empire since the 260s, when Gallienus had declared toleration for Christians following a spurt of persecutions. Christians served in the military and at the imperial court, while Christian communities grew in size and visibility within their cityscapes. Eusebius notes that some Christians had even attained positions of great influence at Diocletian’s court,21 while Lactantius asserts that the Christian community in Nicomedia had constructed a large and imposing church in plain view of the imperial palace.22 In other words, well into the Tetrarchic period, Christianity was an influential sociopolitical entity and enjoyed long periods of toleration, and even integration, in Roman society.

The ‘Great’ Persecutions Notably, for much of the Tetrarchy’s early rule, religious enforcement played no discernible role in governance; no significant legislation that enforced religious conformity survives or is attested in other sources. Moreover, for the first part of the Tetrarchic persecutions, the emperors’ actions can best be characterised as reactive to various circumstances: when the Tetrarchs adopted an anti-Manichaean measure (to which I will return below), they were responding to a letter from a senior administrator. It was not until 303 that they adopted a proactive policy towards religions they regarded as deviant. 19 20 21 22

P.Kell. 1.2, lines 5–6. Edict on Maximum Prices, praef. 5–6; Potter, Roman Empire at Bay, p. 291. Euseb. Hist. eccl. 8.1.1–3 on Christians generally in public. Lactant. De mort. pers. 12.3.

360

Religious Violence in the Later Roman Empire

The earliest recorded anti-Christian actions under the Tetrarchy occurred sometime before the persecutions officially began. According to Lactantius, while Diocletian was in the east, likely at Antioch in 299, he conducted a haruspicy ritual,23 which several Christians in the crowd disrupted with the sign of the cross.24 After trying the ritual again, the chief haruspex explained that the rites had failed because ‘profane persons were present’.25 Consequently, Diocletian demanded that everyone present perform a sacrifice, then turned the order to all those serving at the palace and in the army. All who refused were to be relieved of their posts. Lactantius reports that this purge of court and military officials remained small-scale: Diocletian, he writes, ‘did nothing further against the law and religion of God’.26 Tetrarchic officials also responded on occasion to breaches in military discipline, spurred by religious loyalty. Eusebius records some trials in military camps, wherein Christian soldiers were forced to choose between their faith and their career. The penalty, however, was usually not capital punishment; soldiers were often stripped of their military rank and honours and dismissed from service. A death sentence seems only to have been issued for other crimes.27 The Acts of Maximilian, set in 295, report for example that a military recruit named Maximilian refused his commission because of his Christianity. When the proconsul Dion called Maximilian forward, Maximilian remained steadfast in his conviction that he could not serve in the army. Pleading for Maximilian to take his oath, Dion suggested that Christianity and military service were not in fact incompatible. There were even Christians serving in the imperial bodyguard, argued Dion. If service was acceptable for them, he asked, why was it different for Maximilian? Dion ultimately ordered that Maximilian’s name be removed from the rolls and 23 On haruspicy, see M.-L. Haack, Les Haruspices dans le monde romain (Paris: Diffusion de Boccard, 2003), pp. 177ff. For the time and place, see E. D. Digeser, ‘An oracle of Apollo at Daphne and the Great Persecution’, Classical Philology 99:1 (2004), 57–77, Barnes, Constantine and Eusebius, pp. 18–19, and Kuhoff, Diokletian, p. 251. D. Woods, ‘Two notes on the Great Persecution’, Journal of Theological Studies N S 43:1 (1992), 128–31 argued for the earlier date of 297, but R. W. Burgess, ‘The date of the persecution of Christians in the army’, Journal of Theological Studies 47 (1996), 157–8 firmly rebuts his argument. 24 Lactant. De mort. pers. 10.1–2; Davies, ‘Origin and Purpose’, 90. 25 Lactant. De mort. pers. 10.3. 26 Lactant. De mort. pers. 10.5. Lactantius also offers an account of this episode in Divine Institutes 5.27.3–5, but it is shorter and contains fewer details. Here, however, he notes that Christians preventing pagan rites from working ‘has often been a main cause for bad kings to persecute justice’ (5.27.3). 27 Euseb. Hist. eccl. 8.4.1–4.

361

carl r. rice

sentenced him to death for ‘disloyally refus[ing] the military oath’.28 While he objected to his service because of his religion, Maximilian was punished specifically for refusing his service obligations. As T.D. Barnes has argued, he was a ‘conscientious objector’, executed for his refusal to serve in the Roman army.29 By and large, episodes like this seem to have occurred only rarely. In winter 302/3, however, Diocletian and Galerius met at Nicomedia. Lactantius reports that there they engaged in a series of discussions, where Galerius had come to ‘spur on the foolish old man [i.e. Diocletian] . . . to persecute the Christians’.30 At first, the two emperors met in secret, and no others were allowed in, raising questions of Lactantius’ reliability on the precise content of the discussions. Lactantius suggests that Diocletian initially resisted Galerius’ impulse to persecute. Diocletian seems to have had reservations about introducing discord to the realm and may have believed that Christians would willingly embrace death in pursuit of their religion. Whatever the reason for his resistance, Diocletian sought counsel from other advisors as well.31 Lactantius records that Diocletian, Galerius and the court heard anti-Christian arguments from a philosopher and a judge, as well as other court officials.32 This ‘conference’ brought together a number of 28 Acta Maximiliani 1–3; Acts of the Christian Martyrs, ed. and trans. Musurillo, p. 17. Musurillo calls this a ‘document of great antiquity’. 29 Barnes, Early Christian Hagiography, pp. 106–10. 30 Lactant. De mort. pers. 10.6ff. Leadbetter, Galerius, pp. 130–1 downplays Galerius’ superstition. 31 Lactant. De mort. pers. 11.3–7. 32 Lactant. Div. inst. 5.2.1–12. W. H. C. Frend, ‘Prelude to the Great Persecution: the propaganda war’, Journal of Ecclesiastical History 38:1 (1987), 1–18 details the ‘propaganda war’ leading up to the persecutions. T. D. Barnes, ‘Sossianus Hierocles and the antecedents of the “Great Persecution”’, Harvard Studies in Classical Philology 80 (1976), 239–52 identifies the judge figure as Sossianus Hierocles. E. D. Digeser, ‘Lactantius, Porphyry, and the debate over religious toleration’, Journal of Roman Studies 88:2 (1998), 129–46; J. Schott, ‘Porphyry on Christians and others: “Barbarian wisdom”, identity politics, and anti-Christian polemics on the eve of the Great Persecution’, Journal of Early Christian Studies 13:3 (2005), 277–314; E. D. Digeser, ‘Christian or Hellene? The Great Persecution and the problem of Christian identity’ in Religious Identity in Late Antiquity, ed. R. M. Frakes and E. D. Digeser (Toronto: Edgar Kent, 2006), pp. 36–57; E. D. Digeser, ‘The power of religious rituals: a philosophical quarrel on the eve of the Great Persecution’ in The Power of Religion in Late Antiquity, ed. A. Cain and N. Lenski (Burlington, VT: Ashgate, 2009), pp. 81–92; A. Smith, ‘Philosophical objections to Christianity on the eve of the Great Persecution’ in The Great Persecution: The Proceedings of the Fifth Patristic Conference, Maynooth, 2003, ed. D. V. Twomey and M. Humphries (Dublin: Four Courts Press, 2009), pp. 33–48; and E. D. Digeser, A Threat to Public Piety: Christians, Platonists, and the Great Persecution (Ithaca, NY: Cornell University Press, 2012) examine the role of philosophical debate leading up to and at this conference.

362

Religious Violence in the Later Roman Empire

anti-Christian campaigners in Nicomedia to advocate for the criminalisation and elimination of Christianity. Yet, even the arguments of those at Nicomedia could not persuade Diocletian to abandon his initial reluctance, so he sent a haruspex to the shrine of Apollo at Miletus for further guidance. Apollo, Lactantius writes, replied to the inquiry ‘as one would expect of an enemy of the religion of God’.33 Only when the gods appeared to approve of the proposed persecutions did Diocletian assent to act. Even then Diocletian insisted that no blood be shed. Instead, his actions would focus on the legal rights and social privileges of the Christians. The imperial administration chose a ‘suitable and auspicious day’ to begin their formal persecution: the festival of the Terminalia, dedicated to Terminus, the god who presided over boundaries and ends.34 At dawn on 23 February 303, the prefect of Nicomedia, several military officers and a few financial administrators marched to the city’s prominent church for what Jill Harries has called a ‘display of state terror’.35 There, as Diocletian and Galerius watched from the imperial palace, these men forced their way into the church, looted the property and confiscated and burnt all of the sacred books. Lactantius claims that Galerius argued that the church structure should also be burnt to the ground, but Diocletian feared a fire might spread through the city. Eventually, a troop of praetorian guards arrived and levelled the church to the ground.36 The following day, Lactantius reports that officials posted the first formal anti-Christian edict in Nicomedia. This edict deprived Christians of ‘all official position and status’ and subjected them to tortures according to their social rank; the edict further limited their legal capacities by preventing them from defending themselves in court and from bringing suits against others.37 Eusebius adds that imperial freedmen would lose their freedom and face enslavement once more. He also claims that these letters ordered ‘the churches smashed to the ground’ and ‘writings destroyed by fire’.38 In sum, the edict seems to have ordered that Christians were to be stripped of their social status, put to torture and deprived of their legal rights, while their gathering places and sacred books were to be destroyed. Two texts from Africa provide a glimpse into the mechanics of enforcing this edict. On 19 May 303, Munatius Felix, curator of the African town of Cirta, and his coterie arrived at the Christians’ gathering place to enact the edicts. 33 Lactant. De mort. pers. 11.7. 34 Lactant. De mort. pers. 12.1. 35 Harries, Imperial Rome, p. 88. 36 Lactant. De mort. pers. 12.2–5. 37 Lactant. De mort. pers. 13.1. 38 Euseb. Hist. eccl. 8.2.4.

363

carl r. rice

The city’s bishop responded that the community’s readers had the sacred books with them, but offered the property available at the church, initially surrendering some gold, silver and textiles. A second search of the premises, however, revealed a silver box and lamp hidden in a cupboard. Munatius Felix pressed harder, continuing to demand that books be turned over. Catullinus, a subdeacon, produced a single volume, though he insisted that the readers had the rest.39 The magistrates eventually identified, located and interrogated the readers, collecting a few additional volumes.40 Likewise, a papyrus from Oxyrhynchus, in Egypt, provides another glimpse into the execution of the first edict there. The papyrus, dated 5 February 304, contains an official report wherein a church reader from Middle Egypt, a certain Aurelius Ammonius, responded to government officials: Whereas you (pl.) commanded me in accordance with what was written . . . about the surrender of all the goods in the same former church and whereas I declared that the same former church had neither gold nor silver nor money nor clothes nor cattle nor slaves nor building-sites nor possessions, neither from gifts nor from bequests, apart from only the bronze matter which was found and given over to the logiste¯s in order to be brought down to the most glorious Alexandria in accordance with what was written . . .41

Ammonius’ statement that all goods in the ‘former church’ had been surrendered supports Eusebius’ and Lactantius’ assertions that churches were destroyed or, at the very least, shuttered. Annemarie Luijendijk suggests, moreover, that another claim in this report raise suspicion. First, Aurelius Ammonius claims that the church had only ‘bronze matter’ (chalke¯n hyle¯n) to surrender to imperial officials. The church was probably a small one, but, is it really likely only to have owned some ‘bronze matter’? Probably not. It seems plausible that Ammonius and other church officials hid other types of movable property to prevent it from being confiscated.42 These examples demonstrate that imperial officials sought out churches and clergy to enforce the terms of the edict. They appear to have had 39 Note: the ‘readers’ were a junior grade of the clergy. 40 Gesta apud Zenophilum 2–5, in Optatus: Against the Donatists, ed. and trans. M. Edwards, Translated Texts for Historians 27 (Liverpool University Press, 1997), pp. 150–69, esp. pp. 153–6. 41 P.Oxy. XXXIII 2673; translation in A. Luijendijk, ‘Papyri from the Great Persecution: Roman and Christian perspectives’, Journal of Early Christian Studies 16:3 (2008), 344–5. 42 Luijendijk, ‘Papyri’, 352.

364

Religious Violence in the Later Roman Empire

something like a checklist of materials they were looking for that included gold, silver, money, clothes, cattle, enslaved persons and so on – movable, saleable goods – as well as immovable real estate. When the magistrates arrived, Christians responded in a variety of ways. Some resolutely withstood tortures and faced death as they refused the order, while others seem to have put up little or no defence. Others seem to have found crafty solutions to mitigate the immediate situation. Responses to the surrender of church goods and debates about how to deal with those who had handed over books would, in subsequent decades, ultimately lead to the Donatist Controversy, a schism that rocked the Western Church throughout the fourth and early fifth centuries.43 The consequences of this first edict have elicited a great deal of scholarly discussion. G. E. M. de Ste Croix noted that the first edict ‘forbade collective practice’, while T. D. Barnes emphasised the bans on public worship and enforced performance of a ‘token act of conformity to the established religion’.44 Stephen Williams considered the edict a ‘tactical approach’ that did not create many martyrs, but rather focused on rituals and gatherings, though resistance did result in some executions.45 I will add that the terms of this first edict operated on two fronts: first, the order stripped Christians of certain legal rights and privileges and deprived them of their social rank and status. In short, it alienated them from Roman law, an important punishment for deviant religionists in late antiquity and one that had profound consequences for them. Second, the edict prohibited Christian meetings and ordered the confiscation and destruction of sacred sites and holy texts, targeting the maintenance and transmission of Christian knowledge, an important point to which I will return below.46 43 W. H. C. Frend, The Donatist Church: A Movement of Protest in Roman North Africa (Oxford University Press, 1971) is still the standard treatment, though it is somewhat dated. For more recent perspectives, see B. D. Shaw, Sacred Violence: African Christians and Sectarian Hatred in the Age of Augustine (Cambridge University Press, 2011) and J. Hoover, ‘Cyprianus Plebi Cartagini Consistenti and the origins of Donatism’, Journal of Early Christian Studies 26:3 (2018), 433–61, which draws links to earlier anti-Christian persecutions. N. Lenski, ‘Imperial legislation and the Donatist Controversy: from Constantine to Honorius’ in The Donatist Schism: Controversy and Contexts, ed. R. Miles (Liverpool University Press, 2016), pp. 166–219 considers the imperial interventions in the Donatist Controversy through the fourth century. 44 G. E. M. de Ste Croix, ‘Aspects of the “Great” Persecution’, Harvard Theological Review 47:2 (1954), 77; Barnes, Constantine and Eusebius, p. 22. 45 Williams, Roman Recovery, p. 176. 46 Cf. Ste Croix, ‘Aspects’, 75–6; P. Keresztes, ‘From the Great Persecution to the peace of Galerius’, Vigiliae Christianae 37:4 (1983), 382; S. Corcoran, The Empire of the Tetrarchs: Imperial Pronouncements and Government, AD 284–324 (Oxford: Clarendon Press, 2000), pp. 179–81; and Potter, Roman Empire at Bay, p. 330.

365

carl r. rice

In the days following the first edict, a mysterious fire broke out in the imperial palace at Nicomedia, and a rumour claiming that Christians had started the fire quickly ensued.47 Lactantius suggests that Galerius had employed secret agents to start the fire, blaming Christians to incite Diocletian to take further action against them. A few days later, a second fire struck.48 However the fires started, Diocletian was furious; he tortured members of the imperial household before turning to a series of public trials.49 Eusebius describes the executions that followed: ‘Masses of the whole race of God-fearing people’ met their fates at the end of a sword, tied to a stake or cast into the sea. Others experienced myriad other ‘unheard of kinds of tortures’.50 Whether or not the Christians did in fact set the fires as an act of resistance or were merely framed as scapegoats will remain an unsolved and unsolvable mystery, but the consequences seem to have been severe. The trials, tortures and executions of Christians in the imperial palace and throughout Nicomedia represent a widespread and purposeful intensification of violence against Christians under the Tetrarchic regime. Sometime later in 303, likely in late spring or early summer, reports of political unrest in Melitene and Syria reached the imperial court. According to Eusebius, Diocletian responded with ‘an imperial command [that] went out to put the presidents of the churches everywhere chained in prison’.51 This second edict appears not to have been promulgated or enforced in the western half of the empire. Lactantius claims that in Gaul, Constantius did not act beyond a nominal compliance with the first edict, though Christian authors had good reason to cast the father of the future emperor Constantine I in a good light.52 Christian leaders’ concern in the west with the treatment of traditores (Christians who had handed over sacred books and property under the terms of the first edict) underscores their comparative lack of concern with Christians complying with edicts after the first one. In the east, however, the situation was far more grim. Eusebius’ Martyrs of Palestine describes some clergymen arrested and martyred under the terms of the second edict. For example, Eusebius reports that Procopius, the province’s first clerical martyr, had served as a church reader, a Syriac interpreter and an exorcist for the Lactant. De mort. pers. 14.2; Euseb. Hist. eccl. 8.6.6. Lactant. De mort. pers. 14.1–7 notes the second fire (absent in Eusebius’ account). Lactant. De mort. pers. 14.1–4. Euseb. Hist. eccl. 8.6.7. ‘Unheard of kinds of tortures’ is Lactant. De mort. pers. 15.4, where he confirms the penalties in Eusebius’ account. 51 Euseb. Hist. eccl. 8.6.8. Lactantius does not separately report this edict or the following. 52 There seems to have been regional differences in enforcement throughout the empire. 47 48 49 50

366

Religious Violence in the Later Roman Empire

church at Scythopolis. When he was arrested, in late May or early June 303, he was brought before an imperial administrator in Caesarea and ultimately sentenced to death.53 The mass of arrests under the second edict had an unintended consequence, which resulted in a third edict. Eusebius reports that prisons became so filled with Christian clergy that new orders, which permitted Christians to be freed if they performed a sacrifice but subjected them to various tortures if they refused, arrived shortly thereafter.54 Eusebius records the gruesome tortures some who resisted faced: the rack, lashes, lacerations and myriad other punishments.55 As Stephen Williams has observed, these edicts were ‘striking at the head’ of the Christian communities throughout the eastern empire, by targeting the leaders and teachers.56 The community’s leadership was responsible for both managing its affairs and promulgating the community’s bases of knowledge and culture and sense of self through education, initiation and pastoral work. In other words, these edicts aimed to stifle Christianity’s leadership and prevent it from helping the Christian community maintain and reproduce itself. And, for a time, the mandate to sacrifice was, in fact, confined to the Christian leadership. However, early in the following year a fourth edict was issued.57 The fourth law represents a deliberate escalation in severity of the anti-Christian measures. Eusebius notes that this new edict extended to the entire population what had before then been kept to the clergy: an order for all subjects of the empire to participate in the sacrifice.58 Stephen Williams has suggested that this fourth edict declared Christianity a religio illicita, an ‘impermissible religion’, though this may be an overstatement of the legal scenario.59 Whatever its phrasing, the edict marks the extension of the prior policy of forcing clergy to perform non-Christian rites to the general population. No evidence that the edict was enforced in the western parts of the 53 Eusebius, MP (L) 1.1–2. 54 Euseb. Hist. eccl. 8.6.9. 55 Eusebius, MP (S) 1.3. 56 Williams, Roman Recovery, p. 178. 57 Keresztes, ‘From the Great Persecution’, 383 notes the questions about precise date of issue and publication across the eastern empire. 58 Eusebius, MP (L) 3.1. See also MP (S) 3.1. Lactant. De mort. pers. does not explicitly cite this edict, nor does it appear in Euseb. Hist. eccl. In some ways, this mirrors the so-called ‘Decian persecutions’ of 250. See J. B. Rives, ‘The decree of Decius and the religion of empire’, Journal of Roman Studies 89 (1999), 135–54 and R. Selinger, The Mid-Third Century Persecutions of Decius and Valerian (Frankfurt am Main: Peter Lang, 2004). 59 Williams, Roman Recovery, p. 182. For ‘religio licita’, see Tertullian, Apologeticum 21.1, where he describes Roman attitudes towards Jews in juxtaposition to Roman mistreatment of Christians. This is this phrase’s only attestation in Latin literature.

367

carl r. rice

empire exists.60 The law spread, however, throughout the east, and, as Eusebius reports, many Christians died in Palestine, Egypt and Asia. It was likely in response to this law that the account of soldiers slaughtering a town in Phrygia with which I began this chapter occurred, if it was indeed a historical event.61 There seem to have been many ways of identifying those who defied the edict’s commands. Christians could be singled out in court by refusing to participate in a ‘sacrifice test’.62 Roman authorities seem to have continued to require sacrifice in court under the terms of the first edict and used this device to identify potential Christians they might interrogate under the terms of the fourth edict. In a private letter from Egypt, a Christian named Copres wrote to his wife at home in Oxyrhynchus about his business trip.63 In a glimpse into their personal affairs, he wrote: I want you to know that we arrived on the 11th and it was made known to us that those who appear in court are compelled to sacrifice and I made a power of attorney to my brother and until now we have accomplished nothing but we have instructed an advocate on the 1?th, so that the matter about the arourai might be brought into court on the 14?th.64

Upon arriving at his destination and discovering that all litigants were required to perform a rite, Copres appointed someone willing to perform the ritual – his ‘brother’, though the nature of this relationship is far from clear – to conduct official business on his behalf, thus evading the performance of sacrifice himself and his subsequent identification as a Christian. Through the following year, things seem to have continued apace, until, after suffering an illness, Diocletian abdicated in 305. The following year the newly elevated eastern Caesar, Maximin, renewed the persecution’s vigour when he established a formal protocol for ensuring everyone fulfilled their legal obligations to sacrifice. Eusebius ordered civic magistrates to summon the cities’ inhabitants, to go door to door if necessary, to register the citizenry, and to ensure their participation.65 Maximin had employed the 60 Barnes, Constantine and Eusebius, p. 24. 61 Euseb. Hist. eccl. 8.11.1. 62 See G. E. M. de Ste Croix, ‘Why were the early Christians persecuted?’, Past & Present 26 (1963), 6–38 and G. E. M. de Ste Croix, ‘Why were the early Christians persecuted? – a rejoinder’, Past & Present 27 (1964), 28–33; A. N. Sherwin-White, ‘Why were the early Christians persecuted? – an amendment’, Past & Present 27 (1964), 23–7. 63 Luijendijk, ‘Papyri’, 358ff. 64 P.Oxy. XXXI 2601; translation Luijendijk, ‘Papyri’, 357–8. Note: the question marks represent textual problems like lacunae or illegibility. An aroura was a unit of land about one hundred cubits square (LSJ, s.v. ἂρουρα II). 65 Euseb. MP (L&S) 4.8, 9.1.

368

Religious Violence in the Later Roman Empire

various civic and state administrative apparatuses to enforce the new legal requirement. In time, the persecutions, though still active under the law, ultimately began to stagnate. According to both Lactantius and Eusebius, Galerius took ill in 311 or so. In an attempt to reverse the illness, which Lactantius describes in graphic detail, Galerius issued an edict on 30 April 312 that brought an end to the persecutions.66 In this edict, Galerius, whom Lactantius had named the persecution’s driving force, revoked the earlier laws and restored to the Christians their legal privileges and their right to assemble. In other words, Galerius tried to extend toleration to his Christian subjects in exchange for their prayers for the health and safety of the empire and its emperors.67 Maximin, however, did not strictly adhere to the orders of his senior emperor. Rather than passing the letter along to his own subordinate magistrates, he conveyed its contents orally to them.68 He also took great pains to promote traditional polytheism throughout his territories. He supported temple restorations, and he appointed a centralised hierarchy of priests in each city and province.69 He even solicited various municipalities to petition for the expulsion of Christians from their territories and, throughout the east, civic officials posted their requests and Maximin’s reply on public monuments; Eusebius saw and quoted the one posted at Tyre.70 Eusebius paints a bleak picture of these letters’ aftermath. In this last bout of persecutions, he writes, many Christians suffered, while famine, plague and war devastated the east.71 Moreover, Constantine and Licinius, Maximin’s fellow emperors, quelled a usurpation in the west and turned their sights on Maximin. Maximin met Licinius on a battlefield in Thrace in 313 and suffered a crushing defeat. If Eusebius is to be believed, Maximin took flight and eventually turned away from his traditional paganism to embrace the very Christianity he had so virulently railed against, issuing one last law which finally revoked all previous anti-Christian legislation. He died of a devastating illness shortly thereafter.72 Later that year at Nicomedia, Licinius posted the so-called Edict of Milan, a statement of religious toleration he and Constantine had articulated the year before, bringing to a formal

66 67 69 70 71 72

Lactant. De mort. pers. 33.1ff; Euseb. Hist. eccl. 8.17.2ff. Lactant. De mort. pers. 34.5; Euseb. Hist. eccl. 9.1.1. 68 Euseb. Hist. eccl. 9.1.1–11. Lactant. De mort. pers. 36.4; Euseb. Hist. eccl. 8.14.8–9. Euseb. Hist. eccl. 9.2.1–4.15. See also Mitchell, ‘Maximinus and the Christians’, 105–24. Euseb. Hist. eccl. 9.6.1–4; 9.8.1–15. Euseb. Hist. eccl. 9.10.7ff; Lactant. De mort. pers. 49.5–7.

369

carl r. rice

end the Tetrarchic persecutions of Christians and securing Christianity’s place in the empire’s religious landscape.

The Tetrarchic Persecutions: A Late Roman Genocide? Why should a volume on genocide in world history include a chapter on the Tetrarchic persecutions of the early fourth century? The number of victims killed may have been quite small in comparison to the large-scale atrocities of the twentieth century, and verifying even this comparatively small number of deaths is difficult after so much time has passed. Numbers aside, however, the destruction of the town in Phrygia echoes many other dark moments in history. So, the question stands: should these events be considered a late Roman case of genocide? To answer this question, we must consider what most fundamentally constitutes a genocide. Two surveys show that most definitions hold that genocide involves the intentional destruction of one group by members of another. Each of these keywords – ‘intentional’, ‘destruction’ and ‘group’ – raises significant questions,73 but this core definition enables us to interrogate a particular historical event as a potential genocide; simply put, we ask: did the Roman imperial government intend to eliminate and destroy members of the empire’s Christian community? And, how does asking the question this way add to our understanding of the Tetrarchic persecutions? Previous scholars have spent a great deal of time considering the emperors’ motives, asking what about Christianity drove the emperors to issue their edicts. They have adduced several plausible answers. Many have emphasised what we might call religious motivations: Christian refusal to worship pagan gods and attempts to restore traditional religion or to correct the obstruction of pagan rites, for example.74 Others have seen the persecutions in the context of the Tetrarchy’s role as administrators and keepers of public order.75 Others still have pointed to philosophical disagreements and 73 S. Straus, ‘Contested meanings and conflicting imperatives: a conceptual analysis of genocide’, Journal of Genocide Research 3:3 (2001), 350–5, and A. Jones, Genocide: A Comprehensive Introduction, 3rd ed. (London: Routledge, 2017), pp. 22–9 compile the definitions. At 362–7, Straus critically reviews ‘intent’, ‘annihilation’ and ‘group’ as conceptual categories and potential problems with each. 74 Ste Croix, ‘Why?’, 27; W. H. C. Frend, Martyrdom and Persecution in the Early Church: A Study of Conflict from the Maccabees to Donatus (New York University Press, 1967); Williams, Roman Recovery, p. 175; Davies, ‘Origin and purpose’, 92. 75 Portmann, ‘Zu den Motiven’, 247; M. Humphries, ‘The mind of the persecutors: “By the gracious favour of the gods”’ in Great Persecution, ed. Twomey, and Humphries, pp. 11–32.

370

Religious Violence in the Later Roman Empire

fundamental differences in epistemic world views that decisively othered one group and spurred the emperors to action.76 There is likely an element of truth in each of these motives. Yet, thinking about the Tetrarchic persecutions through a lens informed by Genocide Studies urges us to look carefully at the intent of the Tetrarchs’ actions as well. We must ask not only why the Tetrarchs acted but also what they aimed to accomplish in doing so. The intent behind any past action is notoriously difficult to recover, but one of the most obvious ways to do so is to examine the documents left behind by the actor in question. Yet, the edicts which launched the Tetrarchy’s actions against Christians no longer survive except in paraphrase in Christian sources. A nearly contemporaneous document concerning the Manichaeans does survive, however, and its similarities beg comparison with what we know about the anti-Christian measures. The year before they issued the anti-Christian edicts, the emperors had also issued an anti-Manichaean rescript to Julianus, the proconsul of Africa.77 Several primary sources, and indeed some archaeological evidence, suggest that, by the start of the fourth century, Manichaeism, like Christianity, was a prominent religious community in the Roman east.78 It is unsurprising, then, that this increasingly visible group came to the emperors’ attention. The contents of Julianus’ initial enquiry no longer survive, so we cannot be sure if he was responding to a particular event or generally to Manichaeans’ presence.79 Whatever the reason, Julianus was concerned enough to ask for guidance, and the emperors felt obliged to answer. Their reply was posted in Alexandria in March 302 and a copy survives in a later fourth-century legal compilation known as the Collatio Legum Mosaicarum et Romanarum, which 76 Digeser, ‘Power of religious rituals’, 81–92; Schott, ‘Porphyry’, 277–314; Digeser, Threat to Public Piety. 77 A ‘rescript’ was an imperial response to a formal inquiry. 78 P. Brown, ‘The diffusion of Manichaeism in the Roman Empire’, Journal of Roman Studies 59 (1969), 92–103; S. N. C. Lieu, Manichaeism in the Later Roman Empire and Medieval China (Tübingen: Mohr Siebeck, 1992); and S. N. C. Lieu, Manichaeism in Mesopotamia and the Roman East (Leiden: Brill, 1994) are standard treatments of Manichaeism in the Roman Empire. I. Gardner and S. N. C. Lieu, eds., Manichaean Texts from the Roman Empire (Cambridge University Press, 2004) offers various texts concerning Manichaeans in translation (see especially the excerpts at pp. 114–16). On community in the archaeological record, see I. M. F. Gardner and S. N. C. Lieu, ‘From Narmouthis (Medinet Madi) to Kellis (Ismant El-Kharab): Manichaean documents from Roman Egypt’, Journal of Roman Studies 86 (1996), 146–69. 79 Williams, Roman Recovery, p. 153 suggests a ‘xenophobic atmosphere’ and their purported Persian connections, while Portmann, ‘Zu den Motiven’, 223 argues that Manichaeans may have been causing unrest in Africa.

371

carl r. rice

juxtaposes legal passages from the Hebrew Bible with selections of Roman law.80 In the rescript, the emperors excoriated Manichaeism, decrying the spread of this newfangled religio into the Roman world. Condemning their deviance, the emperors suggested that Manichaeans were guilty of the ‘greatest crime’ as they tried to supplant the Romans’ praiseworthy ‘old religion’ with ‘new belief’.81 As punishment for this ‘crime’, the emperors condemned the recalcitrant Manichaeans to ‘toils and well-earned and fitting penalties’.82 Manichaean ‘leaders and founders’ were to be burnt alongside their ‘abominable writings’, while obstinate followers faced death and their property was confiscated for the imperial fisc.83 Converts to this ‘wicked and entirely infamous sect’ who held imperial office or were people of status or rank lost their patrimonies and were condemned to hard labour.84 These measures were severe and not only physically punished Manichaeans but also deprived them of their place in Roman society. Two significant parallels emerge when we juxtapose these two cases. First, in both cases, the state employed physical violence to punish those who deviated from permissible modes of traditional Romanness and Roman religio. In the Christian case, we can be fairly certain that some Christians died over the course of the persecutions. For the Manichaeans, however, we unfortunately have no data to indicate the extent to which the state carried out the measures. However, the rescript’s rhetoric – which referred to Manichaeism as a ‘crime’ and an ‘evil of worthlessness’ and which associated Manichaeism with its purported Persian origins – clearly indicates that the emperors viewed Manichaeism as fundamentally alien to Romanness. And, indeed, Christians’ refusal to assimilate to certain elements of Roman rites may have inspired similar sentiments about them. Second, the emperors also specifically targeted community leaders and means of knowledge transmission. The rescript ordered that Manichaean leaders were to be burnt at the stake, while the first anti-Christian edict closed churches and the second and third edicts targeted Christian leadership. In each case, the emperors also ordered sacred books to be confiscated and destroyed, as well. Both sacred texts and the religious leadership were essential for the 80 Coll. Leg. 15.3. On the rescript’s date, Brown, ‘Diffusion of Manichaeism’, 92–103 suggests 297. More recently, L. D. Bruce, ‘Diocletian, the proconsul Julianus, and the Manichaeans’, Studies in Latin Literature and Roman History 3 (1983), 336–47 does as well. However, Barnes, ‘Sossianus Hierocles’, 246–50 lays out a compelling case for 302, and I am inclined to agree. On the Collatio in general, see Frakes, Compiling. 81 Coll. Leg. 15.3.2. Frakes’ translation. 82 Coll. Leg. 15.3.5. 83 Coll. Leg. 15.3.6. 84 Coll. Leg. 15.3.7.

372

Religious Violence in the Later Roman Empire

reproduction of Manichaean and Christian communities. Texts transmitted written knowledge from generation to generation, while the clergy mediated interpretation of these texts and its instruction to new members of the community. In the Christian case, churches and their affiliated properties were the locus at which these activities took place. Eliminating clerical leaders, sacred texts and holy places threatened the community’s capacity to reproduce itself as a culturally distinct entity within the Roman world. Simply put, in the case of both the Manichaean and the Christian, it appears that the imperial state apparatus aimed to eliminate these subject populations. This endeavour was not to be accomplished only through killing, but, just as importantly, through the destruction of important sociocultural nodes – communal leaders, books and gathering places – within the victim communities. Most definitions of genocide privilege physical destruction; for example, Article II of the 1948 Genocide Convention enumerates ‘killing’, ‘causing serious bodily or mental harm’, inflicting conditions ‘calculated to bring about [a group’s] physical destruction’, preventing new births and transferring children into another group, all with the ‘intent to destroy’, as the actionable offences of a genocide. This definition results largely from the advocacy of Raphael Lemkin in the aftermath of the Holocaust. In his original conceptualisation of genocide, however, Lemkin also envisioned cultural destruction as one of several possible types of genocide. Unfortunately, these alternative forms of genocidal behaviour were ultimately rejected by the United Nations in their deliberations when member states adopted the Convention.85 Recently, proponents of the study of cultural genocide have begun to emphasise again the important role of social and cultural institutions, relationships and customs for maintaining a group’s identity. Like Lemkin, such scholars have suggested that attempts to hinder a group’s cultural and social reproduction are equally genocidal, shifting away from the juridical focus on acts of killing and physical destruction.86 Recognising that cultural 85 R. Lemkin, Axis Rule in Occupied Europe: Laws of Occupation, Analysis of Government, Proposal for Redress (Washington, DC: Carnegie Endowment for International Peace, 1944), pp. 79ff. 86 Here I follow J. S. Bachman, ‘Introduction: bringing cultural genocide into the mainstream’ in Cultural Genocide: Law, Politics, and Global Manifestations, ed. J. S. Bachman (London: Routledge, 2019), p. 4. See also C. Powell, ‘What do genocides kill? A relational conception of genocide’, Journal of Genocide Research 9:4 (2007), 527–47; D. Short, ‘Cultural genocide and indigenous peoples: a sociological approach’, International Journal of Human Rights 14:6 (2010), 833–48; L. Kingston, ‘The destruction of identity: cultural genocide and indigenous peoples’, Journal of Human Rights 14:1 (2015), 63–83.

373

carl r. rice

destruction can be just as devastating as the physicality of slaughter, the paradigm offered by Cultural Genocide Studies moves beyond mass killing to encompass forms of cultural annihilation, the examination of which is often more readily available in the cultural artefacts that survive from premodernity. Ultimately, by approaching the Tetrarchic persecution from the perspective of Genocide Studies, it becomes clear that Roman imperial officials employed methods of both physical and cultural destruction, with the intent to kill members of subject religious communities as well as to disrupt these communities’ sociocultural reproduction. At the very least, the Tetrarchic persecutions fall squarely under the category of ‘cultural genocide’ and encompass a period of violent physical devastation.

Conclusions I have argued that the Tetrarchic persecutions fit into a longer pattern of genocidal behaviours in world history. The late Roman government enacted harsh policies designed to eliminate from their world particular subject populations as defined by religious affiliation. These policies ordered the deaths of living, breathing individuals; moreover, and just as sinisterly, the state sought to dismantle the systems of knowledge and the cultural practices that supported the Christians’ and Manichaeans’ communal identities. Thinking about the Tetrarchic persecutions in the context of Genocide Studies brings the persecutors’ motives (why?) and intention (to what end?) into sharper relief. Such an analysis stresses the destructive element of the anti-Manichaean and anti-Christian measures for both human bodies and cultural systems. Moreover, rethinking the intent behind actions encourages a reconsideration of the methods the actors employed. Ultimately, they enacted violent injunctions that included deliberate strategies to kill parts of both groups and to destroy the communities’ cultural foundations. In other words, the Tetrarchic efforts to wipe out two minority communities by outlawing and destroying both human life and cultural content constitutes the most central elements of a genocide.

Bibliographic Note On Diocletian and the Tetrarchy generally, Rees, Diocletian and the Tetrarchy provides a good starting place and includes a short chapter on Tetrarchic religion. Harries, Imperial Rome and Potter, Roman Empire at Bay offer easily 374

Religious Violence in the Later Roman Empire

accessible surveys of the period. Somewhat more technical in nature, Barnes, Constantine and Eusebius and Corcoran, Empire of the Tetrarchs explore the Tetrarchy’s administrative structures (including religious and ideological matters). Williams, Diocletian and the Roman Recovery emphasises Diocletian’s reforms, including religious ones, in the wake of the ‘Third Century Crisis’. On the question of the so-called Great Persecution of Christians, Ste Croix, ‘Aspects of the “Great” Persecution’, is still a standard starting point. Kerestzes, ‘From the Great Persecution to the peace of Galerius’, offers a good analytical narrative, and Frend, Martyrdom and Persecution in the Early Church, includes the Tetrarchic persecutions in his overview of early Christian martyrdom and persecution. Frend’s ‘Prelude to the Great Persecution’ also examines the intellectual build-up – the ‘propaganda war’ – to the persecutions. Davies, ‘Origin and purpose of the persecution of AD 303’, critically challenges Lactantius’ account. Elizabeth DePalma Digeser is perhaps the most prolific scholar working on the Tetrarchic persecutions of Christians; her work includes several articles and a monograph (Threat to Public Piety) that examine several aspects of the persecutions. She emphasises the role of philosophical discourses in the build up to and as justification for the persecutions. Brown, ‘Diffusion of Manichaeism’, is an important contribution on Manichaeans in the Roman Empire, as is Lieu, Manichaeism in the Later Roman Empire and Medieval China.

375

14

Genocide, Extermination and Mass Killing in Chinese History victoria tin-bor hui

You have to really go into something before you can understand it . . . I started leafing through a history book . . . [S]crawled this way and that across every page were the words benevolence, righteousness, and morality . . . I read that history very carefully . . . and finally I began to make out what was written between the lines; the whole volume was filled with a single phrase: eat people! Lu Xun, ‘Diary of a madman’, 19181

It may be surprising to include China in a volume on genocide in world history. From premodern emperors to President Xi Jinping, Chinese leaders have claimed to rule in the same Confucian ways that parents treat their own children. Yet, the New Culture writer Lu Xun (1881–1936) pointedly identified the hypocrisy between the ideal of benevolence and the reality of ‘eating people’. Indeed, Lu lived in the early decades of what R. J. Rummel in 1991 called ‘China’s bloody twentieth century’.2 After the fall of the last Qing dynasty, competing warlords slaughtered some 910,000 people outside of battles. The Japanese, Rummel estimated, murdered 4 million Chinese during their invasion. Mass killings were not limited to usurpers and intruders. China’s two most renowned legitimate leaders of the century, Chiang Kai-shek of the Nationalist Party and Mao Zedong of the Communist Party, joined the ranks of Rummel’s ‘mega-murderers’, killing 10.2 million in 1921–48 and 37.8 million in 1923–76, respectively. Chiang’s tally ranked fourth, after Hitler’s 21 million. That of Mao, who famously proclaimed that ‘political power grows out of the barrel of a gun’, was second only to Stalin’s 42.6 million.3 1 Lu, Xun, Diary of a Madman and Other Stories, trans. William A. Lyell Jr (Honolulu: University of Hawai’i Press, 1990), p. 32. Lu Xun used ‘eat people’ metaphorically to refer to the repressive feudal order rather than actual cannibalism. 2 R. J. Rummel, China’s Bloody Century (New Brunswick: Transaction, 1991). 3 Ibid., pp. 3, 7, 8.

376

Genocide, Extermination and Mass Killing in Chinese History

In a related study of pre-twentieth-century ‘massacres, infanticide, executions, genocides, sacrifices, burnings, deaths by mistreatment’, Rummel argued that China’s historical killing toll of 33.5 million put it in the top position among a worldwide total estimate of 133,147,000.4 Second place in world history went to the Mongols, who followed a cold-blooded strategy to terrorise Persians, Arabs, Hindus, Russians, Europeans and Chinese alike into surrender. Their most notorious crime was the genocide of the Tanguts of the Xixia kingdom (in modern-day Gansu) in 1226–33, for their refusal to submit. Among the Mongols’ 29,927,000 estimated victims, some 18,470,000 were Chinese-killed under Khan Kublai’s rule in 1279–94.5 In China as elsewhere, then, mass killing goes back millennia. As Mao once said to party cadres in 1958: ‘What does Emperor Shihuang of the Qin dynasty count for? He only buried 460 Confucian scholars. We have buried 46,000 scholars . . . We have surpassed Qin Shihuang by 100 times’.6 Mao thus both testified to a continuity of brutality since the founding Qin dynasty (221–207 B C E), and boasted that his own killings outdid those of the Shihuang (始皇), the most notorious in Chinese history. In this chapter, I explore the terms ‘genocide’ and ‘extermination’ in the case of China, analyse the primary condition for mass killing, namely state capacity, and then address the validity of statistics in historical sources. The heart of this chapter chronicles the cases of ‘extermination’ that led up to Qin Shihuang’s proclamation of China’s first unified dynasty and immediately after. It also briefly introduces examples of ‘genocide’ and ‘extermination’ in the ensuing Han dynasty.

Definitions: Genocide, Extermination and Mass Killing Ben Kiernan maintains that the conception and practice of ‘genocide’ ‘long preceded our word for it’.7 In the Chinese context, words for ‘extermination’, ‘eradication’ or ‘total destruction’ of the enemy date to early texts and are common in dynastic histories, as Table 14.1 shows.8 4 R. J. Rummel, Death by Government (New Brunswick: Transaction, 1994), pp. 54, 69, 70. 5 Ibid., pp. 51, 49. 6 Zhou Zongqi (周宗奇), Big Awkward Lin Peng: Reflections of a Revolutionary (大聱林鵬: 一 個革命者的反思) (Taipei: Xinrui wenchuang, 2014), p. 173. 7 Ben Kiernan, ‘Is “genocide” an anachronistic concept for the study of early modern mass killing?’, History 99:336 (2014), 530–48, at 542. 8 These words are meant to be representative and not exhaustive. Some are adopted from Alastaire Iain Johnston, Cultural Realism: Strategic Culture and Grand Strategy in Chinese History (Princeton University Press, 1995), pp. 271–3.

377

victoria tin-bor hui

Table 14.1 Common words for ‘extermination’ in dynastic histories Word

Earliest usage

Frequency in dynastic histories*

誅 (zhu) 剿 (jiao) 殲 (jian)

Shiji (Records of the Grand Historian) Shiji (Records of the Grand Historian) Zuochuan (Zuo’s Tradition)

12,166 3,546 1,220

* Frequency counts are based on the twenty-six official dynastic histories in the Chinese Text Project digital library, https://ctext.org: Shiji, Han Shu, Hou Han Shu, San Guo Zhi, Jin Shu, Song Shu, Nan Qi Shu, Liang Shu, Chen Shu, Wei Shu, Bei Qi Shu, Zhou Shu, Sui Shu, Nan Shi, Bei Shi, Jiu Tang Shu, Xin Tang Shu, Jiu Wu Dai Shi, Xin Wu Dai Shi, Song Shi, Liao Shi, Jin Shi, Yuan Shi, Xin Yuan Shi, Ming Shi, Qing Shi Gao.

There are also phrases that normally have harmless meanings but in fact refer to extermination. For example, the Zhanguo ce (Stratagems of the Warring States) offers a strategy to ‘completely’ eradicate ‘evils’ and ‘ills’ when the context suggests extermination of populations. The passage reads: ‘The elimination of evils must be complete (除害莫如盡). Wu did not annihilate Yue and Yue then annihilated Wu; Qi did not annihilate Yan and Yan then annihilated Qi . . . This is because the removal of ills was not complete (除疾不盡)’.9 There are many more words that mean ‘attack’, ‘invade’, ‘kill’ and the like that lead up to ‘extermination’ (Table 14.2). The concept of ‘genocide’ is rendered as ‘extermination’ of an ethnic group (族 or 類) (I will say more on this below). Should we adopt ‘extermination’ or ‘genocide’ for the China case? According to the 1998 Rome Statute, ‘extermination’ is a crime against humanity involving conduct that ‘constituted, or took place as part of, a mass killing of members of a civilian population’ and was ‘committed as part of a widespread or systematic attack directed against a civilian population’.10 It is a crime committed intentionally, not accidentally or without foreknowledge. As for ‘genocide’, the 1948 UN Convention on the Prevention and Punishment of the Crime of Genocide defines ‘genocide’ as ‘any of the following acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such’, including killing its 9 Zhanguo ce, Qin 3, https://ctext.org/zhan-guo-ce/qin-san. 10 Official Records of the Assembly of States Parties to the Rome Statute of the International Criminal Court, First session, New York, 3–10 September 2002 (United Nations publication, Sales no. E.03.V.2 and corrigendum), part II.B; Article 7(1) (b), Crime against Humanity of Extermination, www.icc-cpi.int/NR/rdonlyres/336923D8-A6AD-40EC-AD7B-45B F9DE73D56/0/ElementsOfCrimesEng.pdf.

378

Genocide, Extermination and Mass Killing in Chinese History

Table 14.2 Common words for ‘killing’ in dynastic histories Word

Frequency in dynastic histories

攻 (gong): attack and besiege 擊 (ji): attack, strike at 襲 (xi): attack, strike at 征 (zheng)/征伐 (zhengfa): punish and attack; punitive campaign 伐 (fa): punish and attack 侵 (qin): invade 殺 (sha): kill 斬/斫 (zhan): chop off

21,157 18,407 12,140 14,594 7,247 4,559 25,564 12,898

members, causing serious bodily or mental harm to its members, deliberately inflicting conditions of life calculated to bring about the physical destruction of its members, imposing measures intended to prevent births within the group, or forcibly transferring children of a group to another group.11 The legal definition of ‘genocide’ requires the targeting of a ‘national, ethnical, racial or religious group’ and demands a stronger ‘specific intent’ than does ‘extermination’. Sociological definitions of ‘genocide’ tend not to require this genos component. Since the main difference between ‘genocide’ and ‘extermination’ is the genos, how important is the genos in mass killing in Chinese history? A wellknown case of ‘genocide’ is the Qing emperor Qianlong’s ‘final solution’ of Zünghar Mongols in 1755–8.12 A less known but unambiguous case is ‘Yongzheng’s execution of the Jie and extermination of their kind (永曾之 誅羯士,亦殲其類)’ in 350.13 The Jie were probably ‘Indo-Europeans’ with ‘high noses and thick beards’ (高鼻多鬚).14 Ran Min (also known as Yongzheng) was of Chinese descent and seemed to harbour ‘a bitter and vengeful Han identity’.15 He grew up in a Jie family named Shi because his 11 United Nations, Office on Genocide Prevention and the Responsibility to Protect, ‘Definition of the Crime of Genocide’, www.un.org/en/genocideprevention/geno cide-convention.shtml. 12 Peter Perdue, China Marches West: The Qing Conquest of Central Eurasia (Cambridge, MA: Harvard University Press, 2005), pp. 155, 285. See Volume I I of The Cambridge World History of Genocide. 13 Jin Shu, ch. 107. 14 Charles Holcombe, ‘The Sixteen Kingdoms’ in The Cambridge History of China, vol. I I: The Six Dynasties, 220–589, ed. Albert E. Dien and Keith N. Knapp (Cambridge University Press, 2019), pp. 119–44, at p. 121. 15 David A. Graff, Medieval Chinese Warfare, 300–900 (New York: Routledge, 2002), p. 63.

379

victoria tin-bor hui

father had been taken captive.16 In 350, he seized the Jie throne, reverted to his original Chinese name of Ran, and initiated what the historian David Graff correctly calls ‘a genocidal campaign’ against the Jie (羯) and other central Asians (Hu 胡) in the capital city Ye (鄴城).17 Ran led Han Chinese to massacre these ‘barbarians’ and offered rewards for their severed heads. The Jin Shu records that ‘the dead numbered more than 200,000, with corpses piling outside the city walls where they were eaten by jackals, wolves, and wild dogs’.18 Some managed to escape and sought assistance from the Xianbei Murong clan, who defeated and killed Ran in 352. Genos-targeted killings, however, were far exceeded by mutual slaughter among Han Chinese or different groups of central Asians. A focus on the genos would also essentialise unwarranted racial divisions. It is true that official rhetoric suggests a sharp distinction between civilised Han and uncivilised ‘barbarians’. The early Han dynasty began the tradition of depicting the Xiongnu (another central Asian group) as ‘birds or wild beasts’.19 Central Asians likewise maintained ‘a dual appearance’, using co-ethnics to rule over their own populations, who formed the military core, and Chinese officials to govern the Han, who provided taxes and corvée.20 Despite the official differences, there was much ‘Sino-barbarian synthesis’.21 The very notion of ‘Han Chinese’ was forged out of an immense cultural and ethnic diversity in early China.22 The Qin dynasty was built on ‘a mixed space that intermingled a wide variety of races, ideas, cultures, and regions’.23 Early Han emperors formed diplomatic marriages with the Xiongnu. The Xianbei Tuoba who dominated northern China in the fifth century also cultivated marriage ties with fallen ruling houses.24 The Sui’s and the Tang’s early emperors emerged from this mixed-blood elite and could claim the titles of the ‘Sage Khan’ and the ‘Great Khan’ as well as the ‘Son of Heaven’.25 Whatever their identities, imperial emperors had a long tradition of

16 Holcombe, ‘Sixteen Kingdoms’, 126–7. 17 Graff, Medieval Chinese Warfare, p. 63. 18 Jin Shu, ch. 107, quoted ibid. 19 Sophia-Karin Psarras, ‘Han and Xiongnu: a reexamination of cultural and political relations I’, Monumenta Serica 51 (2003), 55–236, at 137. 20 Holcombe, ‘Sixteen Kingdoms’, 128. 21 Graff, Medieval Chinese Warfare, p. 54. 22 Victoria Tin-bor Hui, ‘Cultural diversity and coercive cultural homogenization in Chinese history’ in Diversity and Its Discontents: Culture and Order in World Politics, ed. Andrew Phillips and Christian Reus-Smit (Cambridge University Press, 2020), pp. 93–112. 23 Zhaoguang Ge, What Is China? Territory, Ethnicity, Culture, and History (Cambridge, MA: Harvard University Press, 2018), p. 101. 24 Graff, Medieval Chinese Warfare, p. 73. 25 Ge, What Is China?, pp. 102–3.

380

Genocide, Extermination and Mass Killing in Chinese History

showering favours on any Chinese and ‘barbarians’ who submitted, while threatening ‘punitive campaigns’ against anyone who resisted. For the purpose of describing general trends in Chinese history, this chapter favours the term ‘extermination’. Chinese developments generally fulfil the legal definition of that term except for one critical aspect: it is difficult to set aside battle deaths and to focus on civilian deaths alone. China developed mass conscription early on, so that the majority of the fighting forces were drafted in wartime from among peasants who normally farmed in peacetime. It was similar for central Asian nomads, who doubled up as warriors. This chapter often retreats to the most general term, ‘mass killing’.

Conditions, Perpetrators and Victims: State Capacity Rummel argued that ‘power kills; absolute power kills absolutely’.26 He examines ‘intentional killing’ by ‘government agents acting in their authoritative capacity and pursuant to government policy or high command’.27 Government agents are broadly defined to include the entire hierarchy from top rulers and generals down to the lowest officers, warlords and rebels fighting against government agents, and non-governmental ‘brigands, pressgangs, or secret societies’ if they enjoy explicit or tacit ‘government approval, aid, or acceptance’.28 This chapter follows Rummel’s focus on ‘intentional government killing’, which means that perpetrators are always government agents, broadly speaking, while victims are the targets they vanquish. Power is as old as human history and ‘intentional government killing’ is as old as government. Both are historical constants; the only variable is the level of restraint on the exercise of power. Restraint on power could take the normative form. In addition to the modern awareness that mass killing should be a punishable crime, Rummel believed that power’s natural tendency to kill can be ‘caged by constitutionally guaranteed political and civil rights’.29 Steven Pinker concurs that democracy is ‘one of the greatest violence-reduction technologies since the appearance of government’.30 Yet, normative restraints were historically weak and have remained insufficient in avoiding ‘genocide’ to this day. 26 Rummel, Death by Government, p. 1. 27 Ibid., pp. 35–6, 42. 28 Ibid., pp. 21–2, 36–7. 29 Ibid., p. 23. 30 Steven Pinker, The Better Angels of Our Nature: Why Violence Has Declined (London: Penguin, 2012), p. 161.

381

victoria tin-bor hui

Restraint could also take the structural form – even if totalitarian rulers have no qualms for mass killing, they could still be limited by their capacities. Mark Levene contends that mass killing is more tied to ‘nation and state building’. If the twentieth century experienced genocide mostly in autocratic countries, it is because these states were belatedly engaged in state and nation building, which Western democracies had similarly experienced, but decades or a century earlier.31 I concur that state building’s critical importance is under-appreciated. A hallmark of the twentieth century is that many states dramatically enhanced their coercive capabilities. This is often attributed to the technology of killing. The invention of ever more lethal weapons of mass destruction, from automatic guns to aerial bombs, facilitated indiscriminate killing. Yet, even more important than the hardware is the technology of rule. The ‘gas chambers’ signified advancement in not just weapons but also social organisation in a ‘systematized, routinized, industrialized conveyor belt’ manner.32 Rwanda is a straightforward case of genocide by the technology of governance – its perpetrators were armed with detailed population registers but low-tech machetes and clubs. If the capacity for mass killing is as much a function of the technology of rule, then ‘genocide’ and ‘exterminmination’ should be more transhistorical than commonly thought. Pinker observes that a major ‘feeder’ for the growing deadliness of war since 1500 has been the invention of conscription.33 Conscription, especially when it is national and compulsory, has facilitated the mobilisation of a larger portion of the same population. It is noteworthy that, in the technology of conscription, China was two millennia ahead of Europe. The technology of rule is compelled by what Charles Tilly calls the ‘warmakes-state’ phenomenon: The building of an effective military machine imposed a heavy burden on the population involved: taxes, conscription, requisitions, and more. The very act of building it – when it worked – . . . tended . . . to promote territorial consolidation, centralization, differentiation of the instruments of government and monopolization of the means of coercion, all the fundamental state-making processes. War made the state, and the state made war.34 31 Mark Levene, ‘Why is the twentieth century the century of genocide?’, Journal of World History 11:2 (2000), 305–36, at 316–17. 32 Ibid., 307. 33 Pinker, Better Angels, p. 255. 34 Charles Tilly, ‘Reflections on the history of European state-making’ in The Formation of the National States in Western Europe, ed. Charles Tilly (Princeton University Press, 1975), pp. 3–83, at p. 73.

382

Genocide, Extermination and Mass Killing in Chinese History

Similar dynamics had unfolded in China’s classical era, which paved the way for the rise of the founding empire of Qin Shihuang. As Mark Lewis observes, In the competition to mobilize ever larger bodies of men, various states began to claim the services of the populations of their rural hinterlands and the lower social classes, which had hitherto played no political role. These reforms involved the allocation of land . . . in exchange for the payment of taxes and the provision of labor and military service. The process was progressively extended in various states, until it culminated in the reforms of Shang Yang in Qin state, where the entire adult, male population was registered, ranked, and allocated land on the basis of military service.35

State building enhances state capacity, which not just empowers government agents but also deprives societal actors of recourse for resistance. As Tilly suggested, state formation ‘produced the means of enforcing the government’s will over stiff resistance: the army’.36 Disgruntled people are ‘normally compliant, rising in resistance only when dramatic windows of opportunity open up’.37 Yet, such opportunities are rare under high-capacity regimes with centralised security forces. They are nearly absent when the central government has the capacity for direct rule, penetrating its society down to the household level to ensure that no one can escape surveillance and punishment.38 Max Weber defined the state as a ‘human community that (successfully) claims the monopoly of the legitimate use of physical force within a given territory’.39 ‘Legitimacy’ may seem to require popular support, but any leaders who succeeded in monopolising physical force could establish a government and claim legitimacy. This is why formally legitimate governments could be ‘mega-murderers’ hiding behind the abstraction of ‘state’, ‘regime’ and ‘government’.40 China developed ‘high modernist’ capacity as early as the fourth century B C E.41 The war-driven nature of Chinese developments is often overlooked because of the widespread presumption of the ‘singularity of 35 Mark E. Lewis, Sanctioned Violence in Early China (State University of New York Press, 1990), p. 9. 36 Tilly, ‘Reflections’, 73. 37 Charles Tilly and Sidney Tarrow, Contentious Politics, 2nd ed. (Oxford University Press, 2015), p. 233. 38 Ibid., pp. 57, 172. 39 Max Weber, From Max Weber: Essays in Sociology, ed. H. H. Gerth and C. Wright Mills (London: Routledge, 1991), p. 78. 40 Rummel, Death by Government, p. 8. 41 Victoria Tin-bor Hui, War and State Formation in Ancient China and Early Modern Europe (Cambridge University Press, 2005), p. 180.

383

victoria tin-bor hui

China’.42 However, China was historically as much a system of states as a unified empire. The Chinese term for ‘China’, zhongguo (中國), originally meant ‘central states’ in the plural form, rather than ‘Middle Kingdom’ in the singular form. The conventional Chinese wisdom also sees division as ‘an aberration’ so that ‘[a]fter each collapse, the Chinese state reconstituted itself as if by some immutable law of nature’.43 Yet, as Jianxiong Ge points out, ‘unity – this sacred term – has been repeatedly associated with war’.44 Before turning to historical examples of mass killings that proceeded from this state military capacity, a note on sources is in order.

Historical Sources Events and statistics for classical China are taken from Gao Rui’s Zhongguo Shanggu Junshishi (Military History of Ancient China), Lin Jianming’s Qin Shi (A History of Qin), Mu Zhongyue and Wu Guoqing’s Zhongguo Zhanzhengshi (History of War in China), Yang Kuan’s Zhanguo Shi (History of the Warring States), the Academy of Military Sciences’ Zhongguo lidai zhanzheng nianbiao (Chronology of Wars in China through Successive Dynasties), the Military Museum of China’s Zhongguo Zhandian (A Military Dictionary of China) and the Armed Forces University’s Zhongguo lidai zhanzhengshi (History of Wars in China through Successive Dynasties).45 These secondary sources are, in turn, based on classical philosophical (Confucian, Legalist, military) and historical texts. They generally have no disagreement on events and figures, and differ only in background information and level of detail. What is more problematic is whether the events and figures in historical texts are accurate. The Chunqiu (Spring and Autumn) Annals is the only text undisputedly written in the Spring and Autumn period (770–453 B C E), but it is extremely terse. The Zuozhuan (Zuo’s Tradition) purports to be about Spring and Autumn events, but it was retrospectively written in the Warring States 42 Henry Kissinger, On China (London: Penguin, 2012), pp. 5, 19, 60. 43 Ibid., pp. 6–7. 44 Jianxiong Ge, Tongyi yu fenlie: Zhongguo lishi de qishi (Unification and Division: Insights from Chinese History) (Beijing: Shenghuo, dushu, xinzhi sanlian shudian, 1994), p. 184. 45 Hui, War and State Formation, p. 249; Rui Gao, Zhongguo Shanggu Junshishi (Beijing: Junshi kexue chubanshe, 1995); Jianming Lin, Qin Shi (Taipei: Wunan chubanshe, 1992); Zhongyue Mu and Guoqing Wu, Zhongguo Zhanzhengshi (Beijing: Jincheng chubanshe, 1992); Kuan Yang, Zhanguo Shi (Taipei: Gufeng chubanshe, 1986); Academy of Military Sciences, Zhongguo lidai zhanzheng nianbiao (Beijing: Jiefangjun chubanshe, 2003); Military Museum of China, Zhongguo Zhandian (Beijing: Jiefangjun chubanshe, 1994); Armed Forces University, Zhongguo lidai zhanzhengshi (Taipei: Liming wenhua chubanshe, 1976).

384

Genocide, Extermination and Mass Killing in Chinese History

period (453–221 B C E). The Zhanguo ce was written as a training manual for persuasion and thus mixed fictional dialogues with historical events.46 Sima Qian’s Shiji (Records of the Grand Historian) relied on the Zhanguo ce and was written in the early Han dynasty (202 B C E – 9 C E). Texts written decades and centuries later run the risk of historical reconstruction. Even for records written at the time, the considerations of compilers differed from those of modern historians. The Chunqiu, for instance, selected events on the basis of their ritual significance to the Lu court rather than for their political significance to the system. Even an event as important as the defeat of the state of Yue by Wu in 494 B C E was not reported.47 Qin’s records highlight military merits; thus they omit losses suffered by Qin and possibly exaggerate casualties on the losing sides.48 Pinker notes that ‘most histories are sketchy when it comes to numbers’.49 Figures are more sketchy than events in Chinese texts. Historians have debated whether the recorded numbers can be taken seriously at all. Michael Loewe maintains that ‘[f]igures that are given for the numbers of troops who were called up, engaged in fighting, or eliminated as casualties are particularly suspect’.50 Cosmically nice numbers (72, 100, 500, 700, 1,000) are intrinsically dubious. The term wan (萬) probably does not literally mean 10,000 but merely ‘myriad’ or ‘a large unit’.51 Mark Lewis contends that such scepticism may make sense for figures such as 1,000,000 or 100,000. However, if the records employ more specific numbers such as 70–80,000, 200,000, 300,000 or 600,000, then they should be seen as rough estimates.52 He judges that various battle death figures are ‘internally consistent’, presenting a ‘picture of an expanding scale of military actions that fits well with what we know of the evolving institutions and practices of the period’.53 Sawyer 46 J. I. Crump, Chan-kuo Ts’e, translation and annotation with an introduction, rev. ed. (1970; Ann Arbor: University of Michigan Press, 1996), pp. 36–40. 47 Yuri Pines, Foundations of Confucian Thought: Intellectual Life in the Chunqiu Period (722– 453 BC) (University of Hawai’i Press, 2002), pp. 250–1 n. 10. 48 Cho-yun Hsu, Ancient China in Transition: An Analysis of Social Mobility (Stanford University Press, 1965), p. 67. 49 Pinker, Better Angels, p. 200. 50 Michael Loewe, ‘The Western Han Army: organization, leadership, and operation’ in Military Culture in Imperial China, ed. N. Di Cosmo (Cambridge, MA: Harvard University Press, 2009), p. 66. 51 Michael Loewe, ‘The campaigns of the Han Wu-ti’ in Chinese Ways in Warfare, ed., Frank A. Kierman Jr and John K. Fairbank (Cambridge, MA: Harvard University Press, 1974), pp. 67–122, at p. 96. 52 Mark E. Lewis, ‘Warring States political history’ in The Cambridge History of Ancient China: From the Origins of Civilization to 221 BC, ed. Michael Loewe and Edward L. Shaughnessy (Cambridge University Press, 1999), pp. 587–650, at p. 627 n. 56. 53 Ibid., 625–6, 628.

385

victoria tin-bor hui

concurs that Warring States battle death figures may be ‘less inaccurate than thought’ because ‘states of the period were certainly capable of mobilising a very high percentage of their male population, particularly when threatened with extinction’.54 The most controversial figure is the burying alive of 400,000 Zhao soldiers in Changping in 260 B C E. Hsu thinks that the number is ‘not necessarily incredible’ and Gao finds ‘no basis to reject it’.55 In all, the sketchy figures are all that we have. The best approach is to follow Lewis in treating them as reflecting ‘orders of magnitude’ rather than absolute numbers.56 Estimates of mass killing in other eras presented here are taken from Sophia-Karin Psarras and Rummel. Psarras’ statistics for the Han dynasty and the Xiongnu are based on dynastic history and archaeological sources.57 Rummel emphasised that pre-twentieth-century estimates reflected ‘the roughest approximations’.58 His reliance on census data could be problematic. Chinese censuses were based on households rather than individuals and thus estimates of population size are necessarily imprecise. More importantly, the purpose of a census was to assess taxation and labour service.59 In times of troubles, villages would band together to form ‘defensive communities’ in inaccessible terrain to hide not just from marauding armies and bandits but also from voracious tax collectors and corvée drafters.60 Thus, population shrinkages could point not only to mass killing but also to underreporting. Let us begin with China’s classical era: the Spring and Autumn (770–453 B C E ) and Warring States (453–221 B C E) periods.

Standards and Critiques Kiernan suggests that studies of premodern ‘genocide’ can avoid the charge of anachronism by holding historical actors to ‘standards . . . and critiques levelled in their day’.61 The case of China demonstrates that awareness of ‘extermination’ as a crime is a transhistorical phenomenon. Fierce interstate 54 Ralph D. Sawyer, with the collaboration of Mei-chun Sawyer, The Tao of Spycraft: Intelligence Theory and Practice in Traditional China (Boulder, CO: Westview Press, 1998), p. 559 n. 38. 55 Hsu, Ancient China in Transition, p. 67; Gao, Zhongguo Shanggu Junshishi, p. 435. 56 Lewis, ‘Warring States political history’, 626. 57 Sophia-Karin Psarras, ‘Han and Xiongnu’, 68. 58 Rummel, Death by Government, pp. 51–3. 59 Diana Lary, Chinese Migrations: The Movement of People, Goods, and Ideas over Four Millennia (Lanham, MD: Rowman & Littlefield, 2012), p. 12. 60 Graff, Medieval Chinese Warfare, pp. 55, 66. 61 Kiernan, ‘Is “genocide” an anachronistic concept?’, 545–6.

386

Genocide, Extermination and Mass Killing in Chinese History

competition gave rise not just to state-building processes but also to the ‘hundred [i.e. many] schools of thought’: Confucian, Mohist, Legalist, military schools and more. The Confucian school is best represented by Confucius (549–479 B C E ) and Mencius (372–289 B C E). According to the Analects (Sayings) of Confucius, when the state of Wei’s ruler Linggong asked about military tactics, Confucius replied, ‘If it is matters of sacrificial vessels, I have heard of them; if it is matters of armies and campaigns, I have never studied them’.62 The Mencius bluntly accuses those who say ‘I am skilled in making formations’ or ‘I am skilled in making war’ as ‘criminals’.63 It further championed the principle that unity should be achieved by ‘the True Monarch’ or ‘the one who has no proclivity to kill’.64 The Mozi, attributed to the founder Modi, most explicitly argues that ‘aggressive war’ is ‘the most heinous of all crimes’. It is scornful of the ‘gentlemen of the world’ who kill in war: If someone kills one man, he is condemned as unrighteous and must pay for his crime with his own life. According to this reasoning, if someone . . . kills a hundred men he is a hundred times as unrighteous and should pay for his crime with a hundred lives. Now all the gentlemen in the world know enough to condemn such crimes and brand them as unrighteous. And yet when it comes to the even greater unrighteousness of offensive warfare against other states, they do not know enough to condemn it. On the contrary, they praise it and call it righteous.65

The Mozi also advocates ‘universal love’: ‘If everyone loves others as he loves himself – there will be . . . no conflicts, and no war’.66 Unfortunately, such critics were overshadowed by military strategists and Legalists, who were concerned with winning victories and maximising state power. The Sunzi’s Art of War reads in the very first line, ‘Warfare is the greatest affair of state, the basis of life and death, the Way to survival or extinction’.67 The Sima fa (Marshal’s Art of War) further asserts that ‘if one 62 The Analects of Confucius, trans. Burton Watson (New York: Columbia University Press, 2007), book 15, 1. 63 The Mencius, trans. Irene Bloom (New York: Columbia University Press, 2009), book 7B4. 64 Cited in Yuri Pines, The Everlasting Empire: The Political Culture of Ancient China and Its Imperial Legacy (Princeton University Press, 2012), pp. 18, 51. 65 Mozi: Basic Writings, trans. Burton Watson (New York: Columbia University Press, 2003), book 5. 66 Ibid., book 4. 67 Sun-tzu: The Art of War, trans. Ralph D. Sawyer (Boulder, CO: Westview Press, 1994), ch. 1.

387

victoria tin-bor hui

must kill men to give peace to the people, then killing is permissible’.68 A Legalist text, the Shang Jun Shu, has this response to the Confucian school: ‘If a state has ritual and music . . . the superiors will not be able to order the people to war, and the state is certain to be dismembered, and even to disappear altogether’.69 Another Legalist text, the Guanzi, similarly rebukes the Mozi: ‘If disarmament theories prevail, the passes will not be defended. If “universal love” theories prevail, the soldiers will not fight.’70 Nevertheless, it is clear that critiques were indeed levelled. More pertinently, extermination was not considered an essential or integral component of Chinese culture.

Spring and Autumn Period, 770–453 B C E In the Spring and Autumn period, the number of political units decreased from about 170 in 770 B C E to about 20 in the mid-fifth century B C E.71 However, during this process mass killing was constrained by limited military and extractive capabilities. The most powerful could seize tiny states, but not peer competitors. Early Spring and Autumn warfare was characterised by elite chariots, with important implications. Charioteers were limited in number – they were hereditary aristocrats with ‘lifelong training in the difficult art of driving a two-horse war chariot and shooting from it with the powerful compound bow’.72 Chariot horses likewise required lengthy training as well as ten acres of good grain land for each team of two horses. Each chariot was supported by approximately ten foot soldiers. The largest states were able to field 600 chariots in the eighth century B C E and about 1,000 by the fifth century. The high costs of war created a strong incentive to maximise territorial or political gains by manoeuvres rather than by direct combat. As the Sunzi calculates, when you send forth an army of a hundred thousand on a campaign, marching them out a thousand [Chinese] miles, the expenditures . . . will be one thousand pieces of gold per day. Those inconvenienced and troubled both within and without the border, who are exhausted on the road or are unable to pursue their agricultural work, will be seven hundred thousand families.73 68 Cited in Bruce Brooks and Taeko Brooks, The Emergence of China: From Confucius to the Empire (Amherst: Warring States Project, University of Massachusetts, 2015), p. 116. 69 Cited ibid., p. 140. 70 Cited ibid., p. 107. 71 Zhongyue Mu and Guoqing Wu, Zhongguo Zhanzhengshi (History of War in China) (Beijing: Jincheng chubanshe, 1992), I I, pp. 23–31 and 132–3. 72 Brooks and Brooks, Emergence of China, p. 19. 73 Sawyer, Sun-tzu, p. 127.

388

Genocide, Extermination and Mass Killing in Chinese History

In this cost-conscious environment, many attacks took the form of ‘unopposed raids’ with limited casualties.74 The first pair of hegemonic rivals, Qi and Chu, never directly fought each other in their brief competition in the mid-seventh century B C E. The next pair, Jin and Chu, clashed only three times (in 632, 597 and 575 B C E ) in their prolonged competition from 632 to 546 75 B C E.

Warring States Period, 453–221 B C E States that survived the Spring and Autumn period made the transition from elite-based chariot warfare to peasant-based infantry warfare in the Warring States period. The conscription of peasant soldiers made for larger armies, which progressively increased from several tens of thousands to 100,000 and more. The technology of land tax further enriched the state and increased incentives for territorial aggression. With all competing states upgrading their bureaucratic, fiscal and military capacities, the Warring States increasingly became war machines. They were no longer restrained by either ambition or capability. The only remaining constraint was the balance of power, or the checking of raw power by raw power. By the mid-fourth century B C E, international competition expanded from struggles over one single city or one particular tract of land to unlimited expansion. The battle of Maling in 343 B C E, in which Qi could travel a long distance to defeat Wei, ‘raised the possibility that one state might conquer a distant rival and thus unify the world’.76 Yet, when the ambitious Qi invaded Yan in 314 B C E, it was expelled by a balancing coalition. When Qi conquered Song in 286 B C E, it was crushed by another alliance and could not recover.

Qin’s All-out Conquest and Extermination Campaigns In the third century B C E, the system ‘shift[ed] from wars seeking advantage in a balance of power to the campaigns of all-out conquest launched by Qin’.77 In a system in which high-capacity states mobilised large portions of their adult male population for war, how could Qin prevail? In 268 B C E a strategist, Fan Ju, articulated a strategy of irrevocable expansion by ‘attacking those nearby while befriending those faraway’, and ‘attacking not only territory but 74 Brooks and Brooks, Emergence of China, p. 24. 75 Hui, War and State Formation, p. 152. 76 Brooks and Brooks, Emergence of China, p. 99. 77 Lewis, ‘Warring States political history’, 627–28.

389

victoria tin-bor hui

also people’.78 Rival states and their populations should be completely annihilated so that they would not be able to recover. Qin’s policy offers a textbook example of the strategic logic of extermination. The Sunzi, written in the fourth century B C E, still advised that ‘preserving [an enemy] army is best, destroying their army second-best’.79 As illustrated by the Mongol conquests, conquerors typically threatened with the warning ‘Surrender or be slaughtered!’ Given such a stark choice, most would opt to surrender and look for future opportunities to rebel. At the same time, conquerors would surely know that acceptance of surrender always incurred the risk of future rebellion. They could rule with an iron fist, but this would involve immense policing costs. The ‘final solution’ was to massacre conquered populations; if left to survive, they would seek to restore independence and annihilate the conquerors in turn. This option was sparingly deployed by even the Mongols because control over people generated greater values than control over territory for most of history. The state of Qin, however, preferred to devastate defeated armies because it had no need to ‘recycle’ unreliable enemy troops. Just as the practice of ‘genocide’ could precede a word for it, so the practice of ‘extermination’ had become commonplace before Fan Ju’s articulation of it as a formal policy. Between 356 and 236 B C E, Qin’s troops slaughtered more than 1.5 million enemy soldiers and civilians in major battles (out of a systemwide population of about 20 million) (Table 14.3).80 One talented commander alone, Bo Qi, was reportedly responsible for killing 240,000 Han-Wei allied troops in 293 B C E, several hundreds of thousands of Chu soldiers and civilians in 279, 150,000 Zhao-Wei allied troops in 273, and 400,000 Zhao forces in 260. Yinhong Shi of the People’s University calls Bo Qi the ‘super butcher’.81 During the final wars of unification from 236 to 221 B C E, the only available toll figure is 100,000 Zhao deaths in 234. But it is certain that Qin killed many adult males of defeated states, so as to minimise the potential for rebellion.82 As mentioned earlier, these figures should be treated as reflecting the magnitude of mass killing rather than exact toll numbers. Measured by counting the severed ears of dead enemies, the statistics were collected for 78 Zhanguo ce, Qin 3; Lewis, ‘Warring States political history’, 640. 79 Brooks and Brooks, Emergence of China, p. 120. 80 Yang, Zhanguo Shi, pp. 112, 484. Table 14.3 is compiled from Yang, Zhanguo Shi, p. 444, and Hui, War and State Formation, pp. 152–3 n. 168. 81 Yinhong Shi, ‘Wuzhuang de zhongguo: qiannian zhanluechuantong jiqi waijiao yiwen’ (‘Armed China: millennia-old strategic traditions and their implications for foreign policy’), Shijie jingji yu zhengzhi 6 (2011), 4–33, at 15. 82 Gao, Zhongguo Shanggu Junshishi, pp. 437–8; Yang, Zhanguo Shi, pp. 189–90.

390

Genocide, Extermination and Mass Killing in Chinese History

Table 14.3 Qin mass killing, 356–236 B C E Year (B C E )

Targeted state/s

Deaths

354 330 317 314 312 307 300 298 293 279

Wei Wei Han, Wei, and Zhao Han Chu Han Chu Chu Han and Wei Chu

275 274 273 264 260 257 256 256 245 244

Han Wei Wei and Zhao Han Zhao Chu and Wei Han Zhao Han Wei

7,000 troops 45,000 troops 82,000 troops 10,000 troops 80,000 troops 60,000 troops 20,000 troops 20,000 troops 240,000 troops several hundreds of thousands troops and civilians 40,000 troops 40,000 troops 150,000 troops 50,000 troops 400,000 troops 20,000 troops 40,000 troops 90,000 troops 30,000 troops 30,000 troops

the awarding of military merit and were ‘doubtless inflated by officers in the field through the collection of additional ears from Qin dead or peasants who strayed too close to the Qin army’.83 The Qin state had an elaborate system of reward for those who slew enemies or commanded victorious units, and to punish those who surrendered to the enemy. It is worth noting that although the Mongols are noted for their cruelty in requiring troops to cut off an ear from each victim as proof,84 apparently they were long preceded by the Chinese – and then succeeded by the Japanese in the sixteenth century (see The Cambridge World History of Genocide, Volume I I , Chapter 6).

Legitimation and Justification In 221 B C E, the victorious King Zheng (246–221 B C E ) of the state of Qin proclaimed the founding of the Qin dynasty and crowned himself First Emperor (Shihuang) (he reigned until 210 B C E). While he had achieved total 83 Lewis, ‘Warring States political history’, 625.

391

84 Rummel, Death by Government, p. 27.

victoria tin-bor hui

Qin c. 300 BCE Qin gains c. 230 BCE Qin gains c. 230 – 221 BCE Qin gains c. 221 – 210 BCE Zhongshan

Yan

Zhao Qi Wei Maling Lu Zhou

Changping

Han

Song

Chu

Map 14.1 Expansion of the Qin Empire, c.300 – c.210 B C E. (Cartography by Bill Nelson)

conquest by extermination, he seemed to understand Weber’s argument that monopoly of the use of force should be seen as legitimate. He ‘declared himself Sage’ and celebrated the success of his ‘punitive expeditions’ against ‘bandit rebels’.85 One of his stone inscriptions reads: . . . The Emperor in his sagacity, benevolence and justice Has made all laws and principles manifest. . . . Great is the virtue of our Emperor Who pacifies all four corners of the earth, Who punishes traitors, roots out evil men, . . . His achievements surpass those of the Five Emperors, His kindness reaches even the beasts of the field; All creatures benefit from his virtue, All live in peace at home.86

85 Pines, Everlasting Empire, pp. 20, 55. 86 Shiji, 6, in Yang Hsien-yi and Gladys Yang, Records of the Historian (Hong Kong: Commercial Press, 1974), pp. 170–2.

392

Genocide, Extermination and Mass Killing in Chinese History

The very success of winning ‘All under Heaven’ represented a virtue; the previous divided era meant chaos, while unity became the synonym for peace. The Shang jun shu (Book of Lord Shang), a Legalist text from Qin, elaborates on the (il)logic: When a sage rules the people, he must attain their hearts; hence he is able to use force. Force generates strength; strength generates awesomeness; awesomeness generates virtue; virtue is born out of force. The sage ruler possesses it exclusively; hence he is able to promulgate benevolence and righteousness in All under Heaven.87

Mass Killing under Unification Nevertheless, universal domination did not stop the mass killing but only changed its character, to killing by other means. The emperor immediately drafted hundreds of thousands of labourers to demolish the pre-existing defence walls between states and to construct standardised major roads linking the capital with major regions of the empire. He also sent some 500,000 troops to conquer the region south of the Ling Mountains in modern Guangdong and Guangxi, which was not part of the Warring States system. The conquering troops had to build a canal linking the Yangtze and Pearl rivers before they finally controlled the region in 214 B C E. In 215 B C E, the emperor charged his heir Meng Tian with 300,000 men to seize territories from the Xiongnu, within the northern bend of the Yellow River. To consolidate the new conquest, the court built new settlements and another major road. In addition to public projects, the First Emperor further enslaved the populations for his own extravagance. He drafted more than 700,000 prisoners to build his private tomb and 700,000 castrated convicts to construct more than 270 imperial palaces. With onerous military service, corvée, land tax and head tax, the ‘death toll too must have been enormous’.88 Resistance Was there resistance to Qin’s extermination campaigns? Confucian thinkers Mencius and Xunzi had provided theoretical justification for the ‘right to rebel’ and even for tyrannicide.89 Before the First Emperor’s grand triumph, 87 Shang Yang, The Book of Lord Shang: Apologetics of State Power in Early China, trans. Yuri Pines (New York: Columbia University Press, 2017), p. 192. 88 Derk Bodde, ‘The state and empire of Ch’in’ in The Cambridge History of China, vol. I: The Ch’in and Han Empires, 221 BC – AD 220, ed. Denis Twitchett and Michael Loewe (Cambridge University Press, 1987), p. 63. 89 Hui, War and State Formation, p. 177.

393

victoria tin-bor hui

subjugated populations in the Shangdang commandery and the non-Sinitic Shu did rebel.90 However, as mentioned above, the extermination strategy of killing defeated forces en masse was designed to decapacitate future rebellions. Moreover, resistance is shaped not just by popular grievances but also by state capacity. The Qin’s capacity for direct rule allowed it to police disgruntled populations with collective responsibility, mutual surveillance and forced migrations. The Qin’s iron grip changed only after the competent First Emperor died in 210 B C E. The Second Emperor imposed even heavier exactions and harsher punishments than the First Emperor had, but he was uninterested in the administration of the empire. When compliance meant not just hard labour but also harsh punishments and even death, the people were left with no choice but to engage in resistance. Rebellions began when Chen Sheng and Wu Guang were delayed by torrential rains on their way to report corvée. Under Qin laws, delay in reporting duty was punishable by death. Even if their lives were spared for this offence, they would still be worked to death doing hard labour. Chen and Wu thus concluded that resistance would involve little additional cost. Not surprisingly, this rebellion was readily crushed by the imperial court’s military might. However, the Second Emperor’s behaviour was so arbitrary and ruthless that he soon alienated his own generals. With the political opportunity structure opened up, multiple rebellions sprang up throughout the empire. The Qin dynasty collapsed in 206 B C E.

From the Qin on The conventional wisdom holds that the Qin state was exceptionally cruel (thus deserving Mao’s conscious comparison with Qin Shihuang) and that Chinese history returned to Confucian benevolence from the Han dynasty (202 B C E – 220 C E) onwards. However, as John Fairbank has pointed out, ‘Han emperors took great pains to claim that their rule was based on the Confucian teachings of social order, even while they used the methods of the Legalists as the basis for their institutions and policy decisions’.91 All ensuing dynasties followed this policy of ‘Legalism with a Confucian façade’.92 90 Ibid., pp. 76, 217. 91 John K. Fairbank, ‘Introduction: varieties of the Chinese military experience’ in Chinese Ways in Warfare, ed. Kierman and Fairbank, pp. 1–26, at p. 11. 92 Kung-chuan Hsiao, ‘Legalism and autocracy in traditional China’ in Shang Yang’s Reforms and State Control in China, ed. Yu-ning Li (Armonk, NY: M. E. Sharpe, Hsiao, 1977), pp. 125–43, at p. 137.

394

Genocide, Extermination and Mass Killing in Chinese History

Emperor Wu (The Martial Emperor) (武) (r. 140–87 B C E ) established Confucianism as the official ideology in 136 B C E but surpassed Qin’s conquests. He expanded to not only what is today’s southern China, as had Qin’s First Emperor, but also the distant Western Regions (the Zungharian and/or the Tarim Basins), southern Manchuria, northern Korea and northern Vietnam. With the Xiongnu Chanyus, early Han emperors had made peace by means of heqing (和親) marriage alliances. Emperor Wu, however, found the practice ‘humiliating’.93 Psarras contends that the only embarrassment was that the Xiongnu were ‘China’s equal’ and challenged China’s preferred representation as ‘the center of all political power’.94 In 133 B C E, Emperor Wu sent out a 300,000-strong expeditionary force against the Xiongnu in an ‘unprovoked aggression’.95 From 133 to 91 B C E he launched twenty-eight expeditions compared with the Xiongnu’s eighteen incursions.96 In all, approximately 489,500 Xiongnu were killed or captured by Han forces.97 Psarras’ estimates of Han casualties range from 286,000 to 477,000, based on 30 per cent to 50 per cent of the mobilised troops.98 As the Xiongnu population was only a fraction of the Han’s, which reached 57.7 million later in 2 C E, the mass killing was far more devastating on the Xiongnu side.99 After Emperor Wu, the Xiongnu also suffered from internal strife. Different Xiongnu groups would surrender to the Han either when they were defeated by Han troops or when they wanted to get an edge over rival Xiongnu. However, if the Xiongnu took ‘surrender’ as a short-term tactical move, the Han court expected ‘complete submission or extermination’.100 For instance, Chanyu Yuchujian surrendered in 91 C E, but then disavowed the agreement and returned to the steppes in 93. The Han court ‘punished’ him by sending 10,000 cavalry to pursue him, ‘exterminating’ ‘not just him but all his people’.101 The Han also launched genocidal campaigns against the Qiang who occupied the strategic Gansu corridor between the interior and the Western Regions.102 The Qiang had agreed to serve as ‘guards of the frontier’ for the 93 Sophia-Karin Psarras, ‘Han and Xiongnu’, 136. 94 Ibid., 60. 95 Ibid., 141. 96 Ibid., 146–7. 97 Ibid., 150. 98 Ibid., 136, 150, 152. 99 Ibid., 150. 100 Sophia-Karin Psarras, ‘Han and Xiongnu: a reexamination of cultural and political relations II’, Monumenta Serica 52 (2004), 37–93, at 74. 101 Ibid., 74, 89. 102 Rafe de Crespigny, ‘The military culture of Later Han’ in Military Culture in Imperial China, ed. Di Cosmo, pp. 90–111, at p. 109.

395

victoria tin-bor hui

Xiongnu

Xianbei Wuhan Joseon

Hexi C orr Qiang ido r

Ye City Chang’an

Maurya Empire

Yelang

Minyue

Dian Nanyue

Former Han Empire c. 140 BCE Former Han expansion by CE 9 Later Han expansion CE 25–220 Territory under Han Protectorate

Map. 14.2 Expansion of the Han Empire, c.140 B C E – 220 C E. (Cartography by Bill Nelson)

Han.103 However, they soon rebelled against ‘misgovernment and exploitation on the part of Han frontier officials’.104 In 160, the ‘Protector of the Qiang’, Duan Jiong, launched an extermination campaign against them. For the next seven years, Duan ‘killed 23,000 [Western Qiang], captured tens of thousands more, seized millions of horses, cattle and sheep, and forced the surrender of over ten thousand camp-fire groups’.105 In 167, Duan turned to the Eastern Qiang. In a ‘final massacre’ in 169, he ‘killed almost twenty thousand and took all their goods and chattels’.106 In 170, Duan returned to the Han capital in triumph with ‘fifty thousand non-Chinese troops’ and ‘ten thousand captives’.107 The Han’s extermination and genocide campaigns imposed immense suffering and death on the Chinese population as well. The celebrated

103 Ying-shih Yu, ‘Han foreign relations’ in The Cambridge History of China, vol. I, ed. Twitchett and Fairbank, pp. 377–462, at pp. 425–6. 104 Ibid., p. 433. 105 Rafe de Crespigny, A Biographical Dictionary of Later Han to the Three Kingdoms (23–220 AD) (Leiden: Brill, 2007), p. 188. 106 Ibid. 107 Ibid.

396

Genocide, Extermination and Mass Killing in Chinese History

Tang poet Du Fu (712–70) lamented of the Emperor Wu’s campaigns in ‘The Ballad of the Army Wagons’ (兵車行): . . . Each marcher at his waist has bow and quiver; . . . They clutch clothes, stamp their feet, bar the way weeping, Weeping their voices rise to darkening Heaven; And when the passers-by question the marchers, The marchers but reply, ‘Levies come often: . . . They take us at fifteen for up the river, To garrison the West, they’ll take at forty, Your Headman has at first to tie your turban, Grey-headed you come home, then back to duty – . . . the blood that’s flowed out there would make a sea, Sir! Our Lord, his lust for land knows no degree, Sir! But have you not heard Of House of Han, its East two hundred regions Where villages and farms are growing brambles? . . . That though a sturdy wife may take the plough, Sir, You can’t see where the fields begin and end, Sir? That Highlanders fare worst, they’re hardy fighters And so they’re driven first, like dogs and chickens? . . . Dare the conscripts tell their wretchedness? How, for instance, only last winter The Highland troops were still in the line When their Prefect sent urgent demands, Demand for tax, I ask you, from where? So now we know, no good having sons, Always better to have a daughter: For daughters will be wed to our good neighbors When sons are lying dead on Steppes unburied! . . . On the Black Lake’s [Kokonor] shore The white bones there of old no one has gathered, Where new ghosts cry aloud, old ghosts are bitter, Rain drenching from dark clouds their ghostly charter?108

Unsustainable conscription, corvée, and taxes were detonators of rebellion. Peasant rebellions were rare in the Warring States era when states had high capacity; they became a recurrent concern from the Han onwards. As unified ‘China’ increased in size, the burden of ruling ‘all under Heaven’ 108 Arthur Cooper, Li Po and Tu Fu: Poems Selected and Translated with An Introduction and Notes (Harmondsworth: Penguin, 1973), pp. 167–70.

397

victoria tin-bor hui

eroded the original capacity for direct rule. The overstretched imperial court came to be increasingly dependent on local leaders to maintain social order.109 The more ambitious among them were then in the position to evade taxes, subvert imperial power, even stage rebellions. While most rebellions could still be easily crushed, some succeeded in carving out their own territory and broke down dynastic unity into another system of independent states. Rummel suggested that ‘genocide’ can be mitigated by the ‘cross-pressures’ of individual and group freedoms.110 But the more transhistorical and effective cross-pressure came from international competition. Michael Loewe asks ‘to what extent the unified empires of Qin and Han maintained easier conditions of living or imposed harsher burdens on the population than the localised kingdoms of China that preceded or followed them’.111 The existence of a multi-state system did not have to mean mass killing if competing states could adhere to shared norms. Indeed, the classical era enjoyed three centuries of stability with elaborate feudal rites, diplomatic protocols, international covenants, alliance agreements and peace settlements. In subsequent eras of division after the Qin and Han, independent kingdoms could maintain a semblance of order with treaties and alliances. Moreover, division always gave rise to the ‘right of exit’, which presented a formidable check against arbitrary power. Nevertheless, power contenders in subsequent periods of division also understood that they could imitate the Qin–Han model to take ‘All under Heaven’ by attacking states and annihilating their populations. Once dynastic founders controlled the Chinese heartland, they would expand outward after Emperor Wu’s example. In court debates, Confucian scholar-officials ‘frequently advocated aggressive policies to “exterminate” the barbarians who had violated the way of Heaven’.112 The Sui dynasty (581– 618) expanded to the Western Regions, southern Manchuria, northern Korea and northern Vietnam. The Tang dynasty (618–907) marched to the Western Regions, Mongolia, eastern Tibet, southern Manchuria and northern Korea. The Yuan dynasty (1279–1368), after conquering the vast Eurasian steppe zone and toppling the Song dynasty, attempted to subdue Korea, Japan, Yunnan, 109 Victoria Tin-bor Hui, ‘How Tilly’s warfare paradigm is revolutionizing the study of Chinese state-making’ in Does War Make States?, ed. Lars Bo Kaspersen and Jeppe Strandsbjerg (Cambridge University Press, 2017), pp. 268–95. 110 Rummel, Death by Government, pp. 22–3. 111 Michael Loewe, ‘Introduction’ in The Cambridge History of China, vol. I, ed. Twitchett and Fairbank, pp. 1–19, at p. 18. 112 Yuan-kang Wang, Harmony and War: Confucian Culture and Chinese Power Politics (New York: Columbia University Press, 2011), p. 138.

398

Genocide, Extermination and Mass Killing in Chinese History

Burma, Vietnam, and even Java. The Ming dynasty (1368–1644) sent expeditionary troops to the Western Regions, Mongolia, southern Manchuria, northern Korea, Burma, Vietnam and beyond in South and Southeast Asia. The Qing dynasty (1644–1911) dominated the Western Regions, Mongolia, Tibet, Nepal and Taiwan. Such campaigns incurred massive deaths and starvation not just among targeted ‘barbarians’ and ‘bandits’ but also among hapless Han peasants. Genocide, extermination, mass killing and human suffering thus recurred with the drive to achieve ‘unity’, expand ‘unity’ and maintain ‘unity’.

Bibliographic Note The Cambridge History of China (multiple volumes) always offers authoritative accounts. For a concise and accessible analysis of how ‘China’ came into being, see Ge’s What Is China? For Qin’s ‘story of world conquest’, see page 15 of Brooks and Brooks’s Emergence of China: From Confucius to the Empire. For a complementary account of wars and mobilisation in ancient China, consult Lewis, Sanctioned Violence in Early China. Hui, War and State Formation in Ancient China and Early Modern Europe is a Political Science account of interstate and state–society relations in the Spring and Autumn and Warring States periods. For an International Relations account of the Song and the Ming’s wars, see Wang’s Harmony and War.

399

part iii *

THE MEDIEVAL WORLD AND EARLY IMPERIAL EXPANSIONS

15

William the Conqueror’s Harrying of the North, 1069–1070 What, If Not Genocide? c. p. lewis ‘The Harrying of the North’ is the name given since the 1870s to a military campaign waged in northern England by William I against English rebels and Danish invaders over a few weeks either side of Christmas 1069. The main focus was Yorkshire and northwards across the river Tees, but the episode is sometimes reckoned also to include a second phase after the king crossed the Pennines and put down rebellion in Mercia (Map 15.1).1 The campaign was the culmination of William’s struggle to secure Norman rule in the North, and a brief account of events over the previous eighteen months is necessary. He had already led armies to York twice, in summer 1068 to take the submission of the Northumbrian magnates, and in February 1069 after the ambush and slaughter of the newly appointed earl of Northumbria and his men at Durham. William built a castle in York during each of those expeditions, and his soldiers looted the city during the second one. In summer 1069 the situation worsened with the arrival of a large Danish fleet in the Humber estuary, doubtless as much to pursue their own ambitions as in support of the rebels. The Danes captured York after fighting which destroyed the castles and burned down the cathedral. In late autumn William again led an army north. He retook York without resistance, the Danes having abandoned the city and moved their ships down the Humber. A Norman detachment kept them at bay while, after Christmas, the king worked northwards to the Tees, where he took the submission of the rebel leaders and then moved on to the Tyne. By then the danger was over and William returned to York and crossed the Pennines, where the Mercian rebels I offer warm thanks to the editor for sound advice in shaping this chapter, and to Dr Bill Aird (Edinburgh) and Dr Mark Hagger (Bangor) for comments on an earlier draft. 1 E. A. Freeman, The History of the Norman Conquest of England, 6 vols., rev. ed. (1871; Oxford: Clarendon Press, 1875–9), I V, p. 287n.2.

403

c. p. lewis

P

R. Tyne

North Sea

Wearmouth

Hexham Durham

N

E

R. Tees

O R

N

lan

d

T

C

e lev

H

N

U M

YORKSHIRE

York

en

R

av

I

ess ern und o Am

B

I

Cr

N

A

Beverley R . H um be r

E

I

A

DERBYSHIRE

R.

CHESHIRE

Tre nt

S

Chester

C E

R

M

STAFFORDSHIRE

SHROPSHIRE N W

E S

0 0

25

50 km 25

Evesham

Map 15.1 Northern England in the time of William I.

404

50 mi

William the Conqueror’s Harrying of the North, 1069–70

melted away. After building a castle at Chester he headed south and disbanded his army. The entire operation lasted from November 1069 to March 1070. A campaign in the depths of winter was unusual and full of danger. William clearly thought that the situation in the North was serious enough to warrant the risk, and it paid off, both objectives being met, since the Danes left and the Northumbrian magnates swore loyalty, for the moment anyway.2 There is no firsthand account of the Harrying, and a cautious approach is essential in assessing narratives written significantly later, such as the tangential evidence of Domesday Book for the condition of Yorkshire in 1086. Domesday Book does not cover the lands north of the Tees, and no medieval author described the Harrying of Mercia, so the evidence for the campaign as a whole is patchy, not least for its effects on the land and people of Northumbria. Anglo-Norman historians writing in the earlier twelfth century had much to say about what happened and how it should be understood. Some of them had access to contemporary sources now lost, but they were all actively shaping an account of William’s reign which spoke to their own perceptions of the course of English history.3 The campaign was covered most fully by two monastic historians writing more than a generation afterwards: Symeon of Durham (at Durham, d. c.1128) and Orderic Vitalis (at Saint-Evroul on the southern borders of Normandy, d. c.1142). They gave independent accounts, graphic and horrifying in different ways. Along with the briefer treatment by another monk, William of Malmesbury, in works completed circa 1125,4 they have set the framework and tone for all modern discussions of the Harrying since John Lingard’s History of England (1819). Lingard, an adopted northerner himself, took the twelfth-century accounts at face value, borrowing and amplifying two of their themes: the king’s quest for revenge on the persistently disloyal northern magnates, and the extermination of the Northumbrians through the systematic wanton destruction of the means of life in northern England.5 2 A. Williams, The English and the Norman Conquest (Woodbridge: Boydell, 1995), pp. 24–44; D. Bates, William the Conqueror (New Haven and London: Yale University Press, 2016), pp. 295–328. 3 A. Gransden, Historical Writing in England, c. 550 – c. 1307 (London: Routledge & Kegan Paul, 1974), pp. 92–185. 4 William of Malmesbury, Gesta regum Anglorum: The History of the English Kings, ed. and trans. R. A. B. Mynors, R. M. Thomson and M. Winterbottom, 2 vols. (Oxford: Clarendon Press, 1998–9), I, pp. 462–5; William of Malmesbury, Gesta pontificum Anglorum: The History of the English Bishops, ed. and trans. M. Winterbottom and R. M. Thomson, 2 vols. (Oxford: Clarendon Press, 2007), I, pp. 324–7. 5 J. Lingard, A History of England from the First Invasion by the Romans, 3rd ed., 14 vols. (London: Mawson, 1825–31), I I, p. 35.

405

c. p. lewis

A similar appalled reaction to the presumed nature of the Harrying is still encountered in both history and fiction,6 though it has rarely gone as far as calling it genocide, except in works aimed at a popular market or derived from them.7 Wikipedia’s assertion (current in July 2020) that ‘Some present-day scholars have labelled [it] a genocide’ is misleading. The Harrying is called genocide in a single scholarly work, W. E. Kapelle’s The Norman Conquest of the North (1979), but only in the opening salvo of an attention-grabbing first paragraph: ‘William the Conqueror . . . made at least two blunders in dealing with the North and . . . rescued himself from the results of these mistakes by committing genocide’. Kapelle makes no attempt to explain or justify the term ‘genocide’.8 Nonetheless, the consensus has been that the Harrying was an exceptionally harsh treatment of the region and its people, and modern labelling commonly falls not far short of genocide. The Harrying is in fact protean, usually standing for Norman subjugation of the English (not just the Northumbrians), but also capable of signalling the south of England’s oppression of the north, or the warrior class’s cruelty to the peasantry. A succession of widely read general histories, biographies of William the Conqueror, and other works published since 1900, representative of others, provides an ever expanding thesaurus: ‘every village . . . a scene of massacre’;9 ‘homes . . . smouldering in the track of the destroying army’;10 ‘a vengeance Turkish in its atrocity’;11 ‘sustained ferocity’;12 ‘terrifying ruthlessness’;13 ‘cold-blooded . . . brutal . . . bestially cruel’;14 ‘reprisals’;15 6 P. Rex, The English Resistance: The Underground War against the Normans (Stroud: Tempus, 2004), pp. 87–106; J. Aitcheson, The Harrowing (London: Heron, 2016); compare the more measured treatment by D. M. Palliser, ‘Domesday Book and the “Harrying of the North”’, Northern History 29 (1993), 1–23. 7 R. Muir, The Lost Villages of Britain (London: Michael Joseph, 1982), pp. 80, 82; Rex, English Resistance, p. 103; A. D. Moses, ‘Empire, colony, genocide: keywords and the philosophy of history’ in Empire, Colony, Genocide: Conquest, Occupation, and Subaltern Resistance in World History, ed. A. D. Moses (New York: Berghahn, 2010), p. 28, citing only Rex, English Resistance. 8 W. E. Kapelle, The Norman Conquest of the North: The Region and Its Transformation, 1000– 1135 (London: Croom Helm, 1979), p. 3. 9 H. W. C. Davis, England under the Normans and Angevins, 1066–1272 (London: Methuen, 1905), p. 22. 10 F. M. Stenton, William the Conqueror and the Rule of the Normans (London: Putnam, 1908), pp. 280, 286. 11 G. M. Trevelyan, A Shortened History of England (1942; Harmondsworth: Penguin, 1959), p. 107. 12 F. M. Stenton, Anglo-Saxon England, 3rd ed. (1943; Oxford: Clarendon Press, 1971), p. 605. 13 A. Bryant, The Story of England: Makers of the Realm (London: Collins, 1953), p. 164. 14 D. C. Douglas, William the Conqueror: The Norman Impact upon England (London: Eyre & Spottiswoode, 1964), pp. 220–1, 373. 15 R. W. Finn, The Norman Conquest and Its Effects on the Economy, 1066–86 (London: Longman, 1971), p. 199.

406

William the Conqueror’s Harrying of the North, 1069–70

‘ruthless and systematic’;16 ‘a merciless, calculated exercise in slaughter and starvation’;17 ‘brutal’;18 ‘appalling’;19 ‘traumatic’;20 ‘a humanitarian disaster’;21 ‘almost unimaginable horrors’;22 ‘starving non-combatants to death’;23 ‘one of the most notorious incidents . . . of English history as a whole’.24 For an exiled Russian professor of jurisprudence, it was ‘a war of extermination’,25 and for others at different times it has recalled a ‘razzia’ of French colonial wars in the Maghreb,26 the Japanese ‘scorched earth policy’ in the Chinese War of Resistance in the 1930s,27 conditions ‘eerily like . . . a modern refugee camp’,28 and Lenin’s treatment of Ukraine in the 1920s.29 Quoting those excerpts prejudges four questions which this chapter addresses. Did the Harrying really constitute the equivalent of a war crime?30 Did it go beyond the norms of contemporary warfare? How extensive was the human suffering it caused? Was that suffering intended, indeed was it the purpose of the exercise? ‘Harrying’ was a contemporary word (Old English hergian, ‘to harry’) for a form of warfare that was both usual and normative. Land was harried to secure resources for oneself and deny them to an enemy. It was inevitably damaging to anyone in the way, especially peasants. During times of unrest, those given sufficient warning and capable of leaving would have tried to 16 P. H. Sawyer, From Roman Britain to Norman England (London: Methuen, 1978), p. 252. 17 S. Schama, A History of Britain: At the Edge of the World, 3000B C – A D1603 (London: BBC Books, 2000), p. 97. 18 D. Crouch, The Normans: The History of a Dynasty (London: Hambledon & London, 2002), p. 105. 19 D. Carpenter, The Struggle for Mastery: Britain, 1066–1284 (London: Allen Lane, 2003), p. 77. 20 F. H. A. Aalen, ed., England’s Landscape: The North-East (London: Collins, 2006), p. 203. 21 M. Wood, The Story of England (London: Viking, 2010), p. 118. 22 N. Vincent, A Brief History of Britain, 1066–1485 (London: Robinson, 2011), p. 71. 23 J. Gillingham, ‘Women, children and the profits of war’ in Gender and Historiography: Studies in the Earlier Middle Ages in Honour of Pauline Stafford, ed. J. L. Nelson, S. Reynolds and S. M. Johns (London: Institute of Historical Research, 2012), p. 61. 24 Marc Morris, The Norman Conquest (London: Windmill Books, 2013), p. 230. 25 P. Vinogradoff, English Society in the Eleventh Century: Essays in English Mediaeval History (Oxford: Clarendon Press, 1908), p. 35. 26 F. Palgrave, The History of Normandy and of England, 4 vols. (London: Parker, 1851–64), I I I, p. 456; for the word, Oxford English Dictionary. 27 H. R. Loyn, Anglo-Saxon England and the Norman Conquest (London: Longmans, 1962), p. 363; M. Chibnall, Anglo-Norman England, 1066–1166 (Oxford: Basil Blackwell, 1986), p. 18; F. Musgrove, The North of England: A History from Roman Times to the Present (Oxford: Basil Blackwell, 1990), p. 59. 28 H. M. Thomas, The English and the Normans: Ethnic Hostility, Assimilation, and Identity, 1066 – c. 1220 (Oxford University Press, 2003), p. 61. 29 Vincent, Brief History, p. 106. 30 D. Bates, William the Conqueror (London: George Philip, 1989), p. 84.

407

c. p. lewis

escape, taking food and saving their animals if they could. That left much vulnerable to armed men on horseback, who might slaughter or drive away plough oxen, flocks of sheep and goats, and the smaller numbers of cattle and pigs that were kept, and burn grain and other produce in store along with cottages and manorial buildings. Anyone who resisted would likely be killed, since by this period the Normans had stopped taking women and children as slaves.31 The particular circumstances of the 1069–70 campaign gave the Harrying some unusual characteristics. First, in order to secure the North by driving out the Danes and bringing the Northumbrian magnates to submission, the Normans needed to feed themselves and their horses in a season when there was nothing growing and at a time when the damage done in York by the earlier fighting probably made it hard to stockpile supplies. So, even more than usual, they would have taken corn and hay from manorial and peasant granaries for themselves and their mounts, rather than torching them, and rounded up livestock for the cooks. We can start a survey of the narrative sources with the Anglo-Saxon Chronicle, the only source truly close in time to 1069–70. There are two surviving versions, worded differently. Version E said only that the king’s army was large and that he ‘altogether harried and wasted the shire’ (Yorkshire), using vocabulary no different from any military campaign.32 Version D added a few words which could be taken for circumstantial detail, but it evidently confused the events of this campaign with William’s earlier expeditions.33 A third version, which has not survived, was evidently available in the 1110s when the monk John of Worcester used it as a source for his Latin Chronicon. He identified 1069 and 1070 as years in which the Normans ‘plundered . . . almost the whole realm, yet principally Northumbria and the adjacent shires’ and described the consequences: ‘famine so prevailed that men, compelled by hunger, devoured horseflesh, dogs, cats, and human beings’.34 31 S. Morillo, Warfare under the Anglo-Norman Kings, 1066–1135 (Woodbridge: Boydell, 1994), pp. 98–102, 124–31; M. Strickland, War and Chivalry: The Conduct and Perception of War in England and Normandy, 1066–1217 (Cambridge University Press, 1996), pp. 258–90. 32 The Anglo-Saxon Chronicle: A Collaborative Edition, 7: MS. E, ed. S. Irvine (Cambridge: D. S. Brewer, 2004), p. 88; The Anglo-Saxon Chronicle: A Revised Translation, ed. D. Whitelock (London: Eyre & Spottiswoode, 1961), p. 149. 33 The Anglo-Saxon Chronicle: A Collaborative Edition, 6: MS D, ed. G. P. Cubbin (Cambridge: D. S. Brewer, 1996), p. 84; trans. Whitelock, p. 149. 34 John of Worcester, The Chronicle of John of Worcester, ed. and trans. R. R. Darlington, P. McGurk and J. Bray, vols. I I and I I I (Oxford: Clarendon Press, 1995–8), I I I, pp. 10–11.

408

William the Conqueror’s Harrying of the North, 1069–70

Famine was a frequent visitor to eleventh-century England.35 Grain yields were low and many peasants had little margin above subsistence. Bad weather, crop blight and cattle plague, as well as war, could easily mean disaster. Between 1066 and 1100 the Anglo-Saxon Chronicle mentioned famine in four separate years, each time saying that it was ‘great’ or ‘very severe’, as if ordinary shortages were not worth mentioning. The only indication of numbers is that ‘many hundreds’ died in 1087.36 Perhaps John’s lost Chronicle for 1070 also said more than the two surviving versions, but it is as likely that John was elaborating on his source, since he was ever a helpful editor whose enthusiasm to tell readers what his sources meant often led him to over-explanation.37 The suspicion must be that his sequence of what the starving devoured is a rhetorical flourish, conducting readers from the taboo but tasty to the unpalatable and finally the horrific. Famine in 1070 is highly plausible as a consequence of harrying in midwinter. By December 1069 the seed corn for next year’s crop of wheat would have been in the ground, but in the North – especially on peasant farms – wheat was not the most important bread grain; the 1069 harvest of barley and oats would still have been in granaries, kept as seed for sowing in spring, or awaiting threshing to make pottage, bread and ale through the months to come; at least some households would have had stocks of bacon salted down in the autumn, and perhaps cheeses and orchard fruit in store, besides hay for the plough oxen. Hunger, then desperate famine, were the likely outcome of a winter harrying. Refugees returning home after the army had passed might well have found nothing to eat and no seed corn. Cannibalism, however, is nearly always suspect in medieval narratives: there were too many biblical models in authors’ minds, and no source which reported it ever claimed eyewitness veracity.38 35 C. Dyer, ‘Did the peasants really starve in medieval England?’ in Food and Eating in Medieval Europe, ed. M. Carlin and J. T. Rosenthal (London: Hambledon, 1998), pp. 53–71. 36 1070, 1082, 1087, 1096: Anglo-Saxon Chronicle, ed. Cubbin, p. 85; ed. Irvine, pp. 89, 92, 95, 107; trans. Whitelock, pp. 151, 153, 160, 162, 174. 37 R. R. Darlington and P. McGurk, ‘The “Chronicon ex chronicis” of “Florence” of Worcester and its use of sources for English history before 1066’, Anglo-Norman Studies 5 (1982), 185–96, at 187. 38 J. Marvin, ‘Cannibalism as an aspect of famine in two English chronicles’ in Food, ed. Carlin and Rosenthal, pp. 73–86; V. Vandenberg, ‘Fames facta est ut homo hominem comederet: l’Occident médiéval face au cannibalisme de survie (Ve–XIe siècle)’, Revue belge de philologie et d’histoire 86:2 (2008), 217–72; J. Rubenstein, ‘Cannibals and crusaders’, French Historical Studies 31:4 (2008), 525–52; T. Bourns, ‘Meat and taboo in medieval Scandinavian law and literature’, Viking and Medieval Scandinavia 14 (2018), 61–80.

409

c. p. lewis

Symeon of Durham wrote his Tract on the Origins and Progress of the Church of Durham between 1104 and 1109, and added local material to an existing History of the Kings over a longer period until his death in the late 1120s. He took John of Worcester’s reference to famine and cannibalism as the starting point for a longer treatment of terrible times. Symeon may well have had some local knowledge of circumstances north of the Tees. Although not an eyewitness, having come to Durham probably in 1091, he may have heard local traditions.39 In the Tract, Symeon says in general terms that King William ‘came with an army to York and devastated everything round about’, almost exactly what the York author Hugh the Chanter said too, writing soon after 1127.40 For Symeon, that statement was background to events in Durham’s territory: the abbot’s flight to Lindisfarne with the body of St Cuthbert during a cold winter, and a vision and miracle of St Cuthbert and St Oswald which did not involve the Normans.41 For the History, Symeon copied John of Worcester word for word on the fighting in York in summer 1069, but provided his own account of hostilities in Durham. His record of the angry King William’s arrival in Northumbria, where he ‘ceaselessly ravaged all winter long, and slaughtered the people’, was again pure John, but rather than stopping at cannibalism, Symeon promoted it to first in the tally of dreadful food, and added an extended passage of his own composition: others sold themselves to perpetual slavery, so that they might in any way preserve their wretched existence; others, while about to go into exile from their country, fell down in the middle of their journey and gave up the ghost. It was horrific to behold human corpses decaying in the houses, the streets, and the roads, swarming with worms, while they were consumed in corruption with an abominable stench. For no one was left to bury them in the earth, all being cut off either by the sword or by famine, or having left the country on account of the famine. Meanwhile, the land being thus deprived of anyone to cultivate it for nine years, an extensive solitude prevailed all around. There was no village inhabited between York and Durham; they became lurking places for wild beasts and robbers, and were a great dread to travellers.42 39 W. M. Aird, St Cuthbert and the Normans: The Church of Durham, 1071–1153 (Woodbridge: Boydell, 1998). 40 Hugh the Chanter, The History of the Church of York, 1066–1127, ed. and trans. C. Johnson, rev. M. Brett, C. N. L. Brooke and M. Winterbottom (Oxford: Clarendon Press, 1990), pp. 2–3. 41 Symeon of Durham, Libellus de exordio atque procursu istius, hoc est Dunhelmensis, ecclesie: Tract on the Origins and Progress of this the Church of Durham, ed. and trans. D. Rollason (Oxford: Clarendon Press, 2000), pp. 184–93. 42 Historia regum, in Symeonis monachi opera omnia, ed. T. Arnold, 2 vols., Rolls Series 75 (London: H. M. Stationery Office, 1882–5), I I, p. 188; Simeon of Durham, A History of the Kings of England, trans. J. Stephenson (London, 1858), p. 137.

410

William the Conqueror’s Harrying of the North, 1069–70

This is powerful and gruesome, and much cited by modern historians keen to stress the dire consequences of the Harrying. But, except for the detail of nine years’ solitude, it is extremely vague. The wilderness or solitude as a place of punishment for God’s people is a theme found throughout the Old Testament and would have been vividly present in Symeon’s historical imagination. His language did not involve direct biblical quotation, but a reason for not taking his account literally is that it lacks the geographical particularity of his later additions to John of Worcester, which described further episodes of harrying. That includes in 1070 the temporary desertion of the land between the Tees and the Tyne by locals who fled to the woods and mountains, Jarrow church destroyed by fire, and Durham crowded with hungry and diseased wretches unable to get further away; and later in the same year a Scots invasion which ravaged Teesdale and Cleveland, and burned Wearmouth church. Alongside such details, Symeon continued to include non-specific atrocities, especially as committed by the Scots: ‘Some aged men and women were beheaded with the sword; others were thrust through with pikes, like swine destined for food; infants snatched from their mothers’ breasts were thrown high into the air, and in their fall were received on the points of lances and pikes thickly placed in the ground.’43 The theme of Scottish barbarity reappeared in other authors’ work later in the twelfth century, often with very similar and equally vague allegations. Symeon believed that his people, the ‘holy people’ of Durham and more widely the Northumbrians, had been caught between two brutal kings, but he made the consequences of William’s actions unintended, whereas the Scottish king Malcolm was cruelty personified. The realities of the 1070 famine are most apparent in a source geographically far removed from the scene, the Deeds of Abbot Æthelwig of the abbey of Evesham (1058–77) preserved in a later history of the abbey. The work was written close to the time by Dominic, evidently an eyewitness and later prior of Evesham.44 The second part covers Æthelwig’s activities as abbot. It starts with an account of his merits as a father to the poor and wretched, especially widows, orphans and strangers, and then provides an illustration in a passage which can only refer to the Harrying of the North. Indeed, in the early years of his reign, King William had some shires in these regions of England laid waste because of the exiles and outlaws who were hiding in the woods everywhere, and inflicting enormous damage upon 43 Historia regum, ed. Arnold, I I, pp. 189–91; trans. Stephenson, pp. 138–9. 44 Thomas of Marlborough, History of the Abbey of Evesham, ed. and trans. J. Sayers and L. Watkiss (Oxford: Clarendon Press, 2003), pp. 172–3; R. R. Darlington, ‘Æthelwig, abbot of Evesham’, English Historical Review 48 (1933), 1–22, 177–98.

411

c. p. lewis

many people, that is, Yorkshire, Cheshire, Shropshire, Staffordshire, and Derbyshire, from where a large number of young and old, and women and children came fleeing to Evesham in great distress from the misery of famine. In his concern for them all, Æthelwig gave them all the sustenance he could. Many who had long been oppressed by severe hunger died through eating the food too ravenously. Throughout the whole town, wretches lay languishing either in the houses or in the street, even in the cemetery itself, being exhausted by hunger before they got there, and therefore when they tasted food . . . most of them died. For this reason there was high mortality . . . so almost every day five or six people, sometimes more, perished miserably and were buried by the prior . . . As these people included many young children, the abbot entrusted to every servant or official of this church, and even to some brethren . . . the responsibility of sustaining one child each . . . Some of these children grew up to be men of high integrity who served the brethren honourably in many spheres of duty.45

Much of this is utterly convincing as an eyewitness description of famine – hunger, flight, exhaustion, malnutrition, death from over-eating (the twentieth century’s Refeeding Syndrome), daily average deaths, care for orphans. Such things would have been imprinted in the community’s memory at Evesham and been carried long down the years by the children who remained servants there. Prior Dominic had no interest in – and did not mention – events in the North in 1069–70, and ‘these regions’, from a viewpoint in Evesham, could hardly include Yorkshire, a hundred miles away. A better fit is with the districts west of the Pennines where William campaigned in February and March 1070, which overlap with the shires where Abbot Æthelwig had special judicial powers, including Shropshire and Staffordshire. Cheshire is missing from the list of the abbot’s shires but may simply have dropped off the end during copying and transmission of that text. Yorkshire and Derbyshire were definitely not under the abbot’s jurisdiction,46 though late eleventh-century Yorkshire stretched west to the Irish Sea and included the districts of Craven and Amounderness through which the king’s army may have passed in early 1070 to cross the Pennines. Evesham was an improbable target for starving refugees from the Vale of York or Teesdale, but from Amounderness, as from Cheshire and Shropshire, it would have been natural to head south. The number of desperate families at Evesham need not have been large to have made the impression conveyed by Prior Dominic, and presumably distant Evesham was only one of the churches where refugees ended up. 45 Thomas of Marlborough, History, pp. 172–3.

412

46 Ibid., pp. 166–7.

William the Conqueror’s Harrying of the North, 1069–70

Further evidence for the displacement of people in 1069–70 comes from Durham and Beverley. Symeon reported that Bishop Æthelwine of Durham and others fled the city with the body of St Cuthbert ahead of William’s army, while the cathedral itself ‘became a den for the poor, the infirm, and the sick, who . . . lay there perishing of hunger and disease’.47 Two different kinds of displaced persons are involved here: the elite removing themselves and their most precious possession (St Cuthbert’s body) from harm’s way, and the poor, compelled to seek safety in the very place that the rich had abandoned as unsafe. Evidence for the minster church of Beverley in the East Riding of Yorkshire is preserved in a late source of around 1150, Alfred of Beverley’s epitome of the History of the Kings of Britain, which relied on Symeon of Durham for the period of the Norman Conquest. When discussing the campaign of 1069, Alfred omitted the flight of Bishop Æthelwine (not of interest at Beverley), and substituted a local miracle story in which St John of Beverley felled the horse and crippled the leader of a party of Norman knights, Thurstan, who was intent on robbing an elderly Englishman trying to gain sanctuary in the church. The Englishman was rich – ‘expensively clothed and wearing a gold bracelet’ – which is why he attracted attention among the crowds of refugees (‘all the population of that region’).48 The story appears at greater length in the 1170s,49 but Alfred provides three details omitted later which point to the authenticity of his account: the Norman camp was seven miles from Beverley (rather than ‘not far from there’), Thurstan’s victim was expensively clothed (not ‘a poor wretch’), and the horse broke its neck when the saint brought it down. The literal truth of what lies behind the miracle is not recoverable; what the story shows is that there was a tradition at Beverley that when William’s army was campaigning against the Danes in the East Riding, the countryside emptied as people of every class fled to the sanctuary of Beverley Minster; this is not a tale of starving and desperate refugees many miles from home, but of families removing themselves and their belongings from the path of a Norman army on campaign. 47 Symeon, Libellus, pp. 136–9. 48 Alfred of Beverley, Annales, sive historia de gestis regum Britanniae, libris ix, ed. T. Hearne (Oxford, 1716), p. 128; J. P. Slevin, ‘The historical writing of Alfred of Beverley’ (PhD thesis, University of Exeter, 2013), 206; J. P. Slevin, ‘Observations on the twelfth-century Historia of Alfred of Beverley’, Haskins Society Journal 27 (2015), 101–28. 49 Slevin, ‘Historical writing’, 44–5; printed in The Historians of the Church of York and its Archbishops, ed. J. Raine, 3 vols., Rolls Series 71 (London: H. M. Stationery Office, 1879– 94), I, pp. 264–9.

413

c. p. lewis

Orderic Vitalis wrote an Ecclesiastical History of the Norman world down to the time of William the Conqueror in the 1120s. For the early years of the Conquest he had two excellent Norman sources written close in time to the events of 1069–70: William of Jumièges’ Deeds of the Dukes of the Normans – which Orderic had amplified earlier in his career – and William of Poitiers’ Deeds of William. Jumièges was vague about the English rebellions, compressing events to the extent of conflating the northern campaigns of summer 1069 and winter 1069/70, and mentioning a massacre in York but not any harrying of the countryside.50 Poitiers, however, was a chaplain in the service of the duke-king during the Conquest years and wrote about military affairs with authority, having been a knight in his youth.51 Orderic took much of his account of the English rebellions and William’s campaigns between 1067 and 1069 direct from Poitiers. Where we can compare the two we can see exactly what Orderic added, but Poitiers’ text is without its ending, being known only from an edition of 1619 which drew on a single manuscript since destroyed by fire. It breaks off in 1067, after which it is a matter of judgement what of Orderic’s history is Poitiers reworded and what is Orderic. Orderic started his account of the English rebellions in his own words, staking out his position in unambiguous terms (‘meanwhile the English were being crushed by Norman pride’), and employed a highly coloured vocabulary: the English were ‘suffering oppressions’; the Normans ‘heaped shameful burdens on them’, while the king’s lieutenants ‘were guilty of plunder and rape’, and committed ‘cruel wrongs’; the English thus ‘groaned aloud’ under the Norman ‘yoke’ and finally were ‘goaded to rebellion’.52 In other words, Orderic had a powerful view of causation and responsibility in English history in these years.53 The history that follows alternated detailed narrative and interpretation, the former taken from Poitiers but the latter in a mode alien to Poitiers’ way of writing. The movements of the king’s army were given with some precision and must all be from Poitiers,54 but Orderic then slid into generalities which do not sound anything like Poitiers’ tone of voice, reverting instead to impassioned and impressionistic rhetoric. The king 50 William of Jumièges, The Gesta Normannorum ducum of William of Jumièges, Orderic Vitalis, and Robert of Torigni, ed. and trans. E. M. C. van Houts, 2 vols. (Oxford: Clarendon Press, 1992–5), I I, pp. 178–83. 51 William of Poitiers, The Gesta Guillelmi of William of Poitiers, ed. and trans. R. H. C. Davis and M. Chibnall (Oxford: Clarendon Press, 1998). 52 Orderic Vitalis, The Ecclesiastical History of Orderic Vitalis, ed. and trans. M. Chibnall, 6 vols. (Oxford: Clarendon Press, 1969–80), I I, pp. 196–201. 53 E. A. Winkler, Royal Responsibility in Anglo-Norman Historical Writing (Oxford University Press, 2017). 54 Orderic, Ecclesiastical History, I I, pp. 230–1.

414

William the Conqueror’s Harrying of the North, 1069–70

himself continued to comb forests and remote mountainous places, stopping at nothing to hunt out the enemy hidden there. His camps were spread out over an area of a hundred miles. He cut down many in his vengeance; destroyed the lairs of others; harried the land, and burned homes to ashes. Nowhere else had William behaved so cruelly. Shamefully he succumbed to this vice, for he made no effort to restrain his fury and punished the innocent with the guilty. In his anger he commanded that all crops and herds, chattels and food of every kind should be brought together and burned to ashes with consuming fire, so that the whole Transhumbrian region might be stripped of all means of sustenance. In consequence so serious a scarcity was felt in England, and so terrible a famine fell upon the humble and defenceless populace, that more than 100,000 Christian folk of both sexes perished, young and old.55

Orderic should be read with care. The passage is noxiously vivid but unspecific. Royal anger was sometimes a stylised performance, an assertion of will rather than a loss of control, and always needs to be interpreted in a nuanced way.56 Some of the Latin is diagnostic of Orderic’s idiolect: cruelly (crudeliter) was a favourite adverb; Transhumbrian (meaning Northumbrian) is an unusual word which he employed elsewhere.57 None of this reads like Poitiers’ military reporting. There is also a non sequitur which undermines credibility, since the destruction of foodstuffs north of the Humber would not have caused scarcity and famine in England at large. The number of famine victims is unreliable: 100,000 is neither an estimate nor even a guess; the figure serves as simply a very big number, pointedly larger than the largest mass slaughters that would have been familiar from the Old Testament.58 Orderic knew perfectly well, from a lifetime on the disturbed frontier of Normandy, what the countryside looked like when a harrying army was in the vicinity. In a passage dealing with Normandy, he evocatively described ‘A great crowd on foot . . . carrying across their necks and shoulders animals and clothes and every kind of furnishing and household goods that raiders usually seize as plunder’, reporting the vision of a local priest who was frightened by the sound of a great army nearby. In his vision, the priest supposed the army 55 Ibid., I I, pp. 230–3. 56 B. H. Rosenwein, ed., Anger’s Past: The Social Uses of an Emotion in the Middle Ages (Ithaca, NY: Cornell University Press, 1998); D. Bates, ‘Anger, emotion and a biography of William the Conqueror’ in Gender and Historiography: Studies in the Earlier Middle Ages in Honour of Pauline Stafford, ed. J. L. Nelson, S. Reynolds and S. M. Johns (London: Institute of Historical Research, 2012), pp. 21–33. 57 Orderic, Ecclesiastical History, I, p. 278; I I, p. 216; I V, p. 94. 58 1 Samuel 6:19; 2 Samuel 24:15; Esther 9:16.

415

c. p. lewis

to be the local magnate’s soldiers, but it turned out to be the legendary ‘Wild Hunt’ led by the mythical Herlechin.59 Orderic concluded his account of the Harrying with a condemnation of King William which was certainly all his own: for this act which condemned the innocent and the guilty alike to die by slow starvation I cannot commend him. For when I think of helpless children, young men in the prime of life, and hoary grey-beards perishing alike of hunger I am so moved to pity that I would rather lament the griefs and sufferings of the wretched people than make a vain attempt to flatter the perpetrator of such infamy . . . such brutal slaughter cannot remain unpunished [by God].60

The immediate juxtaposition shows that Orderic’s presentation of what happened after the occupation of York functions in the narrative as a way of setting up this moralistic judgement. In the following passage, by contrast, Orderic returned to Poitiers’ account for the remainder of the campaign, offering a straight narrative full of specific detail: an attack on an enemy camp along a narrow causeway, pursuit to the river Tees, then on to Hexham, back to York, and finally across to Cheshire. This section included what must be Poitiers’ observations about the difficulties of crossing the Pennines in winter, but nothing at all about the routine foraging of the soldiery, which Poitiers seems not to have mentioned.61 In a later book of the Ecclesiastical History Orderic returned to his theme in a long death-bed speech which he put into the king’s mouth as a way of conveying what he as author thought was the historical and moral significance of the reign. The speech was, of course, invented, and recycled some of the wording of the earlier account: ‘In mad fury I descended on the English of the North like a raging lion, and ordered that their homes and crops with all their equipment and furnishings should be burnt at once and their great flocks and herds of sheep and cattle slaughtered everywhere.’62 None of this tells us anything about what actually happened. Orderic had no personal knowledge of the Harrying and no access to firsthand accounts other than William of Poitiers’. In fact he knew almost nothing about Northumbria beyond what he had read in a handful of identifiable sources. What he had was an opinion about William the Conqueror, founded upon his own half-English origins and strong identification with the English. The fragile factual basis of his over-the-top account is betrayed by a single word 59 Orderic, Ecclesiastical History, I V, pp. 238–9. 62 Ibid., I V, pp. 80–95.

416

60 Ibid., I I, pp. 232–3.

61 Ibid., I I, p. 233n.

William the Conqueror’s Harrying of the North, 1069–70

that he interpolated into Jumièges’ Deeds of the Dukes of the Normans, where he wrongly identified as Cumberland the unnamed county in which the rebel stronghold of Durham stood. Cumberland was in fact on the other side of the country, and not in Norman hands in 1069.63 All the vivid but unspecific details of the Harrying offered by Symeon and Orderic can be dismissed as unverifiable and historically implausible in the circumstances of a short winter campaign by a relatively small force with the clear objectives of containing and driving out the Danes and bringing the Northumbrian magnates to heel. We can safely dismiss the idea that William systematically ravaged a hundred miles of difficult upland countryside and burned piles of crops, animals, chattels and food in order to strip the whole of Northumbria; and we can set aside the notion that the campaign left 100,000 dead of famine and an extensive solitude without cultivation for nine years. None of this was literal description and may not have been meant to be understood as such. Stripping it out of what we can reasonably know about the winter campaign, we are nonetheless still left with harrying, famine, starvation and desperate refugees. We can turn now to the widespread occurrence in Domesday Book (1086) of northern manors described as ‘waste’, a feature on both sides of the Pennines but especially of Yorkshire. This is a large subject which needs fuller discussion than is possible here. It has been widely supposed that the word ‘waste’ literally described land which was uncultivated, uninhabited, empty and deserted, and that it reflects the continuing abandonment of land physically destroyed during the Harrying in 1069–70.64 In fact the term is used in three distinct ways in the materials of the Domesday survey, only one of which fits that definition comfortably. First, the word does occasionally describe a specific act of destruction, normally as a reflex of the verb wastare: as at a small manor on the Fosse Way in Warwickshire ‘laid waste by the king’s army’, perhaps through some act of requisitioning or violence by soldiers passing along the Roman road there.65 Second, ‘waste’ occasionally seems to describe exemption from geld (tax on land), as when a specified 63 Jumièges, Gesta Normannorum, I I, pp. 178–9. 64 T. A. M. Bishop, ‘The Norman settlement of Yorkshire’ in Studies in Medieval History Presented to Frederick Maurice Powicke, ed. R. W. Hunt, W. A. Pantin and R. W. Southern (Oxford: Clarendon Press, 1948), pp. 1–14; Kapelle, Norman Conquest of the North, pp. 158–90; J. Palmer, ‘War and Domesday waste’ in Armies, Chivalry and Warfare in Medieval Britain and France, ed. M. Strickland (Stamford: Paul Watkins, 1998), pp. 256–75. 65 Domesday Book, ed. J. Morris et al., 34 vols. (Chichester: Phillimore, 1974–86), Warwickshire 6,13.

417

c. p. lewis

portion of a larger manor was ‘waste’ but the rest was taxed and had an economic value, as in a series of episcopal manors in Worcestershire.66 That second type is met more fully in the geld accounts of the late eleventh and twelfth centuries.67 The third type comprises the overwhelming majority of citations of Domesday ‘waste’, hundreds of examples rather than handfuls: ‘waste’ as a statement of no value, taking the place in Domesday’s manorial descriptions normally occupied by a sum of money. That third type can be explored further. Domesday manorial values represent the annual income derived from manors by the lord, that is, the surplus over and above what the peasants consumed and kept for themselves. Some such sums must have been actual cash rents, others a notional but realistic statement of what could be expected. They do not represent the total economic activity of the manor. When, for whatever reason, there was no surplus, Domesday hardly ever said (ualet) nichil, ‘(it is) worth nothing’,68 perhaps because ‘worth nothing’ was an elusive concept in a system using roman numerals which had no zero. Instead it said wasta, ‘waste’. Manorial values were a matter of great interest to the Domesday survey. The king and his advisers wanted to know how much each manor brought in annually to its lord. For that reason, the Domesday survey also enquired into each manor’s resources in land, livestock and people, so that treasury clerks and local sheriffs could see in broad terms whether the capabilities of the manor were being fully realised in the seigneurial income generated.69 The vocabulary of Domesday, common to all counties, was fixed at the start of the survey and adjusted in the final stage when the scribe of the presentation copy, Great Domesday Book, began writing. Much of the language is artificial, formulated for the purposes of the survey.70 One late adjustment was to use wasta across the board instead of the variants 66 Domesday Book: Worcestershire 2,2; 2,45; 2,72; 2,76; 2,79. 67 Anglo-Saxon Charters, ed. and trans. A. J. Robertson, 2nd ed. (Cambridge University Press, 1956), pp. 230–7, 481–4; Exon Domesday, entries 2a4, 8b3, 9a5, 9b3, 14b5, 15b3, 16a3, 78b1, www.exondomesday.ac.uk; The Great Roll of the Pipe for the Thirty First Year of the Reign of King Henry I, Michaelmas 1130, ed. J. A. Green, Pipe Roll Society n.s. 57 (London: Pipe Roll Society, 2012), p. 119. 68 An exception: Domesday Book: Sussex 9,25. 69 S. Baxter, J. Crick and C. P. Lewis, Making Domesday: Intelligent Power in Conquered England (Oxford University Press, forthcoming). 70 F. Thorn, ‘Non pascua sed pastura: the changing choice of terms in Domesday’ in Domesday Now: New Approaches to the Inquest and the Book, ed. D. Roffe and K. S. B. Keats-Rohan (Woodbridge: Boydell, 2016), pp. 109–36; S. Baxter, ‘The making of Domesday Book and the languages of lordship in conquered England’ in Conceptualizing Multilingualism in Medieval England, c. 800 – c. 1250, ed. E. M. Tyler (Turnhout: Brepols, 2011), pp. 271–308; C. P. Lewis, ‘The invention of the manor in Norman England’, Anglo-Norman Studies 34 (2012), 123–50.

418

William the Conqueror’s Harrying of the North, 1069–70

employed by different scribes earlier (uasta, uastata, deuastata, uacua, literally ‘vacant, empty’, and guasta, alongside wasta). The Great Domesday scribe may have preferred wasta because it was closest to English weste. The semantic range of weste certainly included ‘waste, uncultivated land’, but more to our point, it appears in the geld accounts and other sources as a category of untaxed land.71 Domesday values may have had something of the character of public information, and it is conceivable that values recorded as waste also implied exemption from geld, logically enough, since a manor not producing a surplus could hardly be expected to pay a tax which was assessed on productive arable land. It may also have followed that the Domesday commissioners were not required to record the resources in plough teams and peasants of manors which were making no annual surplus and paid no tax. That could be true even though on occasion the manorial descriptions noted a real absence of resources, as when a manor of waste value had land for half a plough ‘but it is not there’, or another which ‘when it is stocked is worth 10 shillings’.72 A manor listed as ‘waste’ need not have been uninhabited (and indeed some in Yorkshire and Cheshire did list a few inhabitants making small payments): it was simply that neither the lord’s home farm nor the peasants were producing a surplus. That said, there was a wide range of circumstances which might have caused a manor to produce no annual surplus, all the way from complete ruin and depopulation to some little local difficulty, like the flooding recorded at a manor by the Wash in Lincolnshire.73 The Harrying of 1069–70 was no doubt severe, not, I have argued, as a calculated act of complete destruction but because it was exceptional in taking place in the depth of a cold winter by an invading army which needed to feed itself off the land. We know that people left their manors, suspect that many more died than would have happened at another time of year, and might suppose that in some cases refugees were slow to return. William’s winter war was not the only military campaign in Yorkshire between 1066 and 1086 that disrupted farming and destroyed manorial surpluses. In September 1066 a sizeable Norwegian fleet and army was at large on the way to York, countered first by Northumbrian and Mercian forces, then by Harold. William’s advance on York in summer 1068 will have been a drain of local food supplies, and the ensuing rebellion and the Norman 71 Anglo-Saxon Charters, ed. Robertson, pp. 164–9, 230–7. 72 Domesday Book: Bedfordshire 47,3; Warwickshire 16,4. 73 Domesday Book: Lincolnshire 57,36.

419

c. p. lewis

recapture of the city in late winter and early spring 1069 was probably more disruptive. The Danish force that came in 1069 was also large, lived off the land through the autumn and winter, and was allowed by King William to forage along the coast in late winter and spring 1070 as a means of buying it off. Malcolm, king of Scots, harried the North in summer 1070, and again at harvest time 1079. In 1085, King William had the coasts of England laid waste to deny supplies to another threatened Danish invasion. Besides those military visitations there were natural disasters: a great famine in England in 1082, and bad weather and pestilence among livestock in 1086. Even the most favoured parts of the North were relatively thinly settled and less productive than other regions, with a wetter and colder climate and hence a shorter growing season. The failure of so many northern manors to produce a surplus in 1086 was less likely the long-lasting result of a single winter campaign of ravaging sixteen years earlier than the consequence of repeated military and environmental shocks over twenty years. The most striking evidence of actual depopulation anywhere in Northumbria in 1086 was not in any of the districts known to have been harried by William’s army, but in Amounderness on the west coast. Amounderness comprised the head ‘vill’ (township) of Preston and sixty-one other vills scattered over an extensive territory characterised by poor land: boggy near the coast rising to bleak upland in the east. Domesday is laconic: ‘Of these [vills], 16 are inhabited by a few people but it is not known how many the inhabitants are. The rest are waste.’74 The implication is that 46 of the 62 vills were uninhabited in 1086 (their lands could have been cultivated or grazed by people living in the inhabited vills). Harrying by William on his way from York to Chester in 1070 is possible but unproven, since it is not certain that Amounderness was even on the route taken. The point is that there were plenty of other reasons for depopulation and waste values in Domesday Book. Yorkshire manors with ‘waste’ values had not necessarily been ‘laid waste’ by the Normans in 1069–70. They had no greater propensity to disappear off the map than manors making a profit in 1086. The overwhelming majority of the hundreds of such manors in Yorkshire continued in later centuries. One feature of Yorkshire and Durham is the large number of regularly planned villages. They were long thought to have been laid out by landowners across a landscape completely devastated by the Harrying.75 That explanation is no 74 Domesday Book: Yorkshire 1L1. 75 J. A. Sheppard, ‘Metrological analysis of regular village plans in Yorkshire’, Agricultural History Review 22 (1974), 118–35; B. K. Roberts, ‘Village plans in county Durham: a preliminary statement’, Medieval Archaeology 16 (1972), 33–56.

420

William the Conqueror’s Harrying of the North, 1069–70

longer sustainable or widely held: the chronology of village formation was longer and probably later, and the planned villages, waste Domesday values and likely course of the Conqueror’s campaign do not coincide particularly well even in general distribution, let alone in detail.76 Contemporary kings did sometimes issue an explicit threat to depopulate a land if unable to subdue it,77 but William’s campaign achieved his clearly defined objectives: the Danish army was outmanoeuvred and neutralised, and the Yorkshire and Northumbrian nobility were brought into meek submission. Probably soon afterwards William gave new Norman lords shares of the conquered land.78 They and their agents, and the king’s sheriff and reeves on his behalf, would have been keen to make the land productive and profitable, for which they needed local peasant producers alive, housed and paying rents and taxes. The Norman Conquest, in the North as elsewhere, established new lords and more exploitative lordship on the English peasantry. It was not a ‘settler colony’ which required a landscape cleared of existing settlements. The campaign of 1069–70 was a harrying at the harsh upper end of the scale of such operations. People of all social classes fled in advance of the Normans to whatever sanctuaries they could find. Earlier disturbances around York had depleted food stocks and disrupted supply channels, so that the Normans may have had to forage for everything they needed. Because it was winter, the consequences of foraging and harrying over a large area on both sides of the Pennines were a severe famine which scattered the starving a hundred miles from home. Sick and traumatised, they may have repeated stories they had heard of cannibalism. William’s harrying need not have had genocidal intent to have had extreme consequences. The wide scattering of refugees made the famine far more generally known than most similar military campaigns, which is probably why it struck such a chord with the twelfthcentury historians, and why, for them, it came to stand for the very worst that the Normans did in England. Even so, it is clear that Symeon of Durham thought it was the Scots king who was depraved in his treatment of the Northumbrians, not William.

76 B. K. Roberts, Landscapes, Documents and Maps: Villages in Northern England and Beyond, AD 900–1250 (Oxford: Oxbow Books, 2008), pp. 58–150; J. Blair, Building AngloSaxon England (Princeton University Press, 2018), pp. 309–80, 408–15. 77 Aelred of Rievaulx, Opera historica et hagiographica (Opera omnia, VI), ed. D. Pezzini (Turnhout: Brepols, 2017), p. 59. 78 P. Dalton, Conquest, Anarchy and Lordship: Yorkshire, 1066–1154 (Cambridge University Press, 1994), pp. 19–78.

421

c. p. lewis

When the Anglo-Norman historians of the earlier twelfth century came to think about the failed English rebellions against Norman rule, the northern campaign of 1069–70 struck them as exemplary. With biblical examples in mind, they could easily imagine the intended destruction of a people and a wilderness enduring for years.79 Symeon put the emphasis on famine, Orderic on Norman brutality. How they described the Harrying was more about slotting it into an interpretive framework than about capturing literal historical truth. Interestingly Henry of Huntingdon, whose conceptual framework for insular history positioned the Normans as a plague sent to eradicate the English from the island of Britain, disposed of the entire 1069–70 campaign in seventeen words, not exactly anodyne but very far from apocalyptic.80 None of them saw either the Harrying itself or the great famine that ensued as God’s punishment of the Northumbrians for sin. Nonetheless, a ‘reading’ of the Harrying as the massacre of one people by another was certainly developing a generation or two after the actual event, and it would be a profitable exercise for another time to consider it alongside similar medieval legends.81 ‘Harrying’ as a term has a long history in describing a type of warfare, from the Old English translation of Orosius’ world history in the late ninth century through the Anglo-Saxon Chronicle of the tenth and eleventh centuries to accounts of fighting on the Anglo-Scottish border in the late Middle Ages and the sixteenth century.82 Modern dictionary definitions are more than adequate to describe what William did to the North: ‘Warlike incursion; devastation, laying waste; ravaging, plundering, raiding’.83 It was a severe example of harrying, but there is no evidence for the extreme claims made for its effects: every village between York and Durham destroyed, ‘up to’ 100,000 dead, ‘the country . . . largely uninhabited for a century thereafter’.84 The limitations of the evidence make it hard to test the Harrying of the North against modern legal and sociological definitions of genocide. 79 L. Scales, ‘Bread, cheese and genocide: imagining the destruction of peoples in medieval western Europe’, History 92 (2007), 284–300. 80 Henry, Archdeacon of Huntingdon, Historia Anglorum: The History of the English People, ed. and trans. D. Greenway (Oxford: Clarendon Press, 1996), pp. 396–7. 81 Cf. S. Keynes, ‘The massacre of St Brice’s Day (13 November 1002)’ in Beretning fra Seksogtyvende Tværfaglige Vikingesymposium, ed. N. Lund (Aarhus University Press, 2007), pp. 32–66. 82 A. King and D. Simpkin, eds., England and Scotland at War, c. 1296 – c. 1513 (Leiden: Brill, 2012); G. Phillips, The Anglo-Scots Wars, 1513–1550: A Military History (Woodbridge: Boydell, 1999). 83 Oxford English Dictionary, 1st ed. (Oxford: Clarendon Press, 1898), verbal sense 2a. 84 Moses, ‘Empire’, 28.

422

William the Conqueror’s Harrying of the North, 1069–70

Nevertheless, we can cautiously say that William’s intentions were defined by his military and political objectives, not by a desire to exterminate a people. He will have known the likely effects of stripping the countryside of supplies for his own army, since he was an experienced commander who had spent a lifetime fighting small wars of harrying from the saddle, and burning villages was second nature to him.85 The horror of the Harrying was in its effects, not its intent. It was a winter war in which the armies never engaged, but in which William the Conqueror’s army ravaged and foraged across Yorkshire, Durham and northwest Mercia, triggering a great famine and a refugee crisis.

Bibliographic Note The Harrying makes an appearance in a great many textbooks, historical dictionaries and general works besides those noted in the chapter above, as well as in histories of the north of England and Yorkshire, military history, landscape history and innumerable local histories. Likewise, ‘waste’ is discussed widely in the extensive scholarship on Domesday Book, even when ostensibly not central to the matter in hand. ViIlage histories and archaeological reports often have something to contribute to the origins and dating of individual planned settlements, though Blair’s recent book Building AngloSaxon England sets thinking about the origins of the nucleated village on an entirely new footing. In other words, ‘Writing the Harrying of the North’ would be a large topic in itself. Only recent works which engage significantly with the Harrying, or waste, or both, are listed here. Among recent scholarship, a maximalist view (‘genocide’ or something very close) is taken by Kapelle (Norman Conquest of the North), Rex (English Resistance) and Palmer (‘War and Domesday waste’). They should be read on the understanding that Kapelle was full of new ideas and determined to be iconoclastic, and that Rex was seemingly unaware of modern source criticism. One of the most minimalist positions is taken by Dalton (Conquest, Anarchy), with not quite the focus of this chapter. Palliser (‘Domesday Book’), Williams (English and the Norman Conquest) and Bates (‘Anger, emotion and a biography of William the Conqueror’) all follow the sources closely and explain some of their difficulties; they all conclude that the Harrying was more destructive than most such campaigns, 85 F. Barlow, William I and the Norman Conquest (London: English Universities Press, 1965), pp. 26–34.

423

c. p. lewis

do not take Domesday ‘waste’ at face value but seem a little uncomfortable about saying so, and are not convinced that a catastrophic outcome was intended. The narrative sources for the events of 1069–70 have almost all been translated in good modern editions: Whitelock for the Anglo-Saxon Chronicle, and the Oxford Medieval Texts series for John of Worcester, Symeon of Durham’s Tract on the Church of Durham, Orderic Vitalis, William of Malmesbury and Thomas of Marlborough (the Evesham Deeds of Abbot Æthelwig). The Durham History of the Kings is in an older translation. Domesday Book is easiest to consult in the edition by Messrs Phillimore.

424

16

Genocidal Massacres of Jews in Medieval Western Europe, 1096–1392 maya soifer irish

The lives and religious observances of Jews living in medieval western Europe were protected under St Augustine’s interpretation of the biblical injunction to ‘slay them not lest at any time they forget your law’ (Psalm 59), known as the doctrine of Jewish witness. In every Christian European kingdom, for the duration of the medieval centuries, ecclesiastical authorities officially espoused Augustine’s view that the Jews were destined to survive until the end of times, scattered among the peoples and subjugated by them, in order to serve as a living testimony to the truth of their scripture, and to endure the punishment for rejecting and crucifying Jesus Christ.1 From the purely demographic standpoint, there is no denying that the Augustinian principle, bolstered by the protections extended to Jews by secular powers, was remarkably successful in securing the physical survival of European Jewry in the Middle Ages. Taken as a whole, the Jewish population of Catholic Europe grew steadily between the years 1000 and 1500, its numbers eclipsing those of Jews in the Islamic lands.2 Yet, it is also an indisputable fact that during the same period European Jews endured recurrent episodes of violence at the hands of Christians who targeted them because of their legal and religious status as Jews. Among the most consequential mass killings, which cost Jewish communities hundreds and in some cases thousands of lives, were the slaughter of Jews in the Rhineland during the First Crusade (1096), the massacres in England (1189–90), the Rintfleisch and Armleder massacres in Germany (1298, 1336–8), the Shepherds’ Crusade violence in France and northern Iberia (1320–1), attacks on Jews during the Black Death 1 Jeremy Cohen, Living Letters of the Law: Ideas of the Jew in Medieval Christianity (Berkeley and Los Angeles: University of California Press, 1999), pp. 30–6; Anna Sapir Abulafia, Christian–Jewish Relations, 1000–1300: Jews in the Service of Medieval Christendom (Harlow: Pearson, 2011), p. 408. 2 Robert Chazan, Reassessing Jewish Life in Medieval Europe (Cambridge University Press, 2010), pp. xix–xx.

425

maya soifer irish

Map 16.1 Anti-Jewish violence in western Europe, 1096–1392. (Cartography by Amy Ferguson, GIS/Data Center, Rice University)

epidemic (1348–51) and the anti-Jewish urban riots in Castile and Aragon (1391–2). The regularity and intensity of these attacks suggest that official protections were at best an inadequate bulwark against the outbreaks of Christian hostility. Part of the blame lay with the contradictory message embedded in Augustine’s teaching: while permitting the Jews a legitimate place in Christian society, it enshrined their status as enemies of Christ and Christianity who were condemned to enslavement and exile for the crime of deicide. The rhetoric of medieval theologians reflected this ambiguity inherited from Augustine. Influential ecclesiastical figures like St Bernard and Peter the Venerable could inveigh against the supposed incorrigible 426

Genocidal Massacres of Jews in Western Europe, 1096–1392

wickedness of the Jews, and yet insist that they be protected from violence. It is hardly surprising that many Christians grasped only the first part of the message, and were unable to leave the matter of Jewish guilt to divine judgment and overcome the temptation to exact an immediate and personal revenge on the supposed killers of Christ.3 The link between anti-Jewish preaching and violence helps explain the fact that so many of the assaults happened in the atmosphere of heightened religious fervour, especially during Holy Week or while a new crusade was being preached.4 The attackers’ resentment was also driven by a sense that the Jews, protected by secular and ecclesiastical leaders whom they served, occupied a position in Christian society that seemed out of sync with the lowly status accorded to them by the teachings of the Church. Ever since the Roman Empire’s adoption of Christianity in the fourth century, the Jews’ quasi-legal position had been that of a captive people. Early Christian writers argued that the enslavement of the Jews and the destruction of their Temple by Vespasian and Titus was a just retribution for the rejection of Christ. Subsequent Roman legislation upheld the legality of Judaism, but also emphasised their subjugated position in the now Christian society. Thus, the theological formulation of Jewish inferiority always had political connotations: Jews were supposed to serve Christians, just like the biblical Esau had served Jacob.5 But beginning with the Carolingians, that status became muddled, as kings, princes and major ecclesiastical leaders sought to attract Jews to their domains in an attempt to stimulate local economies, increase revenues from taxation and even (most prominently, in Spain) to utilise the administrative expertise of Jewish notables. The interests of these rulers required that Jewish communities have religious autonomy, enjoy economic privileges and prosperity, and be protected from attacks by hostile neighbours. By the twelfth century a notion developed that the Jews ‘belonged’ to the royal treasury and were dependent on secular rulers.6 The Jews’ close association with authority proved to be a doubleedged sword. While it allowed Jewish communities to prosper and grow, it made them susceptible to fiscal exploitation and, by the end of the Middle 3 Jeremy Cohen, ‘Christian theology and anti-Jewish violence in the Middle Ages: connections and disjunctions’ in Religious Violence between Christians and Jews: Medieval Roots, Modern Perspectives, ed. Anna Sapir Abulafia (New York: Palgrave, 2002), pp. 44–60; Chazan, Reassessing Jewish Life, p. 166. 4 Jonathan Riley-Smith, ‘Christian violence and the Crusades’ in Religious Violence, ed. Abulafia, pp. 3–4; David Nirenberg, Communities of Violence: Persecution of Minorities in the Middle Ages (Princeton University Press, 1996), pp. 200–30. 5 Abulafia, Christian–Jewish Relations, pp. 13–16. 6 Gavin Langmuir, Toward a Definition of Antisemitism (Berkeley and Los Angeles: University of California Press, 1990), pp. 167–94.

427

maya soifer irish

Ages, expulsion by royal edict from England, France and Spain. Moreover, the perception that Jews exercised power they were not supposed to possess led to violent reprisals aimed at putting them back in their inferior place.7 Since the Jews were seen as agents of the monarchy, revolts against royal power and attacks on Jews often happened in tandem.8 The Jews’ position at the nexus of religion and power brought tangible benefits for a time, but ultimately proved to be unsustainable since eliminating the Jews through conversion, violence or expulsions held the tantalising promise of simultaneously correcting political evils and repairing society’s spiritual integrity.9 That said, individual perpetrators of anti-Jewish violence were not always moved by such lofty goals. Medieval commentators generally attributed the violence to the attackers’ venality and greed, describing it as unChristian and contrary to God’s will. Indeed, economic insecurity and indebtedness to Jewish moneylenders were also factors in the violence, which often involved despoliation of Jewish property and destruction of loan contracts. Although driven by long-term anti-Jewish narratives and stereotypes, the violence always played out within local, historically specific contexts.10 Starting in the twelfth century, the dissemination of conspiracy theories about Jews allegedly committing ritual murder or desecrating the Eucharistic host exposed Jewish communities in some regions of Europe to bloody reprisals. The arrival of the Black Death epidemic in 1348–51 also led to regional massacres, when Jews were held responsible for spreading the disease. Contrary to a widespread perception, the perpetrators of anti-Jewish violence were not always lower-class, ignorant masses driven by greed or fear. There is plentiful evidence that Christians from all walks of life – clerics, noblemen, burghers and peasants – participated in the assaults on Jewish communities, and that many of the attackers were under the sway of anti-Jewish ideologies elaborated and disseminated by secular and ecclesiastical elites.

First Crusade Massacres in the German Lands, 1096 The attacks perpetrated by crusading bands in 1096 were one of the first incidents of mass violence against the European Jewish communities, and it left a deep impression on both Christians and Jews. It prompted the issuing of 7 Mark Meyerson, A Jewish Renaissance in Fifteenth-Century Spain (Princeton University Press, 2004), p. 14. 8 Nirenberg, Communities of Violence, p. 48. 9 Cohen, ‘Christian theology’, 53. 10 Miri Rubin, Gentile Tales: The Narrative Assault on Late Medieval Jews (New Haven and London: Yale University Press, 1999), p. 2.

428

Genocidal Massacres of Jews in Western Europe, 1096–1392

the papal bull of Sicut Judeis by Calixtus II that warned against conversion of Jews to Christianity by force.11 The martyrdom (kiddush ha-shem) of some Jews in the Rhenish cities of the Holy Roman Empire, who slaughtered their families and committed suicide rather than abandon their faith, was recorded by the Christian chronicler Albert of Aachen and memorialised in three Hebrew chronicles in the first half of the twelfth century.12 Based on the accounts of the papal speech at the Council of Clermont in November 1095 (all written considerably later), Urban II urged an armed pilgrimage against the Muslim Seljuk Turks to avenge the killings of Christian pilgrims in the Holy Land and to take possession of the Christian holy sites in Jerusalem. The Jews of the Holy Land, let alone European Jews, were never mentioned in the speech.13 Nevertheless, the enthusiastic bands of crusaders that began to gather during the winter of 1095/6 saw little difference between enemies who had crucified Christ centuries ago and the foes who were currently occupying his land, torturing Christians and threatening the eastern Roman Empire. The first violence may have occurred in northern France (Rouen), since the French Jews wrote a letter to their co-religionists in the Rhineland warning them of the danger.14 In early May 1096, a mix of crusaders and local burghers tried to attack the Jews of Speyer, but the local bishop was able to shield most of them from the assaults. In Worms, the Jews were not so lucky. In mid-May a group of crusaders, townsmen and villagers first attacked the Jews who had remained inside their homes, then stormed the local bishop’s palace, where the rest of the Jewish community had found refuge. The violence became more organised and intense in late May at Mainz, where the army of Count Emicho of Flonheim was joined by French, English, Flemish, Swabian and Lorrainer crusaders. The attackers, who seem to have been ideologically committed to annihilating the Jews either through killing or forced conversion, systematically hunted down and murdered every Jew they could find. A part of Emicho’s army then marched to Cologne (June–July), where just as methodically they tracked down and murdered the Jews dispersed throughout the nearby villages by their bishop 11 Solomon Grayzel, ‘The papal bull sicut judeis’ in Essential Papers on Judaism and Christianity in Conflict: From Late Antiquity to the Reformation, ed. Jeremy Cohen (New York University Press, 1991), pp. 235–6. 12 Robert Chazan, European Jewry and the First Crusade (Berkeley: University of California Press, 1987), pp. 38–49. 13 S. J. Allen and Emilie Amt, The Crusades: A Reader, 2nd ed. (University of Toronto Press, 2014), pp. 34–42. 14 Jonathan Riley-Smith, The First Crusade and the Idea of Crusading (Philadelphia: University of Pennsylvania Press, 2009), pp. 50, 55; Chazan, European Jewry, p. 225.

429

maya soifer irish

in a failed effort to save them. Another part of the army possibly slaughtered the Jews in Trier, Metz and Prague.15 Peter the Hermit, a charismatic popular preacher of the crusade, did not directly call for violence against the Jews, but threatened them with his admonitions to avenge the sufferings of Christ, causing the entire Jewish community of Regensburg to convert.16 The social composition of these ad hoc armies that attacked Jews at the start of the First Crusade was mixed: among the killers were barons, townsmen and peasants. At least some of the violence was directed by major nobles whose armies included well-trained foot soldiers and knights.17 As in many episodes of mass violence, the motivations of the killers were complex as well. Most seem to have been driven by a desire to avenge the enemies of Christ at home before embarking on a dangerous voyage to fight his adversaries in the Holy Land. At least some of the participants were influenced by eschatological expectations, and sought to wipe out the Jews in an anticipation of the end of days.18 The local burghers who helped with the killings had their own grievances against the Jews. Relative newcomers to the German lands, the Jews were not well integrated into the local society, and they became a convenient target for violence at the hands of burghers who were upset about the infringements on their autonomy by the emperor and the towns’ lords. Some of the bishops who served as imperial surrogates in the Rhinish towns could not provide effective leadership and protection for their Jewish clients.19 While Albert of Aachen was wrong to attribute anti-Jewish violence entirely to the crusaders’ greed, concerns about provisions and money played a role in the attacks. The first wave of crusaders set out in the spring of 1096, before the official date set by the pope (15 August), and before the summer harvest was collected. Driven by hunger as well as by hatred of Jews, they may have believed that canon law allowed them to expropriate goods of the infidels for the purpose of provisioning a holy war.20 German Jewish communities were devastated and stunned by the violence of the First Crusade. No reliable statistics exist for the number of dead, 15 Riley-Smith, First Crusade, p. 50; Robert Chazan, ‘The anti-Jewish violence of 1096: perpetrators and dynamics’ in Religious Violence, ed. Abulafia, pp. 25–32. 16 Jay Rubenstein, Armies of Heaven: The First Crusade and the Quest for Apocalypse (New York: Basic Books, 2011), p. 49. 17 Riley-Smith, First Crusade, p. 51; Riley-Smith, ‘Christian violence’, 12. 18 Rubenstein, Armies of Heaven, p. 51. 19 Chazan, European Jewry, pp. 36, 83–4; Robert Chazan, God, Humanity, and History: The Hebrew First Crusade Narratives (Berkeley: University of California Press, 2000), p. 1. 20 Riley-Smith, First Crusade, p. 52; Jonathan Riley-Smith, The Crusades: A History (London and New York: Bloomsbury, 2014), p. 50.

430

Genocidal Massacres of Jews in Western Europe, 1096–1392

but Jewish accounts mention that 800 Jews died in Worms and 900– 1100 perished in Mainz.21 The toll in Trier, Metz and Prague is not recorded. An unknown number of Jews were forcibly converted to Christianity, although emperor Henry IV later allowed these forced converts to return to Judaism. It was not the number of victims, however, but rather the manner in which many of them had died that shocked both Christian and Jewish contemporaries. In the words of Albert of Aachen, ‘The Jews, seeing that their Christian enemies were attacking them and their children, and that they were sparing no age, likewise fell upon one another, brother, children, wives, and sisters, and thus they perished at each other’s hands.’22 In the twelfth century, their deaths were commemorated in three Hebrew chronicles, all written in Mainz, that tried to make sense of what had happened and to allay the survivors’ sense of guilt and lingering doubts about the Halakhic permissibility of killing one’s family members.23 Although Talmudic law mandated martyrdom (kiddush ha-shem, sanctification of God’s name) when a Jew faced the choice between practising idolatry and death, homicide was strictly forbidden. Keenly aware of the new and unprecedented character of these acts, the writers of the Hebrew Chronicles constructed an ideology of martyrdom for the benefit of the survivors of the massacres, comparing the suicides and the killings to the sacrifice at the ancient Temple in Jerusalem, and emphasising the heavenly reward awaiting the martyrs. Their deaths were portrayed as a spiritual and moral victory over their Christian slaughterers.24 Subsequently, Ashkenazi rabbinic authorities endorsed the notion that martyrdom was mandatory and the taking of a loved one’s life permissible when the alternative was a forcible conversion to Christianity.25 In material terms, the violence of 1096 had a limited impact on the life of German Jewish communities, whose economic, demographic and cultural development continued unimpeded. But the spectre of crusader violence continued to haunt European Jewry for generations to come. After the fall of 21 Simha Goldin, The Ways of Jewish Martyrdom, trans. Yigal Levin (Turnhout: Brepols, 2008), p. 9. 22 Allen and Amt, Crusades: A Reader, p. 52. 23 Shlomo Eidelberg, trans. and ed., The Jews and the Crusaders: The Hebrew Chronicles of the First and Second Crusades (Madison: University of Wisconsin Press, 1977), pp. 21–115; Jeremy Cohen, Sanctifying the Name of God: Jewish Martyrs and Jewish Memories of the First Crusade (Philadelphia: University of Pennsylvania Press, 2004), p. 22; Abulafia, Christian–Jewish Relations, pp. 142–3. 24 Cohen, Sanctifying the Name of God, pp. 13–14, 22; Goldin, Ways of Jewish Martyrdom, pp. 9, 95–102. 25 Chazan, European Jewry, pp. 146, 157.

431

maya soifer irish

Edessa to Muslims in 1144, Pope Eugenius III proclaimed a new crusade, and soon afterwards popular preachers began to fan out through the Rhineland, exhorting the people to take the cross. As during the First Crusade, some preachers linked this message with an incitement to attack the Jews. According to Otto of Freising, a monk named Ralph travelled through the Rhenish cities preaching the crusade and proclaiming that the Jews ‘should be slain as foes of the Christian religion’. The Jewish writer Ephraim of Bonn thus described Ralph’s message: ‘Avenge the Crucified upon his enemies who live among you. Afterwards you shall journey to battle against the Muslims.’ As a result, violence broke out in some cities, most notably in Wurzburg, where twentytwo Jews were murdered.26 Fortunately, the experience of the First Crusade taught Christian secular and ecclesiastical leaders to be more proactive in defending the Jewish communities under their protection. The bishops and the emperor understood that the townsmen attacking the Jews were also rebelling against their authority. Urged by the bishop of Mainz, Bernard, the bishop of Clairvaux and a strong advocate for the Second Crusade, wrote a letter condemning violence against Jews, and later went to the Rhineland, confronted Ralph, and convinced him to return to his monastery. St Bernard’s reputation for saintliness prevented the violence from spreading further.27

Massacres in England, 1189–1190 Before the end of the twelfth century, English Jewry was spared the mass violence endured by Jews of continental western Europe. The first Jewish communities were not established in England until after the Norman Conquest in 1066, and Jewish immigrants continued to arrive from France as late as the second half of the twelfth century.28 When large-scale antiJewish riots broke out in 1189 and 1190, they had many of the same underlying causes as the violence in the Rhineland a hundred years previously: crusader preaching, apocalyptic expectations, efforts to assert local autonomy, economic insecurity, and resentment against the royal government whose interests the Jews served. Unlike in 1096, crusaders set to depart for the Holy Land played only a minor role in the violence. As in the Rhineland, 26 Ibid., pp. 8, 63, 172; Robert Chazan, Church, State, and Jew in the Middle Ages (West Orange, NJ: Behrman House, 1980), pp. 100–8. 27 Chazan, Church, State, and Jew, p. 107. 28 Robert Stacey, ‘Jews and Christians in twelfth-century England: some dynamics of a changing relationship’ in Jews and Christians in Twelfth-Century Europe, ed. Michael A. Signer and John Van Engen (University of Notre Dame Press, 2001), pp. 340–1.

432

Genocidal Massacres of Jews in Western Europe, 1096–1392

the violence spread over a large territory, affecting London, East Anglia, Lincolnshire and Yorkshire. In York, a French-born rabbi inspired many Jews to commit kiddush ha-shem rather than submit to forced conversion, an act that was memorialised in Hebrew poems penned in Germany and France. The events in England are well documented thanks to the writings of several contemporary Christian chroniclers (Ralph de Diceto, Richard of Devizes and William of Newburgh, among others) and the Hebrew account by Ephraim of Bonn. King Richard I, who had been preparing to fulfill his crusading vow, was crowned at Westminster on 3 September 1189. Jews were prohibited from attending the ceremonies, but during the feast that followed the coronation several of them somehow ended up on the other side of the palace doors, enraging the Christian crowd. A full-scale massacre then commenced, as the Jews who had gathered near the palace were beaten with fists, sticks and stones, and trampled underfoot. After a rumour spread that the king had authorised the killing of Jews, a large Christian crowd marched to the Jewish quarter and set the roofs of houses on fire. At least thirty Jews lost their lives in the London riot. One of them was Benedict of York, who was severely wounded and forced to accept conversion, only to recant and die several days later. According to Ephraim of Bonn, some Jews ‘slaughtered themselves and their children’.29 Before departing on the Third Crusade in mid-December 1189, King Richard sent letters throughout the kingdom placing the Jews under his protection, but the king’s absence created a vacuum of authority that enabled anti-Jewish violence to spread. According to Newburgh, the first attack happened at King’s Lynn, where in January or early February 1190 sailors joined the townsmen in storming the Jewish quarter. The start of the Lenten season fuelled the attackers’ hatred of Jews, and copycat violence spread from town to town. During the first week of February, some Jews in Norwich were murdered in their houses, and the rest found refuge in the royal castle. In early March, young men who had taken the cross led the attacks in Stamford. In Lincoln, some of the town’s most prominent citizens formed an armed band and tried to assault the Jews and plunder their property, but the bishop of Lincoln took decisive action to protect them, and the Jews were able to escape to the walled upper city and its castle. There are also scattered references, some of them unconfirmed, to attacks in Bury St Edmunds, 29 R. B. Dobson, The Jews of Medieval York and the Massacre of March 1190 (York: St Anthony’s Press, 1974), pp. 21–5; Chazan, Church, State, and Jew, pp. 158–61.

433

maya soifer irish

Ospringe, Colchester and Thetford.30 In many of the assaults, the rioters not only aimed to kill the Jews but also to eliminate their presence from the city by torching the Jewish neighbourhood.31 The worst attack by far happened in York in early March 1190. Taking advantage of a fire that was raging through the city, a band of local notables led by an indebted baron, Richard Malebisse, stormed the house of Benedict of York, the victim of the London riot, killing his widow and children, plundering the house and burning it to the ground. Frightened by this incident, most of York’s Jews sought protection inside the royal castle. Meanwhile, Malebisse’s crew proceeded to plunder the house of another prominent York Jew, Josce, killing everyone left behind, then marching through the streets of the city and forcing every Jew they met to choose between baptism or death. The original conspirators were joined by men from the countryside, labourers, townsmen of middling means and clerics. A Premonstratensian hermit delivered fiery anti-Jewish sermons predicting a speedy arrival of the end of days. As the situation escalated, the Jews ensconced in the castle refused to readmit the royal constable, and the sheriff of Yorkshire ordered their removal from the castle by force. Although he later rescinded the order, the crowd interpreted it as a royal approval of their plans to storm the castle. After siege machines were pulled into place on 16 March, Rabbi Yom Tov of Joigny urged the besieged Jews to sanctify God’s name by killing themselves and their families. Many followed his example, and the survivors pledged to accept baptism. However, when they emerged from the castle, the frenzied crowd slaughtered all of them.32 The economic subtext of the violence in York is easy to see: immediately after the massacre, Malebisse and his co-conspirators went to York Minster, seized the bonds documenting Christian debts to Jews, and burned them in the middle of the church. However, the broader picture of the anti-Jewish riots in England from autumn 1189 to spring 1190 suggests that the perpetrators’ aims were more far-reaching. Richard I’s absence from the kingdom created an opportunity for his subjects to openly vent their frustration with the expansion of the Angevin bureaucracy, the royal administration’s impingement on the authority of municipal councils, and an increase in the 30 Joe Hillaby, ‘Prelude and postscript to the York massacre: attacks in East Anglia and Lincolnshire, 1190’ in Christians and Jews in Angevin England, ed. Sarah Rees Jones and Sethina Watson (York Medieval Press, 2013), pp. 43–54. 31 Dobson, Jews of Medieval York, p. 25. 32 Ibid., pp. 26–8; Sethina Watson, ‘The moment and memory of the York massacre of 1190’ in Christians and Jews, ed. Rees Jones and Watson, pp. 5–6; Goldin, Ways of Jewish Martyrdom, pp. 202–7.

434

Genocidal Massacres of Jews in Western Europe, 1096–1392

financial demands imposed by the monarchy. As in the Rhineland in the 1090s, the Jews were relative newcomers, especially in towns like York, where the Jewish community was less than one generation old. They were small in number, poorly integrated into the local community, and perceived as the agents of the Crown. There were expectations that Richard I would emulate the king of France, Philip Augustus, who had recently expelled the Jews from the royal domain.33 Religious motivations also must not be ignored. The overlap between the ongoing preparations for the Third Crusade and the Lenten season heightened Christian animosity towards the Jews.34 Despite the hundreds of deaths, and the psychological impact on the Jewish communities in London and other cities, the massacres had limited long-term consequences. York Jews quickly returned to the city, achieving higher levels of economic prosperity after the outbreak of violence than they had before.35

Rintfleisch 1298 and Armleder 1336–1338 The massacres that took place in Germany in 1298 and 1336–8 were prompted by a novel type of anti-Jewish accusation that had surfaced in England about four decades before the violence in London and York. Allegations that medieval Jews actively conspired against Christians and acted out their hatred by either kidnapping and crucifying Christian children or stealing and abusing the Eucharistic host have been linked by scholars to a new emphasis among Christians on the humanity and suffering of Jesus, the growing popularity of miracle stories about St Mary, and the cult of the Holy Innocents.36 In many cases, however, the anti-Jewish claims and the resultant violence had roots in political divisions, power grabs by local secular and ecclesiastical elites, and the propensity of townspeople and countryfolk to blame the Jews for the oppressive actions of their governments. When it first emerged in twelfth-century England, the ritual murder accusation did not lead to mass violence. It was on the Continent, especially 33 Sarah Rees Jones, ‘Neighbours and victims in twelfth-century York: a royal citadel, the citizens and the Jews of York’ in Christians and Jews, ed. Rees Jones and Watson, pp. 27– 32; Stacey, ‘Jews and Christians’, 341–3; Chazan, Church, State, and Jew, pp. 287–9. 34 Dobson, Jews of Medieval York, p. 25. 35 Rees Jones, ‘Neighbours and victims’, 37–41; Dobson, Jews of Medieval York, pp. 38–43. 36 Langmuir, Toward a Definition, pp. 115–16; Miri Rubin, Gentile Tales: The Narrative Assault on Late Medieval Jews (New Haven: Yale University Press, 1999), pp. 7–27; E. M. Rose, The Murder of William of Norwich: The Origins of the Blood Libel in Medieval Europe (Oxford University Press, 2017), p. 235.

435

maya soifer irish

in the politically fragmented lands of the Holy Roman Empire, that the newly fashioned host desecration narratives acquired lethal potency, causing the loss of thousands of Jewish lives. The Rintfleisch and Armleder massacres are not as well known as the violence of the First Crusade, and yet they claimed many more victims.37 There is now a consensus among scholars that the accusations of ritual murder and host desecration were not a product of the popular imagination but rather the result of an elaboration and dissemination by religious and lay elites. By 1150, the monks at Norwich Cathedral Priory had crafted an allegation that a twelve-year-old named William, found in the woods near the town in 1144, had been crucified by local Jews. Soon after, similar accusations surfaced at Gloucester (1168), Bury St Edmunds (1181) and Bristol (1183), and spread to the Continent. Some secular authorities lent credence to the charges by punishing Jews for their alleged crimes. In 1171, Count Thibaut of Blois burned about thirty Jews to death as a punishment for supposedly drowning a Christian child, even though no children were missing, and there was no body.38 In 1192, King Philip Augustus of France, who as a child had heard rumours that Parisian Jews sacrificed a Christian every year during the Lenten season, condemned about eighty Jews to the flames for killing a Christian in Bray (or Brie).39 Not surprisingly, the first ‘fully documented’ case of a host desecration accusation happened in the ‘much governed, well-policed city’ of Philip IV of France in 1290. A Jew was put on trial and burned at the stake by the royal authorities for allegedly procuring a consecrated host from a Christian woman, striking it with a knife and a hammer until it bled, and throwing it in the fire and water. A cult developed in Paris around the miraculous host, and the desecration narrative was quickly disseminated to the Low Countries, Germany and Italy by the monks of Saint-Denis, preachers, merchants and chroniclers.40 According to various chronicles, starting in April 1298 a wave of anti-Jewish violence swept through the towns of Franconia, a loosely administered region of the Holy Roman Empire. The chroniclers (Prior Rudolph of Schlettstadt and Abbot Peter of Zittau) justified the violence by portraying it as a revenge for a host desecration allegedly committed by the Jews of 37 Cohen, Sanctifying the Name of God, p. 31. 38 Rose, Murder of William of Norwich, pp. 133–51, 187–206, 162. 39 Chazan, Church, State, and Jew, pp. 288–90, 304–6; Abulafia, Christian–Jewish Relations, pp. 180–2. 40 Rubin, Gentile Tales, pp. 40–8; Susan Einbinder, Beautiful Death: Jewish Poetry and Martyrdom in Medieval France (Princeton University Press, 2002), pp. 155–79.

436

Genocidal Massacres of Jews in Western Europe, 1096–1392

Röttingen. However, the indebtedness of many lower- and middle-class townsmen and villagers to Jewish moneylenders also fuelled their animosity. Led by a knight nicknamed ‘Rintfleisch’, bands of disaffected noblemen, burgesses and peasants moved from town to town, burning Jewish houses and indiscriminately killing men, women and children. Unlike in previous episodes of mass killing, the victims were not given the option of baptism, although one account tells of Jews throwing their children and wives into the fire to escape forced conversion.41 From Röttingen and its surroundings, the violence spread to Rothenburg and Würzburg, reaching Nuremberg by late July, and spreading out to the regions of Swabia, Hess and Thuringia. The necrological list compiled by the Jews of Nuremberg mentions 146 communities affected by the violence; between 3,000 and 5,000 Jews lost their lives in these massacres. The German chroniclers vary on the details of the killings, but they all emphasise political grievances of the attackers and their resentment of the protections given to Jews by those in positions of authority. The vacuum of power in the empire also played a role: order was not restored until Albert I of Hapsburg returned to the throne and punished some of the perpetrators.42 By the time another anti-Jewish uprising, the so-called Armleder rebellion, began in 1336, in the same area as the Rintfleisch massacres, the Jews of the empire had become increasingly dependent on the power of individual bishops, princes, counts and towns, and vulnerable to violence associated with local political and social conflicts. The uprising was led by another knight, Arnold of Uissigheim, hailed as ‘King Armleder’ by his followers – townsmen and small farmers. The movement had all the markings of a popular rebellion fanned by grievances against the established authority, resentment of Jewish moneylenders and the urge to avenge rumoured host desecrations allegedly committed by Jews in various locales. Arnold was executed by the end of 1336, but the movement spread from Franconia to Alsace, Hesse, Swabia and Austria. The rebellion targeted the most densely populated areas of Jewish settlement in Germany, destroying sixty-five communities. In 1337, the host desecration accusation was used to justify the attacks on Jews in Lower Bavaria, affecting about twenty communities. A year later, under the same pretext, 150 Jews were slaughtered in Pulkau in Lower Austria, and around thirty communities were destroyed in Upper 41 Goldin, Ways of Jewish Martyrdom, p. 220. 42 Rubin, Gentile Tales, pp. 48–54; Eva Haverkamp, ‘Jews in Christian Europe: Ashkenaz in the Middle Ages’ in The Wiley Blackwell History of Jews and Judaism, ed. Alan T. Levenson (Oxford: Wiley-Blackwell, 2012), p. 187.

437

maya soifer irish

Austria, Bohemia and southern Moravia.43 Overall, the Armleder movement claimed the lives of as many as 6,000 Jews.44

Pastoureaux Violence in France and Aragon, 1320–1321 The impact of violence on Jewish communities in the fourteenth century proved to be much more serious than that of the earlier massacres as it was often exacerbated by expulsions and mass conversions. It may be that the royal expulsion of Jews from England in 1290 set a precedent for dispensing with the Augustinian tradition and establishing an all-Christian society. After the Rintfleisch and Armleder rebellions, followed by anti-Jewish persecutions during the Black Death epidemic, Jews were hesitant to return to the cities and towns in the western part of the empire; many of those who did come back were soon expelled by the local authorities.45 In France, anti-Jewish violence and accusations of well poisoning in 1320–1 served as a justification for the second exile of Jews from the kingdom (the first one had occurred in 1306). And in Spain, the anti-Jewish riots of 1391–2 led to thousands of deaths and conversions, and a permanent disappearance of numerous Jewish communities. All of these anti-Jewish movements were rebellions against the established order that the authorities to varying degrees were unable or unwilling to check. Ultimately, the violence against Jews was triggered by a series of upheavals collectively known as the crises of the fourteenth century: famines, plague, wars, civil strife and an economic downturn. The uprising of the Pastoureaux (the Shepherds) in France and Aragon had some similarities with the earlier anti-Jewish violence associated with the preaching of crusades to Jerusalem. After King Philip V of France took the cross in 1313 and began preparations for his campaign, the prospect of a new crusade was met with a surge of popular enthusiasm.46 It was said that a young shepherd had a vision of a beautiful lady commanding him ‘to go forth to fight with the Moors’. In 1320, bands of mostly poor and young would-be crusaders, whose numbers also included clerics, women and minor nobles, began to assemble in Normandy, then moved on to Paris, where they asked the king to lead their crusade but were rebuffed. The shepherds 43 44 45 46

Haverkamp, ‘Jews in Christian Europe’, 188, 190; Rubin, Gentile Tales, p. 65. Gavin Langmuir, ‘At the frontiers of faith’ in Religious Violence, ed. Abulafia, p. 152. Haverkamp, ‘Jews in Christian Europe’, 192. Jonathan Riley-Smith, The Crusades: A History, 3rd ed. (London and New York: Bloomsbury, 2014), p. 297.

438

Genocidal Massacres of Jews in Western Europe, 1096–1392

continued their march south, into Aquitaine, Provence, Gascony, Navarre and Aragon. Contemporary French chroniclers describe how, disillusioned with the monarchy and by the public order in general, they attacked royal castles, killing royal and seigneurial officials, clerics and Jews.47 Seen as the fiscal agents of the monarchy, Jews had been recently readmitted to France and found themselves in a particularly vulnerable position.48 One famous firsthand account by a Jew named Baruch described the events during the massacre in Toulouse (12 June). Desperate for protection, Baruch appealed to his Christian friends, one of whom was a friar and another a royal official, but they were unable to save him from forced baptism.49 In Aragon, where some French Pastoureaux had crossed in an apparent belief that the king of Aragon needed their assistance in repelling an imminent attack by the Muslims of Granada, the violence was more limited. At the royal castle of Montclus, however, 337 Jews were massacred, and some were forcibly converted. Afterwards, royal officials imposed fines on around seventy local Christians found to be complicit in the violence. While the majority of them were very poor, the list included wealthier individuals and even a few noblemen. Royal archives preserve a rare glimpse into how some Aragonese Jews reacted to the violence. Those who had come to Montclus to bury the dead were so overcome by grief and anger that they rioted, cutting down trees, destroying the town’s bridge and hurling insults at local Christians. Several Jewish aljamas (communities) were punished with fines.50 Meanwhile, municipal authorities in the same areas of southern France that had been affected by the Pastoureaux violence raised accusations that lepers and Jews were conspiring to contaminate wells and poison the entire country. Both groups were associated with spiritual sin and moral corruption: the lepers’ physical illness was considered to be a sign of the disease of the soul, while the alleged evidence of host desecration and ritual murder proved the Jews’ intent to harm Christians. But the fact that much of the violence was legitimised and carried out by municipal officials, rather than by disorderly mobs, suggests that the southern towns were also reacting to the royal encroachments on their jurisdictional privileges. In early 1321, the towns of Toulouse, Albi and Carcassonne petitioned the king to expel the Jews and 47 Nirenberg, Communities of Violence, pp. 43–6. 48 William Chester Jordan, The French Monarchy and the Jews (Philadelphia: University of Pennsylvania Press, 1989), pp. 244–5. 49 Susan Einbinder, After the Black Death: Plague and Commemoration among Iberian Jews (Philadelphia: University of Pennsylvania Press, 2018), pp. 28–9; Nirenberg, Communities of Violence, p. 46. 50 Nirenberg, Communities of Violence, pp. 80–5, 91.

439

maya soifer irish

segregate the lepers. When the petition was not met with approval, municipalities began to take action on their own. In a number of towns, lepers were arrested, tortured and put to death by burning. Attacks on Jews quickly followed. French chroniclers mention the arrests and burnings of Jews in Touraine (160 victims in Chinon), Aquitaine, Toulouse and the diocese of Besançon. In June 1321, Philip, count of Anjou and the future king of France, wrote a letter to the pope, alleging a conspiracy between Jews, Muslims and lepers to poison the wells. In July, feeling the popular pressure and facing the possibility of another rebellion, Philip V issued an edict ordering the arrest of Jews suspected of collaborating with lepers. When the Jews were forced to leave France the following year, this alleged crime was cited as the main justification.51

Black Death Violence, 1348–1351 The Black Death epidemic had catastrophic consequences for the Jews of western Europe, and not only because thousands of them were felled by the disease. The plague’s arrival breathed new life into allegations that the Jews were actively conspiring to harm Christians. Fear of the rapidly spreading and deadly disease fed rumours that foreigners, beggars and Jews, out of sheer malice, contaminated rivers and wells with poisonous powders. Local political conflicts, pre-existing economic tensions between Jews and Christians, as well as the willingness of some secular authorities to take advantage of the persecutions were also factors in the attacks. Anti-Jewish violence was especially virulent in the Holy Roman Empire, but it also affected communities in Catalonia, France and the Low Countries.52 The Black Death arrived in southern France at the start of 1348 and quickly spread south into the lands of the Crown of Aragon, reaching Barcelona in early May and killing about one-third of its population. While there is no contemporaneous evidence of poisoning accusations against the Jews of Catalonia, it is nonetheless clear that some Christians blamed them for causing the epidemic. The accounts of the massacres in Catalonia come from the royal chancery records and the Hebrew narrative penned by Rabbi Hayim Galipapa. Barcelona’s Jewish quarter was attacked on 17 May and about twenty Jews were killed. In the town of Cervera, eighteen Jews suffered the same fate. But the worst violence occurred in Tàrrega, where in 51 Ibid., pp. 51–68. 52 Samuel Cohn Jr, ‘The Black Death and the burning of Jews’, Past & Present 196:1 (2007), 8.

440

Genocidal Massacres of Jews in Western Europe, 1096–1392

early July a crowd armed with axes and other weapons, and shouting ‘death to the traitors!’, broke through the gates of the Jewish quarter, killing its inhabitants and ransacking their homes. At least 300 Jews lost their lives that day. The textual accounts of the massacre were corroborated in 2007 by archaeological excavations of Jewish individual and communal burial pits in Tàrrega. Dated to the mid-fourteenth century, the pits were found to contain the skeletal remains of men, women and children whose bones bore the signs of violent trauma.53 The persecutions of Jews in Germany during the Black Death precipitated a permanent shift of the centres of Jewish population towards eastern Europe. What began as spontaneous outbreaks of violence fuelled by rumours that Jews were spreading the plague, devolved into ‘organised murder’ of the Jews orchestrated by local rulers and municipal authorities.54 In August 1348, the municipal council of Strasbourg sent letters to the main cities in the Rhineland, Savoy and the Swiss cantons, asking for evidence of Jews poisoning wells, springs and food supplies. Many city councils and castellans responded by supplying lists of the accused individuals, and describing their confessions (under torture), trials and punishments.55 For example, in September and October 1348, a number of Jews in the county of Savoy were accused of concocting a plot to put poisonous powder into wells, springs and rivers used by Christians. Judicial inquiries were conducted, and some Jews as well as their alleged Christian ‘co-conspirators’ were tortured and ‘confessed’ to the crime. The accusation quickly spread into German lands. According to the chronicler Heinrich von Diessenhoven, the first outbreak of anti-Jewish violence occurred in November 1348 in Sölden, where all the Jews were burned. In the course of the following year (November 1348 to September 1349), ‘all the Jews between Cologne and Austria were burnt and killed for this crime, young men and maidens and the old along with the rest’: some in their own homes and synagogues, some in the houses constructed specifically for that purpose, some in the fields and pits, and some in the castles where they sought 53 Anna Colet, Josep Xavier Muntané i Santiveri et al., ‘The Black Death and its consequences for the Jewish community in Tàrrega: lessons from history and archeology’ in Pandemic Disease in the Medieval World: Rethinking the Black Death, ed. Monica H. Green (Kalamazoo and Bradford: Arc Medieval Press, 2015), pp. 63–96; Nirenberg, Communities of Violence, pp. 233–41; Einbinder, After the Black Death, pp. 117–25. 54 Jörg R. Müller, ‘Erez gezerah – “Land of persecution”: pogroms against the Jews in the regnum Teutonicum from c. 1280 to 1350’ in The Jews of Europe in the Middle Ages (Tenth to Fifteenth Centuries): Proceedings of the International Symposium held at Speyer, 20– 25 October 2002, ed. Christoph Cluse (Turnhout: Brepols, 2004), p. 257. 55 Cohn, ‘Black Death’, 19.

441

maya soifer irish

Figure 16.1 Representation of a massacre of the Jews in 1349. Antiquitates Flandriae, Royal Library of Belgium, MS 1376/7. (Courtesy Jewish Women’s Archive)

protection. Writing about the massacres of Jews in Germany, the French friar Jean de Venette claimed that ‘mothers would throw their own children into the flames rather than risk them being baptised, and would then hurl themselves into the fire after them, to burn with their husbands and children’ (Figure 16.1).56 The attempts by church and state authorities to halt the persecutions largely failed. In July 1348, Pope Clement VI reissued the Sicut Judeis, following it up with an edict urging the clergy to excommunicate anyone who attacked Jews without a due process. The pontiff was sceptical of the accusations, pointing out that Jews were dying ‘throughout many parts of the world by the same plague, by the hidden judgement of God’. Some secular authorities (the municipal council of Cologne, Duke Albrecht of Austria) also tried to protect Jews, but relented under popular pressure and had the Jews officially condemned and burned.57 Only the Jewish communities on the eastern and southeastern edges of the empire – Bohemia, Slovenia and Austria – mostly escaped persecutions, which continued until 1351.58

56 Rosemary Horrox, ed. and trans., The Black Death (Manchester University Press, 1994), pp. 56, 208–19. 57 Ibid., pp. 219–22. 58 Haverkamp, ‘Jews in Christian Europe’, 191.

442

Genocidal Massacres of Jews in Western Europe, 1096–1392

Anti-Jewish Urban Riots in Castile and Aragon, 1391–1392 Before the fourteenth century, attacks on Jews occurred only sporadically in the Iberian kingdom of Castile, but in 1391 it was Seville, the capital of its southern region of Andalusia, that became the epicentre of horrific violence that rapidly spread to Aragon and northern Castile. About thirty-five Jewish communities were affected by violence in Castile, and about fifty-five in the Crown of Aragon. Like other fourteenth-century anti-Jewish rebellions, the Iberian riots were a broad movement that included noblemen, townsmen, clerics, members of religious and military orders, peasants and sailors, who were driven by a mix of economic grievances, resentment of royal authority and religious hatred. In both kingdoms, the royal authorities’ inability to restrain the violence led to heavy casualties. While the exact number of victims is unknown, estimates suggest that thousands of Jews were killed and many more were forcibly converted to Christianity.59 The massacres of 1391 had a lasting demographic impact on the Jewish communities in Castile and Aragon. In some major urban centres, the Jewish population was substantially reduced or disappeared entirely because of conversions and expulsions. Many Jewish families migrated to smaller towns and villages in search of stability and security.60 The violence of 1391 had its roots in the events of Castile’s civil wars of the 1360s, when Enrique Trastámara used a popular dislike of Jews and Muslims to advance his political ambitions. In an attempt to shore up the support of local municipal councils, Enrique disseminated proclamations accusing his half-brother, King Pedro I, of enriching and empowering Jews and Muslims at the expense of Christians. Even though after killing Pedro in 1369 the now King Enrique II reverted to the centuries-long tradition of putting Jewish officials in charge of Castilian royal finances, his conduct during the civil wars had shown that accusations of favouritism towards non-Christians could be wielded as an effective political tool. As the local elites in Castilian towns moved to consolidate power after the anarchy of the civil wars, their commitment to protecting the royal Jews from violence weakened, just as their anti-Jewish rhetoric at the meetings of the cortes (representative assembly) grew increasingly 59 Benjamin Gampel, Anti-Jewish Riots in the Crown of Aragon and the Royal Response, 1391– 1392 (Cambridge University Press, 2016), pp. 185–9, 352–4. 60 Angus MacKay, ‘Popular movements and pogroms in fifteenth-century Castile’, Past & Present 55:1 (1972), 33–9.

443

maya soifer irish

strident.61 Ferrán Martínez, the archdeacon of Écija, who would later be called ‘the cause’ of the anti-Jewish riot in Seville, began his anti-Jewish preaching campaign shortly after the conclusion of the civil wars. A wealthy landowner and a secular canon at the Cathedral of Seville, Martínez had close ties to members of the municipal elite and enjoyed their support. The leaders of the Jewish aljama warned the king that Martínez’s preaching was going to incite an anti-Jewish riot, but royal letters ordering the municipal council to restrain him went unheeded.62 While the tacit support of Martínez’s activities by Seville’s urban elites paved the way for the anti-Jewish violence, the archdeacon’s preaching struck a powerful chord with the city’s less affluent citizens. Economic hardship, civil unrest, war, famine and recurrent epidemics of Black Death had disproportionately affected the city’s poor. It was easy for Martínez to channel their discontent at the Jewish community, which had a substantial lower class of its own but also included a privileged group of prosperous tax collectors, moneylenders and merchants.63 In October 1390, an eleven-yearold king, Enrique III, and a group of quarrelsome regents were left in charge of the kingdom. In February (or March) 1391, a group of Seville’s citizens enflamed by Martínez’s rhetoric attempted to attack the Jewish quarter, but failed. The royal chronicler Pero López de Ayala wrote that they took advantage of the monarchy’s glaring ineptitude: ‘the people were very aroused, and did not fear anybody, and their eagerness to rob the Jews grew with each passing day’.64 On 4 or 6 June, a large crowd stormed the Jewish quarter. Christian and Jewish chronicles described a scene of violence, death and desolation: the gates of the Jewish quarter set afire, hundreds of Jews killed, others forcibly converted to Christianity and sold into slavery, their houses despoiled by the raging mob. From Seville, copycat violence quickly spread to other cities in Andalusia, including Córdoba and Jerez de la Frontera, and north to Toledo, Cuenca, Madrid, Segovia, Soria and Burgos. Hasdai Crescas, an Aragonese Jewish writer,

61 Maya Soifer Irish, Jews and Christians in Medieval Castile: Tradition, Coexistence, and Change (Washington, DC: Catholic University of America Press, 2016), pp. 221, 227–9. 62 Maya Soifer Irish, ‘Toward 1391: the anti-Jewish preaching of Ferrán Martínez in Seville’ in The Medieval Roots of Antisemitism: Continuities and Discontinuities from the Middle Ages to the Present Day, ed. Jonathan Adams and Cordelia Heß (New York and London: Routledge, 2018), pp. 306–10. 63 Ibid., 311; Mark Meyerson, ‘Under Christian rule’ in The Cambridge History of Judaism, vol. V I : The Middle Ages: The Christian World, ed. Tony Michels and William David Davies et al. (Cambridge University Press, 2018), pp. 158–9. 64 Soifer Irish, ‘Toward 1391’, 307.

444

Genocidal Massacres of Jews in Western Europe, 1096–1392

reported that many Jews converted, but that some ‘died having sanctified His Name’.65 It only took a little over a month for the violence to break out in the neighbouring lands of the Crown of Aragon. Here, too, royal authority proved powerless in the face of the violence. On 9 July, a band of fifty young clerics carrying pennants and crosses made from reeds marched to the gate of the Jewish quarter in the city of Valencia, chanting that ‘the archdeacon of Castile’ was coming to baptise all the Jews. The walls of the Jewish quarter did not provide an adequate defence against large crowds of Christians armed with crossbows, swords, knives and sticks. The attackers sacked Jewish houses, destroyed loan contracts and killed about 230 Jews. Despite the municipal officials’ initial attempts to blame sailors and vagabonds for the violence, it soon came out that noblemen, artisans and master craftsmen had also participated in the massacre.66 Over the next several months, anti-Jewish violence broke out on the island of Mallorca, in Barcelona, Girona, Lleida, Tortosa, Tarragona and other towns in Catalonia and Aragon. In Barcelona, the attack on the Jewish quarter on 5 August evolved into a full-scale urban riot, as lower-class townsmen, sailors, fishermen and peasants from the surrounding countryside attacked municipal and royal authorities, demanding tax relief, abolition of serfdom and greater participation in the city government. The rioters stormed the royal castle where the Jewish survivors of the initial attack had been ensconced. Between 300 and 400 Jews were killed. According to Crescas, when given the choice between baptism and death, ‘many sanctified God in their midst . . . Of these, many slaughtered themselves, others threw themselves from the tower and, already halfway down to the ground, their bodies were broken apart limb from limb. A few left the castle and sanctified the Name in the street. The remainder converted.’67

Conclusion By the end of the Middle Ages a growing number of Christian writers began to opine that Jews represented an existential threat to Christendom and had to be eliminated through forced conversion or expulsion.68 While major theologians never repudiated the Augustinian doctrine or advocated 65 Gampel, Anti-Jewish Riots, pp. 13–22; Ram Ben-Shalom, ‘Kidush ha-Shem and Jewish martyrdom in Aragon and Castile in the year 1391: between Sefarad and Ashkenaz’ (Hebrew), Tarbiz 71 (2001), 229–30. 66 Gampel, Anti-Jewish Riots, pp. 36–9. 67 Philippe Wolff, ‘The 1391 pogrom in Spain: social crisis or not?’, Past & Present 50:1 (1971), 16; Gampel, Anti-Jewish Riots, pp. 102–6. 68 François Soyer, Medieval Antisemitism? (Leeds: Arc Humanities Press, 2019), pp. 39–40.

445

maya soifer irish

violence, many ordinary Christians had no such scruples. From the eleventh century on, their genocidal intent was evident in the indiscriminate attacks on Jewish men, women and children in communities throughout western Europe. Even if in each particular case the assailants had local objectives, and did not aim to eradicate Jews from the entire western Christendom, the cumulative effect of their attacks was profound. As the Middle Ages progressed, the massacres expanded in scope, affecting entire regions and even kingdoms. While medievalists generally avoid modern terms like ‘genocide’ for fear of sounding anachronistic, it seems appropriate to say that these killings bore all the hallmarks of genocidal massacres, described by Leo Kuper and Ben Kiernan as exterminations of specific communities because of their membership in a larger group.69 To distinguish between medieval and modern anti-Jewish hatred, an argument is often made that the former had religious underpinnings whereas the latter was explicitly racial in nature.70 The evidence reported in this chapter shows, however, that the supposed responsibility for the ancient deicide was only one factor, among many, in the attacks on Jewish communities during the Middle Ages. Especially towards the end of the period, Jews were accused of actively plotting to destroy Christianity, inflict physical harm on Christians and undermine their livelihood. In this sense, medieval and modern anti-Jewish violence have far more in common than many scholars are willing to admit.

Bibliographic Note The study of anti-Jewish violence in medieval Europe has been deeply affected by Salo Baron’s admonition to dispense with the ‘lachrymose theory’ of premodern Jewish history and present a more balanced picture of Jewish– Christian relations (Salo Baron, ‘Ghetto and emancipation’ (1929), republished in Leo W. Schwarz, ed., The Menorah Treasury: Harvest of Half a Century (Philadelphia: Jewish Publication Society of America, 1964), p. 61). Historians’ reluctance to narrate Jewish history as a succession of disasters pointing towards the Holocaust probably accounts for the fact that a new comprehensive study of medieval anti-Jewish violence is still waiting to be written. Scholars who examine anti-Jewish discourses and violence are 69 Leo Kuper, Genocide: Its Political Use in the Twentieth Century (New Haven: Yale University Press, 1981), p. 10; Ben Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007), pp. 13–15. 70 Soyer, Medieval Antisemitism?, p. 2.

446

Genocidal Massacres of Jews in Western Europe, 1096–1392

careful to ground their arguments in local contexts. Nirenberg’s Communities of Violence and Rubin’s Gentile Tales remain the gold standard for this type of research. In Living Together, Living Apart: Rethinking Jewish–Christian Relations in the Middle Ages (Princeton University Press, 2007), Jonathan Elukin has pushed the anti-lachrymose position to its logical extreme, arguing that medieval European Jews were not singled out as a target of persecution, and that the occasional episodes of violence against them did not undermine their sense of security (pp. 6–9). In Reassessing Jewish Life in Medieval Europe, Chazan agrees with Elukin’s central idea, pointing out that the demographic and economic growth of Jewish communities belie the popular perception of medieval Christendom as a place of incessant anti-Jewish persecutions (p. xiii). A very different set of assumptions about medieval European society underlies the classic study by R. I. Moore, The Formation of a Persecuting Society (2nd ed.; Malden, MA: Blackwell, 2007). Moore has argued that starting in the twelfth century, in order to gain political influence and consolidate power, church and state authorities developed new mechanisms of persecution targeting Jews, heretics and lepers. In his view, persecutory discourses were deliberately developed, justified and weaponised against minority groups by the Christian elites (pp. 116, 138). In contrast, Langmuir describes the accusations of ritual murder, host desecration and well poisoning as ‘irrational accusations’ that had nothing to do with the official Christian doctrine. In Toward a Definition of Antisemitism, he argues that in the thirteenth century Jews became targets of psychopathological violence perpetrated by paranoid people harbouring doubts about their Christian beliefs (pp. 11–12). Most recently, in Anti-Judaism: The Western Tradition, Nirenberg has put antiJewish discourses back at the centre of European thought and spirituality, showing how in the course of several millennia western peoples used negative perceptions of Jews and Judaism to criticise and transform their societies.

447

17

Crusaders and Mass Killing at Jerusalem in 1099 thomas a. fudge

Crusader violence in the medieval Levant is a commonplace in the literature. In 1954 perhaps the most widely read scholar of the crusades in the anglophone world, Stephen Runciman, famously concluded his three-volume study with the comment: ‘The Crusades were a tragic and destructive episode . . . and the Holy War itself was nothing more than a long act of intolerance in the name of God, which is the sin against the Holy Ghost’.1 Maher Abu-Munshar, a British-trained associate professor of Islamic Studies at Qatar University, summarised the fall of Jerusalem by noting: ‘A horrifying massacre followed during which the crusaders spent a week slaughtering Muslims, killing at least 70,000 people’.2 Based on an unidentified (and perhaps unidentifiable) source on the First Crusade, the German medievalist Martin Erbstösser tells us that a Christian patriarch passed down a Jerusalem street in 1099 murdering every non-Christian he encountered. Arriving at the Church of the Holy Sepulchre, bloodstained hands on sword, he washed and read from the Psalms: ‘The righteous rejoices in the Lord when he sees such vengeance which he takes. He shall bathe his hands in the blood of the godless’. He then celebrated Mass commenting that never had he made a more pleasing sacrifice to God.3 One of the medieval chroniclers, William of Tyre (c.1130–86), graphically described the conquest of Jerusalem: It was impossible to look upon the vast numbers of the slain without horror; everywhere lay fragments of human bodies, and the very ground was covered with the blood of the slain. It was not alone the spectacle of headless bodies and mutilated limbs strewn in all directions that roused horror in all who looked upon them. Still more dreadful was it to gaze upon the victors 1 S. Runciman, A History of the Crusades, 3 vols. (Cambridge University Press, 1954), I I I, p. 480. 2 M. Y. Abu-Munshar, ‘Fa¯timids, Crusaders and the fall of Islamic Jerusalem: foes or allies?’, Al-Masa¯q 22 (2010),˙ 47. 3 M. Erbstösser, The Crusades, trans. C. S. V. Salt (Leipzig: Edition Leipzig, 1978), p. 97.

448

Crusaders and Mass Killing at Jerusalem in 1099

themselves, dripping with blood from head to foot, an ominous sight which brought terror to all who met them. It is reported that within the Temple enclosure alone about ten thousand infidels perished, in addition to those who lay slain everywhere throughout the city in the streets and squares, the number of whom was estimated as no less.4

In consequence, some scholars have denounced Pope Urban II, who called the crusade, as one of history’s mass murderers.5 Such perspectives noted, astonishingly the fall of Jerusalem aroused no contemporary interest or excitement in the Muslim world. As horrific as it sounds, William of Tyre’s narrative must be interrogated keeping in mind that parts of medieval Christian theology included a symbiotic relationship between faith and force, piety and violence, and that crusades were waged for reasons relevant not to our times but to the Middle Ages. Crusades can only be properly understood within that context and William was deeply involved in the political crises occurring in Jerusalem that he recorded. He wrote his chronicle for King Baldwin IV and the royal court at Jerusalem and was thwarted in his ambition to become the patriarch of that city. Like those of practically all medieval chroniclers, the death toll figures are symbolic, signifying magnitude, but they cannot be treated mathematically.6 Runciman’s summation is his opinion. Abu-Munshar’s conclusion is exaggeration supported by no reliable source. Erbstösser’s prose is fiction. One must ponder the evidential bases for these assertions. At the centre of this enquiry is the question of whether genocide occurred at Jerusalem in 1099. Several considerations emerge.

Tumult in the Levant before the Crusades Often overlooked is that for half a century prior to the crusade advent, the Levant had been convulsed with internecine war between contending Sunni and Shi’a factions of Islam and was a cauldron of violent politics. Later Muslim chroniclers admitted as much by pointing out that while the ‘god-damned Franks’ were seizing land from Islam, Muslim rulers were too busy fighting with each other to do anything 4 William, Archbishop of Tyre, A History of Deeds Done beyond the Sea, trans. E. A. Babcock and A. C. Krey (New York: Columbia University Press, 1943), I, pp. 371–2. 5 Noted in B. Z. Kedar, ‘The Jerusalem massacre of July 1099 in the Western historiography of the Crusades’, Crusades 3 (2004), 59. 6 P. Courroux, ‘Remarks on the use of numbers by medieval chroniclers in battle narratives’ in The Medieval Chronicle 13, ed. E. S. Kooper and S. Levelt (Leiden: Brill, 2020), pp. 59–80.

449

thomas a. fudge

about it.7 Political squabbling aside, religious diversity deepened suspicion. The Shi’a divided into three main groups, each with subdivisions. The Sunni were also of three main groups, further belying simplistic notions of unity. In addition to Jews, multiple Christian populations of Copts and Armenians (both mainly in Egypt), Syriac Orthodox, Melkite (Arabic-speaking Greek Orthodox Christians) and Maronites, to say nothing of Byzantine and Latin Christians, abounded. These communities have been written out of many modern histories. Muslims and Christians often collaborated sometimes in fighting other Muslims. In some cases, Muslims were treated better by Christians than by other Muslims, a point noted by the traveller Ibn Jubayr from Spain in the 1180s.8 Latent hostility existed. Many Islamic writers corrupted the Arabic al Kayâmah for the Christian church of the Resurrection to al-Kumâmah, meaning ‘pile of shit’.9 Frequently overlooked or minimised are eleventh-century Turkish invasions that exceeded those of the crusaders in their destruction. Opportunists took advantage of the crusades to advance personal agendas.

Crusade Extermination Policies? The Roman legal maxim vim vi repellere justifies force for defensive purposes. In the 1140s, Bernard of Clairvaux did call for the annihilation of the Wends: ‘aut ritus ipse, aut natio deleatur’, ‘their religion or nation should be destroyed’.10 This was a message of conversion or death. Much later, in 1412, Dominican theologian Johannes Falkenberg deployed loaded terms like ‘pro exterminio Polonorum’, literally, ‘for the extermination of the Poles’. The phrase is followed by the caveat that whoever takes up the initiative must do so ex caritate (from love). It is possible that the ‘extermination’ could refer to the wording of the 1231 anti-heresy papal bull Excommunicamus that Falkenberg refers to. How did contemporaries understand it?11 These are rare 7 I. al-Athı¯r (1160–1223) cited in F. Gabrieli, Arab Historians of the Crusades, trans. E. J. Costello (Berkeley: University of California Press, 1969), p. 18. 8 A. Mallett, ‘The consequences of genocide: Muslim responses to the Crusades’ in A Cultural History of Genocide in the Middle Ages, ed. M. H. Eichbauer (London: Bloomsbury, 2021), p. 139. 9 R. C. Finucane, Soldiers of the Faith: Crusaders and Moslems at War (New York: St Martin’s Press, 1983), p. 156. 10 Letter 457. Latin text: H.-D. Kahl, ‘Crusade eschatology as seen by St. Bernard in the years 1146 to 1148’ in The Second Crusade and the Cistercians, ed. M. Gervers (New York: St Martin’s Press, 1992), pp. 42–3. 11 N. Housley, ‘A Crusade against the Poles? Johannes Falkenberg’s “Satira” (1412)’ in The Templars, the Hospitallers and the Crusades: Essays in Homage to Alan J. Forey, ed. H. J. Nicholson and J. Burgtof (London: Routledge, 2020), pp. 189–90.

450

Crusaders and Mass Killing at Jerusalem in 1099

examples. Neither approaches a policy. Neither is applicable to the Levant. On the other hand, it is certain that crusades were waged as acts of vengeance but this thesis must be understood in its medieval context within its particular religious paradigm.12 The medieval vocabulary of vengeance includes punishing the enemies of God or Allah and is a cornerstone of both crusader and jihad ideology. The Chanson d’Antioche, a song of heroic deeds written soon after the First Crusade, and adapted later in the twelfth century, contains a fully developed concept of the crusade as vengeance. The spiritual benefits of violence lay at the core of crusading. There was an incentive to perpetrate violence and exaggerate it.

Unreliability of Sources Scholars have identified three major twentieth-century developments that have implications for any analysis of the First Crusade: the discovery of the Cairo Geniza letters; the rehabilitation of Albert of Aachen as a valid source; and the connection of ‘blood flowing as high as a horse’s bridle’ for a distance of 1,600 stadia, a reference we find often in Latin accounts of the siege of Jerusalem, to the Apocalypse of John (14:20).13 A number of sources, noted below, make extraordinary claims for deep rivers of blood, borrowing from the biblical image. Taken literally, as stated in the New Testament, a stadion is a Greek measurement roughly equal to 192 metres. The apocalyptic reference to 1,600 stadia amounts to about 300 kilometres, or a distance of 186 modern miles. A horse’s bridle measures about one and a half metres or four and a half feet. These dimensions of blood flowing are obviously impossible, clearly symbolic only. Consequentially significant is the nagging suspicion that not only Latin but Arabic and Jewish sources are characterised by exaggeration, bloated figures, contradiction, bragging and lurid details. There is a tendency to inflate. One contemporary writer, Fulcher of Chartres, compared this proclivity to a disease suffered by many chroniclers. He suggested numbers of casualties were either exaggerated or outright fabrications.14 It is doubtful any medieval chronicler set out to transmit facts in the modern sense. There is, however, broad agreement among Western, Armenian, Arabic and Hebrew sources that after they fought their way into Jerusalem on 12 S. A. Throop, Crusading as an Act of Vengeance, 1095–1216 (Farnham: Ashgate, 2011), pp. 11–71. 13 Kedar, ‘Jerusalem massacre’, 64–5. 14 F. R. Ryan, trans., Fulcher of Chartres: A History of the Expedition to Jerusalem 1095–1127 (New York: W. W. Norton, 1969), p. 280.

451

thomas a. fudge

15 July 1099, the crusaders began a massacre of the city’s Muslim and Jewish inhabitants. Other sources tell us this activity resumed the next day. It is important to note that Christians had left Jerusalem. Letters found in the Cairo Geniza written about nine months after the events (i.e. April 1100) state that crusaders ‘killed everyone in the city . . . and the few who survived the slaughter were made prisoners’. This refers to both Jews and Muslims and the writer also mentions different groups of surviving captives. These letters note a synagogue was destroyed and suggest a three-day massacre, a point confirmed by Albert of Aachen.15 Christian sources report that those in Jerusalem ‘with amazing courage’ fought back with rocks, fire and swords on the walls and in the city. Defenders ‘screamed, blared out with horns, and performed all kinds of acts of mockery’ while desecrating a cross publicly. They fortified the city with ‘extraordinary measures’ and sent spies among the crusaders. These efforts terrified the crusaders, caused many casualties on both sides, and stymied the offensive.16 When the crusaders took Jerusalem, it is impossible to know to what extent Muslims and Jews were involved whether as combatants, rescuers or bystanders. The Muslim sources are too far removed to be of use on this point and the Latin sources have no way of knowing. Modern historians have been affected by the descriptions in the contemporary Christian sources that describe the killing in the most graphic detail. For example, rather than William of Tyre’s figure of approximately 20,000 dead, the claim that crusaders killed over 70,000 in the Al-Aqsa Mosque/Temple of Solomon alone has resonated and the Armenian historian Matthew of Edessa (d. 1144) suggested a figure nearly as high.17 Such massacres led on to the supposed burning of copies of the Quran and to inflated figures of 100,000 dead in the mosque and elsewhere. These figures are impossible. If sources are examined carefully, we find a progression of detail over a number of successive accounts. Jerusalem’s population was 20,000–30,000 and, according to the geographer AlMuqaddasi, a native of the city, Christians outnumbered Muslims.18 Moreover, 15 S. D. Goitein, ‘Contemporary letters on the capture of Jerusalem by the Crusaders’, Journal of Jewish Studies 3 (1952), 162–77. 16 R. Hill, ed., The Deeds of the Franks and the Other Pilgrims to Jerusalem (Oxford: Clarendon Press, 1962), p. 91, J. H. Hill and L. L. Hill, trans., Peter Tudebode: Historia de Hierosolyitano Itinere (Philadelphia: American Philosophical Society, 1974), pp. 115–19, and Ryan, Fulcher of Chartres, p. 120. 17 A. E. Dostourian, ‘The Chronicle of Matthew of Edessa: translated from the original Armenian, with a commentary and introduction’ (PhD thesis, Rutgers University, 1972), p. 311. 18 J. Prawer, The Latin Kingdom of Jerusalem: European Colonialism in the Middle Ages (London: Weidenfeld & Nicolson, 1972), p. 82, and A. V. Murray, ‘A race against time – a fight to the death: combatants and civilians in the siege and capture of

452

Crusaders and Mass Killing at Jerusalem in 1099

there are gross exaggerations of the numbers of Muslim prisoners taken, though the contemporaneous Geniza letters describe these as ‘the few who survived’. An examination of Muslim sources is revealing. These are limited because they are not contemporary with events, but the historiographical trend is instructive. Ibn al-‘Arabı¯ (1076–1148), a Muslim from Seville who studied in Jerusalem between 1092 and 1095, and passed through the city in December 1098, was in Egypt in 1099. He completed a treatise claiming 3,000 were killed in the mosque.19 Noticeably, this figure is unique, without parallel.20 It might be fairly accurate. However, the crusaders are referred to as Byzantines and he provides an incorrect date. The chronicler of Aleppo, Al‘Azimi (1090 – after 1161), simply says crusaders conquered Jerusalem and destroyed the synagogue.21 The rendition of Ibn al-Qala¯nisı¯ (c.1071–1160), a Damascus-based chronicler, is slightly more detailed. He records that crusaders stormed the city, and many tried to escape but were killed. Jews sheltering in the synagogue perished in the fire that destroyed the building. This suggests total extermination of the population. That conclusion is belied by other comments (to say nothing of Latin sources). Some, agreeing to surrender, were allowed to leave.22 As the decades passed, the story of what happened in July 1099 became embellished. The Baghdad Sunni scholar Ibn al-Jawzı¯ (c.1116–1201) advances claims of 70,000 slaughtered Muslims and of crusaders pillaging the Dome of the Rock.23 His account plays a key role in the formation of a tradition. Comparison with the silence of the contemporary chronicles, indicative of no particular political interest in the events at Jerusalem, is striking. The late Ayyu¯bid and early Mamlu¯k-era historian Ibn Muyassar (d. 1278) repeats much of this, also claiming crusaders burned copies of the Quran.24 The historian Ibn al-Athı¯r (1160–1223) omits the fact that few chroniclers at the time appeared concerned about the fall of Jerusalem. He excludes embarrassing details at the coalface of what was at issue in 1099, including

19 20 21 22 23 24

Jerusalem, 1099’ in Civilians under Siege from Sarajevo to Troy, ed. A. Dowdall and J. Horne (London: Palgrave Macmillan, 2018), p. 167. Details in Kedar, ‘Jerusalem massacre’, 73–4. Hirschler, ‘Jerusalem conquest of 492/1099’, 50. C. Cahen, ‘La chronique abrégée d’al-‘Azimi’, Journal Asiatique 230 (July–September 1938), 373. H. A. R. Gibb, ed. and trans., The Damascus Chronicle of the Crusades: Extracted and Translated from the Chronicle of Ibn al-Qala¯nisı¯ (London: Luzac, 1932), p. 48. I. al-Jawzı¯, Al-muntazam fı¯ ta¯rı¯kh al-mulu¯k wa’l-umam, 2nd ed. (Beirut: Dar al-Kutub al˙ I am grateful to Alice Awwad for help with the Arabic texts. Ilmiyah, 1995), p. 108. I. Muyassar, Akhbar Misr: Annales d’Egypte (les khalifes Fatimides), ed. H. Masse (Cairo: Institut Français d’archeologie, 1919), p. 39.

453

thomas a. fudge

the extensive and rampant disunity in the Levant, and details about infighting among Muslim factions. Instead, he exaggerates the sins of the crusaders by manufacturing details, apparently to inflame emotions and take control of an existing but inadequate narrative. In this manner, he constructs a gossipy and breezy piece of journalism that is not a reliable history. Ibn al-Athı¯r also popularised the notion of a week-long slaughter and the killing of more than 70,000 Muslims in the mosque, adding that among the casualties were many Islamic scholars, amid wholesale looting.25 Lastly, the Egyptian historian Ibn Taghribirdi (1411–70) retells events but expands the massacre from the mosque to other locations in Jerusalem and inflates the number of victims to 100,000, who now include the elderly and infirm.26 One wonders who commissioned and undertook the body counts. Ever greater detail and exaggeration characterise these narratives. These include gross inflation of the numbers of prisoners, the alleged severity of treatment of those held, and tales of the absurd. Matthew of Edessa records that crusaders massacred every infidel in Nicaea on 19 June 1097; in fact, the city surrendered and few crusaders actually entered Nicaea.27 Of the fantastic stories, one must add a tale told by the fourteenth-century Egyptian historian Ibn al-Furat, of how Tughtegin, the Turkish ruler of Damascus, captured the lord of Tiberius in June 1108, supposedly hollowed out his still living captive’s skull and drank wine from it while the unfortunate man looked on helplessly for an hour before expiring.28 Early Egyptian and Syrian sources do not assert a large-scale massacre narrative, and suggestive details in the Geniza letters and Albert of Aachen did not attract comment by leading Muslim contemporaries as a major event. Decades passed before it gained political significance. There is no sense in the Syrian or Egyptian sources that the crusades created shock and outrage in the wider Muslim world. Not before the thirteenth century do we find a uniform and hegemonic presentation of events, elements that would later become commonplace in public Islamic accounts. Some sources, especially the Iraqi ones, alter earlier references to Fatimids and Turks, now labelling the latter as ‘Muslims’, thereby creating a metanarrative wherein they become members of a specific group subjected to a large-scale massacre with 70,000 casualties. 25 D. S. Richards, ed. and trans., The Chronicle of Ibn al-Athir for the Crusading Period from alKamil fi’l-Ta’rikh, 3 vols. (Farnham: Ashgate, 2005–8), I, pp. 21–2. 26 I. Taghribirdi, Nujum al-zahira fi muluk Misr wa’l-Qahira (Cairo: Dar al-Kutub alMisriyya, 1939), p. 149. 27 Dostourian, ‘Chronicle of Matthew of Edessa’, 296. 28 Carole Hillenbrand, The Crusades: Islamic Perspectives (London: Routledge, 2000), pp. 549–52.

454

Crusaders and Mass Killing at Jerusalem in 1099

That figure has eschatological significance in Islamic thought.29 It is further included in other sources. The Syriac historian Abu’l-Faraj Ibn al-‘Ibrı¯ (1226– 86), also known as Gregory Bar Hebraeus, insisted the crusaders spent a week in Jerusalem killing Muslims and more than 70,000 perished in Al-Aqsa Mosque.30 Turning to Christian sources, we find even more extravagant narratives which in some respects go as far as lending credence to later Muslim exaggerations, though it must be observed that there are considerable discrepancies in the various accounts. As the siege neared an end, Jerusalem was attacked from two directions: Godfrey of Bouillon and Tancred of Hauteville advanced from the north while Raymond of Toulouse led forces from the south (Map 17.1). Written soon after the fall of Jerusalem, the Gesta Francorum, traditionally regarded as the most important and influential of the Latin accounts, records two massacres. On 15 July, crusaders killed Muslim defenders up to the mosque, where the slaughter caused crusaders to wade in blood up to their ankles. Other forces pursued the defenders to the mosque and the battle lasted all day until the mosque was soaked in blood. Once overcome, crusaders seized survivors, killing them or placing them in captivity. The next day, ‘Saracens’ on the roof of the mosque were attacked (men and women) and some leaped to their deaths to avoid a worse fate. The Gesta claims the entire city was filled with corpses, which were dragged out by survivors and placed in heaps as large as houses, judging the scale of the massacre to be unprecedented. Some non-Christians were allowed to leave.31 Peter Tudebode’s 1101 chronicle account claims Saracens mocked crusaders, desecrated a cross and, instead of surrendering, fortified the city. Once Jerusalem fell, crusaders pursued the defenders, killing them all on the way to the mosque. There was a day-long battle and blood flowed in the mosque. On gaining the upper hand, crusaders indiscriminately killed men and women. Those on the roof of the mosque were initially granted protection. A subsequent order rescinded that guarantee. Men and women were killed while others fell to their death. The city was clogged with corpses accompanied by a great stench. Cadavers were taken outside and piled as high as houses and burned. Despite significant differences, Tudebode concurs with the Gesta that there were many casualties but neither offer specifics.32

29 Hirschler, ‘Jerusalem conquest of 492/1099’, 60–3. 30 I. al-‘Ibri, Grigurius al-Malti, Tarikh mukhtasar al-duwal, 3rd ed. (Beirut: Dar al-Mashreq, 1992), p. 197. 31 Hill, Deeds of the Franks, pp. 90–2. 32 Hill and Hill, Tudebode, pp. 112–20.

455

thomas a. fudge X

Siege Crusade advance

Godfrey’s siege

X

X

X

Herod’s Gate

X

N

X

X X X

St. Stephen’s Gate

X X X

Josaphat Gate

X

X X

New Gate

X X

Quadrangular Tower

X X

Dome of the Rock

Holy Sepulchre

X X

Temple Mount

X X X

Jaffa Gate

X

Tower of David

Al-Aqsa Mosque Temple of Solomon

X X X X X X

Zion Gate

X

Raymond’s attack

X

X

X

Kidron Valley

X

Mount Zion

0 0

1/4 mi 1/4 km

1/2 km

Map 17.1 The siege and capture of Jerusalem, July 1099. (Cartography by Bill Nelson)

456

Crusaders and Mass Killing at Jerusalem in 1099

Raymond of Aguilers’ chronicle of 1105 provides more detail than the Gesta and Tudebode, but makes no mention of a second massacre. He confirms the Saracens greatly fortified the city and claims 60,000 fighting men were in Jerusalem while crusaders had barely 12,000 able to bear arms along with 1,300 knights. The defenders had nine or ten siege engines for every one the crusaders had. The city eventually fell. The ‘amount of blood shed was incredible’. Enemies of the faith were decapitated, shot with arrows or burned. There were ‘piles of heads, hands and feet’ in the streets. In the mosque, men rode in blood up to their knees and bridle reins. The city was filled with corpses and blood.33 Fulcher of Chartres, who arrived at Jerusalem five months after these events, wrote a lurid chronicle account in 1107 that many on both sides were killed. There was a fierce struggle in the mosque. Those fleeing to the roof were assaulted by archers. Others in the Tower of David were allowed to leave unharmed. Nearly 10,000 were beheaded in the mosque alone and blood flowed to the ankles. No one was spared, not even women and children, suggesting total massacre. Evidently, bodies were cut open to check for gold that may have been hidden therein. Great heaps of corpses were burned for the same reason. It appears that Fulcher was the first to record these salacious tales in a narrative he revised over two decades.34 Robert the Monk’s chronicle of c.1110 expands the graphic imagery of the violence by describing waves of blood causing dismembered body parts to float and drift about in the mosque precincts.35 Meanwhile, Guibert of Nogent, writing a chronicle in 1107–8, claimed Saracens prepared booby traps but these failed and crusaders killed everyone they encountered, ‘more in slaughter than in battle’, through the streets up to the mosque. Blood flowed high enough to almost cover ankles. Greek fire was used against Raymond of Toulouse and his men elsewhere in the city. Fighting occurred in the mosque, where ‘a general slaughter of the pagans took place’. No one was spared. The battle in the mosque reduced the Saracens and blood ‘nearly submerged their boots’. Muslims were conscripted to remove the numerous cadavers and thereafter these recruits were summarily killed. In the melee, some defenders escaped to the roof of the mosque. However, the next day those sheltering on the roof were killed, both men and women. The stench was revolting. There were so many bodies one could not easily move 33 J. H. Hill and L. L. Hill, trans., Raymond d’Aguilers: Historia Francorum Qui Ceperunt Iherusalem (Philadelphia: American Philosophical Society, 1974), pp. 116–28. 34 Ryan, Fulcher of Chartres, pp. 119–23. 35 C. Sweetenham, trans., Robert the Monk’s History of the First Crusade: Historia Hierosolimitana (Farnham: Ashgate, 2005), p. 200.

457

thomas a. fudge

within the city without stepping on a cadaver, and so bodies were carried outside the city and burned.36 Baldric of Dol’s 1105 chronicle adds another unique detail. The Muslim conscripts were apparently insufficient to convey the slaughtered outside the city walls and therefore some of the impoverished Christians were paid to provide assistance. This detail is not replicated elsewhere. The Gesta agrees that surviving Saracens were forced to do this work but is silent on the inadequacy of the conscripted removalists or that Christians were employed to help out. William of Tyre includes the detail in his history but relies on Baldric.37 Neither was in the Levant in 1099. Unlike the Gesta, neither Raymond nor Fulcher note that survivors removed bodies. Minor accounts, such as a September 1099 letter from Daimbert, archbishop of Pisa, Godfrey of Bouillon and Raymond of Toulouse to the pope claimed crusaders rode in the blood of Saracens up to the knees of their horses.38 Baldric of Dol and Ekkehard of Aura claim blood approached human knee depth.39 Ralph of Caen, who came to Jerusalem in 1108, relying totally on oral testimony concluded the carnage had no parallel and declared the city awash with blood, insisting one could scarcely see anything but blood.40 The language is exorbitant and cannot be accurate or even possible.41 Writing c.1102, Albert of Aachen is the sole Latin source to extend the massacre to a third day. The discovery of the Geniza letters corroborates him. Tellingly, near contemporary Arabic Egyptian and Syrian sources do not mention the massacre on the roof of the mosque that evidently occurred on the second day. The mosque plays a role in Arabic sources only from Ibn alAthı¯r onwards (written 139 years after the fact).42 None of the Syrian, Egyptian or Iraqi sources refer to corpse desecration, coin searches, a corpsefilled city or the killing of Muslims who dragged dead bodies out of the walled precincts. The Latin accounts are equally untrustworthy on these same details and are embellished apparently to further the vengeance cause of the crusade, while later revisions tend to preserve the more gruesome tales. 36 R. Levine, trans., The Deeds of God through the Franks: A Translation of Guibert of Nogent’s Gesta Dei per Francos (Woodbridge: Boydell & Brewer, 1997), pp. 126–33. 37 Noted in C. Kostick, The Social Structure of the First Crusade (Leiden: Brill, 2008), p. 155. 38 E. Peters, ed., The First Crusade (Philadelphia: University of Pennsylvania Press, 1971), pp. 234–7. 39 Ekkehard, Hierosolymita in Recueil des Historiens des Croisades, Historiens occidentaux, ed. Académie des Inscriptions et Belles-Lettres, 5 vols. (Paris: Imprimerie royale [puis] Imprimerie impériale [puis] Imprimerie nationale, 1844–95), V, p. 24. 40 Gesta Tancredi in expeditione Hierosolymitana in Recueil des Historiens des Croisades, vol. I I I, V I I I, p. 699. 41 T. F. Madden, ‘Rivers of blood: an analysis of one aspect of the Crusader conquest of Jerusalem in 1099’, Revista Chilena de Estudios Medievales 1 (January–June 2012), 25–37. 42 Hirschler, ‘Jerusalem conquest of 492/1099’, 71.

458

Crusaders and Mass Killing at Jerusalem in 1099

Streams of blood increase in depth from ankles to knees to horses’ bridles. Pools of blood in the mosque only seep into the streets of Jerusalem in the thirteenth century whereas by the fourteenth century the rivers of blood seem to encompass the entire city. Given the capacity of the mosque, ankledeep blood therein would require about 93,000 bodies emptied of blood.43 Such horrific images are augmented by Albert of Aachen, who grimly records crusaders smashing the heads of infants against walls and door frames. That, by contrast, is possible. And Albert has recently been demonstrated to be a serious source. He was never in the Levant, but testified: ‘[I] decided to commend to posterity at least some of the things which were made known to me by listening to those who had been there and from their reports’. Writing three years after the events, Albert deliberately strove to create a new literary genre to match a new historical epoch, as he articulated in his own preface.44 Some historians believe Albert’s account is a thoroughly researched and reliable narrative of the massacre.45 The authorship of the numerous Latin sources is overwhelmingly clerical, which is not unusual. Three of the narratives were composed by Benedictine monks, each from northern France, and all three used the Gesta Francorum as a source. It seems obvious that all three sought to improve that urtext by expanding, developing and adding a more fulsome and sophisticated rendition. Four of the other extant accounts were written by participants or eyewitnesses. These include the anonymous Gesta Francorum and narratives written by Peter Tudebode, a priest from Civray, in 1101; Raymond of Aguilers (1105); and Fulcher of Chartres (1107) (only the latter was not in Jerusalem during the siege). The Gesta pretends to be omniscient but incorporates fiction and engages in wild speculation.46 The two accounts of the fall of Jerusalem considered most important are also the shortest (the Gesta and Fulcher) while the two others considered most dependent on the Gesta (Tudebode and Raymond) are twice as long, while Albert of Aachen’s work (1102) is ten times the length. Interpretations of the crusades by modern scholars have often been hyperbolic. Among the more spectacular relate to the Fourth Crusade (1202–4), 43 Madden, ‘Rivers of blood’, 35. 44 Albert of Aachen, Historia Ierosolimitana, ed. and trans. S. B. Edington (Oxford University Press, 2007), p. 3 and pp. xxx–xxxvii. 45 A. V. Murray, ‘The siege and capture of Jerusalem in western narrative sources of the First Crusade’ in Jerusalem the Golden: The Origins and Impact of the First Crusade, ed. S. B. Edgington and L. García-Guijarro (Turnhout: Brepols, 2014), p. 215. 46 Ibid., 200–1. Y. N. Harari, ‘Eyewitnessing in accounts of the First Crusade: the Gesta Francorum and other contemporary narratives’, Crusades 3 (2004), 77, 86 and 91 excludes the Gesta author as an eyewitness.

459

thomas a. fudge

with sweeping conclusions that ‘the sack of Constantinople is unparalleled in history . . . and there was never a greater crime against humanity than the Fourth Crusade’.47 Others claim casualties of up to 6 million.48 Elsewhere, in 1135 Orderic Vitalis claimed on the basis of eyewitness and trustworthy testimony that in one battle 500,000 Christians were ‘violently attacked’ by 1 million pagans.49 Truth or fiction? Some scholars suggest ‘true witnesses were not those who saw but those who understood’.50 Source reliability is crucial and many assertions about crusading fail the test of common sense. Numerous sources report on the Jerusalem massacre. These have been subjected to a brilliant analysis that is a model of forensic source examination.51 What seems irrefutable is that these accounts and their contradictions can be explained principally by a willingness to serve particular convictions. Both sides engaged in ideological power-play tactics endeavouring to subject the July 1099 events to the control of historical narrative in order to further special agendas. Muslim perspectives are not a monolithic unanimity, but seem quite clear when they eventually take up the issue: political capital. The Latin point of view is puzzling at first blush. There is no effort to conceal the cruelty of what happened. Instead, the narratives convey a sense of pride and fulfilment of a larger duty, along with ideological muscle-flexing and good old-fashioned boasting. Is it significant that no medieval Muslim or Arab source mentions crusader cannibalism? The matter is noted by almost all of the dozen or so Latin chroniclers dealing with the crusades up to 1120.52 It is not sufficient to attribute these gruesome Latin narratives to the reputation of the medieval world as an inhumane period marked by atrocitas (brutality), crudelitas (cruelty) and saevitia (savagery) that somehow stood in contrast to a chivalrous Christian view of the Middle Ages. Whereas many medieval texts represent but do not discuss these concepts, some of our First Crusade narratives engage in both. Latin reports of violence in Jerusalem in July 1099 display theological reflection. For example, Baldric of Dol includes 47 Runciman, History of the Crusades, I I I, pp. 123, 130. 48 M. Banitalebi, K. Yusoff and M. Rosalan Mohd Nor, ‘The impact of Islamic civilization and culture in Europe during the Crusades’, World Journal of Islamic History and Civilization 2:3 (2012), 184. 49 M. Chibnall, ed. and trans., The Ecclesiastical History of Orderic Vitalis, 6 vols. (Oxford: Clarendon Press, 1969–80), V, pp. 336–7. On exaggerated army figures, see J. France, Victory in the East: A Military History of the First Crusade (Cambridge University Press, 1994), pp. 122–42. 50 E. Lapina, Warfare and the Miraculous in the Chronicles of the First Crusade (University Park: Pennsylvania State University Press, 2015), p. 36. 51 Kedar, ‘Jerusalem massacre’, 15–75. 52 J. Rubenstein, ‘Cannibals and crusaders’, French Historical Studies 31 (2008), 525–52.

460

Crusaders and Mass Killing at Jerusalem in 1099

a speech in his Historia Hierosolymitana wherein priests address crusaders preparing to assault the city, characterising their armies as the familia Christi. Baldric goes on to draw parallels between the Hebrew Bible children of Israel and the crusader forces, suggesting the ‘household of faith’ ethos of the firstcentury Church. Guibert of Nogent points out that priests and bishops embedded in the crusade armies enjoined Christians to sing liturgical songs, fast, pray and perform other acts of charity and devotion. Robert the Monk’s Historia Hierosolimitana invokes a similar theological propensity. He compares his role in writing a history of crusading to that of Moses, who recorded the creation of the world. The crusade historian is tasked with recording details of the pilgrimage to Jerusalem characterised as the most miraculous undertaking since the events Moses narrates, save for the story of Christ. This is an extravagant claim and, while many more examples might be culled from the pages of the Latin histories, it is compelling to consider that such exaggerations suit perfectly theological purposes.53 One can appreciate the theological euphoria: the crusaders were outnumbered, hemmed in by their enemies, struggling in a foreign environment, short on supplies, forsaken by their eastern counterparts who had asked them to come in the first place, and all of this after undertaking the long and arduous journey from Europe over a period of three years. Now they had succeeded in taking the city of Jerusalem. It is not hard to see why Robert the Monk considered this achievement little short of a miracle. Guibert titled his account The Deeds of God through the Franks (Dei Gesta per Francos), and he spent a decade editing it, arguing that God was the crusading leader of soteriological expeditions.54 It can be argued that many early writers, including Robert the Monk, Guibert of Nogent and Baldric of Dol, believed the Gesta Francorum had been dilatory in explicating the divine aspects in the First Crusade. They undertook to remedy this deficiency by placing the momentous events of 1099 on to a properly considered theological scaffolding wherein readers might easily detect the hand of God. These writers were convinced they were engaging with sacred history, and so when they committed their histories to print they operated from the intellectual platform that the crusade was a divinely inspired miracle. All evidence and testimony was received, interpreted and applied within that hermeneutic.55 The First Crusade was the work of God. 53 Argued persuasively in Kostick, Social Structure of the First Crusade, pp. 51–93. 54 Levine, Deeds of God, p. 28. 55 S. John, ‘Historical truth and the miraculous past: the use of oral evidence in twelfth-century Latin historical writing on the First Crusade’, English Historical Review 130 (2015), 263–301.

461

thomas a. fudge

Are the Latin histories reliable? That depends on for what. As glimpses into the conceptual world of early twelfth-century contemporaries on how the crusades were perceived and understood, they are very reliable. Letters and chronicles of the First Crusade, traditionally accepted as reliable eyewitness accounts, are more the product of twelfth-century imaginations of a miraculous event.56 As sources for a history of these momentous events they are not dependable. These histories are marred by apocryphal distortions caused by theological commitments. The chroniclers were convinced that the meaning of what occurred was more salient than what had been seen or what may actually have happened. Their narratives bear this out. Calculating death tolls is a difficult assignment. In terms of Jews and Muslims killed in Jerusalem, it seems prudent to consider the possibilities. Benjamin Kedar and Joshua Prawer suggest the population could not have exceeded 30,000 and may have been closer to 20,000. The old town of Jerusalem is quite small and even with normal medieval urban overcrowding a higher population number is improbable. In the late eleventh century, Jerusalem was still a largely Christian community. We know the majority Christian population evacuated when the crusaders approached. If we assume a population of 30,000 with half being Christian, and we know that some ‘infidels’ were spared, this suggests a death toll of 10,000–15,000 on the higher end. However, we have no reliable source to inform us of what the population was on 15 July 1099. William of Tyre’s figure of 20,000 killed is possible but improbable. Albert of Aachen suggests 10,000 slain on Temple Mount. This is plausible but unverifiable. Tudebode claims the crusaders were ‘killing some, and sparing others as the notion struck them’ and some of the leaders protected a ‘great number of the infidels’ while Guibert refers to a ‘large crowd’ being sheltered. The Gesta says many prisoners were taken. On the other hand, Fulcher asserts there was a ‘great slaughter’. Raymond claims few survivors found sanctuary in the tower and the crusaders ‘spilled an incredible amount of blood’, with Tudebode concluding: ‘God alone knows the number’. Ibn al-‘Arabı¯ claims 3,000 perished in the mosque. This is probable. There was a massacre in Jerusalem in July 1099, which may have gone on for three days, but it was not on the scale often elaborated. While scholars worry about playing down the scope of the mass killing, they fail to sufficiently observe the role of bias in the narratives, the existing values held by the various authors, their prevailing attitudes and prejudices.57 56 T. W. Smith, ‘First Crusade letters and medieval monastic scribal cultures’, Journal of Ecclesiastical History 71 (2020), 484–501. 57 Kedar, ‘Jerusalem massacre’, 69, 75.

462

Crusaders and Mass Killing at Jerusalem in 1099

Was Crusade Violence Exceptional? Following Augustine, violence was considered ethically neutral though intention remained crucial. The distinction is critical.58 Neither crusading nor its associated violence betrayed central medieval Christian principles, hence crusades were not aberrant. They existed within a culture of medieval communities of violence (Gewaltgemeinschaften). This included displaying prisoners in postures of humiliation, public display of corpses indicating winners and losers, and symbolic mutilation. In 763, Byzantine emperor Constantine V ordered a captive’s arms and legs amputated and a premortem torso dissection. Ritualised violence emerged. Vanquished foes were decollated and heads used as footballs.59 It was customary in war that defeated cities be sacked. This was a principle with near legal status throughout the Middle Ages. The Greek military leader and historian Xenophon (c. 430–354 B C E ) asserted it was a law established universally in perpetuity that when a city was captured in war all persons and property belonged to the victors. In practice, building on the Hebrew Bible, no one should be spared.60 Medieval military codes of conduct following this principle were not condemned. Butchering civilians and raping women was generally condoned.61 Medieval Christian soldiers were entitled to kill and plunder. The early thirteenth-century canon law glossator Joannes Teutonicus affirmed a parallel policy that if heretics remained inside the walls then the entire city could be burned down.62 When defenders of religious faith, whether Muslim or Christian, took up the sword, violence and mass killing ensued. The canonist Hostiensis delineated holy war as conflict between the faithful and the infidel. He understood the infidel as hostes (in the full legal sense, meaning those defeating public enemies assumed absolute power).63 Crusade historians have not found significant evidence for violence beyond normal medieval realities.64 Some disagree. That view has not prevailed.65 Military conflicts in the crusade period generally adhered to 58 N. Housley, The Crusaders (Charleston, SC: Tempus, 2002), p. 178. 59 J. Rogge, ed., Killing and Being Killed: Bodies in Battle – Perspectives on Fighters in the Middle Ages (Bielefeld: Transcript Verlag, 2017), pp. 18, 47, 51–5, 148, 205. 60 W. Miller, trans., Xenophon Cyropaedia Books 5–8 (Cambridge, MA: Harvard University Press, 1914), 7.5.73; Iliad 6.57–62, Herod. 4.202, Diod. Sic. 14.53.1–3, Deuteronomy 20:10–20. 61 M. Keen, The Laws of War in the Middle Ages (London: Routledge, 1965), pp. 119–33. 62 Glos. ord. to C.23 q.5 c.32. Decretum divi Gratiani (Lyon: Ioannes Pitseius, 1554), p. 882, col. A, where the relevant canon, Si audieris, is numbered XXXI. 63 J. A. Brundage, ‘Holy war and the medieval lawyers’ in The Holy War, ed. T. P. Murphy (Columbus: Ohio State University Press, 1976), p. 118. 64 France, Victory in the East, p. 355. 65 Kedar, ‘Jerusalem massacre’, 74.

463

thomas a. fudge

prevailing laws of war. The crusader conquest of Jerusalem was not unparalleled. The Christian garrison at Marash in 1150 was promised safe passage upon surrender. Saladin killed them all. A similar fate befell Templars in 1266.66 Exaggerated massacres occurred on both sides, but these events fail to reflect extermination policies. Both sides engaged in, and marshalled theological support for, holy cruelty (sancta crudelitas). It is impossible to deplore crusading without ‘damning a good deal of the spiritual fabric’ of medieval Christendom.67

Politicising the Fall of Jerusalem It is important to note that the crusades and their historical interpretation has been the subject of significant politicisation, which has shaped their interpretation and problematises the analysis of violence. Modern rhetoric about the crusades is largely useless, anachronistic, fabricated historical conjecture, political propaganda, reflecting false world views imposed on a period for which they do not fit. Contradictions in both Christian and Muslim accounts appear to be the result of exploiting events to serve particular convictions. The gendered nature of language wherein only adult males are taken into account (women and children excluded) also contributes to apparent discrepancies.68 The Latin sources are untrustworthily embellished. Later revisions of the Gesta and other narratives tend to preserve the more gruesome details designed to glorify the necessary holy cruelty of the divinely sanctioned peregrinatio (pilgrimage), expeditio crucis (expedition of the Cross) and negotium Jhesu Christi (business of Jesus Christ). This is unexceptional. In reporting the massacre of Christian prisoners following Hattin in 1187, Ima¯d al-Dı¯n called it ‘pious work’, expressing delight in the slaughter of crusaders. Crusade conquest traditions emerge in Islamic historiography in the twelfth century but not before the thirteenth century, and the chronicle of Ibn al-Athı¯r, is a uniform and hegemonic presentation of events established. This includes a full-scale massacre with assertions that 70,000 innocents were killed in the Jerusalem mosque and crusaders butchered their victims for a full week. Villains can be manufactured on flimsy evidence and massacre details can be advanced to create shock in a manufactured Muslim world. Dichotomous myths, such as Muslims being innocent and peaceful and crusaders guilty and violent, become matters of belief, not evidence. It is simpler

66 Noted in Finucane, Soldiers of the Faith, p. 98. 67 Housley, The Crusaders, p. 181. 68 D. Hay, ‘Gender bias and religious intolerance in accounts of the “massacres” of the First Crusade’ in Tolerance and Intolerance: Social Conflict in the Age of the Crusades, ed. M. Gervers and J. M. Powell (Syracuse University Press, 2001), pp. 3–10.

464

Crusaders and Mass Killing at Jerusalem in 1099

and logical to represent the mass killing as pragmatic military action undertaken to neutralise an existing internal threat within Jerusalem; a military decision aimed at preserving crusader safety when the Fatimid garrison posed a threat from the Tower of David, large numbers of captives were liabilities, Fatimid armies were approaching, and crusaders feared danger on both sides of the city walls. It is tempting to compare the Muslim conquest of Jerusalem in 1187 with Christian calculation in 1099. The circumstances were disparate. In 1099, crusaders eliminated a clear and present internal danger. Albert of Aachen includes a speech from crusade leaders explaining their thinking and provides the only source offering a rational explanation rather than theologically driven justifications.69 Alan Murray has elaborated a convincing argument based on primary sources explaining events at Jerusalem as pragmatic military action in line with medieval practice. In 1187, Saladin deported the entire Frankish population of the city. Both were pragmatically dictated by immediate circumstances achieving the same end: ‘the emptying of the Holy City of enemy inhabitants as defined by their religious affiliation’.70

Considerations of Genocide The term ‘genocide’ originated in 1943 and the United Nations Genocide Convention became international law in 1951. The 1948 United Nations Convention on the Prevention and Punishment of the Crime of Genocide defines the crime as an act intended to destroy, either in whole or in part, a national, ethnic, racial or religious group as such, by killing its members; causing serious physical or psychological harm; intentionally imposing conditions causing total or physical destruction; creating measures intended to prevent births; or forcibly transferring children of the group in question to another group.71 There is no question that concepts of genocide predate twentieth-century formulations. But under some usages today almost any killing may qualify. Notions of extermination go back to antiquity and may equally imply expulsion or destruction. These concepts clearly existed prior to and throughout the Middle Ages. This is not controversial. 69 Albert of Aachen, Historia Ierosolimitana, 440–1. 70 A. V. Murray, ‘The demographics of urban space in Crusade-period Jerusalem (1099– 1187)’ in Urban Space in the Middle Ages and the Early Modern Age, ed. A. Classen (Berlin: De Gruyter, 2009), p. 223; Murray, ‘Race against time’, 176–7. 71 Convention on the Prevention and Punishment of the Crime of Genocide, Paris, 9 December 1948, United Nations, Treaty Series, vol. 78, no. 1021, www.hrweb.org/le gal/genocide.html.

465

thomas a. fudge

There are three options in evaluating the Jerusalem massacres of 1099: the massacres were not on the scale generally assumed and suggested in various histories, or mass killings occurred as strategic military necessity (as per Albert of Aachen), or the violence was a hate-generated extermination event. The language of widespread violence may include terms such as ‘genocide’, ‘genocidal massacre’, ‘genocidal moment’, ‘crime against humanity’, ‘extermination’, ‘mass killing’, ‘collective murder’, ‘hate crime’ and some form of terrorism. Only ‘mass killing’ applies unambiguously to the crusades. In the matter of genocide, we lack what any balanced level of analysis demands, which is ‘not just a reasonable volume but a satisfying variety of surviving evidence’ to make the case.72 ‘Genocide’ has become a politicised and increasingly popularised term, now becoming de rigueur in emerging historiographical discourse, designed to arouse emotions and provoke indignation, and it amounts to the political uses of violence. Christendom and Islam fostered hatred and violence. Saracens rejoiced at the slaughter of crusaders while Raymond of Aguilers declared his delight in seeing an aqueduct in Tripoli in May 1099 choked with corpses of infidels and was happy that the land stank with Islamic blood after the ‘hellish din of battle broke loose’.73 Muslims converted churches to mosques, forced Christians to wear distinctive clothing, encouraged mob attacks, and in 1250s Syria the Christian population declined from 40 per cent to 10 per cent.74 That said, while racism and its cognates had no special place in the crusades, religious prejudice certainly did. Raising the spectre of racism in crusade history diminishes the medieval rationale undergirding these events and in consequence reduces these events to an ahistorical phenomenon. The case for a racial genocide exists on unstable and questionable sources that have been uncritically interrogated. To interpret the fall of Jerusalem as unprecedented mass murder strains the evidence. As noted above, it was a massacre, a mass killing, and it was fuelled by religious prejudice, but it was not unprecedented. Importantly, the demarcation lines between combatants and civilians became blurred on both sides in the course of the Jerusalem siege.75 Civilians could become potential combatants. One must avoid imposing a rigid template on the vast history of crusading because it is possible to regard each crusade as a microcosm of the changing culture of medieval Christianity. 72 N. Housley, Contesting the Crusades (Oxford: Blackwell, 2006), p. 98. 73 Raymond d’Aguilers: Historia Francorum, pp. 105 and 125. 74 Mallett, ‘Consequences of genocide’, 137. 75 Murray, ‘Race against time’, 164.

466

Crusaders and Mass Killing at Jerusalem in 1099

Crusade Mythistory The narratives of the Jerusalem massacre from the Muslim side along with those composed by Latin Christians serve propagandist purposes. Naturally, the crusaders wished to adulate themselves and thought they could do so by exaggerating their bloodthirstiness. Later Muslim narratives appear driven by political agendas and fuelled by revitalised jihadist convictions that emerged as game-changers. These revised narratives actively altered and eventually replaced earlier perspectives. In 1105, the Damascus scholar ‘Ali ibn Tahir alSulami (c.1039–1106) suggested crusades had been divinely ordained of God to allow united Islam to reconquer Jerusalem, Constantinople and Rome.76 Jihad was often a political tool, but notions of unified holy war against crusaders cannot be identified before Nur al-Din and Saladin (1146–93). The former regarded jihad as an instrument to combat crusaders and fortify Sunni Islam. In consequence, fact merged with fiction and history assumed a politically expedient hagiographical dimension. The symbolic significance of numbers – 3,000 in Arabic conquest narratives, 70,000 for Muslims – should not be downplayed. But should they be taken literally? Figures aside, the long-term demographic impact on the minority residents of Jerusalem was significant. Eighty-eight years later the Muslims regained the city. Crusades and counter-crusades went on for centuries. Up to a third of the Jerusalem population perished in 1099 due to military struggle and slaughter. It is worth pondering that while historians ‘must analyse actions and events of the premodern era in the context of intellectual and cultural stipulations discussed at the time’, it is questionable whether to ‘dismiss past episodes of mass killing by explaining them as merely part of a very different society’. This potentially grants ‘historiographical immunity’ which is ‘ahistorical’, and serves to dehumanise ‘individual actors, ignores their decisions, obfuscates the workings of power, distances us from the past, and comes dangerously close to reifying a natural order of things that denies human agency over time’.77 These are cautionary notes for careful consideration. At the same time, endeavouring to modernise the medieval world invites equally serious risks. Some of the most damaging interpretations of the crusades are those ignoring their religious motivation. The result has

76 N. Christie, ed. and trans., The Book of the Jihad of ‘Al ibn Tahir al-Sulami (d. 1106): Text, Translation and Commentary (Farnham: Ashgate, 2015), p. 18. 77 B. Kiernan, ‘Is “genocide” an anachronistic concept for the study of early modern mass killing?’, History 99:336 (2014),442–60, at 460.

467

thomas a. fudge

been the creation of an intellectually vacuous chapter of medieval history no longer historically useful or meaningful. While the crusade source tradition is problematic, and has been subject to exaggeration and elaboration over time, the massacre following the taking of Jerusalem could be placed within modern definitions of genocidal massacre based on religion – though not race – but the argument requires appropriate contextualisation. In particular, the moral connotations associated with the term in the modern period are not appropriate for the medieval world, and the motivational complex could be construed as more pragmatic or strategic rather than ideological. Categorising crusades as genocide is problematic for it is evident that evocative narratives of crusading massacres arise from unreliable historiographical foundations. The aims of the Gesta Francorum and Ibn al-Athı¯r, the arguments of Ibn al-Jawzı¯ and Robert the Monk, or the conclusions of Baldric of Dol and ‘Ali ibn Tahir al-Sulami, need not be ours. These texts might read as genocidal, but their sources lay not in historical fact but in religious imagination and political ambition.

Bibliographic Note The Routledge series Crusade Texts in Translation (35 volumes) is essential. Translated sources include August C. Krey, ed., The First Crusade: The Accounts of Eye-Witnesses and Participants (Merchantville: Martino Publishing, 2006), which lacks explanatory notes, while Peters, ed., First Crusade is a useful anthology. Translations of chronicles include Edgington’s Albert of Aachen, Historia Ierosolimitana, rehabilitating Albert as a serious source; Hill, Deeds of the Franks, which provides Latin text and translation of the Gesta Francorum now challenged as the most important source; Hill and Hill’s Peter Tudebode: Historia de Hierosolyitano Itinere and Raymond d’Aguilers: Historia Francorum Qui Ceperunt Iherusalem, which feature helpful introductions and extensive explanatory notes; Ryan, Fulcher of Chartres, containing a lengthy introduction and extensive notes; Sweetenham, Robert the Monk’s History, which has valuable historiographical chapters preceding the text; while Levine, Deeds of God through the Franks includes only scattered notes, and Steven Biddlecombe, The Historia Ierosolimitana of Baldric of Bourgueil (Rochester, NY: Boydell & Brewer, 2020) has numerous helpful footnotes. All of these translations are of good quality. Essential secondary studies include the journal Crusades, featuring state-ofthe-art scholarship. A straightforward account is Thomas Asbridge, The First Crusade: A New History (Oxford University Press, 2004). The classic study by 468

Crusaders and Mass Killing at Jerusalem in 1099

Carl Erdmann, The Origin of the Idea of Crusade, translated by Marshall W. Baldwin and Walter Goffart (Princeton University Press, 1977), remains usable. Practically anything by Jonathan Riley-Smith is helpful, especially The First Crusaders, 1095–1131 (Cambridge University Press, 1997) and The First Crusade and the Idea of Crusading (Philadelphia: University of Pennsylvania Press, 1986). The recent work of Alan V. Murray is strongly recommended. Muslim perspectives are essential. Hillenbrand, The Crusades, provides a classic guide and analysis of the Arabic sources. More recently Niall Christie’s Muslims and Crusaders: Christianity’s Wars in the Middle East, 1095– 1382, from the Islamic Sources 2nd ed. (London: Routledge, 2020) includes fifty pages of translated documents. Scholarship by Alexander Mallett and Konrad Hirschler is solid. Turning to specific aspects of the First Crusade, Kedar’s ‘Jerusalem massacre of July 1099’ is an excellent teaching tool. Conor Kostick presents compelling Marxist analysis in Social Structure of the First Crusade, while France, Victory in the East argues that crusade success was chiefly military. Harari, ‘Eyewitnessing in accounts of the First Crusade’, and Simon John, ‘Historical truth and the miraculous past: the use of oral evidence in twelfthcentury Latin historical writing on the First Crusade’, English Historical Review 130:543 (2015), 263–301 present well-documented examinations and analyses. Throop, Crusading as an Act of Vengeance demonstrates that crusades were religious and vengeance a clear by-product, while Lapina, Warfare and the Miraculous is worth reading for its source analysis. Thomas F. Madden and Norman Housley are reliable guides.

469

18

The Albigensian Crusade and the Early Inquisitions into Heretical Depravity, 1208–1246 mark gregory pegg

On Wednesday, 22 July 1209, 5,000 horsemen, 10,000 foot soldiers and around 12,000 other ragtag men, women and children surrounded the city of Béziers, between Toulouse and Marseilles (Figure 18.1). All of them, from great lords to beggar boys, noble ladies to washerwomen, wore a cross of cloth on their clothing as the ‘sign of the living God’. They were crusaders in a great holy war against the ‘Provençal heretics’ proclaimed sixteen months earlier by Pope Innocent III. Arnau Amalric, abbot of Cîteaux and papal legate leading this crusade, threatened the people of Béziers with wholesale carnage unless they handed over the heretics he accused them of protecting. The townspeople denied the accusation and prepared for a long siege. Around the middle of the next day, some Bitterois men killed and mutilated a foolhardy French crusader on the bridge leading into the city. This enraged the thousands of boys (runaways, servants, thieves, young monks) in the crusader camp. These ribauds, as these children and adolescents were known, started chanting, ‘Let’s attack! Let’s attack!’ Suddenly, jumping defensive ditches, clambering over one another, they swarmed the walls of Béziers. Scratching and clawing the ramparts, loosening stones and rocks, hundreds of boys irrupted inside the city through cracks and crevices. Terrified by such fury, the people of Béziers abandoned their defences. Mobs of delirious boys surged through the streets, clubbing to death everyone they met. As they slaughtered, they searched for treasure. After an hour or so the crusader knights and soldiers entered the city and, whipping the wild boys like dogs, seized all their prizes. Outraged, the boys howled, ‘Burn it! Burn it!’ Piles of corpses and kindling were set alight. The city was soon an inferno. Knights and soldiers, unable to breathe, unable to see, fled the city. Any survivors from the earlier bloodshed were consumed by fire. Many of the arsonists themselves, blinded by smoke, scalded by ashes, suffocated. Less than three 470

The Albigensian Crusade and the Early Inquisitions, 1208–1246

Figure 18.1 Béziers. (Photograph from Getty Images)

hours elapsed from the first ribaud shout to the last fleeing knight. Béziers and all the people who lived there were annihilated in an afternoon.1 ‘To our wonderment,’ Arnau Amalric wrote to Innocent III a month later, ‘these ribauds of ours spared no order of persons (whatever their rank, sex, or age), putting to the sword almost 20,000 people. After this great slaughter the whole city was despoiled and burnt, as divine vengeance raged marvellously.’2 In his joy, Arnau Amalric exaggerated the number of men, women and children killed by about 7,000. A decade after the massacre, with the crusade far from over, the Cistercian Caesarius of Heisterbach summarised the exemplum of the destruction of Béziers in an all too famous anecdote applauding the wisdom of Arnau Amalric. In this story there were no furious 1 G. de Tudela, La Chanson de la Croisade Albigeoise (La Canso de la Crozada), ed. and trans. E. Martin-Chabot, 2nd ed. (Paris: Les Belles Lettres, 1960), 1, laisses 15–23, pp. 44–65; G. de Tudela, The Song of the Cathar Wars: A History of the Albigensian Crusade, trans. J. Shirley (Farnham: Ashgate, 1996), pp. 18–22. Pierre des Vaux-de-Cernay, Historia Albigensis, ed. P. Guébin and E. Lyon, Société de l’Histoire de France (Paris: Librairie Ancienne Honoré Champion, 1926), I, §§ 84–90, pp. 84–91; Pierre des Vaux-de-Cernay, The History of the Albigensian Crusade, trans. W. A. and M. D. Sibly (Woodbridge: Boydell & Brewer, 1998), pp. 48–50. Now, see M. G. Pegg, A Most Holy War: The Albigensian Crusade and the Battle for Christendom (Oxford University Press, 2008), pp. 71–8. 2 Arnau Amalric, ‘De victoria habita contra hereticos’ in Patrologiae cursus completus: Series Latina (PL), ed. J. P. Migne (Paris: Garnier, 1891), C C X V I, p. cviii, cols. 138–41, esp. cols. 138–9.

471

mark gregory pegg

boys or fiery infernos, only confused crusaders worrying about how to sort out the faithful from the heretics inside the captured city. ‘Lord,’ the crusaders ask Arnau Amaric, ‘what shall we do?’ ‘Kill them!’ he replies. ‘Truly, God will know his own!’3 This much quoted (and much misunderstood) dialogue was imagined by Caesarius of Heisterbach. Nevertheless, he encapsulated two profound truths for Latin Christian intellectuals in the early thirteenth century: first, heretics and ordinary Christians were indistinguishable from each other; and second, a moral imperative for mass murder was an essential principle of the holy war on the ‘Provençal heretics’ (Provincales heretici), or, as it was commonly known to northern French knights, poets, bureaucrats and preachers after 1211, the crusade against the Albigensians, or Albigenses. The men, women and children massacred at Béziers were not an ethnic, national or racial group. This chapter will show that despite being accused of heresy, the victims were not a self-consciously different religious group either. They were, arguably, a regional or possibly a cultural group, but such groups are not covered by the modern legal definition of genocide. They were, nevertheless, deliberately targeted for destruction. I shall argue that what happened at Béziers was a ‘genocidal moment’ and it was not the only such moment during the Albigensian crusade. * On Monday, 14 January 1208, the Cistercian Peire de Castelnau, papal legate and virulent denouncer of heretics and mercenaries in the lands of the count of Toulouse for more than a year, was murdered by an anonymous assassin just before sunrise in the shallows of the Rhône river. Two months later, on Monday, 10 March 1208, Innocent III accused Raimon VI, the count of Toulouse, of orchestrating the legate’s death and so, ‘in the name of the God of peace and love’, he proclaimed a crusade against the count and the ‘Provençal heretics’ supposedly infesting the comital lands. ‘Forward then soldiers of Christ!’ exhorted the pope, and ‘purge those lands of heresy’.4 Fighting between the Garonne and Rhône rivers, a vast region labelled 3 Caesarius of Heisterbach, Dialogus Miraculorum, ed. J. Strange (Cologne: H. Lempertz, 1851), I, V, xxi, p. 302; The Dialogue on Miracles, trans. H. von E. Scott and C. C. Swinton Bland (London: Routledge, 1929), I, V, xxi, pp. 345–6. 4 Innocent III, ‘De caede Petri de Castronovo’, PL, C C X V, p. xxvi, cols. 1354–8, esp. 1355, and Layettes du Trésor des Chartes, ed. Alexander Teulet (Paris: Plon, 1863), I, no. 841, to the ecclesiastical provinces of Narbonne, Arles, Embrun, Aix and Vienne. Innocent III’s letter was quoted verbatim in Historia Albigensis, I, pp. 51–65, §§ 55–65 (History, pp. 31–8). Almost identical letters were sent to the province of Lyon and to Philip II Augustus, PL, C C X V, pp. xxvii–xxviii, cols. 1358–9.

472

Map 18.1 Southern France c.1200. (From M. G. Pegg, A Most Holy War (Oxford University Press, 2008), p. 9)

mark gregory pegg

Provincia (Province) by the medieval papacy and what is now largely modern southern France, was as meritorious as fighting in the Holy Land. ‘Attack the followers of heresy more fearlessly than even the Saracens,’ thundered the pope, ‘since heretics are more evil!’5 The massacre at Béziers occurred at the beginning of a long holy war that lasted until 1229. The Albigensian crusade, as these martial pilgrimages are now known, was the first crusade in which Christians were guaranteed salvation through the killing of other Christians. Yet does the moral imperative to mass murder against ‘Provençal heretics’ sanctioned by Innocent III, and the ‘divine vengeance’ at Béziers that so thrilled Arnau Amalric, qualify as either the general endorsement or a specific example of past ‘genocide’? Raphael Lemkin thought so in his ‘Introduction to the Study of Genocide’, an unpublished manuscript unfinished at his death in 1959. He hoped to prove the ‘practicality’ of the Convention on the Prevention and Punishment of the Crime of Genocide adopted by the United Nations General Assembly in 1948 by showing that ‘genocide has always existed in history’.6 In the Middle Ages he argued for twelve cases of genocide, which he sketched out as (mixing victimisers and victims), ‘Goths, Huns, Vandals, Vikings, Charlemagne, Albigenses, Valdenses, England, Jews, Mongols, Moors, the French in Sicily’, ultimately only writing a long chapter on the ‘Albigenses’ or ‘Catharists’.7 Frank Chalk and Kurt Jonassohn in their History and Sociology of Genocide (1990) argue that the Albigensian crusade and the mass killing of heretical ‘Cathars’ was a premodern transitional case of what they define as a modern nation-state form of genocide, ‘committed to implement a belief, theory, or ideology’, in which ‘its primary victims are citizens of the perpetrator state rather than aliens’.8 For the French king (whom they mistakenly identify as Philip the Fair rather than Louis VIII), ‘it was an 5 Historia Albigensis, I, § 64, pp. 63–5 (History, pp. 37–8). Peire de Castelnau’s murder was also described by Guilhem de Puylaurens in Chronica Magistri Guillelmi de Podio Laurentii, ed. and trans. J. Duvernoy, Sources d’Histoire Médiévale (Paris: Éditions du Centre National de la Recherche Scientifique, 1976), I X, p. 52; The Chronicle of William of Puylaurens: The Albigensian Crusade and Its Aftermath, trans. W. A and M. D. Sibly (Woodbridge: Boydell & Brewer, 2003), p. 27. 6 R. Lemkin, Lemkin on Genocide, ed. S. L. Jacobs (Lanham, MD: Lexington Books, 2012), p. 3. The 1948 Convention on the Prevention and Punishment of the Crime of Genocide is printed as an appendix in R. Gellately and B. Kiernan, eds., The Specter of Genocide: Mass Murder in Historical Perspective (Cambridge University Press, 2003), pp. 381–4. 7 Lemkin, Lemkin on Genocide, pp. 6, 59–80. 8 F. Chalk and K. Jonassohn, The History and Sociology of Genocide: Analyses and Case Studies (New Haven: Yale University Press, 1990), p. 37, and the Albigensian crusade and the Cathars as a case study, pp. 114–34, where they include long extracts from J. R. Strayer, The Albigensian Crusades (1971; Ann Arbor: University of Michigan Press, 1992), pp. 8–10, 60–7, and M. Ruthven, Torture: The Grand Conspiracy (London: Weidenfeld & Nicolson, 1978), pp. 75–81, 90–7.

474

The Albigensian Crusade and the Early Inquisitions, 1208–1246

opportunity to expand his realm’, whereas for the pope ‘it was necessary to eliminate heretics who did not recognize his authority’.9 The crusade was a type four case (out of a fourfold typology) from their general definition that genocide ‘is a form of one-sided mass killing in which a state or other authority intends to destroy a group, as that group and membership in it are defined by the perpetrator’. Other sociologists, anthropologists, political scientists, psychologists and modern historians have similarly argued that the Albigensian crusade was genocidal war against the ‘Cathars’.10 Almost no medieval historian labels the Albigensian crusade a genocidal war.11 Joseph R. Strayer comes close in his Albigensian Crusades (1971) when he refers to the holy war (or rolling series of holy wars for him) as a ‘slash through the history of France and of the Church like a gaping wound’. The crusade and the resulting ‘Inquisition’ were ‘one of the classic examples of repression of dissent and of freedom of thought’. ‘Modern totalitarian governments have made few innovations’ since the crusaders, the French monarchy and the papal inquisitors. Although Strayer never explicitly mentions genocide, Chalk and Jonassohn read him as implying it. Three decades later, Malcolm Barber describes the Albigensian crusade as going ‘far beyond the normal conventions of early thirteenth-century warfare, in 9 Chalk and Jonassohn, History and Sociology of Genocide, pp. 37–8. 10 For example C. Powell, Barbaric Civilization: A Critical Sociology of Genocide (Montreal: McGill-Queens University Press, 2011), pp. 165–83, whose fifth chapter is entitled ‘Genocide of ideological others in Languedoc and Guatemala’; S. Pinker, The Better Angels of Our Nature: Why Violence has Declined (London: Penguin, 2011), p. 141 (‘The reason you have never met a Cathar is that the Albigensian Crusade exterminated them. Historians classify this as a clear instance of genocide.’); G. Lerner, ‘In the footsteps of the Cathars’ in Why History Matters: Life and Thought (Oxford University Press, 1997), pp. 19–32, on comparing the destruction of the Cathars and Jews during the Holocaust; and D. Rieff, Slaughterhouse: Bosnia and the Failure of the West (New York: Simon & Schuster, 1995), p. 27, on the killing of ‘Albigensians’ and the massacres in Bosnia. 11 Some historians, though, have argued for genocides in the medieval West. For example, see James E. Fraser, ‘Early medieval Europe: the case of Britain and Europe’ and L. Scales, ‘Central and late medieval Europe’ in The Oxford Book of Genocide Studies, ed. D. Bloxham and A. D. Moses (Oxford University Press, 2010), pp. 259–79 and pp. 280–303 (esp. brief equivocal comment about the Albigensian crusade, p. 297) respectively. Both see genocides as relating to racial or ethnic cleansing, so while the authors acknowledge the anachronistic perils of ‘genocide’ as a category, they use the equally anachronistic categories of ‘race’ and ‘ethnicity’. The Albigensian crusade does not fit their genocide schemata. Of course, ethnic, racial or national categories may offer explanatory frameworks for some but definitely not the majority of medieval phenomena. Cf. R. Bartlett, The Making of Europe: Conquest, Colonization, and Cultural Change 950–1350 (Princeton University Press, 1993), and Geraldine Heng, The Invention of Race in the European Middle Ages (Cambridge University Press, 2018), where the Albigensian crusade and the inquisitions into heretical depravity are conspicuous by their absence.

475

mark gregory pegg

the scale of the slaughter, in the execution of high-status opponents, male and female, in the mutilation of prisoners, in the humiliation and shaming of the defeated, and in the overt use of terror as a method of achieving one’s goals’.12 Nevertheless, despite seeing a ‘lineal descent’ from the crusade to the savage wars of conquest in the Americas by the Spanish against the Mexica in the sixteenth century, he stops short of discussing genocide.13 R. I. Moore in his famous Formation of a Persecuting Society (1987, 2007) asserts that in the twelfth-century Latin West ‘deliberate and socially sanctioned violence began to be directed, through established governmental, judicial and social institutions, against groups of people defined by general characteristics such as race, religion, or way of life; and that membership of such groups in itself came to be regarded as justifying these attacks’.14 An argument that, while not discussing genocide, has obvious implications. Moore elaborates upon the ‘persecuting society’ in The War on Heresy (2012), outlining the ‘systematic, violent and large-scale repression brought to western Europe’ after 1160, which culminated in the Albigensian crusade and the inquisitions into heretical depravity.15 More than a decade ago, I argued that the Albigensian crusade ‘ushered genocide into the West’.16 I also argued that ‘Catharism’ – the most famous of all medieval heresies and the supposed target of the crusaders – never existed. This last point is an irrefutable (if still controversial) fact.17 Moore and a group of brilliant French scholars continue demonstrating, despite a 12 Malcolm Barber, ‘The Albigensian crusades: wars like any other?’ in Dei gesta per Francos: études sur les croisades deditees a Jean Richard, ed. M. Balard and B. Z. Kedar (Farnham: Ashgate, 2001), p. 55. 13 Ibid. 14 R. I. Moore, The Formation of a Persecuting Society: Authority and Deviance in Western Europe, 950–1250, 2nd ed. (1987; Oxford: Blackwell, 2007), p. 4. J. H. Arnold, ‘AHR reappraisal: persecution and power in medieval Europe. The Formation of a Persecuting Society by R. I. Moore’, American Historical Review 123 (2018), 165–74, while thoughtful, misunderstands the implications of the book. 15 Robert I. Moore, The War on Heresy: Faith and Power in Medieval Europe (London: Profile Books, 2012), p. 8. 16 Pegg, Most Holy War, pp. xiv, 188–9. 17 On the controversy about ‘Catharism’, see the collected essays in Antonio Sennis, ed., Cathars in Question (Woodbridge: Boydell & Brewer and York Medieval Press, 2016). For two perceptive reviews of this volume, see J-L. Biget, ‘Cathars in Question’, Revue d’histoire ecclésiastique 113 (2018), 1258–64, esp. 1258, and A. Trivellone, ‘Cathars in Question’, Cahiers de recherches médiévales et humanistes (2016), 1–8, esp. 8. Now, see the essays responding to Cathars in Question in J-L. Biget et al., eds., Le “Catharisme” en questions, Cahiers de Fanjeaux 55 (Fanjeaux: Centre d’études historiques de Fanjeaux, 2020), esp. J-L. Biget, ‘L’histoire du “Catharisme” occitan: un nœd de questions’, pp. 13– 34 and M. G. Pegg, ‘Le Catharisme en questions: falsifiabilité, vérité historique et une nouvelle histoire du christianisme médiéval’, pp. 331–7.

476

The Albigensian Crusade and the Early Inquisitions, 1208–1246

virulent historiographic revanchism, the non-existence of Catharism.18 As the crusade was a holy war on heresy within Latin Christendom, it is often viewed by historians as an aberration within the overall framework of crusading since the First Crusade in 1095.19 Some scholars circumvent the supposed anomaly of the Albigensian crusade by not only dismissing heresy as a historical irrelevancy but arguing that what happened at Béziers (and elsewhere) was nothing out of the ordinary in medieval warfare.20 The crusade is ‘normalised’ beyond all recognition. An even more egregious scholarly attitude suggests that the only reason anyone thinks the crusade was extraordinarily violent, let alone genocidal, is because it was against European Christians and not ‘against safely othered Muslims’ in the Levant.21 Every scholar of genocide, like the medievalists they read, erroneously assumes the victims of the Albigensian crusade were Cathars. Strayer and Barber presumed as much.22 Such widespread misapprehension about Catharism, while problematic, does not invalidate ‘genocide’ as a way of understanding this holy war. So, then, a few questions. Is it historically accurate or analytically useful to label the frenzied boys and the crusader knights at Béziers as, to borrow Helen 18 See e.g. J-L. Biget, Église, dissidences et société dans l’Occitanie médiévale (Lyon: CIHAMÉditions, 2020); R. I. Moore, ‘Principles at stake: the debate of April 2013 in retrospect’ and J. Théry, ‘The heretical dissidence of the “good men” in the Albigeois (1276–1329: localism and resistance to Roman clericalism’ in Cathars in Question, ed. Sennis, pp. 257– 73 and 79–112 respectively; A. Trivellone, ‘Par-delà les cathares: conflicts de pouvoir et hérésie dans le diocèse de Lucques (XIIe–XIVe siècles)’, Revue Historique 691 (2019), 577– 627; and the two volumes edited by M. Zerner, Inventer l’hérésie?: Discours polémiques et pouvoirs avant l’Inquisition and L’histoire du catharisme en discussion: Le ‘concile’ de SaintFélix (1167) (Nice: Centre d’études médiévales Faculté des lettres, arts et sciences humaines Université de Nice Sophia-Antipolis, 1998 and 2001 respectively). Now, see R. I. Moore, ‘Préface à l’édition française’ in Hérétiques: Resistances et répression dans l’Occident médiéval (The War on Heresy: Faith and Power in Medieval Europe), trans. Julien Théry (Paris: Éditions Belin, 2017), pp. 9–23. 19 Karl Borchardt, ‘Casting out demons by Beelzebul: did the papal preaching against the Albigensians ruin the Crusades?’ in La Papauté et les croisades / The Papacy and the Crusades; Actes de VIIe Congrès de la Society for the Study of the Crusades and the Latin East / Proceedings of the VIIth Conference of the Society for the Study of the Crusades and the Latin East (Farnham: Ashgate, 2011), pp. 77–89, esp. p. 89. 20 L. W. Marvin, The Occitan War: A Military and Political History of the Albigensian Crusade, 1209–1218 (Cambridge University Press, 2008), and L. W. Marvin, ‘The Albigensian crusade in Anglo-American historiography, 1888–2013’, History Compass 11:12 (2013), 1126–38. Cf. G. E. M. Lippiatt, Simon V of Montfort and Baronial Government, 1195–1218 (Oxford University Press, 2017), who, while taking the question of heresy seriously, merely follows the conventional narrative of Catharism, even if he never uses the name. 21 H. Rhodes, The Crown and the Cross: Burgundy, France, and the Crusades, 1095–1223 (Turnhout: Brepols, 2020), p. 191 and n. 75. 22 M. Barber, The Cathars: Dualist Heretics in Languedoc in the High Middle Ages, 2nd ed. (London: Routledge, 2013).

477

mark gregory pegg

Fein’s borrowed nomenclature, ‘génocidaires’ (perpetrators of genocide)?23 Or the crusaders who campaigned and committed atrocities after them, which included Louis VIII and his son Louis IX, kings of France, as génocidaires? Even more problematically, were the massacres and mutilations perpetrated by individuals defending themselves from supposed génocidaires, such as the counts of Toulouse, Raimon VI and his son Raimon VII, also ‘genocidal’ in their brutality? If the ‘Provençal heretics’ were not ‘Cathars’, then who were they? Did these ‘Provençal heretics’ even exist? Should we think of the individual testimonies collected by the first inquisitions into heretical depravity (inquisitiones heretice pravitatis), established in Toulouse in the aftermath of the Albigensian crusade, as analogous to the memories of individuals who witnessed or survived genocides collected by modern tribunals? Is it anachronistic and ahistorical to even speak of ‘crimes against humanity’ and so ‘genocide’ in the medieval world? Is the category of genocide an ethical judgement when applied to the premodern past and so, in the end, is the historian engaging in metaphysical speculation rather than epistemological verifiability? * What do I mean by Catharism never existed? I mean that no ‘living Cathars’, glimpsed however imperfectly, underpin the historiography of Catharism. By contrast, underpinning the historiography of medieval Judaism, no matter what historians argue, is the fact that there were ‘living Jews’ in the Middle Ages.24 (Which is not to say that what defined being Jewish in the twelfth, thirteenth or fourteenth centuries was or is obvious or unchanging.) Catharism does not exist outside of its own historiography. It was invented in the late nineteenth century by brilliant scholars (at first mostly German, then French, Italian, American and British) as they restyled history and religion as modern sciences. It is a fin-de-siècle artefact with no likeness to medieval reality, designed (even down to its name) as much as the historical and religious patterns delineating it were discovered, ingeniously fabricated as a discrete world religion with Eurasian roots. Catharism was constructed by modern scholars as a new way of explaining the obsession with heresy that had gripped Latin Christian intellectuals in the late twelfth century and, most especially, as a more scientific explanation for the Albigensian crusade and the subsequent inquisitions into heretical depravity. 23 Helen Fein, Human Rights and Wrongs: Slavery, Terror, Genocide (Boulder, CO: Paradigm, 2007), p. 133. 24 D. Nirenberg, Anti-Judaism: The Western Tradition (New York: W. W. Norton, 2013), pp. 7, 10.

478

The Albigensian Crusade and the Early Inquisitions, 1208–1246

Until the early nineteenth century the persecution of heretics by the medieval Church was explained by scholars within confessional and national parameters. In the seventeenth and eighteenth centuries the heretics of southern France, including Waldensians, were labelled ‘Albigensians’. They were seen as the ancestors of Protestants and so, depending on denominational and national perspective, were either viciously oppressed or deserving of repression.25 In 1827 an editorial on the ‘Albigenses’ in the Sydney Gazette and New South Wales Advertiser stressed that all Protestants were the ‘descendants of these sufferers for conscience’s sake’.26 A more secular and thus scientific approach, particularly associated with the methods of the Religionsgeschichte (Religious-historical school), emerged at the University of Göttingen after 1850, and consciously endeavoured to escape this confessional and national model (even if most scholars adopting this approach were Protestant).27 Charles Schmidt, who studied at Göttingen, anticipated this methodology when he published his Histoire et doctrine de la secte des Cathares ou Albigeois in 1849. He renamed the ‘Albigensians’ as ‘Cathars’ because this uncommon medieval term had no confessional associations. Scholars influenced by the Religionsgeschichte studied religions, especially Christianity, by comparing seemingly similar philosophies, symbols and myths among different systems of belief. Before such similarities were even discovered, some belief systems were classified as ‘world religions’ or ‘universal religions’ resembling Christianity (and then usually modern Protestantism or, in what initially seems contradictory, medieval Catholicism). World religions possessed clerical hierarchies, evangelical missionaries, fixed rituals, foundational sacred texts and distinctions between the secular and the religious. Hinduism, Confucianism and Buddhism, for instance, were constructed as world religions, but so too were late antique paganism, Gnosticism, Manichaeism and, by 1900, ‘Catharism’.28 25 Sandra Pott (Richter), ‘Radical heretics, martyrs, or witnesses of truth? The Albigenses in ecclesiastical history and literature (1550–1850)’ in Heresy in Transition: Transforming Ideas of Heresy in Medieval and Early Modern Europe, ed. I. Hunter and J. C. Laursen (Farnham: Ashgate, 2005), pp. 181–94. 26 ‘The crusades against the Albigenses’, Sydney Gazette and New South Wales Advertiser, Wednesday, 14 February 1827 (25, no. 1309), p. 4 (the last page out of four large folded pages making up the newspaper). Now, see M. G. Pegg, ‘Albigenses in the Antipodes: an Australian and the Cathars’, Journal of Religious History 35:4 (2011), 577. 27 On Religionsgeschichte see S. Marchand, German Orientalism in the Age of Empire: Religion, Race, and Scholarship (Washington, DC: German Historical Institute and Cambridge University Press, 2009), esp. pp. 259–67. 28 Ibid., pp. 259, 262; on Buddhism, pp. 270–9, 298, 319–20; on Hinduism, pp. 193, 317–18; on Confucianism, pp. 372, 476; on Gnosticism, pp. 268, 286–7. See also M. G. Pegg, ‘Innocent III, les “pestilentiels provençaux” et le paradigme épuisé du catharisme’ in

479

mark gregory pegg

It is rare to say without reservation that a field of study axiomatic for more than a century is false, so it is not said lightly. The falsifiability of Catharism necessitates judging all modern histories of the Albigensian crusade which maintain it was a holy war against the Cathars (no matter how exact their descriptions of battles and sieges) as erroneous. It is worth noting that contemporary analogies between Cathars and Jews, and so the Albigensian crusade and the Holocaust, is an implicit continuation of what the nineteenth-century scholars who invented Catharism also tacitly assumed. In belle époque France these scholars not only viewed Catharism as a cosmopolitan world religion threatened by church and state; they explicitly saw the persecution of the Cathars as a lesson from the past affirming their vision of a more tolerant and less clerical nation – not surprisingly, they were all Dreyfusards.29 Regrettably, a modern loathsome variation among some French writers is the accusation of négationnisme against any historian who even doubts the existence of Catharism, explicitly linking such an individual with Holocaust denial.30 * It may be argued that even if nineteenth-century scholars invented Catharism, what they thought they discerned really did exist, no matter what we call it. This is not the case. Just as there were no ‘living Cathars’ so there were no ‘living heretics’ either, at least not before around 1220. If by ‘heretics’ what we mean were individuals who self-consciously said and did things that they knew would lead to their imprisonment, confiscation of property, banishment, persecution and death, then there were no ‘living heretics’ before the thirteenth century. (No individuals or groups accused of heresy in late antiquity or the early medieval west self-consciously adopted the accusations of their accusers as an integral part of their identity.) It is this heretical self-consciousness that scholars argue for when they argue for the existence of Catharism and against what they see as the ‘constructedness of heresy’, where even suggesting Innocent III et le Midi, ed. J. Théry, Cahiers de Fanjeaux 50 (Fanjeaux: Centre d’études historiques de Fanjeaux, 2015), pp. 281–5, and M. G. Pegg, ‘The paradigm of Catharism; or, the historians’ illusion’ in Cathars in Question, ed. Sennis, pp. 22–5. 29 Pegg, ‘Paradigm of Catharism’, 33–4. 30 M. Roquebert, ‘Le “déconstructionisme” et le études cathares’ in Les cathares devant l’histoire: mélanges offerts à Jean Duvernoy, ed. M. Aurell (Quercy: L’Hydre, 2005), pp. 105– 33, accuses scholars, especially Monique Zerner, of négationnisme in even doubting Catharism. Zerner replied to Roquebert (and others) in ‘Mise au point sur les Cathares devant L’historie et retour sur L’histoire du Catharisme en discussion: le débat sur le Charte de Niquinta n’est pas clos’, Journal des Savants (2006), 253–73. Alessia Trivellone more recently has been accused of négationnisme on account of a travelling exhibition in universities in southern France detailing the historical falsehood of Catharism.

480

The Albigensian Crusade and the Early Inquisitions, 1208–1246

that society plays some role in the formation of identities in the past is, somehow or other, to deny that those past identities were real and meaningful.31 Let me be clear about something that should be obvious but all too regrettably is not – denying the reality of Catharism does mean denying medieval reality, and most especially, it does not mean denying the reality that thousands of men, women and children were accused of heresy, murdered by crusaders and persecuted by inquisitors. There were widespread accusations of heresy by Latin Christian intellectuals (monks, priests, nuns, bishops, canons, popes, university teachers and students) in the twelfth century, especially after 1130. Initially, they accused one another of heresy, mostly in learned polemics and lecture hall sermons, but eventually they directed such accusations at ordinary lay Christians outside the cathedral schools and nascent universites, counts and peasants alike. An accusation of heresy functioned as a reprimand, criticism, admonition, act of spiritual guidance, rhetorical threat and, for the first half of the twelfth century, an affirmation of the historical continuity of the Church. ‘For I do not recall having heard anything new or extraordinary in all their assertions,’ Bernard, Cisterican abbot of Clairvaux, preached about heretics and their doctrines in 1144, ‘but only trite commonplaces long vented amongst the heretics of old.’32 There had always been heretics within the Church, they were platitudinous and not that threatening. A year later, though, he warned Anfoz-Jordas, count of Toulouse, that a ‘wild beast’ named Henri was poisoning Toulousains with heretical preaching. He advised the city of Toulouse not to ‘receive any outside or unknown preacher, unless he be sent by the supreme pontiff, or have permission to preach from your bishop’. False preachers always mixed poison with honey – the words of heretics, no matter how trite, putrefied innocent Christians, ‘imperceptibly, just like cancer’.33 What infuriated Bernard was the sheer indifference of the count to freelance holy men like Henri. Anfoz-Jordas dismissed the admonition, but this accusation of nonchalance towards one vagabond ‘heretic’ was the beginning of threatening ultimatums against lords 31 Cf. J. H. Arnold, ‘The Cathar Middle Ages’, in Cathars in Question, ed. Sennis, pp. 53–78, esp. 73–6. On the ‘constructedness of heresy’ see his ‘Voicing dissent: heresy trials in later medieval England’, Past & Present 245:1 (2019), 3–37. 32 Eberwin of Steinfeld, Ep. 472, PL, C L X X X I I, cols. 676–80; and Sancti Bernardi Opera II: Sermones super Cantica Canticorum 36–86, ed. Jean Leclercq, C. H. Talbot and H. Rochais (Rome: Editions Cistercienses, 1958), sermon 65, pp. 172–7. For translations of both see Heresies of the High Middle Ages: Selected Sources Translated and Annotated, ed. Walter Leggett Wakefield and Austin P. Evans (New York: Columbia University Press, 1991), pp. 126–38. 33 Wakefield and Evans, Heresies, Ep. 242, pp. 128–9.

481

mark gregory pegg

and counts between the Garonne and Rhône that, within four decades, had escalated into extremism. ‘In Gascony and the lands of Albi and Toulouse, and in other places,’ decreed the last canon of the Third Lateran Council in March 1179, ‘those heretics, whom some call Cathars, others the Patarenes, others the Publicani, and others by different names, has grown so strong that they no longer practise their wickedness in secret.’34 These evil heretics were the same as (and often in alliance with) the sinful mercenaries ravaging these lands, known as Brabançons, Aragonese, Coterels, Basques and Triaverdines.35 Any lord defending or supporting heretics and mercenaries was ‘placed under anathema’.36 The council saw no difference between heretics, mercenaries and the lords who sustained them – they all wounded the body of Christ. The privileges of fighting heretics and mercenaries in the Toulousain were the same as fighting Saracens in the Holy Land. Later that same year when Pons d’Arsac, archbishop of Narbonne, relayed what he had heard at the council to his suffragan bishops, he stressed how the increasing ‘madness of heresy’ in the Agenais, Toulousain and Albigeois was caused by ‘foreign mercenaries and thieves’. He recited the list of mercenaries, but never mentioned Cathars, Patarenes or Publicani, as these names were too obscure, Italian and erudite, with no relation to local knowledge.37 The heretics denounced by the Third Lateran Council were literally figures of speech, stereotypes from the schoolroom, allowing Roman canonists and Parisian intellectuals to make sense of and so classify the supposedly war-torn and fractious world between the Garonne and Rhône.38 Thirty years later, in April 1207, Peire de Castelnau excommunicated Raimon VI for employing mercenaries and his unwillingness to swear and maintain peace along the Rhône.39 Innocent III, when he confirmed this excommunication, warned the count that if he continued his endless small wars against disobedient village nobles and peasants with evil foreign mercenaries, ‘then we enjoin all the princes around you to rise against you as an 34 Decrees of the Ecumenical Councils, 2 vols., ed. and trans. Norman P. Tanner (Washington, DC: Georgetown University Press, 1990), I I , pp. 211–25, esp. pp. 211, 224–5. On the Third Lateran Council, see Danica Summerlin, The Canons of the Third Lateran Council of 1179: Their Origins and Reception (Cambridge University Press, 2019), esp. pp. 58–62, for a discussion of this canon; unfortunately, she accepts the reality of Catharism. 35 Ibid. 36 Ibid. 37 Histoire générale de Languedoc avec des notes et les pièces justificatives, Histoire générale de Languedoc avec des notes et les pièces justificatives, ed. C. Devic and J. Vaissette (c.1730), 2nd ed., ed. A. Molinier (Toulouse: Édouard Privat, 1889), V I I I , n. 37, cols. 341–4. 38 Pegg, Most Holy War, pp. 52–3. 39 Historia Albigensis, I, § 27, pp. 30–1 (History, p. 21).

482

The Albigensian Crusade and the Early Inquisitions, 1208–1246

enemy of Christ … lest they become even more infected by the stain of heresy under your rule’.40 Ten months later ‘one of those mercenaries of Satan’ murdered the papal legate. * ‘Before the coming of the crusaders,’ Bernart Amielh confessed to the Dominican inquisitors Bernart de Caux and Johan de Sant Peire on Monday, 3 July 1245, in the Romanesque cloister of the abbey church of Saint-Sernin in Toulouse, ‘I saw heretics publicly walking through the streets.’41 For Bernart Amielh these ‘heretics’ (heretici in Latin; eretges in Occitan) were village ‘good men’ (boni homines, probi homines in Latin; bons omes, prozomes in Occitan). Less than a year later, the knight Bertran de Quiders told the same inquisition on Monday, 12 March 1246 that ‘before the crusaders first came’, around 1209, ‘when I was five years old, or thereabouts, I frequently ate nuts, bread, and other foods given to me’ by na Garsenda and na Gualharda, his widowed grandmother and aunt, both ‘heretics’.42 For Bertran de Quiders his grandmother and aunt were village ‘good women’ (bone femine in Latin; bonas femnas in Occitan) or ‘good ladies’ (bone domine in Latin; bonas domnas in Occitan). Almost 6,000 other persons (girls over twelve, boys over fourteen) were interrogated at Saint-Sernin between May 1245 and August 1246.43 It was the largest inquisition in the Middle Ages. As far as Bernart de Caux and Johan de Sant Peire were concerned, the ‘good men’, ‘good women’ and their ‘believers’ (credentes in Latin; crezens in Occitan), along with a few Waldensians, were the heretici infesting the lands between the Garonne and the Rhône. Courtliness, the famous cortezia of troubadours, shaped the lives of everyone living between the Garonne and Rhône after 1140. Daily village courtesies (bows, greetings) allowed individuals to survive and navigate their way through the complex system of land tenure; a swirling tempest of rights and claims, at once moral and material, and all too vulnerable to dispute and violence.44 The ‘honour’ (onor) of a man lay in the accumulation and communal recognition of his proprietary claims and rights. It was only through day-to day cortezia that questions of honour were answered and a man’s identity affirmed. Conflicts of honour between village men and women resulted in the relentless small wars denounced by Church councils 40 29 May 1207, PL, C C X V, cols. 1166–8, and RHGF 13, p. 140. 41 Toulouse, Bibliothèque municipale, MS 609, fols. 18v–19r. 42 Ibid., fol. 29v. 43 M. G. Pegg, The Corruption of Angels: The Great Inquisition of 1245–1246 (Princeton University Press, 2001), pp. 45–51. 44 On the complex issue of land tenure, see Pegg, Most Holy War, esp. pp. 29–31.

483

mark gregory pegg

and papal legates as the breeding ground of heresy. Sometimes, though, these conflicts were adjudicated by village nobles and peasants acting as ‘good men’. It was a temporary honorific. However, one or two village men were always ‘good’ as they were the holy ‘good men’. Holiness was as capricious as honour, ebbing and flowing through (and around) all humans. Questions of holiness as much as questions of honour were simultaneously answered and realised through cortezia. The holy good men were local manifestations of imitatio Christi, the developing notion in Latin Christendom that ordinary humans through habitual words and actions emulated Christ. Village men and women, noble and peasant, constantly offered the good men greetings and bows, and these courteous gifts were known as the melhoramen (melioration). The good men offered consolamen, ‘consolation’ or ‘comforting’, when an individual lay dying, transforming him or her into a holy good man or good woman before death. Before the Albigensian crusade almost all holy good men were widowers, although some little boys were ‘good men’ (often living with their widowed fathers).45 The good women before the Albigensian crusade were mostly older noble matrons beyond the years of fertility, no longer able or willing to marry, sometimes widows, sometimes separated from elderly husbands, living together in twos or threes in village houses. They rarely left these houses, never preached, and were given few if any gifts of cortezia, unlike the very public courtliness for good men. Remarkably, thousands of noble girls as young as four, although no older than twelve, lived with these old good women as prepubescent ‘good women’. What was similar about all these houses of women at the beginning and end of their lives was their intense isolation from the moral and social instability of the world. Inside these houses the good women did not offer each other decorous speech or genuflections; granddaughters did not bow to grandmothers; nieces did not ask for benedictions from aunts.46 * ‘Readers should know that the heretics of Toulouse and other cities and villages, as well as their defenders, are generally called “Albigensians” in this book,’ the youthful Cistercian Pierre des Vaux-de-Cernay clarified when he dedicated his history of the Albigensian crusade (much of which he witnessed 45 Ibid., pp. 34–8; M. G. Pegg, ‘“Un démon dans mon ventre”: Cortezia, amour, hérésie avant la Croisade albigeoise’ in L’Église et la chair (XIIe–Xve siècle), ed. J. Théry, Cahiers de Fanjeaux 52 (Fanjeaux: Centre d’études historiques de Fanjeaux, 2019), pp. 73–4. 46 Pegg, Corruption of Angels, pp. 92–103.

484

The Albigensian Crusade and the Early Inquisitions, 1208–1246

as a participant) to Innocent III, ‘since this is the name that other nations came to use for all Provençal heretics.’47 He described only two sects of ‘Provençal heretics’. The first preached the falsehood of two creators: ‘one of things invisible whom they called the “benign” God and one of things visible whom they named the “malign” God’. The second were ‘very much less perverted’ and went by the name of Valdenses, from ‘Valdes, a citizen of Lyon’. The worst error of the ‘Waldensians’ was (out of four) wearing ‘sandals in imitation of the Apostles’. Pierre des Vaux-de-Cernay himself preferred using ‘Albigensians’ only for heretics who were not Waldensians, which, for him, were the good men.48 In 1208 when northern French knights announced they were marching against the ‘Albigensians’, the name did not signify heresy, only meaning ‘southerners’, deriving from Albi being the southernmost diocese within Bourges, the southernmost archdiocese of France. This changed around 1210; for example, when Herbert Perier de Chaudion became a crusader and gave away half the revenue of his lands to the abbey of Signy in the Ardennes, the abbot Gautier de Chaumont lauded him as inspired by fervore fidei christianae contra Albigenses incredulos, ‘the fervour of the Christian faith against Albigensian unbelievers [or disobedients]’.49 After 1211 Albigenses encompassed heretics, heretical believers, mercenaries, refugee southern nobles, criminal mutilators, good men and good women, in short, anyone who opposed or was accused of opposing the holy war.50 ‘Indeed, such pestilential Provençals not only strive to devastate all we possess, but they strive to annihilate us!’ declared Innocent III when he proclaimed the crusade in 1208. ‘Not only are they sharpening their tongues to annihilate our souls, they are raising their hands to annihilate our bodies! Truly, they have become perverters of souls and putrefiers of bodies!’51 When the pope used disease metaphors to conjure up images of heretics between the Garonne and the Rhône, it was not (just) bombast, it was a precise and 47 Historia Albigensis, preface, § 4, pp. 3–4 (History, p. 6). 48 Historia Albigensis, I, §§ 10–19, pp. 9–20 (History, pp. 10–15). 49 R. T. Pippenger, ‘Crusading as a Family: A Study of the County of Champagne, 1179 to 1226’ (PhD thesis, Princeton University, 2018), pp. 163–4, and generally as a brilliant analysis of crusading. 50 D. Power, ‘Who went on the Albigensian crusade?’, English Historical Review 128:534 (2013), 1047–85, is excellent on French nobility using ‘Albigensian’ around 1208. He argues this undermines J-L. Biget’s Hérésie et inquisition dans le Midi de la France (Paris: Picard, 2007), pp. 142–69, and Pegg, Most Holy War, pp. 21–2, 117, that the term was only later associated with heresy. He never demonstrates that ‘Albigensian’ was a synonym for heresy before the crusade for the northern French, he merely assumes it. 51 Historia Albigensis, I, § 61, p. 60 (History, p. 35).

485

mark gregory pegg

methodical way of articulating what he saw as the threat to Christendom.52 These metaphors simultaneously evoked the spatial specificity of heresy growing within the lands of the count of Toulouse, the temporal continuity of heresy within the Church, and the insidious way in which heresy secretly poisoned careless Christians so that they (and their neighbours) were unaware of their own pestilence. What made this threat so apocalyptic was that all Christians (including popes) would eventually be corrupted by heresy unless it was exterminated. Innocent III imagined himself in the middle between God and humanity, ‘below God but above man, less than God but greater than man, who judges all things but who no one judges’ – and he judged a crusade was the only way of cleansing Latin Christendom of this plague of the ‘pestilential Provençals’.53 ‘The errors of the Albigensians,’ argued Caesarius of Heisterbach a decade or so after the crusade proclamation of Innocent III, ‘infected more than a thousand towns’ between the Garonne and Rhône and, without the ‘swords of the faithful’ hacking it back, then unquestionably, ‘I think this infection would have corrupted the whole of Europe’.54 Nevertheless, despite such apocalypticism after the fact, it was mostly the appeal of crusading in the Toulousain that attracted so many martial pilgrims in 1209, rather than the horror of heresy, which, while very real for Innocent III and other learned clerics, was not yet a fear permeating Latin Christendom. It was the Albigensian crusade itself, especially after the massacre of Béziers, and then the inquisitions into heretical depravity, that transformed heresy from an accusation of bookish intellectuals into a widespread reality for all Christians.55 * Fein’s ‘generic definition’ of genocide, supplementing the 1948 Convention, ‘focuses on collectivities – persisting identity groups that include, but are not limited to, ethnic, religious, racial, and national groups’.56 The purpose of the Convention ‘was to protect nonviolent collectivities, real groups that are the source of identity and exist apart from the invention of the state perpetrators, 52 Pegg, ‘Innocent III’, 225–79, and R. I. Moore, ‘Heresy as disease’ in The Concept of Heresy in the Middle Ages (11th–13th C.): Proceedings of the International Conference, Louvain May 13– 16, 1973, ed. W. Lourdaux and D. Verhelst (The Hague: Leuven University Press and Martinus Nijhoff, 1976), pp. 1–11. 53 Innocent III, ‘Sermo II’, PL, C C X V I I, col. 658. 54 Caesarius of Heisterbach, Dialogus Miraculorum, I, V, xxi, p. 301 (Dialogue, p. 345). 55 In Most Holy War I did not fully appreciate that it was the appeal of crusading and not the fear of heresy that attracted so many martial pilgrims in 1209. 56 Fein, Human Rights and Wrongs, p. 132.

486

The Albigensian Crusade and the Early Inquisitions, 1208–1246

whose members are persecuted for who they are, not what they have done’.57 She argues (against Chalk and Jonassohn) that ‘fictive groups’ like ‘enemies of the state, subversives’ (often created by ‘the arbitrary labelling of dictators’) cannot be ‘victims of genocide’, even if real individuals suffer as ‘victims of state terror’ because of such invented accusations. Her emphasis upon genocide applying only to real groups whose identity exists apart from the imagination of génocidaires is, paradoxically, useful for the Albigensian crusade. ‘Catharism’ relates to no real group in the twelfth and thirteenth centuries, being an invention of nineteenth-century scholarship. ‘Heresy’ related to no real group in the twelfth or early thirteenth centuries either; nevertheless, it was not an invention of Latin Christian intellectuals, as this diminishes the seriousness, sincerity and eventual horror of the accusation. For Innocent III, the heretici were not a fictive group, but very real. (Just as for many dictators, ‘enemies of the state’ were and are a real group.) Nevertheless, the men, women and children massacred during the Albigensian crusade were not an ethnic, national, racial or religious group, as defined by the Convention. Fein further argues that in most modern genocides ‘the génocidaires either have been the state or have seized the state’.58 She does not speculate on premodern precedents. Chalk and Jonassohn, following Strayer, do argue that the papacy and the French crown were medieval versions of modern states implementing particular beliefs and ideologies in the execution of genocide.59 Ignoring their error about real Cathars being murdered, their type four motivation for genocide is partially applicable to the Albigensian crusade. What is less pertinent is the notion of the state, whether the papacy or the French monarchy, as the impresario of crusading or inquisitorial génocidaires. The Convention, though, does acknowledge that a state is not necessary for acts of genocide to be perpetrated. In 1204 Innocent III implored the king of France, Philip II Augustus, to wage a holy war in all but name against ‘heretics and faithless clerics’ in the lands of Raimon VI, ‘so that the material sword can be seen to compensate for the deficiency of the spiritual sword’. He even offered ‘the same indulgence of sins which we grant to those who cross over to aid the Holy Land’.60 The king rebuffed the pope. Four years later he again declined to go south against Raimon VI, even disputing the pope’s right to declare a crusade into 57 Ibid. 58 Ibid., p. 137. 59 Chalk and Jonassohn, History and Sociology of Genocide, p. 37. 60 Innocent III, ‘Ut assistat legatis apostolicis contra haereticos’, PL, C C X V, p. lxxix, cols. 361–2.

487

mark gregory pegg

the county of Toulouse as the count himself was not specifically accused of heresy.61 In the end, though, Philip II Augustus reluctantly allowed the duke of Burgundy and the count of Nevers to go on crusade with only 500 knights. No crusader was acting on behalf of any state (kingdom, empire) when they first campaigned between the Garonne and Rhône.62 What adds to the (stateless) complexity of that first summer was that Raimon VI, through an act of repentance beneath the tympanum of the church of Saint-Gilles in June 1209, became a crusader himself, joining the ‘soldiers of Christ’ at Montpellier (whose overlord was King Pere II of Aragon). Around 2,000 Provençal nobles and peasants, many of them sometime village good men, also became crusaders. Most northern crusaders distrusted the count, as they had signed themselves with the cross weeks and months earlier with the fierce intention of attacking him and expunging heretics from his territories. Raimon VI as a crusader did not change the fact that a landscape of pestilence still existed between the Garonne and the Rhône that needed to be cleansed through holy war. Arnau Amalric surveyed the horizon and, gauging what stood between him and eventual victory over the ‘Provençal heretics’ in the lands of the count of Toulouse, decided to attack the neighbouring territories of the 24-year-old Raimon-Roger Trencavel, viscount of Béziers, Carcassonne, Razès and Albi, whose overlord (as count of Barcelona) was Pere II and whose uncle was Raimon VI.63 * After the lands of Raimon-Roger Trencavel were targeted by Arnau Amalric, the martial pilgrims marched two days along the Mediterranean coast until they reached Béziers on the Orb river. The 12,000 camp followers, including the ribaud boys, were an astonishing millennial mob – only what we call the First Crusade of 1096 had inspired so many ordinary Christians. This is important. In 1208 Innocent III promised that all Christians wearing the ‘life-giving sign of the cross’ were to ‘know that remission of sins has been granted by God and His vicar to all who, fired by zeal for the orthodox faith, take up arms for this work of piety’.64 The same privileges as an armed pilgrimage to the Holy Land were granted to all crusaders. The family and property of a man who ‘yielded to Christ’ were protected from the time he took the cross until he died or returned – so that all claims to his property 61 ‘Responses faites par le roi Philippe-Auguste au pape, touchant l’affaire d’Albigeois’ in Histoire générale de Languedoc, ed. A. Molinier, V I I I , p. 138, cols. 558–9. 62 Pegg, Most Holy War, p. 60. 63 Ibid., pp. 68–70. 64 Historia Albigensis, I, § 64, pp. 63–5 (History, pp. 37–8).

488

The Albigensian Crusade and the Early Inquisitions, 1208–1246

were suspended, so that all interest on his debts ceased – and, if funds were needed for his journey, lands and other possessions could be pledged (without interest). Clerics who took the cross were able to fund their pilgrimages by mortgaging two years’ worth of revenue from their benefices. By becoming a pilgrim on this crusade, even if you were a beggar boy, noble lady, monk or knight, was to ‘walk like Christ’ as if you were actually in the Holy Land.65 This was the pope’s magnificent redemptive gift. Like the First Crusade, where the goal of capturing and cleansing Jerusalem only became the divine purpose of the crusaders when they reached the Levant, it was only by mid-summer 1209 that the fear of heresy destroying Christendom, while never widespread except among ecclesiastical intellectuals before 1208, and certainly not why so many joined the crusade a month earlier, was zealously embraced (if only at first by liminal pilgrims like the ribauds) by the soldiers of Christ. The massacre at Béziers in its haphazard beginning, escalating feral fury and whirlwind ferocity was at once the result and justification of this millennial exhilaration of imitating Christ by killing His supposed enemies. Although the crusading exhortations of Innocent III exalted the necessity of expurging and exterminating ‘Provençal heretics’ from Christendom, it was not until Béziers that this homicidal ethic was fully adopted by those doing the killing. This moral imperative to mass murder starkly differentiated the expedition of 1209 from all crusades (past and future) to the eastern Mediterranean or Iberian peninsula. Pierre des Vaux-de-Cernay compared purged Béziers to redeemed Jerusalem; except, he was not thinking of crusaders in 1099, he was recalling when the Roman emperor Vespasian and his son Titus destroyed the Temple and expelled the Jews a thousand years earlier.66 Innocent III and his legate Arnau Amalric warned, over and over, that most persons who lived in and around the county of Toulouse were wholly or partially diseased with heresy and did not know it. Unless men and women ostentatiously announced themselves opposed to heresy, in the manner of Raimon VI and hundreds of lesser Provençal nobles becoming crusaders, then by default they were either heretics or supporters of heretics – and even then that was ultimately not enough for the count. The crusaders never thought they were fighting a ‘counter-Church’ of heretics (as the nineteenth-century invention of Catharism would have it). What added to their ardour, and so their murderousness, was the conviction that if heresy were not erased then, sooner or later, they 65 Pegg, Most Holy War, p. 60.

66 Historia Albigensis, I, § 91 p. 93 (History, p. 51).

489

mark gregory pegg

too would be corrupted by this plague.67 It was only through the precautionary killing of all and sundry in a city like Béziers that the crusaders avenged (and imitated) Christ and so saved themselves. What happened at Béziers was obviously not a state-sanctioned massacre; indeed, much of the redemptive and transformative efficacy of the martial pilgrimage (like any pilgrimage) derived from its statelessness. Neither does it fit the 1948 Convention and Fein’s ‘generic definition’ derived from it. The slaughter at Béziers had nothing to do with ethnic, racial or national collectivities. These categories offer no explanatory framework for understanding the Albigensian crusade and the early inquisitions.68 Religious ‘group identity’ is more problematic, especially as those killing and those killed at Béziers all assumed they were part of the same religious group, that is, Latin Christianity. (Accusations of heresy always presupposed the accused were Christians, just misguided and dangerous, or had had the revelation of Christ and turned away from Him, which was why Jews and Muslims were soon accused of being heretics like the ‘Albigensians’.) More useful is Ben Kiernan’s emphasis upon ‘genocidal massacres’ (after Leo Kuper, Chalk and Jonassohn, and Fein) or the concept of ‘genocidal moments’, which are often spontaneous, unexpected and intensely communal, even if they are manifestations of previous hatreds and fears, or more disturbingly, brutal exemplifications of recent accusations and tirades about supposedly long-standing hatreds and fears.69 Clearly, the slaughter at Béziers was not a riot or a pogrom, possessing more than just ‘a protogenocidal quality about it’.70 It certainly occurred within the institutional framework of what defined a crusade and a crusader, which the papacy was formulating throughout the twelfth century, and which only became systematised in large part due to the Albigensian crusade. Nevertheless, it was a maelstrom within that framework, appearing as if out 67 Jacques Semelin, Purify and Destroy: The Political Uses of Massacre and Genocide, trans. Cynthia Schoch (New York: Columbia University Press, 2007), where killing as an act of purification for the murderers and those they murdered is extensively discussed. 68 C. Caldwell Ames, Medieval Heresies: Christianity, Judaism, and Islam (Cambridge University Press, 2015), p. 264, ‘If Pegg is correct [about genocide], we can see this crusade as anticipating late-medieval associations of heresy and race as objects of persecution.’ Also, see her ‘Christian violence against heretics, Jews, and Muslims’ in The Cambridge World History of Violence, vol. I I: 500–1500 CE, ed. Matthew Gordon and Richard Kaeuper (Cambridge University Press, 2020), pp. 470–91. 69 B. Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007), pp. 13–14. See also L. Kuper, Genocide: Its Political Use in the Twentieth Century (New Haven: Yale University Press, 1981), pp. 10, 32, 60; and H. Fein, Genocide: A Sociological Perspective (London: Sage, 1993), pp. 18–19, 86–8. 70 D. L. Horowitz, The Deadly Ethnic Riot (Berkeley: University of California Press, 2001), p. 20.

490

The Albigensian Crusade and the Early Inquisitions, 1208–1246

of nowhere, engulfing everyone within it, before subsiding. What happened at Béziers was a ‘genocidal moment’ specific to a particular time and place. The French knights and the ribauds were momentarily holy génocidaires. ‘As rumours of so miraculous and terrifying event steadily disseminated far and wide,’ wrote Arnau Amalric to Innocent III, stressing how the holocaust at Béziers was a momentous wonder, ‘everyone thereabouts sought refuge in the mountains and inaccessible places.’71 Like a miracle, a ‘genocidal moment’ is rare and unforeseen, fleetingly uniting heaven and earth for murderers and murdered alike, giving the world frightening new meanings and models. ‘If it can be shown that some heretics are in a city,’ lectured the Bolognese lawyer Johannes Teutonicus around 1217, ‘then all the inhabitants can be burnt.’72 * A ‘genocidal moment’ unequivocally erupted at Béziers in July 1209. Indeed, the miraculous terror of this moment only dissipated at the end of summer. When Carcassonne was captured a month later all the citizens, rather than being slaughtered – although the crusaders thought about it – were expelled into the scorching countryside as refugees, dressed only ‘in shirts and breeches’.73 Most of the crusaders then wanted to loot and burn the city.74 Arnau Amalric decided Carcassonne deserved preservation (unlike its citizens) as he envisioned it as a citadel from which present and future crusaders would fight heresy. The crusade was a war of years and not months. Arnau Amalric and the French lords appointed the 44-year-old Simon IV, lord of Montfort, as the new viscount of Béziers, Carcassonne, Razès and Albi. (Raimon-Roger Trencavel was imprisoned, dying soon afterwards.) The millennialism that had inflamed the crusaders largely vanished when Simon de Montfort became viscount and nominal leader of the holy war. The apocalyptic edge that was so sharp in the ribaud boys at Béziers, transforming them into génocidaires, and still fierce among other crusaders at Carcassonne, was dulled and finally blunted by Arnau Amalric’s conviction that the crusade would not exterminate heresy in one summer. The realisation that heresy, 71 Amalric, ‘Victoria habita contra hereticos’, p. cviii, cols. 138–9. 72 Glossa ordinaria (to the Decretum), ed. A. Caravita and P. Caravita (Venice: Apud Iuntas, 1605), C. 23 q. 5 c. 32 ad v. omnes. Now, see Brundage, ‘Holy war and medieval lawyers’, pp. 99–140, esp. pp. 123 and 139n167. 73 Crozada, 1, laisses 31–3, pp. 76–83 (Crusade, pp. 25–6); Historia Albigensis, I, §§ 98–9, pp. 99– 100 (History, pp. 54–5); Chronica, X I V, p. 62 (Chronicle, p. 34), and PL, C C X V I, p. cviii, col. 139 (Chronicle, p. 128). 74 Amalric, ‘Victoria habita contra hereticos’, p. cviii, cols. 139.

491

mark gregory pegg

while restrained, was still widespread, diluted the fervour of the crusaders. Holy wars, especially where pilgrims were momentarily génocidaires, are surprisingly fragile endeavours. ‘The Lord in His Compassion did not wish this most holy war to end soon,’ wrote Pierre des Vaux-de-Cernay as he explained why the crusade did not end in 1209, ‘because it providently provided forgiveness for sinners and the enhancement of grace for the just.’75 Over the next twenty years, among the seemingly endless moments of viciousness and bloodlust by both crusaders and those they attacked, further savage acts were ambiguously ‘genocidal moments’. Simon de Montfort and his small crusader army blinded and sliced off the noses of a hundred villagers from Bram, leaving just one man one eye to shepherd his mutilated friends to another village during the bitter winter of 1209/10. They did this in retaliation for two crusaders similarly mutilated (eyes, ears, noses, upper lips) a few weeks earlier in Minerve.76 In 1211 at Lavaur, Simon de Montfort burnt 400 townspeople accused of heresy in a colossal funeral pyre, arguably a ‘genocidal massacre’. Na Girauda, the lady of Lavaur, was held over a deep well, shrieking and screaming, before being dropped in and stones thrown on top of her.77 In February 1211, Arnau Amalric excommunicated Raimon VI for employing mercenaries and Jews, exacting high tolls, not going to the River Jordan as a pilgrim, and harbouring heretics.78 Simon de Montfort immediately invaded the Toulousain, seizing the village of Les Cassés and incinerating sixty accused heretics. Even if the ‘plague of heresy’ was finally receding, Arnau Amalric informed Innocent III in January 1213, ‘in the city of Toulouse and a few villages, like filth sinking into a bilge-hold, the residue of heretical depravity has collected’.79 At the 1215 Fourth Lateran Council the pope named Simon de Montfort as the new count of Toulouse.80 Raimon VI vigorously resisted. In late summer 1217 he entered Toulouse by wading across the Garonne, hidden by the morning mist. Men and women, giddy at the count’s return, ran through 75 Historia Albigensis, I, § 109, pp. 113–14 (History, p. 60). On the views of Innocent III and continuous holy war, see ‘De negotio comitatus Melgoriensis’ and ‘De negotio fidei’, PL, C C X V, pp. cccxxxii–cccxxxiii, cols. 1546–7. 76 Historia Albigensis, I, §§ 122–36, 146–50, pp. 125–40, 151–4 (History, pp. 68–74, 80–2). 77 Crozada, 1, laisses 62–71, pp. 154–75 (Crusade, pp. 39–43); Historia Albigensis, I, §§ 213–16, 223–30, pp. 211–16, 223–30 (History, pp. 110–12, 115–18); Chronica, X V I, p. 68 (Chronicle, pp. 38–9); and Alberic (Aubri) de Troisfontaines, ‘Chronica Albrici monarchi Trium Fontium, a monaco novi monasterii Hoiensis interpolate’ in Monumenta Germaniae Historica Scriptorum 23, ed. P. Scheffer-Boichorst (Hanover: Hahnian, 1874), p. 892. 78 Pegg, Most Holy War, pp. 111–12. 79 Historia Albigensis, I I, § 392–8, pp. 65–96 (History, pp. 181–4), and ‘Super eodem’, PL, C C X V I, p. xli, cols. 836–9. 80 Pegg, Most Holy War, pp. 141–9.

492

The Albigensian Crusade and the Early Inquisitions, 1208–1246

the streets crying ‘Toulouse! Toulouse!’ as they clubbed to death all northern Frenchmen.81 Simon de Montfort, caught flat-footed outside the city walls, besieged Toulouse. An anonymous troubadour remembered a garden ‘burst forth and blossomed’ every day before the walls. ‘It was sown with lilies; but the white and the red which budded and flowered were flesh and blood and weapons and the slosh of splattered brains. Spirits and souls, sinners and saved, the freshly killed replenished hell and paradise.’82 Simon de Montfort’s skull was crushed by a rock flung from a mangonel inside Toulouse worked by noble ladies, merchant wives and little girls in June 1218. The anonymous troubadour sang with bitter irony of the epitaph above the count’s tomb at Carcassonne (all but labelling him a génocidaire): The epitaph says, for those who can read it, that he is a saint and a martyr and that he shall breathe again and inherit and flourish in marvellous joy and wear a crown and be seated in the Kingdom. And me, I have heard it said that this must be so if, by killing men and by spilling blood and by squandering souls and by sanctioning deaths and by trusting evil counsel and by setting fires and by destroying barons and by dishonouring paratge and by seizing lands and by nourishing pride and by lauding evil and mocking the good and by massacring ladies and by slaughtering children, a man can win over Jesus Christ in this world, then the count of Montfort wears a crown and shines in heaven.83

In May 1219, Prince Louis, son of Philip II Augustus, marched south as a crusader, soon capturing the town of Marmande, whose 5,000 men, women and children were immediately butchered. ‘Barons and ladies and little children and men and women, their clothes slashed, their bodies naked, were hacked and cut to pieces by sharp-edged swords,’ sang the anonymous troubadour. ‘Flesh and blood and brains and torsos and limbs and faces were carved and ripped. Livers and guts were torn out and tossed about the ground as if they had rained down from above.’ Marmande was then set on fire. The

81 Crozada, 2, laisses 181–3, pp. 264–83 (Crusade, pp. 119–23). 82 Crozada, 3, laisses 192–8, pp. 62–127 (Crusade, pp. 144–55). 83 Crozada, 3, laisse 208, p. 228 (Crusade, p. 176). Now, see D. Power, ‘The Albigensian crusade after Simon of Montfort (1218–24)’ in Simon de Montfort (c. 1170–1218): le croisé, son lignage et son temps, ed. M. Aurell, G. Lippiatt and L. Macé (Turnhout: Brepols, 2020), pp. 161–78.

493

mark gregory pegg

frenzied slaughter took Louis by surprise and the sweet-smelling smoke of the burning corpses sickened him.84 Unlike more ambiguously genocidal carnage, such as the ‘killing fields’ before the walls of Toulouse, the mass murder at Marmande was a ‘genocidal moment’ as terrifying miracle, where the crusaders were suddenly transformed into génocidaires. This maelstrom was inspired as much, if not more so, by the sacred and secular legitimacy conferrred upon such violence by the French monarchy. Seven years after Marmande the prince as King Louis VIII – Philip II Augustus had died in 1223 – embarked on a grand crusade into the ‘Albigensian lands’. The new king thought these southern lands were by their very nature an equal threat to both his kingdom and to Christendom. A 1226 Paris sermon compared Louis VIII claiming his earthly kingdom when he descended into the Albigensian lands to Christ claiming His heavenly kingdom when He ascended into heaven.85 The holy war on heresy was now a royal crusade.86 Although Louis VIII died soon after the Marmande massacre, the crusade continued. In late 1228 the new count of Toulouse, Raimon VII – his father Raimon VI had died in 1222, and, as an excommunicant, remained for another five centuries unburied in a coffin on a pedestal in Toulouse – was offered peace, and, fighting a losing battle against the French crown, he accepted. The next year, on Holy Thursday, 12 April, the 31-year-old count stood barefoot in a loose shirt in the great square of Notre-Dame cathedral in the presence of the almost fifteen-year-old king of France, Louis IX, and thousands of French nobles, prelates and Parisians. Raimon VII was reconciled to Jesus Christ and His Church; he did homage to the king. The Albigensian crusade was over.87 Four years later Pope Gregory IX, worried that heresy was spreading in France and resurfacing in Provincia, issued two letters (20 and 22 April 1233) asking the Friars Preachers, the Dominicans, to eliminate this insidious serpent through inquisitions into heretical depravity.88 Unlike early modern institutional Inquisitions, medieval inquisitions were always ad hoc (and so should not be capitalised). Apart from the great inquiry at Saint-Sernin very few testimonies exist from other inquisitorial tribunals before 1250. The 84 Crozada, 3, laisse 212, pp. 282–91 (Crusade, pp. 186–9); and Chronica, X X X, pp. 106–9 (Chronicle, pp. 64–5). 85 C. T. Maier, ‘Crisis, liturgy, and the crusade in the twelfth and thirteenth centuries’, Journal of Ecclesiastical History 48 (1997), 628–57, esp. 652–3. 86 Pegg, Most Holy War, pp. 176–8. 87 Chronica, X X X V I I, pp. 134–5 (Chronicle, p. 80), and Pegg, Most Holy War, pp. 178–81. 88 Y. Dossat, Les crises de l’Inquisition Toulousaine au XIIIe siècle (1233–1273) (Bordeaux: Imprimerie Bière, 1959), pp. 118–21, and his edition of both letters on pp. 325–9.

494

The Albigensian Crusade and the Early Inquisitions, 1208–1246

confessions that have survived (usually as seventeenth-century copies of lost originals) reveal how thousands of men, women and children lived their lives before, during and after a brutal holy war. * History is not a metaphysical discipline, it is an epistemological one. It is not about how the world could have been or could be, it is about what can be known of the actual world, as it is and was.89 Any discussion of genocide, human rights or crimes against humanity in the past, especially the premodern past, inevitably shades into metaphysics.90 Although classifying any of the murderous episodes during the Albigensian crusade as crimes against humanity is anachronistic and metaphysically ahistorical, it still comes as a surprise that the category of genocide, which initially seems as anachronistic, if not more so, actually provides epistemological clarity about these episodes. The holy war itself was not a sustained genocidal enterprise over twenty years, despite the constant ecclesiastical exhortations to mass murder, but two massacres unquestionably transcended the more routine butchery as ‘genocidal moments’. Violence as a redemptive act, the escalating terror of heresy, and the holiness of being signed with the cross all converged in these moments. A miraculous and horrific metamorphosis occurred at Béziers (especially) and Marmande, when already holy killers became holy génocidaires. It is the underlying sacrality of genocides, rarely discussed by modern historians and theorists, which differentiates them from other mass murders. Although we can identify the génocidaires in the genocidal moments of Béziers and Marmande, we cannot and never will be able to identify the supposed ‘heretics’ among the dead, as real heretics as a real group did not yet exist. The crusaders and ribaud boys were not exterminating ‘Provençal heretics’ as a religious collectivity in 1209; instead, they were eliminating heresy as an insidious plague within Christians like themselves. This fear was enhanced for northern French nobles by the insidiousness of the ‘Albigensian plague’ within the kingdom of France in 1219. It is not a coincidence that ‘living heretics’ – individuals self-consciously saying and doing things they knew were ‘heretical’ and risking death – only existed at the very earliest after 1220. Finally, in the two decades after the crusade, inquisitors transformed the surviving and now fugitive holy good men and good women into a ‘heresy’ 89 This definition of metaphysics and epistemology is adapted from S. Kripke, ‘Identity and necessity’ in Philosophical Troubles (Oxford University Press, 2011), I, pp. 14–15. 90 I. Clendinnen, Reading the Holocaust (Cambridge University Press, 1999), pp. 10–16: S. Moyn, The Last Utopia: Human Rights in History (Cambridge, MA: Harvard University Press, 2010), pp. 1–43.

495

mark gregory pegg

by transforming them into a ‘religion’, or more correctly, transfiguring them into a sect or an ecclesia, a ‘Church’. All the good men and good women had become heretics not only to their accusers but to themselves. It is perhaps one of the terrible ironies of history that it was only on account of the genocidal violence of the Albigensian crusade and then the inquisitions into heretical depravity that the medieval heretic – self-aware, authentic, full of agency – can be said to have come into existence, to have become real.

Bibliographic Note Wakefield and Evans’ Heresies of the High Middle Ages is still exemplary. The introduction and the preambles to sixty documents from the twelfth and thirteenth centuries are deeply learned and the endnotes are informative and exhilarating to read. Even taking into account that there was no serious debate five decades ago about the reality of heresy or Catharism, a reader would have come away fully informed about the state of the field in 1969. Walter Wakefield’s own Heresy, Crusade, and Inquisition in Southern France, 1100–1250 (London: Allen & Unwin, 1974) has six appendices covering sixty pages of translations related to heresy, crusade and inquisition; indeed, the translations came first and what was meant to be an introduction became an splendid historical monograph in itself. Regrettably, this book is not in print. R. I. Moore, The Birth of Popular Heresy (London: Edward Arnold, 1975) and Edward Peters, Heresy and Authority in Medieval Europe: Documents in Translation (Philadelphia: University of Pennsylvania Press, 1980) are excellent collections by great historians. The Siblys’ translation of Pierre Vaux-deCernay’s History of the Albigensian Crusade is good, with outstanding appendices of translations from other documents related to the crusade. Similarly good is their translation of The Chronicle of William of Puylaurens. Guilhem de Tudela’s Song of the Cathar Wars, translated by Janet Shirley, is useful (if mediocre), with limited notes and apparatus. More recently, there is the accomplished collection of sources on heresy and the Albigensian crusade by Catherine Léglu, Rebecca Rist and Claire Taylor, The Cathars and the Albigensian Crusade: A Sourcebook (London: Routledge, 2013). John H. Arnold and Peter Biller’s Heresy and Inquisition in France, 1200–1300 (Manchester University Press, 2016) is a marvellous and field-changing collection (fifty-seven documents) consciously complementing Wakefield and Evans’ Heresies of the High Middle Ages. Unlike the older volume, though, this sourcebook does not introduce and footnote sources with current scholarship in a critical and open-minded manner, especially misrepresenting or ignoring historians who do not think Catharism 496

The Albigensian Crusade and the Early Inquisitions, 1208–1246

existed. Finally, there is a superb and ongoing online edition and English translation of the great inquisition of Bernart de Caux and Johan de SantPeire undertaken by Jean-Paul Rehr at the Registry of the Great Inquisition at Toulouse, 1245–1246: Edition and Translation of Bibliothèque municipale de Toulouse MS 609, http://medieval-inquisition.huma-num.fr.

497

19

Mongol Genocides of the Thirteenth Century timothy may

In 1202, Temüjin, khan of the Borjigid Mongols and the future Chinggis Khan, defeated the Tatars in eastern Mongolia at the battle of Dalan Nemürges. After taking the survivors prisoner, the Mongol leader ordered his men to execute all males taller than the linchpin of a wagon wheel.1 The linchpin, in the axle, was no higher than the wheel’s radius, thus exempting few Tatar men or boys from the massacre. The massacre of the Tatars by the Mongols was the first general massacre conducted by Temüjin’s Mongols. It took place even prior to the rise of the Mongol Empire, for in 1202 Temüjin was still the servitor of Toghril OngKhan, khan of the Kereit confederation (which included the Borjigid Mongols) in central Mongolia. Temüjin’s actions, however, were purely his own. Toghril had not been involved in the destruction of the Tatars, hereditary enemies of the Mongols. Yet the massacre became the hallmark of the Mongols as they became an imperial power. Was the destruction of the Tatars a genocide as defined by the 1948 United Nations Genocide Convention?2 Although applying a modern legal definition to a medieval event is anachronistic, it still bears consideration. Should this massacre be defined as genocide, or alternatively, should it be categorised under what scholars have called genocidal massacres, compensating for the rigidity of the UN’s definition, in which a mass killing takes place in a localised setting, but without intent to annihilate the entire group?3 Or should it be considered 1 Igor de Rachewiltz, ed. and trans., The Secret History of the Mongols: A Mongolian Epic Chronicle of the Thirteenth Century (Leiden: Brill, 2003), p. 77. 2 United Nations, Office on Genocide Prevention and the Responsibility to Protect, ‘Genocide: Background’, www.un.org/en/genocideprevention/genocide.shtml (accessed 16 July 2020). 3 Ben Kiernan, ‘Is “genocide” an anachronistic concept for the study of early modern mass killing?’, History 99:336 (2014), 530–48; Frank Chalk and Kurt Jonassohn, The History and Sociology of Genocide: Analyses and Case Studies (New Haven: Yale University Press, 1990), p. 26.

498

Mongol Genocides of the Thirteenth Century

anything beyond a massacre? But what differentiates a massacre from a genocidal massacre? A war crime? Intent? As with all things, the devil is in the details. As indicated, the 1202 event was neither the first nor the last massacre that the Mongols committed. Yet, was there a difference between the massacres in central Asia and Iran when Chinggis Khan invaded the Khwa¯razmian Empire in 1219–23 or in other campaigns and the much more specific massacres conducted against the Tatars, the Tangut in 1227, or the Isma¯ʿı¯lı¯s in 1256? Throughout their conquests, the Mongols perpetrated what might be considered genocidal massacres, but it is unclear if they committed genocide. In light of assertions that atmospheric carbon levels dropped considerably during the period of the Mongol Empire, suggesting a significant reduction in population, it is fair to wonder if genocidal practices became the standard operating procedure under the Mongols.4

Background The Mongol Empire arose through a series of wars between the pastoral nomads of the Mongolian plateau in the late twelfth and early thirteenth centuries. Before Temüjin, who became Chinggis Khan in 1206, unified the various tribes of Mongolia into the Yeke Monggol Ulus (Great Mongol Nation/ People), no single power had dominated the steppes since the Liao Empire (906–1125) collapsed after being overwhelmed by the Jurchen people of the Manchuria Empire, who went on to to establish the Jin Empire (1125–1234) in northern China. The Jin Empire did not seek to extend their direct control into Mongolia. Rather, they sought to influence events there through proxies and focused on preventing a confederacy from rising that might threaten their power. The Jin government did so by a time-honoured practice of playing tribes off against each other.5 When one became threatening, the Jin secured the support of another through the extension of gifts, titles and prestige for the leaders, as well as favourable trade terms. In return, the nomads worked with the Jin Empire to undermine the rising power. In the vacuum left by the collapse of Liao influence in the steppe, a Mongol khanate arose in northern Mongolia. The nascent khanate raided the Jin 4 Julia Pongratz et al., ‘Coupled climate-carbon simulations indicated minor global effects of wars and epidemics on atmospheric CO2 between AD 800 and 1850’, The Holocene 21:5 (2011),849. 5 Thomas Allsen, ‘The rise of the Mongolian Empire and Mongolian rule in north China’ in The Cambridge History of China, vol. V I: Alien Regimes and Border States, 907–1368, ed. Denis Twitchett and John K. Fairbank (Cambridge University Press, 1994), p. 324.

499

timothy may

borders persistently from 1135 to 1147. The Jin became sufficiently concerned that they supported the Tatars, who dwelled closer to the Jin Empire, to counter the growing Mongol threat. The Tatars had their own feuds with the Mongols; thus this alliance was not simply the Jin using ‘barbarians to fight barbarians’.6 For their part, the Tatars used the vast resources of the Jin to execute their own goals. Dates are vague for this period, but in the 1150s or 1160s, the Mongols led by Ambaghai Khan sought some reconciliation with the Tatars through a marriage alliance. En route to the wedding, a group of Tatars serving as part of the Jin jüyin forces took him captive. The jüyin were groups of nomads and others who served as frontier guards for the Jin. As there were several Tatar tribes, not all served as jüyin, nor were all jüyin Tatars or even nomads. The Jin tortured and executed the captive Ambaghai.7 While animosity with the Jin continued, the Mongols focused their ire primarily on the Tatars as a tangible enemy not only for their perfidy but also due to their proximity, as the Mongols lacked the capability to truly threaten the Jin emperor beyond raiding the frontier.8 After the death of Ambaghai, Qutula became the Mongol khan. Their wars of vengeance did not enjoy resounding success, although they experienced some minor victories. One Mongol leader who established himself in this period was Yisügei, who captured a Tatar leader named Temüjin Üje.9 Yisügei’s son was born at this time, in 1162. Yisügei viewed it as an auspicious moment and named his son Temüjin, after this Tatar prisoner. The Mongol– Tatar feud did not subside, but there were moments of peace, and during one such, in around 1170, Yisügei sought to find a bride for Temüjin. While the nomads did not marry so young, due to the vast distances of the steppe and exogamous customs, it was not unusual to arrange a marriage and enter an agreement many years before the bride and groom were married (around age fifteen). Yisügei found a suitable bride, named Börte, among the Onggirad. Leaving Temüjin with her father, Dei Sechen, Yisügei and his retainers rode back to their camp, several days’ journey away. On the way back, Yisügei and his company rested at another camp. On the steppes, hospitality was the custom for travellers, as one never knew when the situation might be 6 Rashı¯d al-Dı¯n, Ja¯miʿ’ al-tawa¯rı¯kh, ed. B. Karı¯mı¯ (Tehran: Iqbal, 1983), p. 59; Rashiduddin Fazlullah, Jami’u’t-tawarikh: Compendium of Chronicles, trans. W. M. Thackston (Cambridge, MA: Department of Near Eastern Languages and Civilizations, Harvard University, 1998), p. 45; hereafter citations will be RD/Mu¯savı¯, RD/Karı¯mı¯ and RD/ Thackston respectively. 7 Rachewiltz, Secret History, p. 10; RD/Karı¯mı¯, p. 179; RD/Thackston, p. 121. 8 RD/Karı¯mı¯, p. 60; RD/Thackston, p. 45. 9 Rachewiltz, Secret History, p. 13; RD/Karı¯mı¯, p. 210; RD/Thackston, p. 139.

500

Mongol Genocides of the Thirteenth Century

reversed. Although the camp proved to be among the Tatars, Yisügei could still expect food and shelter. The Tatars, however, violated the cultural norms. Although they dared not openly attack him, they poisoned his food so that he sickened and died shortly after arriving at his camp.10 Before his death, Yisügei sent for Temüjin, and the boy was reunited with his family. Afterwards, the family suffered hardships, and Temüjin did not forget the many grievances against the Tatars. He slowly rose in power, beginning his journey as a nökör of Toghril, the khan of the Kereit, a powerful tribe in central Mongolia. Further, Toghril and Yisügei had an anda (blood brother) relationship; thus, Temüjin had a form of familial tie to the Kereit leader. With Toghril’s assistance, Temüjin rose in prominence among the Mongols, but he ran afoul of a rival, Jamuqa, who had similar aspirations and also served Toghril. The two rivals fought in 1186–7, and Temüjin was defeated at the battle of Dalan Baljut. Toghril did not take part, possibly because he had been ousted from the throne by his own kinsmen. After this defeat, Temüjin disappears from the sources for approximately a decade. While the Tatars had been allies of the Jin Empire for much of the twelfth century, in the 1190s the relationship soured, resulting in raiding and warfare. The Jin turned their support to Temüjin, who had probably fled to them after Jamuqa defeated him,11 and assisted not only in restoring him as khan of the Borjigid Mongols but also actively supporting him in defeating the Tatars in 1196. Temüjin clearly saw the opportunity to strike at a hereditary enemy, which took precedence over any lingering feelings about the Jin’s execution of Ambaghai Khan.12 This joint operation, which also included Kereit troops, inflicted a significant defeat on the Tatars and garnered Temüjin and Toghril titles, indicating that the Jin recognised them as their vassals, which also entitled Toghril and Temüjin to trading opportunities with the Jin.13 After this, Temüjin and Toghril continued to accumulate power by defeating a number of enemies. Although Temüjin remained Toghril’s subordinate, his position and standing grew. Indeed, Toghril may have even considered Temüjin to be his successor, something that did not sit well with Toghril’s son Senggüm and others.14 It is against this setting that Temüjin made his final move against the Tatars. He was now a powerful 10 Rachewiltz, Secret History, pp. 16, 337. 11 Zhao Hong, Meng-Da Bei-Lu (Polnoe opisanie Mongolo-Tatar), trans. N. Ts. Munkuev (Moscow: NAUKA, 1975), p. 49. 12 Rachewiltz, Secret History, p. 56. 13 Ibid. 14 Ibid., p. 84; Timothy May, The Mongol Empire (Edinburgh University Press, 2018), pp. 36–7.

501

timothy may

figure in the steppes, supported by both Toghril and the Jin Empire. With this in mind, he decided to inflict a final and decisive defeat upon the Tatars.

Tatars After the 1196–7 campaign, the Tatars played an insignificant role in the steppes for several years. The Secret History does not mention them again until 1201, when a new confederation led by Jamuqa arose to counter the growing dominance of Toghril and Temüjin. Jamuqa’s confederation consisted of a wide variety of tribes, including the Tayichi’ud Mongols and other Mongols who did not accept Temüjin as their khan. In addition, the Turkic Naiman from western Mongolia as well as the Oirat from the forests north of Mongolia, in Siberia, as well as others joined Jamuqa’s campaign. Among them were a contingent of Tatars led by Jarin Buqa.15 The fact that they were not a significant force indicates that the Tatars had not fully recovered from their defeat in 1196–7. Furthermore, when Toghril and Temüjin routed Jamuqa’s confederation, Temüjin pursued the Tayichi’ud rather than the Tatars, viewing the former as a greater threat. Yet, in 1202, with the Tayichi’ud defeated, only the Tatars remained to challenge Temüjin’s domination of eastern Mongolia. As mentioned earlier, Toghril did not take part in this campaign as he marched against the Merkit in northern Mongolia. It is not clear if Toghril’s operation was coincidental or if the two allies planned to eliminate two enemies at once. Temüjin marched against the Tatars and encountered them at Dalan Nemürges, located on the western slopes of the Khingghan mountains near Nömrögiin Gol, a tributary of the Khalkha river.16 There they encountered the Cha’an, Alchi, Duta’ut and Aruqai Tatars, suggesting the Tatars had gathered their full might to resist Temüjin.17 The battle lasted for a few days, and ended in a Mongol victory. Afterwards, the Mongols pursued the fleeing Tatars to their camp along the Ulqui Silügeljit stream. With the Tatar warriors in disarray, the Mongols plundered their camp and took the majority prisoner.18 Meeting in council with his commanders, Temüjin decreed the execution of any Tatar male taller than a linchpin of an ox cart: ‘And kill them to the last man, / We shall utterly slay them / The rest we shall enslave.’19 15 Rachewiltz, Secret History, p. 62. The text indicates Jalin Buqa, but Rachewiltz notes that the proper name based on other evidence should be Jarin. 16 Ibid., p. 567. 17 Ibid., p. 76. 18 Ibid., pp. 76, 137. 19 Ibid., p. 77.

502

Mongol Genocides of the Thirteenth Century

However, advance word of the massacre got out when Temüjin’s halfbrother, Belgütei, taunted the Tatars with their fate shortly after the council, an indiscretion which later caused him to be banned from participation in important councils.20 The forewarned Tatar leader, Yeke Cheren, rallied the Tatars and barricaded the site where they were being held, forcing the Mongols to storm it and suffer casualties. Then, even after being forced from their prison, the Tatars resisted and inflicted further injury upon the Mongols, using knives they had concealed in their clothing.21 One might question why the Mongols had not disarmed them, but all men carried these knives for routine tasks such as trimming arrows or cutting meat. While they could be used as weapons, they certainly were not designed nor viewed as such. Furthermore, in the voluminous deels or long robes worn by nomads, such tools could be kept undetected. The resistance failed to save the Tatars, but it cost the Mongols numerous casualties while the Tatars died as warriors rather than as sheep – an important distinction in the ethos of the steppes, although it did not win the Tatars any mercy. The Mongols’ Secret Historian recorded: ‘In this way the Tatar was finally measured against the linchpin of a cart and exterminated’ (‘ci’ün-tür ülijü kiduju alaju ögüya’).22 Yet, the destruction was not complete. Measuring against a linchpin meant that small male children were spared, along with the women. They were then incorporated into the community of the Mongols as slaves, although some of the women became wives. Temüjin, the future Chinggis Khan, took the sisters Yisui and Yisügen as his wives.23 It is probable that many of the young children were adopted as sometimes occurred among orphaned children among the steppe tribes.24 Chinggis Khan had four adopted brothers, each one an orphan from defeated tribes, including Shiqi Quduqu, whom Chinggis Khan had found among the wreckage of the Tatar camp after the 1196–7 campaign.25 By dispersing the women and children among the Mongols, Chinggis Khan sought to erase the Tatar tribal identity. We must question, however, whether this depiction in the Secret History is accurate. It is unlikely that all males were killed in this seeming act of gendercide. While we do know that some men who escaped the earlier massacre were killed upon discovery, Yisüi’s former husband being the prime example, as Igor de Rachewiltz indicates, in reality the massacre 20 Ibid., pp. 77–8. 21 Ibid., p. 77. 22 Ibid. See also Igor de Rachewiltz, Index to the Secret History of the Mongols (Richmond, UK: Curzon Press, 1997), p. 72. 23 Ibid., p. 78. 24 Ibid., pp. 43, 46, 58, 60. 25 Ibid., pp. 58–60.

503

timothy may

eliminated the men of the leading clans.26 Thus, while the Mongols eradicated the elites – they carried out eliticide – the commoners, including men, survived. By destroying the leadership of the Tatars, Temüjin attempted to destroy the identity of the Tatars. While the name might linger, without the elites, the bonds of tribal identity weakened, thus making it easier for the Mongols to incorporate the commoners. While the ethnonym ‘Tatar’ might continue to exist, as a separate identity distinct from the Mongols, it faded. The elimination of tribal leadership had become standard practice for Temüjin in his rise to power beginning in 1196–7. During that campaign against the Tatars, another Mongol clan, the Jürkin, had raided his camp in his absence; they were supposed to have joined him in his attack on the Tatars. Not only did he seek vengeance against them, but Temüjin again perpetuated eliticide and the rest of the Jürkin clan became incorporated into the Borjigid Mongols.27 The same occurred with the Tayichi’ud after Jamuqa’s defeat in 1201.28 Other tribes that voluntarily submitted to Temüjin escaped the same fate, but their aristocracy did not share an equal status as the altan uruq, for example the family of Chinggis Khan. Nomadic confederations consisted of numerous polities or tribes that operated under the identity of the confederation’s leader. For instance, while Temüjin became Chinggis Khan, a title meaning firm, fierce or resolute leader, for most of his early career he was a nökör, or bondsman, of Toghril, the khan of the Kereit.29 As such, he and his followers were part of the Kereit confederation. When Jamuqa organised a confederation against Toghril, he took the title of Gur-Khan (Universal Ruler), a prestigious title also used by the ruler of the Qara Khitai Empire (1125–1218). With Jamuqa’s defeat in 1201, his Gur-Khanid confederation proved ephemeral, but had he have proven successful perhaps history would have recorded it much like other confederations such as the Xiongnu, Xianbei and others. Nomadic tribes were fluid bodies that, as Rudi Lindner has indicated, could expand and contract as members joined.30 They did so voluntarily to seek protection or better access to goods, or forcibly when the defeated submitted to the leadership of the dominant tribe, or in some cases were dispersed among a tribe or confederation as slaves, or simply to disrupt their former identity. While the ruler was powerful, he was not an autocrat and had 26 Ibid., pp. 78–9, 571. 27 Ibid., pp. 58–60. 28 Ibid., pp. 68–70. 29 See Igor de Rachewiltz, ‘The title Cˇ inggis Qan/Qaγan re-examined’ in Gedanke und Wirkung: Festschrift zum 90. Gerburtstag von Nikolaus Poppe, ed. Walter Heissig and Karl Sagaster (Wiesbaden: Harrassowitz, 1989), pp. 281–98. 30 Rudi Lindner, ‘What was a nomadic tribe?’, Comparative Studies in Society and History 24:4 (1981),701.

504

Mongol Genocides of the Thirteenth Century

to listen to the shared interests of his subordinates, as members of the tribe could vote with their feet and depart or even overthrow the ruler.31 If a ruler proved too autocratic, coups were not uncommon, as Toghril experienced as khan of the Kereit when his reach exceeded his grasp.32 Chinggis Khan, however, avoided this through repeated successes. To be certain, there were times when his authority was challenged, but he also never abandoned consulting his subordinates. Over time, the authority of his bloodline became unrivalled due to his success, which was seen as divine qut, or fortune. After his death, the ideology of Tenggerism developed, which held that Chinggis Khan and his descendants were bequeathed the earth and those who resisted their authority were in rebellion against the will of Heaven (Tenggeri).33 This and the charisma of the altan uruq secured for decades leadership for the descendants of Chinggis Khan and ensured that only those of his bloodline could rule (Figure 19.1). As Chinggis Khan, Temüjin also created a new tribal identity known as the Yeke Monggol Ulus or Great Mongol Ulus. It is difficult to define ulus; it means ‘people’, ‘appanage’, ‘nation’, or ‘state’. When Chinggis Khan first adopted the term, he certainly viewed it as ‘people’ and it eventually came to mean ‘state’ with the formation of the Yeke Monggol Ulus. It is quite clear that the

Figure 19.1 Chinggis Khan statue at the Chinggis Khan Statue Complex in Tsonjin Boldog, Mongolia. (Photo: Timothy May) 31 Ibid. 32 Isenbike Togan, Flexibility and Limitations in Steppe Formations: The Kerait Khanate and Chinggis Khan (Leiden: Brill, 1998), pp. 69–70, 80–2. 33 Sh. Bira, ‘Mongolian Tenggerism and modern globalism: a retrospective outlook on globalisation’, Inner Asia 5:2 (2003), 110.

505

timothy may

Mongols reserved ulus for those they considered Mongols; that is, those nomads who became part of the Yeke Monggol Ulus, which included the surviving Tatars. Other defeated tribes also saw their aristocracy eliminated with the commoners then incorporated into the Yeke Monggol Ulus. For all others, namely nomads outside of the Mongol entity or sedentary populations, the Mongols used irgen, another term for ‘people’.34 As a result, the Mongols developed a true sense of the other and it defined much of their world view.

The Tangut The kingdom of Xi Xia (982–1227), located in the Ordos Loop and modern Gansu and Ningxia provinces of China, was dominated by the Tangut, a Tibeto-Burmese people. They ruled a polyglot and polyethnic state of Tangut, Han, Turks, Tibetans, Uighurs and Khitans.35 The Mongols first encountered them in 1205 while hunting for Senggüm, Toghril Ong Khan’s son. He fled there after the Mongols defeated the Kereit in 1203. Thereafter, the Mongols raided Xi Xia and eventually undertook a full-scale invasion. Although the kingdom resisted, the Mongol successes and their determination in siege warfare daunted the Tangut. Eventually, Xi Xia submitted in 1210, resolving that paying tribute, which included a princess who became one of Chinggis Khan’s wives, was a far better proposition than suffering the continual invasions.36 The Mongols did not occupy Xi Xia and remained satisfied with the tribute agreement, which also obligated the Tangut to provide troops when Chinggis Khan requested. As the first primarily sedentary or agricultural realm that the Mongols conquered, they demonstrated no interest in ruling Xi Xia directly or even occupying it. The request for troops soon manifested itself after the Mongols invaded the neighbouring Jin Empire, which straddled Manchuria and northern China, bordering Xi Xia. One can imagine that the Tangut did not object to the request as the Jin Empire had rebuffed Xi Xia’s earlier attempts to gain Jin support against the Mongols.37 At the height of Xi Xia’s involvement, they

34 Lhamsuren Munkh-Erdene, ‘The Mongolian nationality lexicon: from the Chinggisid lineage to Mongolian nationality (from the seventeenth to the early twentieth century)’, Inner Asia 8:1 (2006), 57. See also Lhamsuren Munkh-Erdene, ‘Where did the Mongol Empire come from? Medieval Mongol ideas of people, state and empire, Inner Asia 13:2 (2011), 211–37. 35 Ruth Dunnell, ‘The Hsi Hsia’ in Cambridge History of China, V I ,pp. 154–6. 36 Ibid., 208. 37 Ibid., 207.

506

Map 19.1 The Mongol Empire, 1250–60. (Cartography by Mapping Specialists, Fitchburg, Wisconsin)

timothy may

contributed approximately 50,000 Tangut troops to the Mongol invasion of the Jin Empire and proved to be loyal vassals.38 This changed in 1223. At the time, Chinggis Khan and the majority of the Mongol army were in central Asia and Iran, destroying the Khwa¯razmian Empire in a war initiated in 1219. The Khwa¯razmsha¯h, Sultan Muhammad II (r. 1200–20), had permitted a Mongol-sponsored caravan to be massacred in Otra¯r by his governor’s forces, and then killed Mongol envoys who sought to resolve the issue diplomatically. At the time, Chinggis Khan had little interest in another war as his forces had the Jin Empire on the ropes. Yet with this offence, Chinggis Khan had little choice but to respond. Thus, in 1219 he marched over 1,600 kilometres and invaded the central Asian empire. Nonetheless, Chinggis Khan still hoped to reduce the Jin Empire and left the general Muqali with a small army augmented by Jin deserters, those who had rebelled against the Jin and the Tangut troops from Xi Xia. As the Jin now began to hold their own, the Tangut sensed an opportunity and rebelled against the Mongols. Muqali then focused his efforts against Xi Xia, but he died in 1223. His death strengthened the Tangut’s resolve. News soon reached Chinggis Khan in Afghanistan of the Tangut’s perfidy. Having destroyed the Khwa¯razmian Empire, Chinggis Khan withdrew his army north of the Amu Darya river. Leaving only a small force behind, he led his army back to Mongolia. He did not invade Xi Xia immediately. Instead, he engaged in diplomacy to resolve the matter, demanding hostages from the Xi Xia royal family to ensure their behaviour. Negotiations soon broke down as the Tangut refused to send the hostages. The time spent was not wasted, however; it allowed the Mongols to rest and regroup their armies. Undoubtedly, the Tangut fortified their kingdom as well. Two Tangut officers in the employ of the Mongols, Chahan and Xili Gambu, continued to attempt negotiations, but at best they could only mitigate the severity of the Mongol reprisal, as their kinsmen refused to submit.39 Nonetheless, the Mongols invaded in 1226 and systematically destroyed the Tangut fortresses. The Tangut emperor escaped the Mongols and fled to Xiliang. Chinggis Khan continued to Zhongxing, the capital, and began a siege of the city. During the process, he sustained injuries while hunting, falling from his horse after qulan (wild asses) startled it.

38 Ruth Dunnell, ‘The fall of the Xia Empire: Sino–Steppe relations in the late 12th – early 13th centuries’ in Rulers from the Steppes: State Formation on the Eurasian Periphery, ed. Gary Seaman and Daniel Marks (Los Angeles: Ethnographic Press, 1991), p. 174. 39 Dunnell, ‘Hsi Hsia’, 211–13; Dunnell, ‘Fall of the Xia Empire’, 177.

508

Mongol Genocides of the Thirteenth Century

Despite his injuries and the urgings of his commander to return to Mongolia, Chinggis Khan insisted on continuing the siege and destroying Zhongxing. He gave orders to keep his death secret and continue the siege even if he died. Even when Emperor Xian (r. 1226–7) came from Zhongxing to offer his submission, he was not admitted to Chinggis Khan’s tent. Undoubtedly, the Mongol ruler did not want to give the Tangut king an opportunity to see his illness and possibly renege on his submission or give the Tangut hope. When Chinggis Khan died in 1227, the siege continued. The Mongols soon breached the city’s defences and followed Chinggis Khan’s last orders to destroy the city and massacre the population, although Chahan restrained the Mongols from conducting a complete massacre.40 With the destruction of Zhongxing, the Mongols erased Xi Xia from the map and the region formally became part of the Mongol Empire. This has been described as a genocide.41 Yet, the Tangut survived the destruction of Xi Xia. Even into the fourteenth century, edicts in the region were issued not only in Mongolian and Chinese but also in Tangut, which had a written language based on Chinese characters even though the spoken language is related to the Tibeto-Burmese linguistic family. Furthermore, as mentioned above, Tangut officers served with the Mongols during the destruction of Xi Xia. Chahan personally prevented the complete massacre of the population at Zhongxing as well as at other cities. In the case of Ganzhou, all but thirty-six were spared despite resisting the Mongols.42 So did a genocide occur? As with other instances of genocide, total annihilation is not necessary to qualify it as such. Certainly, a large-scale massacre took place at Zhongxing and, as with the Tatars and many others, the royal family was eradicated. The Mongols concluded that the royal family of Xi Xia would never truly accept Mongol rule and thus ceased to serve a useful purpose. As long as they were willing to remain subordinate to Chinggisid rule, their position was secure. Failure to do so, however, meant the ruling family was no longer needed. Thus, eliticide once more occurred. On other occasions, the Mongols often replaced a recalcitrant vassal with a relative. In this instance, they did not. It must be remembered that although Xi Xia had a substantial nomadic population, it was the first realm with 40 Dunnell, ‘Hsi Hsia’, 213. 41 See e.g. René Grousset, The Empire of the Steppes: A History of Central Asia (New Brunswick: Rutgers University Press, 1970), p. 248: ‘The fields were covered with human bones . . . [Of] Ningsia, the enemy capital . . . in accordance with the Conqueror’s last commands, the whole population was slain’. 42 Dunnell, ‘Hsi Hsia’, 211–13.

509

timothy may

a significant sedentary population that the Mongols conquered. While they initially had no interest in ruling it directly and were content to extract tribute, which included not only wealth in terms of silk, gold and silver but also falcons, camels and other livestock and goods, in the almost two decades since Xi Xia’s initial submission the Mongols had now developed a broader world view. Furthermore, they also came to understand the sedentary perspective of rule. From the nomadic perspective, authority and rule was based on control of populations. Just as a nomad’s wealth was based on ownership of livestock, a nomadic khan’s rule was based on the people under his rule. Land was important for pasture and sentimental attachment to one’s homeland, but nomads could not afford to place sentiment above necessity. If the pastures failed, then the herds and flocks suffered and one must find new ones. While good pastures were worth protecting, one could always find new ones. Not being mobile, sedentary societies have an opposite view. While people remained an important element of a state, the importance of the land rises in significance. Borders are defined and what exists on the other side becomes the Other. Ownership of land often defined power, and authority and wealth came from those who owned the land. Thus, the Tangut royalty could view Xi Xia only as their kingdom and not as part of the Mongol Empire. While they had previously paid tribute to the Mongols and before that, at different points, to the Jin and Song empires, they always remained independent because they ruled their own land and retained sovereignty over their affairs and institutions.43 With the growing demands by the Mongols for troops and other resources, Xi Xia began to lose its autonomy even without Mongol occupation. With these increasingly irreconcilable views, the likelihood of continued Tangut obeisance dimmed and thus, in the Mongol mind, necessitated their destruction. The sources have apocryphally attributed to the Mongols the view that they would rule the world by the Will of Heaven, but this idea did not exist at the time of Chinggis Khan. Undoubtedly, his actions against the Jin and Khwa¯razmian empires as well as the destruction of Xi Xia fostered the concept, which most likely arose during the reign of his successor, Ögödei Khan (or Qa’an).44 Thus, we cannot assert that the Mongols destroyed the Tangut out of a particular ideology. Still, we must return later to the question of whether or not the Mongols committed a genocide in Xi Xia.

43 Dunnell, ‘Fall of the Xia Empire’, 161.

44 May, Mongol Empire, p. 97.

510

Mongol Genocides of the Thirteenth Century

Nizari Ismailis Our final case study involves the Nizari Ismailis, a branch of Sevener Shi’i Islam.45 This group often enters Western literature as the Assassins, and indeed, it is from this group that the word ‘assassin’ entered the English language. Finding that they could not compete on the battlefield against the Seljuq Empire or the Khwa¯razmian Empire (the pre-eminent military powers in Iran in the eleventh and twelfth centuries), the Nizari Ismailis used assassination and mountain fortresses to maintain their independence. The term ‘assassin’ was derived from hashashiin, or ‘hash-eater’, as it was believed that their imam used marijuana to induce loyalty and train his devoted assassins.46 Prior to the reign of Möngke Khan (r. 1251–9), the Mongols maintained either an alliance or a client relationship with the Isma¯‛ı¯lı¯s, largely based on mutual hostility towards the Khwa¯razmsha¯hs.47 Although the Mongols had ravaged much of eastern Iran during their invasion of the Khwa¯razmian Empire in 1219, they did not attack Ismaili possessions in Quhista¯n or elsewhere. Indeed, the strongholds of the Nizari Ismailis even became havens for refugees as the Mongol storm raged outside its borders. This is likely due to the appearance of Nizari Ismaili envoys in Chinggis Khan’s camp in Balkh or Ta¯liqa¯n in Afghanistan in 1221, probably offering submission.48 In 1228, ˙ Badr al-Dı¯n, an Ismaili, visited the Mongol court. Then in 1230, the Mongol general Chormaqan again invaded Iran. The exact relationship with the Nizaris remains vague, but nonetheless during this period the Mongols attacked neither Quhista¯n nor the Ismaili possessions around Alamu¯t, again suggesting that the Nizaris maintained their tributary status.49 Chormaqan 45 The best introduction is Farhad Daftary, The Isma¯ʿı¯lı¯s: Their History and Doctrines, 2nd ed. (Cambridge University Press, 2007). 46 Ibid., pp. 19–24; Marshall Hodgson, ‘The Isma¯ʿı¯lı¯ state’ in The Cambridge History of Iran, vol. V : The Saljuq and Mongol Periods, ed. J. A. Boyle (Cambridge University Press, 1968), p. 443. 47 Muhammad ibn Ahmad Nasa¯wı¯, Sirah al-Sutltan Jala¯l al-Dı¯n Makubirtı¯ (Cairo: Dar alFikr˙Arabı¯, 1953), p.˙255; ‘Ala-ad-Din ‘Ata- Malik˙ Juvaini, Genghis Khan: The History of the World Conqueror, trans. J. A. Boyle (Seattle: University of Washington Press, 1997), p. 703; ʿ Ala’ al-Dı¯n Ata Malik ibn Muhammad Juwaynı¯, Ta’rı¯kh-i-Jaha¯n-Gusha, vol. I I I , ed. ˙ Brill, 1937), p. 248; Ibn al-Athir, Al-Kamil fı¯ al-Tarı¯kh Mı¯rza¯ Muhammad˙Qazvı¯nı¯ (Leiden: (Beirut: Da˙¯ r Sadr, 1979), X I I , p. 495; Bernard Lewis, The Assassins: A Radical Sect in Islam (Oxford University Press, 1987), p. 86; Marshall Hodgson, The Order of the Assassins (The Hague: Mouton, 1953), p. 252. 48 Daftary, Isma¯ʿı¯lı¯s, p. 382. 49 Minha¯j Sira¯j Ju¯zja¯nı¯, Tabaqa¯t-i-Na¯sirı¯, 2 vols., trans. H. G. Raverty (New Delhi: Oriental ˙ (hereafter Ju¯za¯nı¯/Raverty); Minha¯j Sira¯j Ju¯zja¯nı¯, Book Reprint Corp.,˙ 1970), p. 1197 Tabaqa¯t-i-Na¯sirı¯ (Lahore: Markazi Urdu Bord, 1975), I I ,p. 313; Lewis, Assassins, p. 90; C. ˙ ˙ Edmund Bosworth, ‘The Isma‘ilis of Quhistân and the Maliks of Nîmrûz or Sîstân’ in

511

timothy may

may have even benefitted from Ismaili intelligence as his troops hunted Jala¯l al-Dı¯n, the last Khwa¯razmsha¯h.50 Their amiable relationship decidedly ended sometime after 1240, when Chormaqan became incapacitated due to an illness, though the roots of discord may have come earlier. After his illness, Chormaqan was replaced by one of his lieutenants, Baiju. The event that triggered the conflict was the assassination of Chaghadai Qorchi, one of Baiju’s lieutenants.51 Although the assassination of a Mongol general occurred in the 1240s, the Nizari Ismailis may have reconsidered their relationship with the Mongols earlier. In 1238, a Nizari ambassador arrived in the courts of the kings of England and France, telling them of the Mongols and the threat they posed to the world. This embassy was conducted in co-operation with the Abbasid Caliphate, which had become equally concerned about the growing Mongol threat.52 Previously, both the Abbasids and the Nizari Ismailis had viewed the Mongols as a better option than the Khwa¯razmians, who had threatened both of them. In time, though, it became apparent that the Mongols, having destroyed Khwa¯razm, had no intention of leaving the Middle East. The fact that these two rivals, not only for territory but also in religious interpretation and ideology, had now found common ground on which to approach the monarchs of Europe is a clear indication of their concern. The Nizari ambassador warned them that if Islam failed, then so too would Christendom. Neither the French nor the English king, however, offered assistance. Indeed, in England, the bishop of Winchester, Peter de Roches, took delight in the idea of the Mongols and Muslims annihilating each other, so that then Christian armies could come in and mop up the remnants.53 Despite attempts to broker a European alliance as well as issues with the Mongol commanders in Iran, the Nizari Ismailis still sought amiable relations for the time being. In 1246, when Güyük succeeded his father Ögödei (r. 1229– 41) as qa’an (khaghan, emperor), visiting Nizari Ismaili ambassadors accompanied those of Caliph al Mustaʿsim (r. 1242–58) and other Muslim leaders. ˙ While Güyük welcomed the others, he treated the Nizari Ismaili envoys

50 51 52 53

Medieval Isma‘ili History and Thought, ed. F. Daftary (Cambridge University Press, 1996), p. 224. Timothy May, ‘A Mongol–Isma¯ʿı¯lı¯ alliance?: thoughts on the Mongols and Assassins’, Journal of the Royal Asiatic Society 14:3 (2004), 238–9. RD/Karı¯mı¯, p. 56; RD/Thackston, p. 42; Juwaynı¯, Ta’rı¯kh-i-Jaha¯n-Gusha, I I I, p. 277; Juvaini, Genghis Khan, p. 724. The exact date of his death is uncertain. Daftary, Isma¯ʿı¯lı¯s, p. 388. Matthew Paris, English History, vol. I, trans. J. A. Giles (New York: AMS Press, 1968), p. 131.

512

Mongol Genocides of the Thirteenth Century

Shiha¯b al-Dı¯n and Shams al-Dı¯n with contempt.54 Deciding to deal with the Nizari Ismailis, Güyük replaced Baiju with his own commander, Eljigidei. Eljigidei received explicit orders to destroy the Ismailis, although there is no evidence that Eljigidei executed these orders. Güyük died in 1248, which likely pre-empted any major operations. Tensions did not dissipate after Güyük’s death in 1248. The next Mongol ruler, Möngke, took the qa’an’s throne in 1251. Möngke (r. 1251–9) restored Baiju as the military governor in the region. Baiju also reported to Möngke that the Mongols’ two most ‘obstinate enemies were the Caliph and the Ismailis’.55 Others, such as the qa¯d¯ı of Qazvı¯n, Shams al-Dı¯n, also urged Möngke to crush ˙ ¯n appeared before Möngke wearing a chainmail shirt the Isma¯ʿı¯lı¯s. Shams al-Dı beneath his robes. He explained to Möngke that he wore it as a precaution against the daggers of the Ismailis and that the entire region lived in fear of their assassins and that their religion was contrary to that of Muslims and Christians, as well as the Mongols. He finished by saying that if the Mongols ever lost power, the Ismailis would surely replace them.56 The Mongols had additional reasons for wishing to end any friendly relations with the Nizari Ismailis. William of Rubruck, a Franciscan friar who ventured to the Mongol court on behalf of King Louis IX of France, reported that 400 Assassins had entered the empire in an attempt to assassinate Möngke.57 Due to this, the Mongols carefully investigated any foreigners who arrived at Karakorum in addition to rejecting the Ismailis’ envoys. As a response to the alleged attempt, Möngke ordered his brother Hülegü to ‘kill them all’.58 Although Hülegü received his orders, it took several years for him to assemble his massive army and march to Iran. Indeed, he did not cross the Amu Darya river until 1256. Despite, Hülegü’s leisurely arrival, which involved not only assembling an army but securing its logistics, the Mongols did not delay their operations as they had armies already stationed in the Middle East.59 To augment these, 54 Daftary, Isma¯ʿı¯lı¯s, p. 388. 55 Lewis, Assassins, p. 91. 56 Ju¯zja¯nı¯/Raverty, p. 1189. 57 Guillelmus de Rubruc, ‘Itinerarium Willelmi de Rubruc’ in Sinica Franciscana: Itinera et Relationes Fratrum Minorum Saeculi XIII et XIV, ed. P. Anastasius van den Wyngaert (Florence: Apud Collegium S. Bonaventurae, 1929), p. 286 (hereafter Rubruc); William of Rubruck, ‘The journey of William of Rubruck’, translated by a nun of Stanbrook Abbey, in The Mongol Mission: Narratives and Letters of the Franciscan Missionaries in Mongolia and China in the Thirteenth and Fourteenth Centuries, ed. C. Dawson (London: Sheed & Ward, 1955), p. 184 (hereafter Rubruck/Dawson); William of Rubruck, The Mission of Friar William of Rubruck: His Journey to the Court of the Great Khan Mongke, 1253–1255, trans. Peter Jackson (London: Hakluyt Society, 1990), p. 222 (hereafter Rubruck/Jackson). 58 Rubruc, 287; Rubruck/Dawson, 185; Rubruck/Jackson, p. 222. 59 See John Masson Smith Jr, ‘Hülegü moves west: high living and heartbreak on the road to Baghdad’ in Beyond the Legacy of Genghis Khan, ed. Linda Komaroff (Leiden: Brill, 2006), pp. 111–35.

513

timothy may

Hülegü also sent his vanguard force, commanded by Ket-Buqa Noyan (d. 1260). Ket-Buqa attacked the regions of Ru¯dba¯r and Quhista¯n in 1252. He attacked Girdku¯h as well, but failed to capture it. Leaving a lieutenant, Büri, to continue the siege, Ket-Buqa then shifted his attention to raiding Quhista¯n. The Nizari Ismailis actively resisted. Girdku¯h received reinforcements from Alamu¯t, their primary stronghold. Still, there was little they could do to counter the mobility of the Mongols and thus the Nizari Ismailis mainly resisted from behind their mountain fortresses, a successful strategy in previous wars against the Seljuqs and the Khwa¯razmians. Yet, the Nizari Ismailis may have been hampered by internal disagreements. While their imam, Ala al-Dı¯n Muhammad III, sought ˙ to resist the Mongols, many of his advisors as well as his son, Rukn al-Dı¯n Khwursha¯h, believed reconciliation with the Mongols was possible. If nothing else, they sought to mitigate the bloodshed. While their fortresses could resist, villages burned. As the Mongol attacks continued and word of Hülegü’s approach arrived, a coup took place and brought Khwursha¯h to power.60 In 1256, Rukn al-Dı¯n Khwursha¯h, now the Ismaili leader, wrote to Yasa’ur, the Mongol commander in the ‘Ira¯q ‘Ajamı¯ region, stating that he wanted to submit. Yasa’ur instructed him to go before Hülegü in Khura¯sa¯n. Meanwhile, the process of negotiations did not stop the Mongol attacks. Yasa’ur invaded Ru¯dba¯r. Again, it appeared to be more constant raiding rather than a sustained attempt at conquest. If the Mongols met too much resistance, they simply went elsewhere to ravage the countryside.61 Ket-Buqa also took advantage of Khwursha¯h’s negotiations in 1256. He overran most of Quhista¯n. As had happened since the Mongols irrupted from Mongolia, he killed most of the population that resisted, excluding the artisans. Meanwhile, most of the fortresses continued to hold out, even after orders from Khwursha¯h instructed them to surrender. Khwursha¯h did agree to demolish several fortresses in Ru¯dba¯r, but when the Isma¯ʿı¯lı¯s delayed the demolition, Hülegü sent Mongol observers to ensure that his decree was carried out. Despite their negotiations, Khwursha¯h never went to Hülegü, sending proxies instead.62 Eventually Hülegü promised him complete protection from retribution if he would destroy all of his fortresses and come in person. As an act of good faith, Hülegü also ordered Yasa’ur’s army to cease its operations in Ru¯dba¯r and return to ‘Ira¯q ‘Ajamı¯.63 When Khwursha¯h did not 60 Qutb al-Dı¯n Shı¯ra¯zı¯, ‘Akhba¯r-i Moghu¯la¯n’ in The Mongols in Iran: Qutb al-Dı¯n’s ˙ ¯ r-i Moghu¯la¯n, trans. and ed. George Lane (New York: Routledge, 2018), ˙ Akhba p. 51; Daftary, Isma¯ʿı¯lı¯s, pp. 391–2. 61 Hodgson, Order of Assassins, p. 266. 62 Shı¯ra¯zı¯, ‘Akhba¯r-i Moghu¯la¯n’, p. 52. 63 Hodgson, Order of Assassins, p. 266.

514

Mongol Genocides of the Thirteenth Century

fulfil his demands, Hülegü began to execute the Ismaili notables that fell into his hands as well as intensifying his attacks on Ismaili strongholds. Alamu¯t surrendered after a brief siege, although the great fortresses of Lammasar and Girdku¯h continued to resist (Figure 19.2). Lammasar eventually fell in 1257, but after Hülegü bypassed it. He isolated the fortress and left a force to besiege it. After Khwursha¯h formally submitted, the Mongols used him to secure the surrender of numerous other fortresses. He then sent Khwursha¯h to Möngke Qa’an in Mongolia. While Khwursha¯h travelled under escort to Mongolia, Hülegü had most members of the leading families, including Khwursha¯h’s family, executed or sold into slavery to ensure that the Nizari Ismailis would remain leaderless.64 The population was not spared either. Whereas the Mongols typically massacred those who resisted, once peace was established they did not molest the population. With the submission of the Khwursha¯h, however, the Mongols assembled the population and slaughtered them indiscriminately in Quhista¯n, including those who had submitted.65 Khwursha¯h reached Mongolia, but Möngke did not deign to see him and sent him back to Hülegü, or that is what Khwursha¯h believed was happening. In western Mongolia, his escort took Khwursha¯h and his retinue off the road and killed them all.66 Whereas Shı¯ra¯zı¯, who wrote of the events decades later, left a terse and direct account of the destruction of the Nizari Ismailis, Juwaynı¯, who witnessed the events, took great delight in writing about the destruction of the Nizari Ismailis as he viewed it as eradicating a plague. To be sure, Shı¯ra¯zı¯’s account is also hostile, but one also detects a hint of wistful pity for them. For Juwaynı¯, who accompanied the Mongols as an official in Hülegü’s army and eventually served as governor of Baghdad, the destruction of the heretical Ismailis had become a raison d’être. Of course, his chronicle ends before the Mongols destroy the Abbasid Caliphate and sack Baghdad in 1258, but it is clear that he views Hülegü’s actions against the Nizari Ismailis as just. Yet one cannot criticise Juwaynı¯ too much. He is correct in that Hülegü had explicit orders to destroy the Nizari Ismailis, which coincided with Juwaynı¯’s own rationalisation of the Mongols as an instrument of God’s will.67 Whereas the Tatars and Tangut were isolated cases of depictions of genocide in the sources, the eradication of the Nizari Ismailis in Iran was not. 64 Ibid., p. 269; Daftary, Isma¯ʿı¯lı¯s,p. 397. 65 Daftary, Isma¯ʿı¯lı¯s, p. 397. 66 Shı¯ra¯zı¯, ‘Akhba¯r-i Moghu¯la¯n’, pp. 52–3. 67 See Timothy May, ‘The Mongols as the scourge of God in the Islamic world’ in Violence in Islamic Thought from the Mongols to European Imperialism, ed. Robert Gleaves and István T. Kristó-Nagy (Edinburgh University Press, 2018), pp. 32–57.

515

timothy may

Figure 19.2 The siege of Alamu¯t, from Rashı¯d al-Dı¯n’s Jami’ al-Tawarikh. (Bibliothèque nationale de France)

516

Mongol Genocides of the Thirteenth Century

It truly was a genocide. As with the Tatars and Tangut, the Mongols executed the leaders of the Nizaris, but they also massacred most of the men. Furthermore, this was a military operation that took place over much of eastern Iran. While the Mongols conquered Xi Xia, their final attack on Zhongxing was not followed by wholesale massacres across Xi Xia. Once a city submitted, it was spared. Only resistance or rebellion instigated Mongol reprisals of massacres. Zhongxing was not only was the last stronghold of the Tangut ruling family, but it was also the site of Chinggis Khan’s death. One might consider the massacre not only as a reprisal for the rebellion but also as a sacrifice to Chinggis Khan.68 The acts against the Nizari Ismailis were nothing less than the intended destruction of a specific group.

Analysis It is sometimes argued that genocide is a modern concept that cannot be applied to premodern societies.69 The term ‘genocide’ was not coined until 1943–4 and the 1948 United Nations Genocide Convention became international law in 1951.70 While the Armenian genocide and the Holocaust are the usual examples for studying genocide, and indeed the basis for the UN Genocide Convention, earlier examples exist. With their numerous massacres, the Mongol conquests of the thirteenth century often garner much attention. Scholars specialising in the history of genocide seem to be divided as to whether or not these should be considered genocide, although often it is due to the adherence to the idea that genocide is purely a modern concept.71 Based on the UN Genocide Convention definition, it does become questionable in part whether modern laws and moral values apply to the past. We have seen that in some cases the Mongols did exterminate in the sense of total extirpation or utter destruction, as well as in the older sense of ‘driving out by force’.72 Indeed, the argument that genocide is only a historically recent (twentieth-century) or even modern phenomenon falls apart upon close scrutiny. As Kiernan demonstrates, other terms and concepts with connotations similar to genocide existed well prior to the mid-twentieth century.73 68 Dunnell, ‘Hsi Hsia’, 213–14. 69 Kiernan, ‘Is “genocide” an anachronistic concept?’ 70 Ibid. 71 Daniel Feierstein, Genocide as Social Practice: Reorganizing Society under the Nazis and Argentina’s Military Juntas (New Brunswick: Rutgers University Press, 2014), pp. 11, 13, 36; Chalk and Jonassohn, History and Sociology of Genocide, pp. 24, 33, 94–113. 72 Kiernan, ‘Is “genocide” an anachronistic concept?’, 446. 73 Ibid., 442–3.

517

timothy may

Chalk and Jonassohn argued that should an archaeology of genocide ever be developed, ‘we may acquire more conclusive proof of what happened to the population of cities that have disappeared’.74 They are certainly correct, but not necessarily in the way they meant, which pertained to the act of genocide. In their discussion of this possibility, they connected it with the thirteenth-century Mongol devastation of Iran, more specifically (though not stated by the authors) Khura¯sa¯n, which the Persian chronicler Qazvı¯nı¯ said that ‘would not recover for 1,000 years’.75 While, indeed, pyramids of skulls were found, Peter Jackson has conclusively demonstrated that while a great deal of death and destruction occurred, nonetheless, it was far less than the contemporary historians recorded.76 That said, we should accept that it was a terrifying event, particularly in Khura¯sa¯n. Similarly, in regard to why certain groups disappeared, one needs to look no further than the process of tribal and confederation formations among the steppe nomads. The name of the group typically originated from the dominant group. It did not necessarily mean that another group had been eradicated. While that did happen, usually only the leading lineages disappeared as the others would simply be absorbed into the new confederation. Some older names lingered, but new identities also emerged. Although the Mongols employed massacres frequently, they were not genocidal on most occasions. Massacres typically served as a form of psychological warfare as well as an object lesson to others, which some would qualify as a genocidal massacre, as indicated below. While it can be argued that the Mongols indiscriminately massacred, it can also be argued that the they committed genocide. In our three case studies, the Mongols eliminated the Tatars in 1202, the kingdom of Xi Xia in 1227 and the Nizari Ismailis in 1256. Do any of these qualify as a genocide? The Mongols certainly engaged in what is now termed ‘genocidal massacre’, as termed by Leo Kuper. In this, one group engages in shorter, limited episodes of killing directed at a specific local or regional community targeted because of its membership in a larger group but at the hands of perpetrators who may not (or cannot be proven to) harbor an intent to destroy the group ‘as such’ (which the Convention requires). Genocidal massacres often serve as object lessons for other members of the group.77

74 Chalk and Jonassohn, History and Sociology of Genocide, p. 33. 75 See Peter Jackson, The Mongols and the Islamic World: From Conquest to Conversion (New Haven: Yale University Press, 2017), p. 155. 76 Ibid., pp. 169–73. 77 Kiernan, ‘Is “genocide” an anachronistic concept?’, 453n48.

518

Mongol Genocides of the Thirteenth Century

The Mongols practised a form of psychological warfare in which they would massacre a population of a city in order to induce others to submit without resistance. When the Mongols attacked, they typically sent an ultimatum, which boiled down to ‘surrender or die’. If their target surrendered, the Mongols would demand tribute and the destruction of their defences, but if they resisted, then destruction and a massacre were guaranteed. In most instances, not all were killed – artisans and women were selectively spared, and those who would be forced to work for the Mongols as a levy for sieges were spared.78 In these instances, the massacres were punitive as retribution for the cost of Mongol lives or other offences such as the killing of envoys. A complicating factor in studies of genocide is that most have focused on sedentary societies. The Mongols do receive attention, but the scholars who include them often approach their case with a sedentary-centric bias and fail to recognise that the culture of pastoral nomads was decidedly different in its implications for populations. This is not to say that nomads did not ever commit genocide, but rather that scholars often make inappropriate assumptions about pastoral nomads.79 While pastures were important, the nomads were mobile, not tied to particular tracts of land, and they typically sought to incorporate people into their tribes, bringing them along as they moved. Land use to them was fluid. One used pasture land, but one did not own a specific plot of land, and labour could be incorporated into the tribe. In assessing the roles of race, religion, gender and territorial ambition in the Mongols’ motivations for committing their genocidal massacres, race per se was not a factor. The Tatars were hereditary enemies but they were not of a different race. Additionally, the Mongols did not massacre all Tatars, but a specific group of them. The same could be said of the Tangut, as that massacre targeted one specific city. As for the Isma¯ʿı¯lı¯s, while they certainly were not Mongols or Turks, even the Ismailis were not targeted for their ethnicity. Some Iranian soldiers served with the Mongols. With both the Tangut and the Ismailis, the concept of the other came into play. The cultural differences between the Mongols and these two societies contributed to their irreconcilable differences. Religion factored only in an individual way. For the Tatars, it had no bearing on their situation. For the Tangut, the fact that the Tangut rulers viewed themselves as Bodhisattvas (emergent Buddhas) rulers may have 78 Timothy May, The Mongol Art of War (Barnsley, UK: Pen & Sword, 2016), pp. 79–80. 79 See Chalk and Jonassohn, History and Sociology of Genocide, p. 32.

519

timothy may

influenced their decisions to resist and reject Mongol rule. For the Nizaris, religion had everything to do with it. For the Mongols, the Nizari defiance violated the will of Heaven. For the Nizari Ismailis, their own ideas of God’s favour and support also meant that they had had to resist once the threat became clear. Furthermore, they could not be ruled by infidels. While other Muslims had found ways to accommodate the rule of the infidel over the Believers, the Nizaris’ doctrine made it difficult to reconcile. Additionally, the resistance continued in part due to their views on the virtue of martyrdom, making peaceful resolution less likely. Yet, as demonstrated by the coup that brought Rukn al-Dı¯n Khwursha¯h to power, others saw an accommodation, as so many Muslim intellectuals elsewhere had determined, to be possible. Furthermore, the Mongols were undoubtedly encouraged by Muslim advisers who viewed the Nizaris as a heretical plague in need of eradication. The role of gender in the Mongol use of massacre varied. In the case of the Tatars, men and boys were massacred. Women and children, including very young males, were spared. For the Tanguts, there is no indication that gender had any bearing on the massacre at Zhongxing. With the Nizari Ismailis, women and children were sold into slavery, and the men were killed. To be certain, many women also died in the process, particularly in the early phases of the campaign in which Ket-Buqa raided Quhista¯n indiscriminately, but the focus of the massacres was on the men and not just elites. This process of gendercide, however, also contributed the genocide in an effort to end the Nizari Ismaili identity. In all cases, the goal was to destroy not only the political institution but also any military capability as well. Futhermore, with the destruction of family units, whether by the physical death of men or the enslavement of women and children, the Mongol Empire strove for the cultural death of the Nizari Ismailis, thus removing the possibility of a revival. As Nadia Eboo Jamal has demonstrated, even the Mongols could not achieve the complete destruction of the Ismailis, and they continued to exist, albeit in a greatly truncated form.80 Of course, the Mongols had territorial ambitions. In 1202, the Tatars were the final obstacle to Temüjin’s domination of eastern Mongolia. In the case of the Tangut, Mongols had yet to embrace the will of Heaven (they would do so only after the death of Chinggis Khan). Again, land itself did not matter to the Mongols. They viewed the Tangut as subjects who should pay tribute. 80 Nadia Eboo Jamal, Surviving the Mongols: Niza¯rı¯ Quhista¯nı¯ and the Continuity of Ismaili Tradition in Persia (London: I. B. Tauris, 2002).

520

Mongol Genocides of the Thirteenth Century

When the Mongols first conquered the region and gained obeisance, they did not occupy Xi Xia, the first sedentary region they acquired. The Tangut had a different view; as the elite transitioned into a more sedentary society, they saw their kingdom in terms of territory; the Mongols were foreign interlopers even though they did not occupy it. Although the Tangut paid tribute, as they had to the Jin and Song empires, the fact that the Mongols increasingly made demands on the military institutions of Xi Xia meant that the Tangut viewed the Mongols as violating their sovereignty. As a Tangut emperor still sat on the throne, this violation became an encroachment on his authority and perhaps even a threat to the territorial integrity of Xi Xia. Realising that the Tangut would not and could not abandon this view, the Mongols destroyed the Tangut elite, which was identified with the territory. The Tangut as a people remained and were incorporated into the Mongol Empire; they were, however, not incorporated into the Yeke Monggol Ulus, the Mongol identity. The same could be said of the Nizari Ismailis, whereas the Tatars became part of the Mongols. Thus, the key part of the genocidal massacres of the Tangut and Ismailis was the conflicting perspectives of the nomadic and sedentary worlds. Yet, at the same time, one cannot attribute the massacres to mono-causal explanations. In the final analysis, on the basis of modern terminology, the Mongol actions against the the Tanguts at Zhongxing (and many others) were genocidal massacres, and their war against the Nizari Ismailis was unquestionably genocidal, executed from the start with a deliberate intent to eradicate them, not simply destroy the ruling elite, as with the Tangut and Tatars. The genocidal massacre at Zhongxing was in response to the resistance (the siege lasted six months) to the Mongols, and probably intensified as a result of Chinggis Khan’s death. Nonetheless, Chahan also saved many of the inhabitants from retribution, as indicated earlier. The Tatars, however, should be excluded. While the Tatar elites were eliminated and ceased to exist as an independent entity, the fact that the Tatar name continued and even became synonymous with that of the Mongols is clear evidence that their destruction was limited to eliticide. One must be careful with the term ‘genocide’, as scholars continue to refine and define it, or develop new terminology. One can make a mistaken case that every massacre constitutes genocide, yet in most instances a massacre is just a massacre. Horrible events do spiral out of control despite the intent or will of a commander. It often has nothing to do with racial or ethnic tensions, politics or religion, but anger that seeks an outlet. Indeed, the massacres committed by the Mongols after a city resisted, systematic as they were, often fit this pattern. While the term 521

timothy may

‘genocide’ certainly can be applied to premodern events, we must also be careful not to view all events through the same lens.

Bibliographic Note on Sources in English The Secret History of the Mongols is a crucial source for any study of the Mongols as it is one of the few primary sources from the Mongol perspective. While it chronicles the rise of Chinggis Khan, it also informs the reader of the Mongol view of the world – what was important to them. For the purposes of genocide, it discusses the Tatars and Xi Xia. It is available in several translations, although scholars tend to prefer Igor de Rachewiltz’s translation not only for its clarity but also for its copious notes. Another source is the Shengwu Qinzhenglu (Campaigns of the Holy Warrior). Although this is in Chinese, its Mongolian idiom demonstrates that the original manuscript was Mongolian. An English translation does not exist as of the time of writing, although it has been demonstrated by Christopher Atwood that Rashı¯d al-Dı¯n’s Jamiʿ al-Tawa¯rı¯kh (Compendium of Chronicles) copies large sections of it verbatim (Christopher P. Atwood, ‘Rashı¯d al-Dı¯n’s Ghazanid chronicle and its Mongolian sources’ in New Approaches to Ilkhanid History, edited by Timothy May, Dashdondog Bayarsaikhan and Christopher P. Atwood (Leiden: Brill, 2021)). Rashı¯d al-Dı¯n also relied heavily upon Juwaynı¯’s Tarikh-i Jahan-i Gusha (History of the World Conqueror). Both he and Juwaynı¯ worked for the Mongols and provide a Persian perspective. Rashı¯d al-Dı¯n covers all three of the events described here, while Juwaynı¯ discusses the events in Xi Xia and concludes his work with a detailed account of the assault on the Nizaris. Other Persians sources such as the Akhbar-i Moghulan (Mongol News) and Ju¯zja¯nı¯’s Tabiqat-i Nasiri add additional detail to the attacks on the Nizaris. Most of the Islamic sources are hostile to the Nizaris on theological grounds, so the reader must be wary. One may find some additional information regarding them in The English History by Matthew Paris, a monk in England who included information about the Mongols and Assassins when their actions involved western Europe. While there are a number of Chinese sources relating to these events, particularly Xi Xia and the Tatars, most are not in English, although I have used unpublished translations of two Tangut generals who served in the Mongol military through the courtesy of Paul D. Buell. The Syriac monk Bar Hebraeus also provides an account of the destruction of the Nizaris, although it follows largely the narrative of Juwaynı¯.

522

20

Viêṭ Nam and the Genocide of Champa, 1470–1509 george dutton

Between 1470 and the first decade of the sixteenth century Vietnamese armies carried out a series of attacks upon the neighbouring realm of Champa, killing tens of thousands of Cham and irreversibly eroding fundamental elements of Cham culture and society. In the aftermath of their attacks, the Vietnamese also laid claim to a significant portion of Cham territory, which was permanently integrated into the Vietnamese realm, crippling the Cham polity. Although there had been a long pattern of aggression between the two kingdoms, the fifteenth-century Vietnamese invasion represented a substantial breach in what had been an even longer dynamic of diplomatic and cultural exchange between the two states. These two realms had long interacted, both along their common border and at their respective capitals. In the political arena, Viets and Cham had frequently crossed their mutual border, at times as soldiers but also as envoys, refugees or exiles. Vietnamese chronicles describe frequent tribute missions from the Cham, offering elephants and other gifts as part of the process of diplomatic relations. There were also cultural encounters that shaped each side. Cham cultural forms, both in performance and the decorative arts flowed into the Vietnamese territories. Furthermore, the earliest manifestations of Buddhism to appear in the Vietnamese kingdom had initially passed through Champa before reaching the Red River region. The Vietnamese military campaigns of the later fifteenth century thus represented, in some respects, a definitive repudiation of this cultural sharing and an end to the long diplomatic relationship. While the Cham peoples and their polity survived after these attacks, 1471 marked the beginnings of a political, demographic and cultural decline from which they would never recover. This chapter will, to the extent possible, describe the nature of the genocide that was perpetrated against the Cham. However, much of the discussion will be contextual rather than explicit

523

george dutton

because the amount of surviving data regarding this event is extremely limited. Furthermore, the evidence that does survive is that preserved by the perpetrators, the Vietnamese. These sources are primarily in the form of official chronicles compiled by court historians, with some subsidiary, and possibly contemporaneous, accounts that either reproduced or, at best, paralleled the official histories. With respect to specifically genocidal actions – the killing of Cham peoples – the records are terse, including very few statistics and only a general reference to the manner of killing (beheading) and some description of the political outcome of the Vietnamese aggression. Consequently, any discussion of a genocide is, necessarily, inferential rather than unequivocal, and is almost entirely lacking in external corroboration. What is clear, though, is that the Vietnamese waged an aggressive and highly destructive military campaign, and that the number of Cham killed in the course of the invasion was substantial.

The Perpetrators The Vietnamese realm was one that had coalesced as a political state over the first millennium of the common era. During this period it was, to varying degrees, under direct Chinese administrative and military control, an experience that profoundly shaped the Vietnamese world. While the Vietnamese people maintained their essential sociocultural and linguistic distinctiveness, their elites felt the influence of Chinese cultural and political norms. Thus, when the Vietnamese state shook off the last vestiges of Chinese colonial control in the later tenth century, it bore significant traces of this extended encounter with the Chinese. These traces could be measured in Vietnamese political structures, in their use of administrative written languages, their religious and philosophical landscape, and, more broadly, their world view. While Vietnamese rulers continued to pay periodic tribute to successive Chinese dynasties, they regarded themselves not as mere supplicants but as political and cultural equals. Chinese cultural influences upon the Vietnamese were periodically refreshed through ongoing contact, particularly in the exchange of embassies and cultural artefacts (primarily written materials), but were later substantially reinvigorated during a two-decade-long Chinese military occupation between 1406 and 1427. During this period, in which a vigorous young Ming dynasty sought to impose cultural order upon the reabsorbed Vietnamese territory, Vietnamese elite interest in and commitment to Confucian political ideals and cultural norms was significantly 524

Viêṭ Nam and the Genocide of Champa, 1470–1509

Map 20.1 The Vietnamese attack on Champa, 1471. (Cartography by George Dutton)

525

george dutton

strengthened.1 The Ming reconquest of Vietnam only ended with the emergence of a resistance force led by an aristocrat turned general, Lê Lơ i. Lê Lơ i, aided by his advisor and strategist Nguye˜̂ n Trãi, slowly ˙ ˙ assembled an army, one increasingly equipped with captured Chinese weapons, and in 1427 he finally succeeded in ousting the Chinese. Victory accomplished, a new dynasty, the Lê, was proclaimed. Lê Lơ i ˙ and his successors set about building a new, stronger and more centralised state.

The Victims The Cham peoples, the victims of the Vietnamese attacks, are a culturally and linguistically distinct population, related to the Malay and other Austronesian peoples of Southeast Asia. They originally arrived on the Indochinese peninsula via sea, possibly from the Philippine islands, marking them as an oceanoriented people.2 With their arrival and expansion up and down Southeast Asia’s eastern mainland coastal region, the Cham language became the predominant linguistic medium across the coastal area of what is today central Vietnam. In this setting, the Cham found themselves sandwiched between the powerful, and significantly more populous, Vietnamese realm to their north and the Khmer kingdom to their south and west. While Vietnamese and Khmer power varied substantially over the centuries, each, at times, represented a significant threat to the Cham polities. Meanwhile, to their west lay upland peoples with close linguistic and cultural affinities to the Cham. These peoples regularly engaged in trade, which was sometimes undergirded by marriage alliances. Such engagements reflected a further complex reality of the Cham states as being comfortably polyethnic in ways that were not always true of the Vietnamese. In addition to being linguistically distinct from their Vietnamese neighbours to the north, the Cham were also different in other important cultural measures. Although both Vietnamese and Cham had historically practised Buddhism, that of the Cham was distinctive for being more heavily Indian influenced, like that of their Khmer neighbours and unlike the more Siniticly imprinted Buddhism found among the Vietnamese. This was reflected in religious iconography, in

1 Details can be found in J. Whitmore, ‘The development of Le government in fifteenthcentury Vietnam’ (PhD thesis, Cornell University, 1968). 2 M. Vickery, ‘A short history of Champa’ in Champa and the Archaeology of My Son (Vietnam), ed. A. Hardy et al. (Singapore: NUS Press, 2009), p. 45.

526

Viêṭ Nam and the Genocide of Champa, 1470–1509

ritual practices and in pilgrimage routes, which tended to be oriented more to the west and to India rather than north and to China. As a polity, Champa was considerably less centralised than its Vietnamese neighbours. Historians have come to recognise that the Cham realms did not constitute a monolithic kingdom with a traceable lineage of rulers, as in Vietnam. Rather, and partly dictated by geographical considerations, Champa consisted of contending polities based in a series of fertile east–westoriented river valleys. The ascent to supremacy of one such ruler, with power typically derived from a combination of spiritual and economic authority, would result in a temporary geospatial shift of the centre of Cham influence to that ruler’s capital. Consequently, the Cham realm was adaptable to changes and its political centre would shift in response to pressures from external as well as internal forces. This Cham centre moved repeatedly over the centuries between sites that included Panduranga, Indrapura, Vijaya and Kauthara. Champa’s agglomerated strength varied considerably over time as well, reflecting the relative power of the rulers who emerged as temporary leaders and the larger circumstances of the strength of its Viet and Khmer neighbours. Another key feature of the Cham polities was their rather fragile economic base, with a relatively small population and constrained amounts of agricultural land. Thus, the Cham substantially relied on trading both up-river /down-river with their upland affines and with foreign vessels that called at the Cham coastal ports. Smaller-scale agriculture and trade, however, were insufficient to support the Cham economy and the ambitions of its rulers. Consequently, the Cham leaders also relied to a degree on occasional raidand-plunder activities in which their ships would engage in coastal piracy, or their armies would attack their Khmer and Viet neighbours.3 These military attacks were rarely designed to seize and hold territory, but rather to appropriate movable goods of value, including people for its populationpoor territories. The goods and labour that were captured propped up the Cham economy and enhanced the prestige of the leaders who conducted the raids. While a necessary part of the Cham economic structure, this plunder strategy had inevitable consequences, namely friction with neighbours that prompted periodic counter-attacks of varying intensity, including clashes with the Vietnamese. 3 J. Whitmore, ‘The last great king of classical southeast Asia: “Che Bong Nga” and fourteenth-century Champa’ in The Cham of Vietnam: History, Society and Art, ed. Tran Ky Phuong and Bruce M. Lockhart (Singapore: NUS Press, 2011), pp. 186ff.

527

george dutton

Historical Background The Lê dynasty’s brutal attacks against the Cham in 1471 must be understood in the context of the historical dynamic between these two realms, which had been rivals for centuries. The most relevant precursor to the fifteenth-century wars was the large-scale Viêt–Cham battles of the mid-fourteenth century. ̣ This was a period in which the two kingdoms saw an inversion of their relative historical strength. While the Vietnamese had historically been the stronger of the two powers, the middle of the fourteenth century saw a fading Vietnamese realm being confronted by a resurgent Cham kingdom. The fourteenth century saw a slow decline in Vietnamese power as the influence and political coherence of the Trà ̂ n dynasty (1225–1400) waned in the aftermath of its triumph in the wars against the Mongols.4 These wars had occupied the Trà ̂ n court for most of the second half of the thirteenth century, and while the Vietnamese had ultimately prevailed, a certain fatigue had set in from which the Trà ̂ n rulers never fully recovered. Just as the Trà ̂ n dynasty saw itself in decline, the Cham realm was in the ascendance. By the early 1300s the Vietnamese chronicles no longer record the regular arrival of Cham tribute missions, a clear indication that Champa’s rulers saw the Vietnamese as being in decline, and unable to enforce tribute requirements. Indeed, the rather more balanced nature of their relationship is reflected in the fact that in 1306 the Vietnamese agreed to send one of their royal princesses to marry the Cham ruler, in exchange for the Cham cession of two tracts of its territory. While the deal was cancelled after the Cham ruler’s unexpected death, it speaks to the closeness of the two states at this time, and the sense of relative political balance. Indeed, as John Whitmore notes, at this time ‘the Vietnamese did not regard the Cham as barbarians and inalterable enemies, but neighbors of a not too different cultural tradition’.5 While the Cham gained in strength over the course of the fourteenth century, the apogee of their might came with the rise to power, in 1368, of the militaristic ruler, Po Binasor (aka Che Bonguar), more commonly known by the name Vietnamese historians assigned him, Chè̂ Bô ́ng Nga.6 The Cham had already been carrying out periodic cross-border raids against Vietnamese territory in the 1360s, though the Vietnamese had been able to repel these. Likely in response to these attacks, in 1367 the Trà ̂ n sent an army to attack the northern Cham territories, but were pushed back. This offensive marked the 4 J. Whitmore, Vietnam, Ho Quy Ly, and the Ming (1371–1421) (New Haven: Yale Council on Southeast Asia Studies, 1985), pp. 2–7. 5 Ibid., p. 4. 6 Whitmore, ‘Last great king’, 185.

528

Viêṭ Nam and the Genocide of Champa, 1470–1509

last time that the Vietnamese military would challenge the Cham realm for the next four decades. The following year, 1368, the Cham felt confident enough in their strength to demand that the Vietnamese return a portion of territory that had been seized during an earlier conflict. This was the same year in which Po Binasor ascended to the Cham throne. The new ruler was strong and self-confident, and, emboldened by the current balance of power, had aspirations of military achievements. Furthermore, the Cham assiduously began to court China’s new rulers, the Ming dynasty, sending emissaries almost immediately after the Ming took the Chinese throne, hoping to secure their support or at least non-interference during the course of Cham military campaigns.7 Under Po Binasor’s leadership, the loosely knit Cham lands were again consolidated and strengthened. Po Binasor’s first attack against the Vietnamese took place in 1371, and its resounding success seemed to bear out the Cham’s assessment of the weakened condition of the Vietnamese kingdom. The Cham armies had not only crossed the border into Đai Viêṭ largely unchallenged, but had then pushed ˙ In an unprecedented attack, the Cham penetheir campaign northwards. trated the heart of the Red River region, and entered the Vietnamese capital Th a˘ ng Long (modern Hanoi), looting its palaces and laying waste to much of the city.8 This was arguably the most demoralising and humiliating Vietnamese defeat since that kingdom had gained its autonomy from the Chinese in the latter half of the tenth century. Moreover, the 1371 invasion was merely the first of ten separate attacks the Cham launched upon the Vietnamese state over the course of two decades. And yet, while the Cham could conduct such raids with some regularity, they did not have the capacity or the manpower to occupy the land they had so easily overrun, and would always withdraw to their own territory. This power dynamic, which so strongly favoured the Cham beginning in the early 1370s, was not altered until the last decade of the fourteenth century. The key development was the death in battle of Po Binasor in 1390, which dealt a crippling blow to Cham military confidence and political coherence. After decades of absorbing repeated Cham incursions, in 1396 the Vietnamese finally launched their own successful invasion of Champa. Then, from 1400 to 1403 Vietnamese armies carried out annual attacks against their southern rival, that of 1402 forcing the Cham to surrender the region known as 7 Ibid., p. 188. 8 Đai Viê ̣t Sử Ký Toan Thư (Complete Compiled Record of the History of Dai Viet) ˙ (DVSKTT), trans. Ngô Đu´̛ c Tho (Hanoi: Nhà Xuâ ́ t Bả n Khoa Hoc Xã Hô i, 1998), I I, ˙ ˙ ˙ p. 154.

529

george dutton

Amaravati (roughly covering the modern Vietnamese provinces of Quả ng Ngãi and Quả ng Nam). A 1403 Vietnamese attack using 200,000 troops penetrated deep into Cham territory, reaching the capital at Vijaya, though its siege ultimately failed.9 Not long thereafter, Ming armies invaded the Vietnamese state which they then occupied from 1407 to 1427 (during which period the Cham were able to regroup and rebuild while the Vietnamese were busy trying to oust their Chinese occupiers). The Vietnamese attacks upon their Cham neighbours that followed later in the fifteenth century were very much fuelled by circumstances that grew out of the Vietnamese responses to the Chinese occupation and the warfare that followed. These factors were demographic, military and philosophical. Demographically, the Vietnamese state faced the sobering reality that after seeing population stagnation and even decline during the fourteenth century, its population grew rapidly over the fifteenth century. According to some estimates, the population may have tripled in less than a century between 1417 and 1490, even as Vietnam’s territorial expanse had remained largely unchanged since before independence.10 Increasing shortages of land and the capacity to provide food for its populations meant that the Vietnamese faced a serious demographic crisis. Territorial expansion was the only viable response to this crisis, but this option was geographically curtailed because mountains rising to the west of the capital region made movement in that direction difficult, while the sea presented an absolute limitation to the east. This left only southward movement towards Champa as a demographic safety valve. Growing clashes between the two sides were almost inevitable. The second key development fuelling the crisis with Champa was the emergence of a powerful Vietnamese military. The Vietnamese state and its armies reached the apogee of their power in the second half of the fifteenth century. Since its founding in 1428, the Lê dynasty had quickly grown in strength and authority. It issued new legal codes, compiled histories, drew maps, created regulations, standardised weights and measures; in short, it was a regime with significant capacity for action, for exercising control and for mobilising populations in pursuit of tasks both civilian and military. It was also a militarily powerful state with a generation of battle-hardened soldiers in possession of a range of significant military technologies. Indeed, as 9 L. Sun, Chinese Military Technology and Dai Viet: c. 1390–1497, ARI Working Paper 11 (Singapore: Asia Research Institute, 2003), p. 20. 10 V. Lieberman, Strange Parallels: Southeast Asia in Global Context, c. 800–1830 (Cambridge University Press, 2003), I , p. 386.

530

Viêṭ Nam and the Genocide of Champa, 1470–1509

Laichen Sun has shown, much of the Vietnamese technological capacity had been acquired in the course of their battles with the Ming. The Vietnamese had captured a great many firearms of various descriptions, but they also reverse-engineered some of these to manufacture their own. Moreover, when the Chinese were finally defeated in 1427, the Lê generals allowed their defeated soldiers to retreat unmolested, but insisted that they leave their weapons behind.11 Consequently, the Lê state was both strongly centralised and in possession of a large, well-trained and well-equipped military. The only question that remained was what it would do with these capabilities, a question soon answered in the Cham campaigns, as the Lê court began to flex its muscles and test the potential of its newly amassed strength. Champa was, in essence, the first target of the new Vietnamese military might, and the first test of its capacity. At the same time, while demographic realities dictated the necessity of southward expansion, and new organisational and military techniques delivered the firepower, Confucian ideals provided the political and moral legitimation of such an action. The period in which the Vietnamese attack took place marked, at some level, the apogee of Vietnamese Confucianist fervour, one highlighted by a self-righteous commitment to virtue and cultural propriety. While none of these factors made a genocide inevitable, their combination enhanced the likelihood of a violent clash.

The 1470–1 Genocide and Its Aftermath Even before the 1471 campaign, tensions had been rising between the two sides. The early 1440s saw a series of Vietnamese attacks against the Cham, most of a punitive nature in retaliation for Cham incursions into Vietnamese territory. The last and largest of these attacks occurred in 1446, when a Vietnamese army said to have numbered 600,000 soldiers succeeded in capturing and sacking the Cham capital. The northern troops then seized the Cham ruler and more than 30,000 of his subjects, forcibly marching them as captives back to Vietnam.12 The 1446 attack, however, was not the key moment in fifteenth-century Viet–Cham relations. For that we need to look to the year 1460, which marked the moment at which each kingdom saw the emergence of a new 11 Sun, Chinese Military Technology, pp. 11–12. 12 DVSKTT, I I, p. 356; 61b. The size of the attacking force was likely a considerable exaggeration as it is significantly larger than any other historical citations for Vietnamese troop strength.

531

george dutton

ruler. In Champa, a man known as Ban-la Tra-toan ascended to the throne after his father had ousted the sitting Cham ruler. His rise to power saw a commensurate increase in the number of Cham protests to the Chinese about Vietnamese harassment and tributary gift demands. Ban-la Tra-toan regarded Vietnamese requests for tributary gifts as nothing more than crude extortion, and complained to the Ming, urging the Chinese court to rein in its nominally tributary state. For their part, the Vietnamese saw Ban-la Tra-toan as a menace whose armies were constantly harassing Vietnamese borderlands.13 Meanwhile, Vietnam’s new monarch was a man whom the history books would later name as Lê Thánh Tông (r. 1460–97). At thirtyseven years, the new emperor’s reign would be the second longest in Vietnam’s imperial era, surpassed only by the earlier Lý Nhân Tông’s more than half a century on the throne (1072–1127). The emperor ruled for his first decade using the reign name Quang Thuâ n, or ‘Glorious Harmony’ (1460– ˙ 70). But, in 1470, deciding that his realm had entered a new phase, the emperor announced that he was changing the name of his reign era to Hò ̂ ng Đu´̛ c, ‘Immense Virtue’, which reflected the emperor’s now more activist ambitions. It also underscored the emperor’s sense of his own charismatic power and capacity to bend people and even states to his will. Tensions between the two states rose steadily during the first decade of these two new monarchs’ respective reigns. The Vietnamese were the primary provocateurs in these clashes, both diplomatically and militarily, but the Cham were certainly also testing Vietnamese resolve and military capacity. At the same time, each made regular appeals and protests to the Ming court, though the Cham were the more vocal of the two. In 1460 they first complained about Vietnamese aggression, and later in that decade protested yet again at Vietnamese ‘extortion’ of not only elephants but now rhinoceroses as well.14 In 1469, the Cham launched a raid across the strategic Hả i Vân pass, which had, historically, represented a significant natural barrier between the Vietnamese and the bulk of the Cham populations. This attack had been the farthest thrust into Vietnamese territory since possibly the Po Binasor attacks of the 1380s. This was followed, in the autumn of 1470, by an even larger Cham attack, in which 100,000 soldiers reached the 13 Thiên Nam Tu´̛ Chí Lô Đò ̂ Thư (Book of the Four Route Maps of the South of Heaven), in ˙ Hong Duc Reign Atlas) (Saigon: Bô Quô c Gia Giáo Duc, 1962), Hò ̂ ng Đu´̛ c Ban̉ Đò ̂ (The ́ ˙ ˙ p. 100. 14 G. Wade, trans., Southeast Asia in the Ming Shi-lu, e-book (National University of Singapore Press, 2005), http://epress.nus.edu.sg/msl/reign/tian-shun/year-8-month3-day-7 (accessed 15 January 2020).

532

Viêṭ Nam and the Genocide of Champa, 1470–1509

Vietnamese outpost at Hoa Châu, near the modern city of Huê ́ . The Vietnamese quickly counter-attacked with 260,000 soldiers of their own, driving back the Cham forces.15 Although this Vietnamese attack featured impressive numbers of troops, it was merely the prelude to a massive punitive invasion that the Hò ̂ ng Đu´̛ c Emperor was already organising. The Vietnamese prepared for their campaign both by planning practical elements of the attack and by addressing its political and moral dimensions. At the practical level, the Hò ̂ ng Đu´̛ c Emperor began by recruiting his troops, drawing on eligible male villagers as young as twelve. He then carried out extensive supply requisitions. Thereafter, the emperor began drilling his troops, planning attack routes and preparing troop deployments. In addition, he collected intelligence while arranging for local experts to draw detailed maps of the terrain the Vietnamese would be crossing in the course of their campaign.16 Of these preparations, the mapping project was particularly significant, producing numerous route maps that survive to the present. This was not simply a logistical initiative but a political one as well, for in mapping the Cham territory the Vietnamese sought to transform and reimagine these lands. In the course of their cartographic project, Vietnamese map-makers and military commanders created campaign maps and notes that largely disregarded existing Cham nomenclature. Instead, they employed newly coined place names reflecting topographical attributes recognisable to Vietnamese military leaders and administrators.17 As they conducted this mapping project, it is clear that the Vietnamese reimagined the Cham landscape, overlaying on it their own administrative view. While not fully eradicating Cham place names or local understandings of them, this Vietnamese renaming process furthered their claims to this territory while conveniently dispensing with any former naming associations. Such actions might not necessarily suggest a calculated strategy of erasure as well as conquest and subjugation, but the net result was surely to marginalise and reduce the Cham’s imprint and presence upon the land. 15 DVSKTT, I I, pp. 441–2; 54b–55a. 16 For details of the preparations see ibid., I I, pp. 441–2; 54b–55a. See also Tây Nam Biên Tai Luc (Record of the Frontier Passes to the West and South), microfilm, HM 2132, pp. 17b–18a. 17 G. Dumontier, ‘Étude sur une portulan annamite du 15e siècle’, Bulletin de geographie historique et descriptive (1896), 142. J. Whitmore notes that this text is similar to the Binh Nam Chı̉ Chư ở ng Nhâ t Trình Đò ̂ (The Daily Journal and Map of the Command of the ˙ Pacification of the South), which he describes in ‘Vietnamese historical sources: for the reign of Le Thanh-tong (1460–1497)’, Journal of Asian Studies 29:2 (February 1970), 378.

533

george dutton

At the political level, the key consideration for the Vietnamese ruler was to ensure that the Chinese court could be persuaded to remain on the sidelines during the impending campaign. The Vietnamese were aware of the ongoing Cham public relations efforts in Beijing, and knew that they needed to present their own side of the story. In November of 1470, a pair of Vietnamese officials was sent to China, where they complained of Cham border harassment and incursions.18 Although neither the Ming Shi Lu nor the Đai Viê ̣t Sử Ký Toan Thư discusses the specific Ming response to the ˙ Vietnamese complaints, it is likely that the Chinese reiterated their longstanding expectation that the two sides would settle their differences without resorting to warfare, and that the Vietnamese in particular should be careful about overstepping their authority. In any case, Vietnamese plans were hardly contingent upon the Chinese court’s reply. The envoy’s complaints were merely part of the political groundwork for the impending attack. But, while planning the logistics of the campaign and making a case to their nominal Chinese overlords were each important elements of Vietnamese preparations, equally important, from the emperor’s perspective, was to present a compelling moral case to his domestic audience. This audience included not only civilians and court officials but also particularly the soldiers who were being recruited and trained for the impending invasion. To present his case, the Vietnamese emperor, like some of his predecessors, articulated his rationale in the form of a proclamation, an edict on the need to ‘Pacify the Chams’.19 Vietnamese rulers were sticklers for underscoring the legitimacy of their military actions, and the ‘Pacify the Chams’ edict is a classic example of this genre of political discourse, and one of the core texts relating to the Vietnamese genocide upon their southern neighbours. It is, equally importantly, a document fundamentally grounded in the new, highly Confucianoriented political landscape crafted by the Lê dynasty in the aftermath of the Ming occupation. The edict reflected a new Vietnamese world view in which the kind of tolerant cultural relativism that had shaped early Viet–Cham relations was now replaced by a high-handed sense of Vietnamese moral superiority to their southern neighbours. In the edict, the Hò ̂ ng Đu´̛ c Emperor emphasised the Vietnamese view that the fault for the breakdown in relations between the states lay squarely with the Cham. As such, the Cham were, themselves, responsible for having 18 DVSKTT, I I, p. 441. 19 A translation of this edict is found in G. Dutton et al., eds., Sources of Vietnamese Tradition (New York: Columbia University Press, 2012). The original is in the DVSKTT, I I, pp. 441–5; 55a–58a.

534

Viêṭ Nam and the Genocide of Champa, 1470–1509

provoked the Vietnamese invasion and its consequences. The edict described the Cham as perpetrators of ‘cruel crimes’. They were depicted as killers, as schemers, as cruel, immoral animals. Indeed, a notable element of the edict’s language was its routine representation of the Cham as subhuman beasts. The emperor noted that ‘like animals, they eat their fill and forget their moral debt’, and he declared that ‘in the south, we see a pig wallowing in the mud’.20 During the course of the edict, Cham are likened to rabbits, bees, foxes, silkworms, dogs, ants, crows, fish and pigs. In short, the text might be regarded as a classic, and literal, example of the dehumanising process necessary for and often a prelude to acts of genocide. At the same time, the edict also emphasised questions of political legitimacy and the need to punish those who, unlike the Vietnamese, did not respect proper royal succession. Thus, the acting Cham ruler was accused of having rebelled against his legitimate king and usurping the kingdom’s throne. Once again using animal imagery, the Hò ̂ ng Đu´̛ c Emperor described this political upheaval as ‘taking the stink of a fish, the father of a dog, the mother of a pig’ to ‘seize and kill their king, expelling the offspring of King Bo De and usurping his throne’. But it was not merely the fact of such internal political rebellion that drove Vietnamese actions. The emperor’s account suggested that, fundamentally, the Cham people were agents of disorder and had to be dealt with accordingly. ‘They dare to grow hearts that love chaos and act arbitrarily; it is precisely such crimes that deserve to be attacked and must be killed. Heroes hear and believe, gnash their teeth, and are furious. Those who are loyal understand the situation and, in their guts, are saddened.’21 Paradoxically, the edict, like many in this genre, also represented the Vietnamese invasion as a rescue mission. The emperor highlighted the oppression he claimed the Cham people were suffering: ‘Males and females alike serve and labor hard; officials apply cruel punishments . . . consequently, the people of Champa are heavily taxed and pitifully punished’.22 In short, this portion of the edict conveyed not the sense of a campaign to annihilate Champa because of its inherent cultural inferiority or because of its annoying military aggression, but one calculated to inflict punishment and moral rectification. In any case, the edict represents an important element in regarding the Vietnamese actions as genocidal, for its dehumanisation of the Cham and its view of the enemy as morally illegitimate both served to render the Vietnamese military campaign more than a mere answer to Cham

20 Dutton et al., Sources, p. 142.

21 Ibid., pp. 141–2.

535

22 Ibid., p. 141.

george dutton

incursions; it was predicated on the need to eliminate the moral and cultural rot the Vietnamese ruler had identified in Cham society. Having issued this proclamation to rally his troops and delegitimise Cham claims to moral or political standing, the Vietnamese army was ready to act. The Vietnamese launched their campaign in late November of 1470, beginning a two-month southward march. The first contingent, of 100,000 soldiers, left on 28 November, while the remaining 150,000 troops, led by the emperor himself, set off on their march on 8 December. Meanwhile, the Hò ̂ ng Đu´̛ c Emperor’s naval forces landed in Champa on 8 January 1471, a month after leaving Th a˘ ng Long. Having penetrated Cham territory, the Vietnamese forces took their time consolidating their positions in preparation for an assault on the Cham capital. For most of January, the emperor spread his forces across portions of the Cham realm, resupplying his troops and ordering local assistants to draw new campaign maps. Unnerved by the patient Vietnamese preparations, and perhaps hoping to use the element of surprise, the Cham king sent his younger brother to lead 5,000 troops in an attack upon the Vietnamese troop positions. The Cham forces struck on 26 February, but their surprise attack was no match for the hardened Vietnamese position. Forced to flee a counter-attack, Cham troops found Vietnamese soldiers blocking all potential escape routes. The peril of his position having been made clear, the Cham ruler, Tra-toan, tried to initiate surrender negotiations with the Vietnamese emperor. The Vietnamese leader, however, chose to ignore the request and instead began his inexorable advance towards the Cham capital of Vijaya.23 Vietnamese troops, their emperor leading the way, finally began their direct assault on the centres of Cham power in the middle of March. The attack began with a raid on the Cham port of Thi Nai near the modern city of ˙ ˙ Qui Nhơ n. In this attack the port’s citadel was destroyed and more than 100 Cham were reported beheaded. From Thi Nai, the emperor led his ground ˙ ˙ armies on the short journey to surround Vijaya. Having reached the citadel on 18 March, the Vietnamese forces encircled the city’s walls and began their siege. Unlike the 1403 campaign, in which Vijaya had successfully resisted a sustained Vietnamese siege, this time the city was no match for the northern army. Vietnamese specialists scaled the wall, breaching Cham defences, while their main troops poured into the fortress after either destroying the main gates or having had them opened by a Cham official. The northern soldiers overran the citadel and killed, by beheading, 40,000 23 Whitmore, ‘Development of Le government’, 211.

536

Viêṭ Nam and the Genocide of Champa, 1470–1509

people. Indeed, the Malay annal Sejarah Melayu adds that Vietnamese soldiers who entered the citadel ‘ran amok’, suggesting a violent and chaotic scene.24 The Vietnamese captured another 30,000 people, none more important than the kingdom’s ruler, Tra-toan. The Cham king became the Vietnamese prize captive, and plans were made to escort him to Th a˘ ng Long as the conquering army retreated. Noteworthy in these descriptions of the Vietnamese violence is their use of beheading (斬首) to kill their Cham enemy. Throughout the war, the primary sources all repeatedly speak of killing by beheading. This would appear not to be a generic substitute for ‘killing’ as there are other terms that appear elsewhere in the Vietnamese historical records describing the killing of enemy forces. Indeed, other references in the chronicles to military killing by beheading are exceedingly rare. Prior to the anti-Ming campaigns, beheading is almost always mentioned as an action carried out against a single individual, the chief exception being an earlier slaughter of 30,000 Cham in 1044. Thus, the act of beheading carried out by Vietnamese troops seems to be a deliberate act, one calculated as a form of ritual punishment meted out against the Cham populations collectively and also, as not merely enemies of the Vietnamese generically, specifically as Cham. This suggests the overtly punitive nature of the Vietnamese attacks against the Cham, and lends further credence to an argument for considering the Vietnamese actions as constituting genocide rather than killings incidental to early modern warfare. Underscoring the Vietnamese strategy of beheading their Cham rivals was their treatment of the captured Cham ruler, King Tra-toan. Captured alive, Tra-toan fell ill while being transported back to Đai Viêt.̣ At this point, whatever respectful treatment the Vietnamese had ˙extended to the ruler evaporated. Not long after taking ill, the ruler died, whereupon his Vietnamese captors immediately beheaded him and displayed his bloodied head on their boat as a trophy and, presumably, a gruesome warning to any who viewed it. With the death of the Cham ruler, Vietnamese atrocities against the Cham largely came to an end. There was still, to be sure, the matter of the 30,000 captives, some portion of whom was transported to Vietnamese territory to serve as virtual slave labour on farms near the northern capital. These captive Cham were ordered to take Vietnamese names and marry into local villages, essentially Vietnamising themselves to erase their Cham identities. This 24 Sejarah Melayu, The Malay Annals, translated from the Malay Language, trans. John Leyden (London: Longman, Hurst, Ress, Orme & Brown, 1821), p. 211.

537

george dutton

Figure 20.1 Cham Tower still standing at Vijaya (1). (Photo: George Dutton)

further suggests the Vietnamese sense of both their cultural superiority vis-àvis the Cham and the desirability of the Cham’s cultural transformation. This exile community fell victim to the final act of Vietnamese vituperation against the Cham, which occurred in 1509. It was triggered when one of the captives, a scion of the Cham royal family, fled back to his homeland in 1505 after the Vietnamese ruler was toppled in a coup. The political chaos in Th a˘ ng Long that had prompted his flight soon encouraged a large number of the enslaved Cham to flee southwards as well, in what amounted to a mass escape. The situation came to a head when, in 1509, a Vietnamese official reported plots for an uprising among the remaining Cham settlers. Incensed, the Vietnamese emperor responded with force, ordering the complete

538

Viêṭ Nam and the Genocide of Champa, 1470–1509

Figure 20.2 Cham Tower still standing at Vijaya (2). (Photo: George Dutton)

extermination of the remaining Cham captives, a command which was soon carried out.25 This massacre, which killed an unknown number of people, marked the last act of large-scale Vietnamese atrocities against the Cham. A final tally of the genocide’s victims based upon the only statistics available to us – those contained in the Vietnamese historical chronicles – suggests a minimum of 40,400 Cham were killed, a figure which does not include the indeterminate number killed in 1509. What percentage of the Cham population

25 DVSKTT, I I I, pp. 45–6, 47b–48; Khâm Đinh Viê ̣t Sử Thông Giám Cư ơ ng Muc (The ˙ Imperially Ordered Mirror and Commentary on the History of the Viêt), ̣ ˙2 vols., trans. Hoa Bằng, Pham Trong Điè ̂ m and Trà ̂ n Van ˘ Giáp (Hanoi: Nhà Xuâ ́ t Bả n ˙ 32–3. ˙ Giáo Duc, 1998), I I, pp. ˙

539

george dutton

this figure might have represented is impossible to determine for lack of any remotely usable statistics.26

Victim Accounts Simply put, there are virtually no contemporary Cham sources describing the Vietnamese aggression or acts of killing between 1470 and 1509. While there are Cham sources, many of them palaeographic, that detail aspects of the polity’s earlier centuries, and then Cham written documentation from the seventeenth century onwards, the fifteenth century is a virtual documentary black hole. As such, any account of what happened to the Cham is almost entirely dependent upon the surviving Vietnamese sources, with all of the limitations that this entails. The only scrap of indigenous testimony is the Cham envoys’ presentation of the situation to the Chinese court, which is preserved in the Ming dynasty’s Veritable Records: During the second month of the seventh year of the Cheng-hua reign (Feb/ Mar 1471), the Annam troops arrived and attacked our country’s capital, carrying off the king of the country Pan-luo Cha-quan (VN: Ban-la Tra-toan) and his family members, a total of over 50 persons. They also seized the seals, burnt dwellings and killed or carried off innumerable troops and civilians, both men and women.27

While brief and clearly couched in diplomatic terms, the Cham description of the episode is largely in agreement with Vietnamese accounts of the attack. It mentions the seizure of the ruler and members of his immediate family, the destruction of buildings and the killing or kidnapping of large (though indeterminate) numbers of both military personnel and civilians.

Aftermath and Long-term Consequences The primary evidence of a Vietnamese genocide against the Cham people lies in the devastating 1471 attacks. Although this chapter surveys a larger period stretching into the early sixteenth century, most Vietnamese actions against the Cham after that year were relatively minor by comparison. Indeed, most of the clashes that stretched from the 1471 attack to the dawn of the sixteenth century were diplomatic rather than military. Thanks to the same Ming 26 Anthony Reid estimates a combined Cham and Khmer population in 1600 of 1,230,000, which is the nearest recorded measure. 27 Wade, Southeast Asia in the Ming Shi-lu.

540

Viêṭ Nam and the Genocide of Champa, 1470–1509

Veritable Records that hinted at the Cham reaction to the 1471 attacks, we have a thread of historical evidence that both the Vietnamese and the Cham continued their diplomatic bickering with the Chinese courts over the period spanning 1470 to the 1490s. The records make clear the considerable Chinese frustration with both sides, for causing disorder and uncertainty and for trying to drag the Chinese into the fray. It seems quite evident, however, that the Chinese tended to side with the Cham, and to be quite sceptical of Vietnam’s claims regarding Cham actions and its own military responses. What is clear, however is that after their 1471 victory the Vietnamese now had definitive control over vast swaths of conquered Cham territory, rewriting the political landscape of the Indochinese coastal lands. In the aftermath of their overwhelming victory and the unconditional surrender of the Cham leaders, the Vietnamese asserted their authority over Cham territory. The Vietnamese claimed a substantial amount of what had been the northern and central Cham territories. This land, stretching from Phú Yên and Bình Đinh, ˙ the Vijaya region, to Quả ng Ngãi in the north, was formally constituted as the circuit (dao) of Quả ng Nam. The Vietnamese established the border between the two kingdoms at the Cả Pass in what is today Phú Yên province, the point at which the Đá Bia mountains spill into the sea. South of that point the Vietnamese left the Cham a substantially reduced territory comprising the modern Vietnamese coastal provinces of Bình Thuâ n and Khánh Hòa and the upland interior provinces of Gia Lai and ˙ Đă ́ k Lă ́ k. The Vietnamese, furthermore, divided the rump Cham kingdom into three parts, each of which was formally put in the charge of a Vietnamese-recognised ruler.28 While the tripartite partitioning of the Cham territories created a geopolitical circumstance common in Cham history, this and the territorial diminution ensured that Champa would no longer be capable of posing a serious challenge to the Vietnamese. There were several minor skirmishes in the sixteenth and seventeenth centuries, but these amounted to little more than justification for further Vietnamese expansion southwards, creating new military colonies there and resettling additional Vietnamese populations into the newly pacified territories. As the Vietnamese expanded their political and demographic reach, there was a simultaneous and significant curtailment of cross-cultural exchange. The period following 1471 marked the end of what had been the long spell of mutual cultural interaction seen in earlier centuries. In particular, the now 28 Nguye˜̂ n Đình Đà ̂ u, ‘The Vietnamese southward expansion, as viewed through the histories’ in Champa and the Archaeology of My Son (Vietnam), ed. A. Hardy et al. (Singapore: NUS Press, 2009), p. 69.

541

george dutton

Figure 20.3 Cham Hindu temples at Tháp Chàm, near Phan Rang, built in the reign of Champa’s King Jaya Sinhavarman III (r. 1288–1307). (Photo: Ben Kiernan, Jan. 1975)

heavily Confucian-oriented Vietnamese rulers and their courts no longer regarded the Cham as an appropriate and acknowledged source of cultural borrowing. It is true that Vietnamese settlers moving into Cham territories adopted elements of Cham cultural practices, partly for practical considerations in areas such as farming and fishing, but the extent of such borrowing was much more modest. The Vietnamese also borrowed Cham religious elements, but transformed the deities and sacred sites to their own requirements (Figure 20.3). This form of cultural appropriation was yet another obscuring of Champa. In many respects, the events of 1471–1509 represented not the culmination of Vietnamese genocidal actions against the Cham populations but rather one aspect of a protracted process of physical and cultural marginalisation of these people. The Vietnamese conquests and territorial seizures in the wake of the 1470–1 attacks set the stage for a centuries-long expansion of territory and influence that saw the Cham systematically pushed southwards, their culture eroded, and their population set in decline. This was, in short, not a single moment of violent and utter destruction but rather a kind of statement of Vietnamese intent, a harbinger of what was to follow. Despite the Vietnamese violence and destruction visited upon the Cham in the 542

Viêṭ Nam and the Genocide of Champa, 1470–1509

fifteenth century, Cham kingdoms survived and occasionally even posed modest military challenges to the Vietnamese, particularly in the seventeenth century. However, the post-1470 trajectory was clearly one of the steady diminution of Cham political and cultural influence. It also marked a period of significant Cham demographic retreat. While population figures are very difficult to come by, the declining amount of Cham territory in conjunction with growing numbers of Vietnamese settlers certainly constituted an inescapable reality. The Cham fled in the aftermath of the Vietnamese attacks and certainly in the face of subsequent Vietnamese expansion, with most settling in Khmer-controlled territory, and smaller numbers settling along the Malay Peninsula. In sum, Vietnamese genocidal actions against the Cham in the latter decades of the fifteenth century represented only one particularly violent episode in a long-running contest between the two states. While the two had been rivals for centuries dating back to the first millennium of the Common Era, the fifteenth century marked a new moment in which a Vietnamese sense of cultural superiority helped to fuel the attacks of 1470–1. Although a sizeable number of Cham were killed, perhaps as many as 40,000, the larger consequence of the Vietnamese attack was to set in motion an inexorable decline in Cham political power and demographic stability. Brief moments of Cham defiance to Vietnamese polities did flare up in the seventeenth century, but the reality was that Champa was never the same after 1471.

Bibliographic Note One inherent challenge in attempting to represent and assess Vietnamese actions against the Cham is a severe shortage of contemporaneous sources. The material that does exist is virtually all in Vietnamese. There are no known direct Cham accounts of the Vietnamese atrocities perpetrated between 1470 and 1509, nor any accounts from third-party eyewitnesses. And yet, despite our reliance upon Vietnamese sources, it is, in fact, these sources themselves that lend credence to labelling the attacks upon the Cham as genocidal. The Vietnamese historians of the later fifteenth century were hardly shy about describing their army’s brutal attacks against the Cham, as their references to killing by beheading would suggest. The most important source, indeed the basis for virtually all accounts, is the Đai Viê ̣t Sử Ký Toan Thư , the Complete Compiled Record of the History of Dai Viet˙ (Toan Thư ). This text was initially compiled in the second half of the fifteenth century, subsequently expanded to cover later events, and printed in 1697. 543

george dutton

Unsurprisingly, it offers the Vietnamese imperial perspective, with the Lê Emperor serving as the focal point. He is the actor who assembles the army, directs his generals, trains his troops, orders maps to be drawn, and then determines strategy. The chronicle offers some limited, and at times vague, statistical information regarding Cham killed or taken captive, though there are no independent sources against which to cross-check its numbers. We also have several other Vietnamese sources that may date from the fifteenth century, though all are extremely similar to the Toan Thư in their descriptions of the Champa campaigns. There is also an account of the campaign against the Cham, the Binh Nam Chı̉ Chư ở ng Nhâ t Trình Đò ̂ (The ˙ Daily Journal and Map of the Command of the Pacification of the South), which appears to have been created during the course of the attack itself and was likely produced at the behest of the emperor himself. Another text, the Thiên Nam Dư Ha Tâ p (Collection of the Leisures of Thien Nam [the South of ˙ ˙ 1, A334.), appears also to date from around the same Heaven]) (microfilm time and includes discussion of the Cham campaigns, largely echoing the narrative in the Toan Thư . A more narrowly geographical text, the Thiên Nam Tu´̛ Chí Lô Đò ̂ Thư (Book of the Four Route Maps of the South of Heaven) was ˙ compiled in 1490 and describes geographical details of the route from Th a˘ ng Long to the Cham capital. A somewhat later text, the Tây Nam Biên Tai Luc ˙ (Record of the Frontier Passes to the West and South), covers Vietnamese relations with peoples to the west and south, including the Cham, between 1418 and 1527. This text is very similiar to the Toan Thư account, including the full text of the emperor’s ‘Pacifying Champa Edict’ and similar statistics on numbers of soldiers and battlefield deaths and captives taken. Finally, later Vietnamese court histories, notably the nineteenth-century Khâm Đinh Viê ̣t ˙ Sử Thông Giám Cư ơ ng Muc (The Imperially Ordered Mirror and Commentary ˙ on the History of the Viet) typically contain condensed versions of the account found in the Toan Thư . The language and references are almost completely identical. Thus, while one can find an assortment of Vietnamese historical texts discussing the Viet–Cham wars, the Toan Thư is clearly the critical ‘urtext’. Alongside the Vietnamese texts we have two external records of the conflict. Particularly useful is the Ming Shi Lu (The Ming Veritable Records), which represents the official Chinese view of the Ming court and its world. This text includes numerous references to the clashes between the Vietnamese and the Cham, though it records only the reports presented to the Chinese by the two antagonists. Thus, the Ming Shi Lu provides us with summaries of Cham reports about being attacked by the Vietnamese and, in 544

Viêṭ Nam and the Genocide of Champa, 1470–1509

turn, by the Vietnamese justifying their attacks upon their neighbour. As one might suspect, such reports offer distinctly one-sided representations of the conflict, featuring varying degrees of hyperbolic language from both parties. Finally, while its description of the events is extremely brief and somewhat vague, the Sejarah Melayu (Malay Annals), originally composed during the fifteenth and sixteenth centuries, contains a short account of the Vietnamese attack and its circumstances. This account was likely based upon contemporary or near contemporary reports from Cham who had fled the attacks and resettled in peninsular Southeast Asia. Secondary Sources There are no detailed secondary source materials on the Vietnamese attacks against the Cham, which typically merit only brief discussion in broader historical narratives. Whitmore’s unpublished 1968 PhD thesis, ‘Development of Le government’, remains the most comprehensive examination of the events of the fifteenth century and the Lê dynasty more broadly, though even it only offers a few pages on the Cham war. Ben Kiernan’s Viê ̣t Nam: A History from Earliest Times to the Present (Oxford University Press, 2017) is among the most recent and detailed discussions of the Cham genocide, making considerable use of the available Vietnamese and Chinese sources. Kiernan also discusses the broader historical context of Viet–Cham clashes in Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007). Keith Taylor’s A History of the Vietnamese (Cambridge University Press, 2013) briefly describes the Vietnamese attacks, focusing on Vietnamese characters, notably the emperor. Taylor does not cite any figures for Cham deaths; nor does he characterise the attacks as genocidal. Christopher Goscha’s recent Vietnam: A New History (New York: Basic Books, 2016) also discusses the Vietnamese attacks in brief, describing the episode largely as ‘territorial conquest’ in which the Lê imperial armies ‘crushed’ the Cham (pp. 415, 32). Lieberman’s Strange Parallels is useful for synthesising multiple accounts while situating the Viet–Cham clashes in a much broader historical and regional context. Lieberman emphasises the cultural dimension of the Vietnamese conquest – the desire to transform and Vietnamise the Cham (pp. 379–81). A more detailed discussion of the military context of fifteenth-century Viet–Cham interactions is Sun’s Chinese Military Technology and Dai Viet. Sun focuses on the implications of military technology, and particularly firearms, in the course of Vietnamese–Chinese clashes, and briefly discusses the Vietnamese attacks against Champa. He emphasises the enormous technological advantage the Vietnamese enjoyed, while pointing to the Vietnamese attack as a major reason for an early modern 545

george dutton

Cham diaspora. Finally, while Vietnamese scholars’ work on this episode has primarily been published in Vietnamese, Nguye˜̂ n Đình Đà ̂ u’s recently translated ‘Vietnamese southward expansion’ is a notable exception. While broadly discussing the Vietnamese political and demographic migration southward, the article does address the fifteenth-century clash. Đà ̂ u provides a distinctly bloodless rendering of the event, which he blames on Cham provocation, and for which he describes no casualties, cites no taking of Cham captives, and makes no reference to the Hò ̂ ng Đu´̛ c Emperor’s edict or the rhetoric surrounding the Vietnamese aggression (69).

546

21

Genocidal Massacres in Medieval India raziuddin aquil This chapter is written in India at a time of a huge and desperate struggle for peace in the context of widespread violence perpetrated by extreme rightwing majoritarian fanatics. The justification for this violence is based on claims of massacres against Hindus perpetrated by Muslim rulers in medieval India, which supposedly need to be avenged by a systematic modern ethnic cleansing of Indian Muslims. The people in power now have access to all the technologies of violence controlled by the state. Political leaders and ideologues inspired by an early twentieth-century fascistic project deny that the violence they plan and perpetrate constitutes crimes against humanity. They argue that they are merely correcting wrongs of thousands of years ago, euphemistically referred to as slavery, to rehabilitate Indian culture to its imagined ancient glory through the political emancipation of the longsuppressed Hindu people. Much of it is based on crude political propaganda peddled as history. In large measure, they draw on the murky picture of medieval India constructed by British colonial administrators and writers since the early nineteenth century. That justified British colonial conquest as liberation from violence and murders perpetrated by Muslim rulers who preceded them. There are still professors with commonwealth chairs who believe that colonialism was good for the colonised, exposing them to a new enlightened modernity and bringing them out of their dark middle ages. This view ignores the massive changes that had occurred in India in several spheres during three or four centuries of the early modern era.1 Professional historians must return to examine medieval conquests to judge whether the violence involved might have been genocidal or political. Historians of medieval and early modern India, often self-identified as left of 1 For an analysis of the continuing justification of imperialism and colonialism, see Richard Drayton, ‘Where does the world historian write from? Objectivity, moral conscience and the past and present of imperialism’, Journal of Contemporary History 46 (2011), 671–85. For an early discussion of the connected history of the early modern world and India’s place in it, see J. F. Richards, ‘Early modern India and world history’, Journal of World History 8:2 (1997), 197–209.

547

raziuddin aquil

centre, secular or liberal scholars, and condemned by Hindu communalists as anti-nationals, have tended to de-emphasise the nature and scale of violence under Muslim rulers, sanitising it and even justifying major massacres as simply consequences of political conquests. They argue that reference to religion was made by rulers to justify and legitimise violent attacks, whereas the actual intent or motive was political conquest and economic aggrandisement. Early eleventh-century Sultan Mahmud of Ghazna’s attacks on Hindu places of worship have been characterised by even the best of historians as mere loot and plunder, which did not involve any large-scale ethnic cleansing on religious grounds as part of the conquest. There is something of a consensus among secular historians on political conquests and governing principles to emphasise peaceful community relations marked by ethnic and religious diversity in medieval India. Religious tolerance was the hallmark of Turkish, Afghan and Mughal rulers who built and ruled over vast subcontinental empires for the major part of medieval and early modern eras, the assertion goes. Medieval India witnessed a broad and inclusive framework of political ideas, theories and practices, later adopted by Mughal emperors from the sixteenth century.2 Violence was permissible in the wake of conquests and empire-building, but medieval works on political theory advocated minimum use of force under all circumstances. To cite a well-known example, Abu’l Fazl’s Akbarnama and Ain-i-Akbari articulated Mughal political ideology on how to govern a vast subcontinental country of great diversity, especially for resolving the question of religious assertions through emphasis on broadbased governing principles, conceptualised as peace with all.3 It was understood that political conquests might involve a lot of violence, but governing principles of the state had to be inclusive and just, offering space for all the diversity to respectfully coexist. Intersecting between the political and cultural domains, the Sufi–Bhakti complex also preached peaceful coexistence and tolerance of other people’s religious beliefs and practices, which further explains India’s amazing diversity as a historical reality. The spiritual domain saw Muslim Sufis and Hindu Sants bringing people together, especially the significant role played by such figures as Sant Kabir and Guru Nanak in 2 For an introduction to Mughal history, J. F. Richards, The Mughal Empire (Cambridge University Press, 1991) is still useful. For the longer period of precolonial history, see Catherine B. Asher and Cynthia Talbot, India before Europe (Cambridge University Press, 2006). 3 Abu’l Fazl, Akbarnama, trans. H. Beveridge, vols. I – I I I (Delhi: Oriental Reprint, 1972–3); Abu’l Fazl, Ain-i-Akbari, vol. I , trans. H. Blochmann, vols. I I and I I I , trans. H. S. Jarret (Delhi: Oriental Reprint, 1977–8).

548

Genocidal Massacres in Medieval India

creating modus vivendi for a vast diversity of people, though rigid markers of identities were also carved by the followers who constituted sparring communities.4 In this context, there is a need to take cognisance of cases of aggressive sectarianism, complex identity formation, questions of conversion and desecration of places of worship, and occasional attempts to abuse political power in the name of religion. Medieval and early modern Indian rulers continued to seek religious legitimacy for their political acts. Thus, historians have attempted to explain religious justification of political violence in the wake of battles of conquest, to examine governing principles and political frameworks (since the conquerors needed to come down from their horses and establish their rule as well as law and order), and to understand continuous attempts at resolving religious and cultural conflicts. Moving away from the usual binary – either celebrating or condemning political violence and bloodshed, or completely denying the cases of violence – some writings have stressed the struggle for peace as proof of peaceful community relations. Mass violence would not have been the order of the day, but politics and society were not free from religious and communal tensions either.5 This chapter turns to specific evidence of political violence and massacres during the medieval Delhi Sultanate (1206–1526), for most of which period roughly the whole of the Indian subcontinent was ruled from Delhi by powerful sultans with Turkish antecedents and exposure to the Persianate political culture of central Asia and Iran.6 In view of the abundant contemporary sources, the focus will be on moments of violence under the three most despotic rulers who brooked no criticism, resistance or protest against their rule, which tended to be arbitrary and in complete disregard of existing conventions of politics and governance. They were Sultan Ghiyas-ud-Din 4 For a wide-ranging examination of medieval Bhakti devotional literature deployed in the service of the country’s political and cultural unity, see John Stratton Hawley, A Storm of Songs: India and the Idea of the Bhakti Movement (Cambridge, MA: Harvard University Press, 2015); for Sufis’ social and cultural roles in Indian history, including the period covered in this chapter, see Raziuddin Aquil, Lovers of God: Sufism and the Politics of Islam in Medieval India (London: Routledge, 2020). 5 Some of the critical issues have been discussed in Raziuddin Aquil, The Muslim Question: Understanding Islam and Indian History (New Delhi: Penguin, 2017). For an internal critique of the limitations of politically dictated secular history, see Neeladri Bhattacharya, ‘Predicaments of secular histories’, Public Culture 20:1 (2008), 57–73. 6 For Dehli Sultanate history, see Aquil, Lovers of God, ch. 1. See also Peter Jackson, The Delhi Sultanate: A Political and Military History (Cambridge University Press, 1999); Sunil Kumar, The Emergence of the Delhi Sultanate, 1192–1286 (New Delhi: Permanent Black, 2007).

549

raziuddin aquil

Balban (r. 1266–86), Sultan Ala-ud-Din Khalji (r. 1296–1316) and Sultan Muhammad bin Tughluq (r. 1325–51). Each of these three rulers expanded and consolidated their power over larger areas than they had inherited, though the last phase of the Tughluq Sultan also witnessed a crumbling of the empire in outlying regions such as the Deccan and Bengal, which went on to establish regional sultanates in southern and eastern India.7 Much information on this era is available in contemporary court chronicles, the histories and travelogues of such credible authorities as Amir Khusrau, Ziya-ud-Din Barani, Izz-ud-Din Isami and Ibn Batuta, among others.8 Each wrote from a different perspective, often corroborating the others’ accounts and providing justifications or offering criticisms from their varied locations in the political spectrum of the period, from the middle of the thirteenth century to the middle of the fourteenth. We shall focus here on Barani’s detailed accounts of cases of large-scale killing not only during battle but also of those captured alive and hanged in spectacles of violence. Barani offers opinions on how violence was either justified or condemned, and how occasional attempts to intervene and prevent cold-blooded massacres also worked. The chapter therefore is both about both mass murders under three of the most powerful Delhi sultans and about Barani’s analysis of their rule and of those killings, identified here as genocidal massacres.9 A quick introduction to Barani’s life (1285–1357), career and writings will help. Barani’s mid-fourteenth-century work Tarikh-i-Firuz Shahi is one of the finest examples of Indo-Persian historiography, a text much abused in modern times by colonial authors, Hindu nationalists and other interested 7 For Deccan, see R. M. Eaton and Philip B. Wagoner, Power, Memory, Architecture: Contested Sites on India’s Deccan Plateau, 1300–1600 (Oxford University Press, 2014). For Bengal, see Syed Ejaz Hussain, The Bengal Sultanate: Politics, Economy and Coins (AD 1205 – 1576) (New Delhi: Manohar Books, 2003). For the subsequent history of Bengal, see Raziuddin Aquil and Tilottama Mukherjee, eds., An Earthly Paradise: Trade, Politics and Culture in Early Modern Bengal (London: Routledge, 2020). 8 Amir Khusrau, ‘The campaigns of Alauddin Khalji, being the English translation of “The Khaza’inul Futuh” of Amir Khusrau’ in Politics and Society during the Early Medieval Period: Collected Works of Mohammad Habib, vol. I I, ed. K. A. Nizami (New Delhi: People’s Publishing House, 1981); Ziya-ud-Din Barani, Tarikh-i-Firuz Shahi, ed. Sir Syed Ahmad (Aligarh: Sir Syed Academy, 2005) and English translation by Ishtiyaq Ahmad Zilli (Delhi: Primus Books, 2015), which is used and cited in this chapter; Futuh-us-Salatin or Shah Namah-i Hind of Isami, trans. Agha Mahdi Husain (Bombay: Asia Publishing House, 1967); Ibn Batuta, Safarnama Ibn Batuta, Urdu trans. Syed Rais Ahmad Jafri (Karachi: Nafis Academy, 1986). 9 For some valuable theoretical insights on what constitutes genocides, genocidal massacres or mass killings of people on political grounds or ethnic difference, see Ben Kiernan, ‘Is “genocide” an anachronistic concept for the study of early modern mass killing?’, History 99:336 (2014), 442–60.

550

Genocidal Massacres in Medieval India

parties.10 Dismissed in secular histories as a bigoted theologian whose views smacked of communalism, Barani was a fine historian, aperceptive political theorist, a well-informed personal advisor of Sultan Muhammad Tughluq, a sincere disciple of the Chishti Sufi Nizam-ud-Din Auliya and a close friend of the noted Sufi poets and sultanate officials, Amir Khusrau and Amir Hasan. Barani had nothing to do with theology and he dismissed as impracticable the claims of Islamic shari’a as state law, advocated by the Sunnite ulama, religious scholars. Instead, he articulated an elitist power discourse: the supremacy and domination of Islam could be established only in contrast to the inferiority and subordination of non-Islam, since power cannot be enjoyed in a vacuum. Conversion to Islam was not recommended and even Indian converts to Islam were to be treated as inferior people who should not have been converted in the first place. As the cliché goes, history is written by the conquerors sometimes on the bodies of those they decimate, but the conquered people often survive to write their own histories. With a certain degree of democratisation of the sultanate polity, a large number of Indians were able to break into the system to change the course of history. Gradual political and social change had meant that one did not even need to formally convert to Islam to rise in the sultanate’s nobility and bureaucracy – a process initiated as much by the emergence of the Khalji and Tughluq Sultans as by the presence of the Chishti Sufis with their inclusive cultural practices. These new people were giving a tough time to Delhi’s power elite, previously entrenched in the system. Certainly, Barani demeans himself by condemning them using a variety of invectives: calling them unworthy, ignoble, low-born and bazaar people who can have no sense of history!11 Misfit in the changing scheme of things and sidelined from the court of Sultan Firuz Shah Tughluq, Barani refashioned himself to start his career as a historian at the age of about seventy, realising all along how the times had changed from a strong Ghiyas-ud-Din Balban’s use of sophisticated violence to consolidate his position to a ruthless Muhammad Tughluq who built an empire geographically comparable to those of the ancient Mauryas; it was another matter that the Tughluq Sultan could not control and govern that 10 Important new works on Barani include Vasileios Syros, ‘Indian emergencies: Barani’s Fatawa-i Jahandari, the diseases of the body politic, and Machiavelli’s accidenti’, Philosophy East & West 62:4 (2012), 545–73; Blain Auer, ‘Pre-modern intellectual debates on the knowledge of history and Ziya’ al-Din Barani’s Tarikh-i Firuzshahi’, Indian Economic and Social History Review 52:2 (2015), 207–23; Aquil, Muslim Question, ch. 3. 11 Tarikh-i-Firuz Shahi, pp. 7–12.

551

raziuddin aquil

empire to anyone’s satisfaction. In his introduction to the Tarikh-i-Firuz Shahi, Barani emphasised the value of acquiring historical knowledge, outlined his method and stressed that what he had written was a matter of truth, arguing that his work could be read as a chronicle of the rule of the sultans, covering nearly a century, or as political analysis of regulations regarding matters of governance, or as counsel and advice for rulers – historical insights as guiding principles in the politics of the present. According to Barani, historians should be able to rise above the limitations of fear or favour in discussing the virtues and merits of those in power, not sweeping under the carpet their failings, demerits and cruelties. If it was not possible to openly criticise a despot, then historians could deploy their linguistic skills in such a way that they were able to speak by means of allusion or euphemism; and, if historians found it altogether impossible to write about a contemporary ruler, they could be excused, but as regards people of the past, they were expected to write truth alone. These views have been articulated by Barani in his work on political theory or ideas relevant for the Delhi sultans, Fatawa-i-Jahandari.12 * Late in the twelfth century, the Ghurid forces of Mu‘iz-ud-Din bin Sam, referred to in the early sources as the lashkar-i-islam (‘army of Islam’), overran Ghaznavid Punjab. The Rajput resistance was smothered in the second battle of Tarain, in 1192. The Muslim army went on to occupy large parts of upper north India. It eliminated the symbols of Rajput power and prestige there, but no general massacre or major demographic dislocation ensued. Much as the chroniclers celebrated the conquest of new territories, the conquerors preferred minimum use of force. Though iconoclasm may have motivated some soldiers, places of worship were generally plundered for their wealth. Alternatively, their despoliation was aimed at hammering home the point that the old regime was overthrown; it could no longer protect the people and their religious places. The general public was thus made aware that the Turks and their sultan had established a new, Islamic order. Indeed, the minaret attached to Delhi’s congregational mosque, later known as the Qutb Minar, was subsequently perceived as a victory tower (Figure 21.1). Among the prime targets of Ghurid campaigns were Muslim Ghaznavids as well, but the Muslims of the Sultanate period particularly liked to remember the

12 Ziya-ud-Din Barani, Fatawa-i-Jahandari, ed. Afsar Salim Khan (Lahore: Research Society of Pakistan, 1972).

552

Genocidal Massacres in Medieval India

Figure 21.1 Qutb Minar, Delhi. (Photograph from frentusha / Getty Images)

crushing of the infidels (kafirs), especially the army of the Rajput ruler Prithviraj Chauhan. Significantly, the enthronement of Sultan Qutb-ud-Din Aibak (r. 1206–10) at Lahore coincided with the election of Chengiz Khan as the great leader of the Mongol hordes. The raids of the Mongols wrought large-scale devastation in central and western Asia in the next fifty years. Major centres of Islam such as Bukhara and Baghdad were sacked. The subcontinent was as yet protected, though Punjab and Sindh were exposed to the threat of a possible onslaught. Escaping the wrath of the Mongols (also referred to as Tartars), Islam prospered in the Delhi Sultanate, with the name of the caliph still being 553

raziuddin aquil

mentioned in the khutba (Friday sermons) and the sikka (coins). In decades to come, rising from the ranks of slave soldiers of Sultan Shams-ud-Din Iltutmish (r. 1211–36), Ghiyas-ud-Din Balban ruled for over forty years from about the middle of the thirteenth century, acting first as naib (deputy) of the puppet Shamsi sultan Nasir-ud-Din Mahmud (r. 1246–66) and subsequently ascending the throne of Delhi as Sultan Balban (r. 1266–86) in his own right. Balban took the throne at a time when the Mongols had ravaged a major part of the Muslim world. Having sacked Punjab, they were threatening to take Delhi by storm. Balban’s aggressive-defensive Mongol policy protected Delhi from the depredation of the Mongol hordes. Celebrating the destruction of the Abbasid seat of power at Baghdad, the Mongol leader Halaku Khan had sent his envoys to Delhi in 1259–60. The naib, Balban, himself a recent captive of the Tartars, welcomed the guests in typical Mongol fashion. The route that the visitors’ cavalcade took was ornamented with the severed heads of rebels from the neighbourhood of Delhi. Their hides were stuffed with straw and was displayed. Besides, 200,000 footmen and 50,000 horsemen were posted along the way. The chronicler Minhaj-us-Siraj records in his Tabaqat-i-Nasiri that nothing in particular occurred on the occasion, but exposed to the spectacle of violence, both visual and suggestive, the envoys must have returned with a sufficient idea of Delhi’s power and aggressive intentions.13 Further, Balban not only rebuilt and fortified the cities in Punjab and Sindh that had suffered at the hands of the Tartars, but also constructed large forts on the route to the northwest in order to block their advance into the sultanate’s territories. The Mogols, dreaded as ‘fire from hell’, were to be checked on the frontier. Huge forces were also garrisoned in the forts en route to Punjab, which prevented the penetration of the invaders into Hindustan. Delhi was protected from the Mongol menace, but at home the Mewatis or Meos were accused of indulging in all kinds of crimes in the city, including robbery and molestation, after which they disappeared in the neighbouring forests. Ziya-ud-Din Barani records in his Tarikh-i-Firuz Shahi that the Meos had plundered all the inns in the vicinity, and frequently attacked the city at night. Unable to check the problem, a weak Delhi administration ordered the city gates closed before sunset. No one had the courage to go out of the walled city after that time. Those wishing to visit the tombs of the saints or to enjoy the pleasant environs of Hauz-i-Shamsi went at their own risk. The 13 Minhaj-us-Siraj, Tabaqat-i-Nasiri, trans. Major H. G. Raverty (Delhi: Oriental Books, 1970), I I , pp. 854–9.

554

Genocidal Massacres in Medieval India

Tarikh-i-Firuz Shahi notes that Meos robbed water-carriers and molested slave-girls who went to fetch water from the tank, even during the day. Balban took a whole year to suppress the miscreants. The forests around Delhi were cut down and an estimated 100,000 Meos put to death. Many military posts (thanas) were established and assigned to Afghan warlords and soldiers. Land allotted for thier maintenance was made tax-free. The city was secured from the Meo onslaught. Of course, Barani’s account reflects the viewpoint of the Delhi administration of the time; a Meo narrative would tell a different sordid story. Barani, who recommended violence to establish the authority of Muslim sultans over completely subjugated non-Muslims, narrated in his voluminous Tarikh-i-Firuz Shahi several other violent campaigns undertaken by Balban. Here he writes on the authority of his father, grandfather and other trustworthy people such as Amir Khusrau and Amir Hasan, who were his friends and senior contemporaries in the sultanate service. The example of the sultan’s Bengal campaign against the rebellious Tughril Khan may be presented here in its lurid details. Barani’s maternal grandfather Husam-ud-Din was part of the campaign and was appointed by Balban as the administrator of the capital city of Bengal, Lakhnauti, as the local governor, when Tughril Khan vacated it along with a large contingent of his soldiers and people dependent on him. Beginning as a slave, Tughril was appointed as governor by Sultan Balban himself and had then declared his independence from the control of Delhi. Barani opines, as a general rule, that as long as awe, fear and terror does not get imprinted in the heart of the masses and among the select living in regions far and near, the affairs of kingship and the state could not be conducted in a proper manner. If the king is negligent in upholding the majesty of the government, then the people, both those who live close and those who live in faraway regions, will not be kept under the fear of the authority and majesty of the king. This leads to chaos in the affairs of the state and the subjects tend to become rebellious. The rebelliousness of subjects makes the body of the state sick.14 Accordingly, when the news arrived of Tughril declaring his independence – adopting the royal umbrella, getting coins struck in his name, ordering insertion of his name in Friday sermons, and not sending elephants and other spoils from a recent campaign in Jajnagar (Odisha) – an enraged Balban sent huge armies to suppress him. When two successive attempts failed to crush Tughril’s rebellion, Sultan Balban was overwhelmed by shame and wrath. 14 Tarikh-i-Firuz Shahi, p. 21.

555

raziuddin aquil

Determined to destroy Tughril, Balban decided to march to Bengal. The charge to defend a larger region from Mongol attacks was given to Prince Muhammad. The younger son Mahmud (later known as Bughra Khan), who was administering parts of Punjab, was asked to join the sultan’s Bengal campaign, and the responsibility to maintain law and order in the capital and adjoining regions was assigned to the most senior official, styled as Malik-ul-Umara, Kotwal of Delhi, with orders that all departments including finance and military would be reporting to him in Balban’s absence, which lasted for nearly three years. As the sultan reached the vicinity of Lakhnauti, Tughril decided to vacate the place with his war elephants and riches, selected troops, eminent associates, their women and children, and others who could prove useful to him. Some people accompanied him fearing punishment of the sultan, while others joined in the hope of reward, should Tughril succeed in resisting Balban’s onslaught – hoping the Delhi sultan would not remain in Lakhnauti for long and that the intervening period could serve for another campaign against Jajnagar, and they could return stronger and with fresh booty.15 Within a few days of reaching Lakhnauti, Sultan Balban made arrangements for hot pursuit of Tughril on the way to Jajnagar. An advance party of between 8,000 and 10,000 cavalrymen were ordered to march 32–40 kilometres ahead of the army as a vanguard. A select band rode ahead of the vanguard, scouting and reporting on Tughril’s army. Accordingly, Malik Muhammad Sher Andaz, administrator of Kol, and his brother, Malik Muqaddar, and an officer called Tughril Kush (the man who eventually killed Tughril) – men known for their bravery and intent on revenge – were sent from the vanguard along with thirty to forty cavalrymen to gather information. Based on information forcibly extracted from a group of merchants travelling from the other direction, one of whom was murdered to terrorise others into speaking up regarding Tughril’s army and their sale of grain to it, the scouting party was able to locate the army, which was seen safely encamped on a vast patch of land by the side of a stone-built pond. Water was available for the soldiers to bathe and wash their clothes, horses and other cattle were let loose in the pastures and elephants had trees on which to forage. Instead of relaying the news to the vanguard, leaders of the scouting party decided to attack Tughril’s camp with the small number of cavalrymen they had on the assumption that Tughril’s forces would think that the army of the sultan had arrived and everybody would try to escape. As it turned out, 15 Ibid., pp. 52–3.

556

Genocidal Massacres in Medieval India

when the small detachment charged with swords in hands, shouting for Tughril, and rushed directly into his camp in a blitzkrieg of sorts, his army took to flight due to fear of the sultan as a great panic descended on them. Muqaddar and Tughril Kush went in pursuit of Tughril. Spurring his horse, Tughril Kush was able to spot Tughril and struck him with a half-pointed arrow, throwing him on the ground. Before he could recover, Muqaddar came down from his horse, severed his head, concealed it under the helm of his garment and busied himself in washing his hands and face. The bodyguards and armour bearers were still searching for Tughril and calling for him, but the arrival of the vanguard of the sultan’s army led by Malik Barbak also meant Tughril’s force was in no position to recover. They took to flight. Malik Muqaddar and Tughril Kush presented Tughril’s head to Malik Barbak. The wives, sons and daughters of Tughril along with elephants, treasures, the confidants, close associates and officers with their women and children fell into the hands of the army. Barani appreciates the report that the troopers of the vanguard secured immense riches, horses, weapons, slaves and maids, sufficient for them and their children for years to come.16 Later, 3,000 to 4,000 able-bodied men and women were sent to Balban, who had camped at the place where he had received the news of the victory and the beheading of Tughril. Malik Barbak presented himself before the sultan with all the booty captured along with prisoners of Tughril’s army and submitted a detailed report. After admonishing the leaders of the advance party for their impatience, which could have turned out to be disastrous, the sultan nonetheless gave to all the scouts robes of honour and rewards according to their status. Their status was raised and the one who had struck Tughril was given the title of Tughril Kush (‘killer of Tughril’). The fathnama (composition to celebrate the victory) composed to mark the occasion served as a model for times to come, and when it reached Delhi there were celebrations all around. The awe and prestige of Sultan Balban in the hearts of the people of his dominions increased a hundredfold, according to Barani. However, for Balban, the campaign was not over yet. Upon his return to Lakhnauti, orders were issued for gallows to be erected on both sides of the big marketplace of the provincial capital, stretching for more than three kilometres. The sons, wives, sons-in-law, officials, local administrators, slaves, confidants, army commanders, bodyguards and renowned foot soldiers of Tughril – all those associated with the former governor – were hanged. A Muslim mystic Qalandar held in great respect by Tughril and 16 Ibid., pp. 54–5.

557

raziuddin aquil

called Sultan by the Qalandars, who had no role in the rebellion as such, except perhaps offering blessings and legitimacy to Tughril’s rule, was similarly executed, along with his close associates, and hung on the gallows. Speaking for the victim, as it were, Barani, whose maternal grandfather Husam-ud-Din had been made the administrator of the city during the interim march in pursuit of Tughril towards Jajnagar, writes that during the two to three days of carnage the Sultan carried out such large-scale massacres that many elderly spectators fell unconscious and died from the horrors of these killings. He writes that he has heard from old people that the genocide that the Sultan carried out in Lakhnauti was such that he did not allow even the name or any sign of Tughril, and those associated with him, to remain alive. Nobody remembers such a bloodbath to have ever taken place under any ruler of Delhi, nor anywhere in Hindustan.17 A new and more inclusive political system under Sultan Jalal-ud-Din Khalji (r. 1290–6) ended with his brutal murder by a more ambitious and despotic nephew, Ala-ud-Din Khalji, who took over as the sultan of Delhi for twenty years. Ala-ud-Din’s conquest of newer areas in Rajasthan, Gujarat and Deccan involved a lot of bloodshed, and he ruled with a heavy hand to eliminate any opposition to his rule. He is also known for his market reforms, needed to maintain a huge army in Delhi to protect the capital from the Mongols. * When Ala-ud-Din Khalji seized power in 1296, our historian Barani was just a young boy. He was able to witness political events at first hand and also through both his father Muayyadul Mulk, administrator of Baran (thus ‘Barani’), and maternal uncle Alaul Mulk, a close confidant of Ala-ud-Din and partner in all his early crimes and administrative head (Kotwal) of Delhi. Barani set down in the introduction to his book that he would record only the truth – both the merits and the demerits of the people whose history he would be writing. If he avoided mentioning their wickedness and mentioned only the positive things about them, his writings would not have any credibility in the eyes of readers and ‘no hope of salvation before God either’. Therefore, ‘in keeping with this precondition that I wrote about the days of Sultan Ala-ud-Din when he was a noble whatever I saw relating to his killing of his benefactor. Similarly, whatever I have witnessed relating to his governance and conquest that too I have recorded.’18 17 Ibid., pp. 55–6.

18 Ibid., p. 146.

558

Genocidal Massacres in Medieval India

Early in the sultan’s reign, Barani’s uncle Alaul Mulk, Kotwal of Delhi, had advised him against thinking of starting his own religion with the help of his four friends – as according to him, the Prophet of Islam had done – and also to control his desire to conquer the world as Alexander the Great had done – styling himself as Sikandar Sani, Alexander the Second. This title was inserted in sermons and also struck on his coins. Alaul Mulk had, instead, advised the sultan to pay attention to matters of governance, crush any opposition within his territory, subdue local kingdoms and protect the sultanate from the Mongol menace that had continued to threaten large parts of upper north India and Delhi. The sultan governed with a heavy hand.19 Of those cases which can be identified as cold-blooded genocidal massacres, repeated treatments meted out to a large population of neo-Muslims (families of enslaved Mongol soldiers who had embraced Islam and settled down in Delhi to serve as mercenaries for the sultanate army) may be mentioned here. One example is from the third year and another from towards the end of the sultan’s long rule. Early in his reign, the sultan sent a huge army to conquer Gujarat. Reporting on the Khalji conquest of Anhilwara, the then capital of Gujarat, Amir Khusrau writes in his Khaza’inul-Futuh that when the imperial army reached the city of that land, the sword of the righteous monarch completely conquered the province and much blood was shed. A general invitation was issued to all the beasts and birds of the forest to a continuous feast of meat and drink. In the marriage banquet, at which the Hindus were sacrificed, animals of all kinds ate them to their satisfaction.20 Barani informs us that having ransacked and pillaged the territories of the local Hindu king, the leaders of the sultanate army began their return with enormous booty. They also subjected members of their own army to stringent investigation and severe punishment, demanding onefifth of the war booty (khums) and recovering from them large amounts of gold and silver extracted through much torture. Exasperated with the severity of the investigation and torture, some 2,000 to 3,000 neo-Muslim cavalrymen mutinied, and mistakenly killed the brother of one of the commanders of the army, Nusrat Khan. Sensing that they would be overpowered, the rebels fled in different directions to seek refuge with local rulers.21 The sultanate army returned to Delhi with all the treasure, elephants, slaves and booty seized from the pillage of Gujarat. Since news of the mutiny had already reached Delhi, an outraged Sultan Ala-ud-Din ordered that the 19 Ibid., pp. 160–5. 20 Khusrau, ‘Campaigns of Alauddin Khalji’, 181–2. 21 Tarikh-i-Firuz Shahi, pp. 154–5.

559

raziuddin aquil

wives and children of the mutineers be apprehended and imprisoned. Barani opines that the seizing of women and children for the crimes of men began from that date. Earlier, in Delhi, women and children had never been rounded up and imprisoned for the crimes of men. Indeed, a greater outrage was introduced by Nusrat Khan, who initiated many tyrannical ways of exercising power in Delhi. He took revenge on those who had killed his brother through great disgrace and ignominy by handing over their women to sweepers to have them raped, while their small children were killed in front of their mothers. A thoroughly disgusted Barani, whose own family was deeply entrenched in the system, comments that in no religion was that kind of cruelty ever committed. According to him, the people of Delhi expressed their astonishment and surprise, and trembled among themselves.22 Barani further reports that during the last years of his reign, Sultan Ala-udDin achieved wide-ranging victories and shaped affairs of the state as he wished. This was the period when he was also busy rooting out the Mongols. It was brought to his notice that some neo-Muslim amirs, or nobles, were complaining among themselves about their penury and impoverishment. They had been jobless for many years and their livelihood was greatly reduced because of new rules introduced by the government. The sultan was told that out of ill will they were saying that he was keeping a stranglehold over the people and had filled his treasury through means of extortion and tyranny. That they were criticising the sultan for prohibiting wine, hemp and other intoxicants, fixing arbitrary taxes and in the country putting the people under great distress. That a rebellion could be initiated with 200 to 300 neo-Muslim infantry and cavalry, ambushing the sultan when he went out for recreation. Other people would be pleased to join the rebellion for the common good of relief from the sultan’s oppression.23 It is possible that no such conspiracy was hatched, but the reports, fake or otherwise, of popular discontent and murmurs of protest were enough to incur the sultan’s wrath even in his last years. Barani notes that owing to his harsh temper, disposition towards brutality and habit of extreme punishment, the sultan was driven only by the interests of his kingdom and did not care for any other considerations whatsoever, be they religious or family or any other obligation. He issued orders that the neo-Muslims living within the dominion were to be massacred. They were to be killed during the span of a single day and, rhetorically, in such a way that not one neo-Muslim was to be left on earth. As a result of this order, in which, according to Barani, the 22 Ibid., p. 155.

23 Ibid., p. 205.

560

Genocidal Massacres in Medieval India

sultan ‘acted like Pharaoh and Nimrud’, 20,000 to 30,000 neo-Muslims, a great majority of whom had no knowledge of the so-called conspiracy, were massacred. Their households were ransacked and their women and children devastated.24 Around the same time, Ibahati libertines and Buddhists appeared in the city, mentions Barani. Sultan Ala-ud-Din ordered that thorough investigations were to be carried out into their activities; all of them were to be traced, subjected to the worst possible treatment and then killed. For Ala-ud-Din, a new method of punishment was introduced: the accused were sawed into two from the head down. After this, talk of permissiveness (ibahat) did not come to the tongue of anyone in the city.25 In terms of methods of punishment, Ala-ud-Din also initiated the dumping of wine sellers in dry wells dug up specially for the purpose. Since his counsellors had advised against the consumption of liquor – which was considered a source of his weakness and of the rebellious activities of political opponents, and a general sin and crime – it had to be banned, though he was advised that he himself could consume it in private. Accordingly, the selling of wine was banned. The sultan stopped organising drinking parties and ordered that all the flasks of wine, goblets made in Ma’bar or imported from China, besides drinking cups inlaid with gold and glasses of the imperial drinking chamber, were to be broken and the pieces dumped on the main city square. The illegal production and sale of any kind of wine was to be stopped. Large quantities of wine were sent to elephant stables to be fed to the animals. Those who were responsible for selling it, instrumental in bringing it inside the city and those who consumed it were subjected to severe beatings and chained and imprisoned for some days. But when incidences increased, wells were dug up by the side of the main thoroughfare, and sellers and drinkers were thrown into these prisons. Barani writes that due to their narrowness, many died of suffocation while others were brought out half dead. The outer limits of the city and suburbs within a radius of thirteen and sixteen kilometres were declared completely dry. Yet in keeping with the prevalent position in Islamic law, Ala-ud-Din had ordered his officials not to harass those who privately distilled wine and consumed it inside their houses, who did not create a ruckus in public or show any rebellious intent.26 Thus, while the public selling of wine was banned, private consumption was not. This brings us to the question of shari’a and ambiguity in Islam as practised in the Delhi Sultanate. The sultan wanted a discussion on this with 24 Ibid.

25 Ibid., p. 155.

26 Ibid., pp. 172–3.

561

raziuddin aquil

theologians to confirm that his rules were within the ambit of Muslim law. The nature of the punishments meted out to the rebels or sellers and drinkers of wine was among the topics he discussed with a leading jurist, Qazi Mughis of Bayana. Barani presents a detailed account of the dialogue between the sultan and the qazi, even if much of it might be Barani himself speaking at the behest of the latter. It reveals that even though men of religion were happy with the thoroughly irreligious sultan speaking for Islam, the latter needed to be educated about what was recommended in Islam and what was prohibited. Qazi Mughis characterised most of the sultan’s actions as un-Islamic, with continuous reiteration and qualification that what he was saying was written in the law books and not his personal opinion, which was that the sultan was doing a great job. This concluding comment was needed – as Barani reminds us, the sultan listened to all that with a sword ready in his hand to cut the qazi to pieces! After listening to the qazi’s opinion on several matters relating to conquest and governance, such as the imperial rights on war booty, collection of revenue and corruption in administration, Sultan Ala-ud-Din enquired about the permissibility of his methods of punishment. Barani quotes the sultan as saying that he had ordered the deduction of three years’ salary of those cavalrymen who do not report for muster. He also threw the drunkards and wine sellers into the pits; men were castrated for adultery with someone else’s wife, and women put to death. Insofar as rebellions were concerned, he ordered a general massacre irrespective of good or bad, young or old, and also destroyed or reduced to destitution their children and wives. Outstanding revenues were extorted by blows, pincers and beatings, and as long as even one jital remained unpaid defaulters were kept imprisoned, in chains and under guard. Political offenders were held under permanent incarceration. The sultan wanted to know: would you say all this is unlawful?27 The qazi submitted that whether the king of the world let him live or ordered him to be killed then and there, he should still know that all that was illegal; neither the traditions of the Prophet nor the observations of ulama could endorse a ruler resorting to any such measure for conducting his administration. On hearing this, the sultan left the hall in a hurry. In the meeting which took place the following day, the sultan generously rewarded the qazi for speaking truth as exemplified in the law books, despite the threat to his life. The qazi himself had come prepared to be killed by the sultan, who was not known to respect any laws, customs or traditions.28 27 Ibid., pp. 176–9.

28 Ibid., pp. 179–80.

562

Genocidal Massacres in Medieval India

According to Barani, the sultan defended his actions by admitting that he had not acquired any knowledge or studied any books, and knew only a few verses of the Quran, but he was born a Muslim and was brought up among those who had been Muslim for several generations. He insisted that what he had decreed was in the interests of the state and the people, to ensure there was no rebellion and that people would not get killed. He had found that some people were indifferent to his orders, so it was necessary to adopt harsh measures to make them obedient. He added that he would not care whether his rules and regulations were lawful or not; he would decree whatever he found to be in the interest of the state and expedient, whatever his fate would be on the Day of Judgment.29 Elsewhere, Barani also noted that the superintendents of the markets ensured a strict compliance of the measures imposed by the sultan to control prices and check corrupt practices such as weighing less than the actual weight – in order that lowering of price did not mean exorbitant profit could be accrued by sleight of hand. The punishment for such cheating included something barbarous and beyond the imagination. Punishment for a quantity that was less than the actual weight was as follows: double the quantity was deduced from the shop owner who had cheated the buyer in flesh cut from the buttocks and other body parts. Market officials were supposed to be honest men, men of confidence, but they were also harsh, illtempered and merciless, used to much beating by feet and sticks, and other kinds of torture, besides chaining and imprisonment.30 Qazi Mughis had already pointed out that the cutting of hands of corrupt officials in the sultan’s administration was unlawful; they could perhaps be fined or imprisoned.31 Towards the end of his discussion on Ala-ud-Din’s career, Barani notes that, despite the sultan being a strange kind of man, and in spite of the indifference that was ingrained in his nature, God had adorned his kingdom with rare talents and outstanding personalities. Barani also highlights the fact that the capital city of Delhi, with its vast suburbs, was also blessed by the Chishti Sufi saint, Khwaja Nizam-ud-Din Auliya, residing as a living legend with his many accomplished disciples, counting himself and seniors Amir Khusrau and Amir Hasan Sijzi. However, according to Barani, when worldly fortune turned its face from the sultan and felicity felt tired from his attendance, time demonstrated its old tradition of perfidy on him and the treacherous sky took out its dagger to overthrow him. The sultan also committed

29 Ibid., p. 180.

30 Ibid., pp. 193–4.

31 Ibid., p. 178.

563

raziuddin aquil

Figure 21.2 Tughlaqabad fort. (Photograph from Getty Images)

certain acts of omission and commission that turned out to be the cause of the destruction of his kingdom.32 Within four years of the death of Ala-ud-Din, his Khalji dynasty came to a shocking end when one of his sons who had usurped power, Sultan Qutbud-Din Mubarak Shah (r. 1316–20), was brutally killed by a close Hindu associate. The latter declared himself sultan but was soon defeated and killed by an army led by a Tughluq Turkish officer, who was known for his successful defence of the northwestern frontier region of the Delhi Sultanate against the Mongol onslaughts. This man went on to establish his own Tughluq dynasty, ascending the throne of Delhi as Sultan Ghiyas-udDin Tughluq (r. 1320–5) (Figure 21.2). However, Ghiyas-ud-Din died under mysterious circumstances, ostensibly due to the collapse of a newly constructed building in a village near Delhi. The building had been constructed on the orders of his ambitious son and successor, Sultan Muhammad bin Tughluq (r. 1325–51), who was to become known in history as a mad ruler who arbitrarily issued hundreds of orders and killed thousands of people for not agreeing to his rule, which lasted for over a quarter of a century. * 32 Ibid., p. 226.

564

Genocidal Massacres in Medieval India

Sultan Muhammad Tughluq’s outrageous farmans (orders) are invoked every time people in power come up with ideas that radically change the system and adversely affect the populace. The sultan’s eventful rule offers important insights on how brutal measures adopted to implement even some very good ideas can get completely discredited. The people rebelled against the despot because of the severity of his actions; and he severely punished those who defied his authority – the more the people resisted, the more they were chastised. At the height of his power, Muhammad Tughluq presided over a large subcontinental empire. Condemned as somewhat devilish, the charges levelled against him include: the all-knowing sultan would not listen to any advice, no matter how good and well-meaning; he would brook no criticism nor tolerate any opposition; opponents could be suppressed and removed in the most ingenious ways; perhaps he had no qualms in killing his own father to capture power, though no judge could have proved the charge; finally, the grand ambition that goaded him to pursue outlandish projects proved to be his undoing, both because he was thinking ahead of his time and also trying to emerge as a world conqueror. Having subdued the whole of India, the monarch wanted to conquer the Himalayas – the entire mountainous region between India and China. This would enable him to break into central Asia, conquering Khurasan and west as far as Iran and Iraq. A huge amount of money was spent in mobilising the army, but the project proved unfeasible. Hill chiefs who controlled the territories now in Himachal, Kashmir and Ladakh foxed the Delhi army into a deathtrap; hundreds of thousands of horsemen were deployed, but few returned to the capital to break the news of the massacre. The money for the disastrous campaign was collected through two related measures: implementation of a heavy tax regime and introduction of token currency of brass and copper, instead of usual gold and silver. Both hard-working agriculturalists and cautious traders were badly hit by the move. A couple of years of poor monsoons further aggravated the crisis, with famine raging and scarcity of food and fodder causing starvation and death. Minting and circulation of a large quantum of fake copper coins also adversely affected trade and commerce. The failures of these schemes brought widespread unrest. The somewhat neglected southern India demanded special attention, which the king thought could be tackled better from the formidable fort of Devgiri – renamed Daulatabad and announced as the new capital. Delhi’s power elite was ordered to be relocated, causing much anxiety and hardship. Like other schemes, this failed. Delhi’s power was considerably diminished, even as 565

raziuddin aquil

the despot could not hold on to his control over Tamil country; soon, the Kannada- and Telugu-speaking regions were usurped by the emerging empire of Vijayanagara and northwestern Deccan saw the rise of the Bahmani kingdom. By the end of his rule, the sultan faced revolts in Maharashtra and Gujarat – his last bastions. Bengal had already gone its own way. Through all this, Muhammad Tughluq continued to placate religious leaders, philosophers, foreign dignitaries and the hitherto marginalised, besides seeking testimonials from the caliph in Egypt – for a while showing great deference to his moral authority, which baffled the public for they thought the sultan was beyond any control. When disgusted, he even contemplated vacating his seat of power in favour of his cousin Firuz (who eventually succeeded him), and making a pilgrimage to Mecca, but he was not one to give up easily. Disobedience to the autocrat was interpreted as defiance of the authority of God, which demanded severe chastisement. Therefore, instead of focusing on policy and governance, he applied himself excessively to the business of punishment; none could escape his wrath. Treatment ranged from simple beheading to more spectacular skinning and cutting into pieces to make veritable meat pilaf for serving to elephants, who were expected to revolt in disgust, for elephants were supposed to be vegetarians. Our author Ziya-ud-Din Barani served Muhammad Tughluq as a boon companion (nadim) for over seventeen years, and thus not only witnessed political activities at close quarters but was also party to privileged conversations with the sultan on his military campaigns, statecraft and governance. The Tarikh-i-Firuz Shahi was written within five years of the death of Sultan Muhammad, so Barani had the freedom to say what he could not say to his ruler in person. He was however careful not to antagonise his cousin and successor, Sultan Firuz Shah Tughluq (r. 1351–88). Firuz Shah was brought up like a son by his uncle Sultan Ghiyas-ud-Din, and cherished the memories of his close association with Sultan Muhammad. Yet Barani wrote enough in his Tarikh-i-Firuz Shahi to illustrate the large-scale killings and massacres under Sultan Muhammad. Much of it can be corroborated by two other contemporary writers, Izz-ud-Din Isami in his Futuh-us-Salatin and Ibn Batuta in his travelogue, Rehla.33 Barani’s account is important as it also presents recollections of his conversations with Muhammad Tughluq, his justification of his 33 For Ibn Batuta’s account of Hindus hiding in forests and hills to protect themselves, see the Urdu translation of his travelogue, Safarnama Ibn Batuta, pp. 147–8.

566

Genocidal Massacres in Medieval India

actions and Barani’s own advice to him, though he has also qualified that the sultan did not listen to him, or indeed to none of his advisers. That is not entirely true. While reporting on the widespread incidents of sedition and revolts which arose in different regions, which required the sultan personally marching to suppress them, Barani reports that after crushing a revolt in Multan the ruler wished to mete out dire punishment to the entire population of the frontier town. The local Suhrawardi Sufi saint, Shaikh Rukn-ud-Din, interceded with the sultan on behalf of the inhabitants of Multan. The sultan accepted the intercession of Shaikh Rukn-ud-Din in favour of the people of Multan and did not order their punishment.34 However, such intervention was not easily available. Large parts of upper north India witnessed chastisement of rebels who were chased down to the last man while running to hide in jungles. Such actions appear to be mainly directed against rebellious governors and their soldiers, though sometimes the general population resisting imperial regime also suffered. The example of widespread popular resistance against excessive revenue demands in Doab, large tracts between the rivers Yamuna and Ganges, may be mentioned here. The Hindus set the grain harvest on fire and drove their cattle out. The sultan ordered local administrators to ransack and plunder the region, and they killed some village officials and blinded others. Those who were spared got together and sought refuge in the jungles. In Barani’s understated criticism, the region of Doab was ruined. According to him, during those days the sultan went on a hunting expedition to Baran and on his orders the entire region of Baran was ransacked and plundered. The heads of Hindus were brought and hung from the parapets of the fort of Baran.35 Barani offers some insights into the workings of the mind of the sultan through glimpses of his conversations with him. Indeed, he notes that the sultan once consulted him and enquired: you have read much history, have you seen anywhere, in what kind of situations, earlier kings had resorted to killing? Barani replied that he had read in Persian histories that the king cannot rule over the people without taking resort to capital punishment. If the king did not inflict ultimate punishment when needed, God alone would know what kind of calamities may occur due to the recalcitrance of the headstrong and unruly people and what forms of impiety and sinfulness would appear from the subjects. Barani also added that, in this regard, a confidant of the ancient Iranian ruler Jamshed had asked the king what kind of crimes deserve capital punishment. Jamshed replied that there were 34 Tarikh-i-Firuz Shahi, p. 295.

35 Ibid., p. 295.

567

raziuddin aquil

seven crimes for which capital punishment was appropriate, but going beyond them would lead to disorder and chaos, eruption of calamities and troubles, and damage to the state.36 It may be worth listing them here to illustrate how the sultan ignored Barani’s advice. Barani quoted Jamshed mentioning the first kind of crime as apostasy from the true religion; a person who persisted with that should be killed. Second, one who kills a law-abiding subject would also be killed. Third, one who has a wife and commits adultery with another woman. Fourth, one who contemplates treason against the king and their treason is proved. Fifth, one who rebels and remains engaged in rebellion. Sixth, a subject of the king who enters into friendship with enemies of the king. Seventh, one who commits disobedience to the king that causes damage to the state would likewise be killed.37 The sultan enquired how many of these cases were approved in the tradition of the Prophet and how many concerned the discretion of the king? Barani submitted that three of the seven crimes which warranted capital punishment were mentioned in the tradition of the Prophet: apostasy, killing of a Muslim, and adultery by a married man. The other four concerned the discretion of kings. Barani also mentioned that Jamshed observed that the reason why rulers appointed wazirs, raised them to high status and entrusted their kingdom and administration to them was that they implemented rules and regulations, so that rulers did not need to personally stain their hands with the blood of any creature.38 The sultan replied to Barani that the use of capital punishment about which Jamshed had spoken applied to earlier times. In his own era, the sultan asserted, there were very large numbers of wicked and disobedient people and he killed them under suspicion of rebellion, treason, wickedness and treachery. He would continue to kill them till he perished or until they stood corrected and renounced their disobedience. Their hostility had to be checked by him personally, as he had no wise wazir who could invent necessary regulations and implement them strictly.39 Barani continued in his service, even though he regularly witnessed the execution of large numbers of ulama, Sufis, Syeds and Qalandars, besides clerks and troopers on mere suspicion of hostility and disobedience. Barani admitted not only being greedy for all the gold he received; he was also afraid of dying by speaking the truth before the sultan. The bigger sin he committed was connivance with the sultan and render help in contravention of religious sanctions and narrating inauthentic religious traditions in support of his 36 Ibid., p. 313.

37 Ibid., pp. 313–14.

38 Ibid., p. 314.

568

39 Ibid.

Genocidal Massacres in Medieval India

actions. The author cursed himself for his fall from grace under the next sultan when he was writing his books, away from the court, as punishment for what he had done under Muhammad Tughluq, waiting for the bigger punishment that lay in store for him after his death.40 Barani also wrote an autobiographical narrative, and appropriately entitled it an account of unfulfilled desires, Hasratnama.41 As a nadim of the sultan, recollecting the extraordinary qualities of the latter, Barani refers to his learning (not only traditional knowledge but also rational sciences including philosophy), ambition, discernment, sagacity, valour, generosity, skill and wisdom. According to Barani, these were contradicted by the sultan’s arbitrary decisions, extreme irascibility and hardheartedness, which led to the killings of so many people, including chaste and innocent Muslims, and the decline of the kingdom and a general sentiment of aversion against him. The large empire he had inherited, conquered and ruled came crumbling down by the end of his reign.42 Within months of his death, during his last campaign to suppress rebels in Gujarat and Sindh, Muhammad Tughluq explained his position in clear terms. Barani quoted him as saying (in a comment which also sums up the sultan’s actions all through his reign) that the people had fully understood his temperament and he was acquainted with their conduct. He was tormented and they were aggrieved. Whatever treatment he tried, the medicine simply did not work. For him, the cure for rebels, the disobedient, opponents and illwishers was the sword. As long as the people resorted to opposition, he would resort to greater punishment.43 The matter ended with the sultan’s peaceful end, caused by an illness which could not be cured during a campaign to suppress rebels at Thatha, in Sindh.44 * Indian medieval conquests and empire-building involved rampant violence, whether in pitched battles or drawn-out wars. Medieval historian Ziya-udDin Barani’s accounts of the reigns of three powerful Delhi sultans – Ghiyasud-Din Balban, Ala-ud-Din Khalji and Muhammad Tughluq – show not only that opponents resisting their conquests or regimes were brutally killed in large numbers during armed combat, but also that many of those captured alive were later massacred in cold blood. The women and children of soldiers 40 Ibid., p. 287. 41 Aquil, Muslim Question, p. 51. 42 Tarikh-i-Firuz Shahi, pp. 286–7. 43 Ibid., p. 321. 44 Ibid., pp. 322–3. For a recent study of failures of Muhammad Tughluq and relatively successful reign of his successor Firuz Shah, see Vasileios Syros, ‘State failure and successful leadership in medieval India’, Studies in History 37:1 (2021), 7–25.

569

raziuddin aquil

from the other camp were not only imprisoned and enslaved but also bodily harmed, often killed; women were also raped as punishment for the alleged crimes of their male relatives. The perpetrators of such violence were led by conquerors and rulers of Turkish Muslim antecedents, who invoked the Quran, traditions of the Prophet Muhammad and the contemporary caliphs in Iraq and Egypt in whose names they ruled. Their armies were called lashkar-i-islam, and campaigns against non-Muslim opponents were referred to as the holy war, jihad. If the victims were Muslims, whether of the same Turkish race or neoMuslims of Mongol background, Muslim intellectuals and writers such as Barani would point out that shedding of such blood was un-Islamic. On such occasions, the rulers presented themselves as working above the dictates of Islam and in doing so sought legitimation from examples from the careers of ancient Greek and Iranian rulers. Clearly, these rulers put themselves above the law, even as they continually sought endorsement of their violence from the law to which they themselves wanted their people to adhere. This meant that conventional Muslim law, shari’a, which was frequently referred to, was set aside by rulers and ideologues to justify political actions. Instead, new arbitrary regulations, zawabit or uslub, were invented, the failures of which provoked rulers to cynically punish the people who were perceived as disobedient. Following dictums such as ‘obey God’, ‘obey the Prophet’ and ‘obey those in authority among you’, and adopting titles such as Shadow of God on Earth, Zillullah, the ruthless sultans interpreted any defiance of their authority as defiance of God’s will, which was considered both a major sin and a serious criminal act. In dealing with any violations of their authority, the rulers also claimed to be following the Hanafi school of Sunni jurisprudence. When opponents were non-Muslims, say Hindus, they could be condemned as kafirs and thus any amount of brutal action against them would be considered lawful. In Islamic theology, kafirs resisting Muslim political authority, peacefully or otherwise, were to be either killed or forcibly converted to Islam or forced to migrate. Yet none of these options were systematically exercised by the Delhi sultans. From the time of the Arab conquest of Sindh (711–12 C E), Hindus of the Indian subcontinent were treated under Islamic regimes as protected people of the book, zimmis, who could live peacefully by paying a humiliating and arbitrary tax, called jizya. The political processes of the Delhi Sultanate in the thirteenth and fourteenth

570

Genocidal Massacres in Medieval India

centuries also indicate that large numbers of non-Muslims occupied powerful positions with or without conversion to Islam. In terms of support or opposition to a particular regime, conversion to Islam did not matter. Muslim rebels, their soldiers and families were suppressed in the most brutal manner and in complete disregard of suggestions that a sultan ruling in the name of Islam shedding so much Muslim blood was completely unjustifiable. The rulers sometimes listened to the intervention of Sufis and other men of religion for mercy, but in general the rebels, whatever their religion, race or sectarian backgrounds, were ruthlessly punished, and new methods were continually devised. Such action served to chastise opponents and as a deterrence to others. Those seeking to justify violent actions of the sultans would also refer to the contradictory qualities of God, both beneficent and wrathful. In being what they were, for Barani, the sultans were either divinely inspired or magically or diabolically dictated. In the end even someone like Barani, from a politically endowed family who enjoyed proximity to Sultan Muhammad Tughluq and advocated violence and massacres of rebels, in his last years regretted his complicity in the terrible violence wrought out of greed for power. He feared retribution for his justification of all the violence, both in this world and in the hereafter. Violence could be either justified or condemned, but peace was considered to be the most recommended practice for an agreeable resolution of political crises. Amid all the violence or cries for peace, the need for justice was continuously invoked. Without justice, there would be anarchy. Medieval intellectuals and ideologues feared the dismantling of law and preferred a hundred years of despotism over a day of anarchy on the streets of the city. It was this fear of anarchy that erroneously forced people to submit to the authority of tyrants who thrived on violence. Those who resisted or rebelled were subjected to genocidal massacres.

571

22

Mass Extermination in Prehistoric Andean South America danielle kurin

Striking moments of genocidal mass extermination have punctuated the lengthy prehistoric sequence of Andean South America, an area that encompasses modern-day Ecuador, Peru, Bolivia and northern Argentina and Chile (Map 22.1).1 Yet the societies that flourished in this region for more than eight millennia left no written record detailing the motives and modus operandi of their deadly confrontations. Accurately gauging the intensity and intent of violence – in particular mass extermination – among ancient peoples requires an approach sensitive to forensic skeletal evidence that reveals the cause and manner of death, as well as the nature of corpse disposal or burial, which informs us of how bodies were treated after death. Arguably, the most sobering precepts of genocide include causing serious bodily harm and death to members of a targeted group, and the infliction of conditions calculated to bring about a group’s destruction. These precepts were familiar to people of the ancient Andes. Among terms used by the Inca to describe annihilation via excessive violence was huañuchiyta atipani, a Quechua phrase that connotes the utter and absolute power to manufacture death.2 For the Inca and their ancestors, the creation of destruction was calculated and comprehensive. Complete death was produced through slaughter of healthy and strong fighters, as well as the torture and murder of civilian noncombatants, including the very old and the very young, the sick and frail, and the unarmed and incapacitated. Outright obliteration was also wrought by pillaging rival villages, through abductions and captive-taking, and by ravaging village stores and herds. More insidious tactics included pulling crops from the root with the understanding that having no food 1 E. Arkush and T. A. Tung, ‘Patterns of war in the Andes from the Archaic to the Late Horizon: insights from settlement patterns and cranial trauma’, Journal of Archaeological Research 21:4 (2013), 307–69. 2 A. Durston, Pastoral Quechua: The History of Christian Translation in Colonial Peru, 1550– 1650 (South Bend, IN: University of Notre Dame Press, 2007).

572

Mass Extermination in Prehistoric Andean South America

Map 22.1 Andean South America showing sites of prehistoric mass extermination. 1. Pacatnamú; 2. Túcume; 3. Huanchaquito-Las Llamas; 4. Chan Chan; 5. Kuelap; 6. Huaca de La Luna; 7. Cerro Sechin; 8. Lima; 9. Punta Lobos; 10. La Paloma; 11. Amato; 12. Huari; 13. Andahuaylas; 14. Cusco; 15. Mt. Ampato; 16. Tiwanaku; 17. Chinchorro; 18. Mt. Llullaillaco. (Cartography by Danielle Kurin)

imperils livelihoods in the long-term. All told, violent episodes of eradication in the Andes may best be understood as genocidal moments, outbreaks of local communal strife, limited in both place and time.3 These moments capture the collective murder of large numbers of people who are specifically targeted 3 B. Kiernan, Blood and Soil: A World History of Genocide and Extermination from Sparta to Darfur (New Haven: Yale University Press, 2007), pp. 13–15.

573

danielle kurin

because they are members of a larger group. Using the physical and material record as a lens can help us identify catastrophic death assemblages, reconstruct the profiles of victims and perpetrators, discern the methods of murder and even deduce underlying motivations for attack.

Types of Evidence We Can and Cannot Access Investigating mass exterminations among societies that left no written record presents unique challenges. The archaeological record itself is incomplete. There are glaring geographic and temporal gaps in the literature. Many epochs – especially those deepest in time – remain under-studied, if at all. While Peru is among the most heavily investigated countries in the world, the cultural histories of nations of the more peripheral Andes, including Colombia, Ecuador, Bolivia, Chile and Argentina, and all of ancient Amazonia, are mostly unknown. A range of factors, including social instability in host regions, uneven academic support and even differential preservation, contribute to patchy coverage. And even the archaeologist’s most reliable tools have their limitations. For instance, radiocarbon dates are often not prevalent or sensitive enough to parse out history in years versus decades or centuries. There is also a strong temptation for archaeologists to over-rely on potentially problematic historic chronicles that purport to be the authority on prehistoric peoples. Finally, while no less harmful than physical assault, manifestations of deprivation owing to structural violence leave only a faint skeletal footprint that can be hard to define in the prehistoric past. This chapter focuses instead on the highly conspicuous, direct, material and corporeal evidence of mass murder and shared victimology that has survived the passage of time.

Skeletal and Burial Indicators of Mass Extermination Proof of genocide rests on the intentions of the perpetrators. Yet inferring criminal intent and deliberate actions among prehistoric groups engenders certain challenges. Many of these uncertainties may be overcome using a bioarchaeological approach. Bioarchaeology is the study of human remains in contextualised prehistoric settings; it is deployed to reconstruct the lived experience of individuals, social groups and even entire populations.4 Human 4 D. L. Martin, R. P. Harrod and V. R. Pérez, Bioarchaeology: An Integrated Approach to Working with Human Remains, Manuals in Archaeological Method, Theory, and Technique (New York: Springer, 2013).

574

Mass Extermination in Prehistoric Andean South America

mummies and skeletons arguably provide the most direct evidence of the lived experience because the congealed history of a person is literally sedimented into their bones and teeth. The size and shape – and even the elemental and genetic composition of bones – are used to determine a skeleton’s sex and age at death, where they were born, what they ate, to whom they are related, the diseases from which they suffered, the ways they intentionally modified their bodies, and the physical trauma they experienced during their lifetimes and at the time of death (Figure 22.1). Following the cessation of biological life, bodies take on a social life. The dead do not bury themselves; the manner in which decedents are treated and the way corpses are disposed of reveals much about the attitudes of those still living. Combined, complementary avenues of skeletal and burial data are well suited to addressing instances of mass extermination in the prehistoric past. Archaeologists remain alert for violent patterns that hint at genocidal moments in particular.5 Foremost is evidence that death was catastrophic: neither random nor the result of natural population attrition over time. The deadly trauma on mangled bodies is the result of attacks and not accidents. Genocidal death assemblages consist of large numbers of individuals who are intentionally killed. The humanity of the dead fades further as funerary rituals and crucial burial customs are eschewed. The victims are individuals pursued because of a perceived affiliation or identity with a targeted corporate group. The archaeologist’s ability to identify and isolate groups of people based on their shared victimology informs us about experiences of interpersonal violence that range from domestic abuse to raids and ambushes, to human sacrifice, to all-out warfare. Likewise, violence perpetrated on a body after death – be it mutilation, trophy-taking or cannibalism – sheds light on the ethos of assailants regarding their attitude towards the victim in particular, and of culturally acceptable treatments of human beings and bodies in general. Clarifying violence in the past relies on the ability to identify trauma – broken and fractured bones – related to the circumstances of injury or death.6 Identifying the cause of death (e.g. blunt force trauma) and the manner of death (e.g. accident versus homicide) relies on several factors. A major one is trauma timing and lethality, which is classified using the 5 C. J. Knüsel and M. Smith, eds., The Bioarchaeology of Human Conflict: Traumatised Bodies from Earliest Prehistory to the Present (New York: Routledge, 2013). 6 V. L. Wedel and A. Galloway, eds., Broken Bones: Anthropological Analysis of Blunt Force Trauma, 2nd ed. (Chicago: C. C. Thomas, 2014).

575

danielle kurin

Figure 22.1 Diverse manifestations of deadly head wounds. Clockwise from top left: (A) successive assaults: healed ante-mortem trauma (black arrow) and lethal peri-mortem trauma (white arrow); (B) peri-mortem blunt force trauma with bone hinging; (C) ring fracture: execution blow to the cranial base; (D) a radiating fracture line (arrow) emanates from point of deadly blunt force impact; (E) overkill: symbolic, destructive, deadly polytrauma; (F) peri-mortem wounds from coincident blows (arrows). Note: skull elongation results from intentional cranial modification (head reshaping).

576

Mass Extermination in Prehistoric Andean South America

terms ante-mortem, peri-mortem or post-mortem. Ante-mortem trauma occurs before death and is identified by the presence of new bone growth, which allows broken bones to fuse back together, and signs of healing. Perimortem trauma, which occurs at or around the time of death and may be the cause of death, lacks any signs of healing. Instead, these lethal wounds are identified by the inward deformation of affected bone and internal bevelling at the point of impact. Like a spider’s web, radiating fracture lines emanate from the centre of a wound. Bits of bone remain hinged around the impact point, and there is colour continuity between the pieces of broken and intact bone. Finally, when the soft tissue that envelops a living skeleton is no longer present, bone behaves differently. Postmortem trauma, the type that occurs after death, can be identified because bones broken after death are of a different (usually lighter) colour than intact bone. Post-mortem fractures are irregular and defy the living laws of biomechanics. When possible, trauma is also characterised by the type of force applied: sharp, blunt or projectile.7 Sometimes fracture patterns will correspond in size and shape to the type of force and weapon – bullet holes from smallcalibre ammunition are a good example of this phenomenon. Aside from force, wound location is crucial to infer whether injuries were accidental or intentional. For example, fractures that involve the collarbones, knees, ankles, elbows or wrists tend to be caused by accidents. In contrast, wounds to the head and face tend to result from violent attacks. Other signs of intentional harm to victims include broken ribs, nightstick fractures of the lower arm bones indicative of defence and cut marks on bone consistent with torturous mutilation or murder. Finally, in exceptional cases – such as finding an arrowhead embedded in a shoulder blade – there is clear evidence of violent (versus accidental) trauma. But more often than not trauma patterns are not perfect mirrors of violence in society. Cases of peri-mortem trauma are certainly undercounted because many people die of wounds that impact only the soft tissue, leaving no trace on the skeleton. Similarly, as broken bones heal, the bony evidence of injury itself may disappear completely over time. Nevertheless, skeletons from contextualised (versus looted or unprovenanced) contexts provide a uniquely personal perspective on the expression and frequency of violence among past peoples. 7 N. C. Lovell, ‘Trauma analysis in paleopathology’, Yearbook of Physical Anthropology 40 (1997), 139–70.

577

danielle kurin

An Overview of Eight Millennia of Violence Instances of mass killings in the South American Andes have punctuated prehistory,8 although the assailants, victims, methods of attack and motivations for assault have differed. Some of the earliest direct evidence for violence consistent with interpersonal conflict is from the coastal deserts of northern Chile, among the Chinchorro people (7000–1600 B C E). In burial sites there, a quarter of adults – mostly males – showed healed skull fractures tending to the left side of the head, a result of assault by a right-handed opponent. Mass killings also have deep roots. Some 7,000 years ago at a site called La Paloma, on Peru’s central coast, the skeletons of viable infants were intentionally entombed beneath house floors in one of the earliest known case of ritualised infanticide. The earliest indirect or circumstantial evidence of mass extermination was uncovered farther north on the Peru coast, at a site called Cerro Sechín. Six thousand years ago, skilled craftsmen created bas-relief carvings on dozens of gigantic stone slabs. These monoliths formed the facades of monumental temple pyramids. The engravings vividly depict bound and naked captives being violently dismembered. Headless bodies, decapitated heads and dissected body parts are flung about as victors make trophies of mutilated corpses. However, this sensational spectacle may be more myth than history, as we have yet to find strong skeletal signatures of mass murder at Cerro Sechín. New forms of violence emerged as large chieftaincies began to coalesce 2,000 years ago.9 The earliest direct skeletal evidence of mass killings dates to before 100 C E. This formative era in the Andes witnessed raids, ambushes and occasional village massacres. The broken skeletons of victims, bludgeoned, mutilated and left where they fell, attest to the intent of the perpetrators: the extermination of an enemy group. The underlying motive: usurpation of lands and resources. By the third century C E, Moche warriors were targeting men from rival tribes for capture and sacrifice. The killings were an outcome of politically competitive warfare. Finally, around 600 C E, the multitude of small chiefdoms that dotted the region were subsumed by the expansive highland Andean sister-states of Wari and Tiwanaku. Conquest, revolt and the suppression of rebellions motivated imperial armies to displace entire

8 M. Moseley, The Incas and Their Ancestors (London: Thames & Hudson, 2001). 9 M. S. Murphy and S. L. Juengst, ‘Patterns of trauma across Andean South America: new discoveries and advances in interpretation’, International Journal of Paleopathology 29 (2020), 35–44.

578

Mass Extermination in Prehistoric Andean South America

villages; nimble war parties were organised therein to fully exterminate groups of unruly subjects. The dramatic collapse of these archaic empires after around 1000 C E was a prelude to four centuries of internecine warfare in Ecuador, Peru, Bolivia and throughout northern Chile and northwestern Argentina. The dawn of this Andean dark age was characterised by social instability and exacerbated by a prolonged drought. On the coast, polities such as the Chimú deployed warriors who executed and unceremoniously disposed of rivals, sacked and burned towns, and defiled sacred sites. Contemporaneously, high in the Andean range, tribes including the Chanka engaged in widespread systematic killing sprees that overwhelmingly targeted ethnically-marked men, women, children, the infirm and the elderly. And in the cloud forests of the Amazon, fearsome groups like the Chachapoyas often fought among themselves. Violence throughout the Andean region subsided only in the fifteenth century, as regional polities, including the Chimú, Chankas and Chachapoyas, were vanquished by an unstoppable Inca army, led by a series of beguiling and ambitious emperors. Historical chronicles relate that, over the next century, the Inca implemented policies congruent with contemporary notions of genocide. The Inca deliberately destroyed sanctuaries and settlements, instituted a complex system of intentional mass displacement, controlled reproduction among women, murdered and transferred children from subservient groups, and orchestrated the maiming and murder of both formidable rivals and civilian non-combatants alike. Tragically, acts of annihilation increased in orders of magnitude after 1532, when an unscrupulous crew of Spanish ruffians routed what was to be the western hemisphere’s last autochthonous empire.

Contexts of Killing We find evidence of genocidal moments throughout the Andes – up and down the coast, from the cool southern highlands to the northeastern reaches of the Andean cordillera, which gently rolls down into the Amazon basin. In most cases, mass extermination campaigns were sporadic and highly localised, but in a few instances they were more systemic and widespread. In every case, there was indeed a sustained, purposeful action by perpetrators to physically destroy a particular collectivity, regardless of the surrender or lack of threat offered by the victims. So too, small groups of perpetrators made an outsized impact. Although the nature of mass extermination varied in the Andes, it was most often made manifest in the context of powerful sacrificial 579

danielle kurin

practices, politically motivated massacres and moments of endemic ethnocide.

Human Sacrifice: Mass Murder and Extermination Human sacrifice – death by ritualised and intentional killing – took many forms in ancient South America. Many of the best preserved and clearest cases of mass human sacrifice are from the north coast of Peru.10 Individuals were killed and buried as dedicatory offerings to consecrate monumental temples and pyramids; blood and bodies were offered as gifts to appease local deities. Victims of sacrifice varied. Sometimes, they were women of some status, who were sacrificed as retainers entombed alongside important priests and leaders to serve as their vassals in the afterlife. In other cases, they were captives taken in raids and battles. Prisoners were sacrificed by skilled executioners who murdered victims in standardised ways as part of highly choreographed ritual pageants. The corpses of sacrificed victims were the subject of curation, manipulation, and mutilation before being tossed into mass graves. Moche Temple of the Moon

Beginning around 300 C E, the Moche chieftaincy of north coastal Peru began to expand in earnest.11 The society was predicated on militarism. Moche soldiers sought to seize new lands in neighbouring valleys and subjugate the people living there. Consequently, concerns about community defence lead to decisive action. Every settlement and refuge site in the region was fortified with high walls, parapets and caches of sling stone weapons. Even the Moche artistic canon is flush with images of armed fighters in battle. Mass-produced, mould-made portrait jars and finely painted vessels commemorated renowned warriors and victorious religious leaders. Distinct patterns in combatants’ dress and face paintings suggest distinct social groups. Painted scenes of combat include weapon bundles comprised of mace heads, shields and crescent-shaped Tumi knives. Other tropes include images of limbs and trophy heads suspended by rope, accompanied by scenes of dancing skeletons and prisoners of war. Hapless captives are depicted as nude prisoners with ropes about their necks, presumably awaiting the executioner’s knife. 10 H. D. Klaus and J. M. Toyne, eds., Ritual Violence in the Ancient Andes: Reconstructing Sacrifice on the North Coast of Peru (Austin: University of Texas Press, 2016). 11 J. W. Verano, M. Tufinio and M. Valle Lund, ‘Esqueletos humanos de la Plaza 3C de Huaca de la Luna’ in Proyecto Arqueológico Huaca del Sol y de la Luna: Investigaciones en la Huaca de la Luna 2001, ed. S. Uceda, E. Mújica and R. Morales (Trujillo: Universidad Nacional de la Libertad, 2008), pp. 217–46.

580

Mass Extermination in Prehistoric Andean South America

Artistic portrayals of violence sketched on pottery and pyramid walls were not just propagandistic or idealised representations of ritual combat, but depictions of actual historic events. For instance, within the inner sanctum of the imposing Moche Temple of the Moon, more than 100 young adult males were periodically tortured, sacrificed, mutilated and discarded. Prior to their unceremonious disposal in the mud of temple floors, these men had been fighters. Broken bones throughout their bodies – war wounds – in various stages of healing are direct evidence that these young troops experienced episodes of conflict prior to – and possibly just after – their capture. Their deaths were executed by priests impersonating Moche deities and the objective was to draw blood. Some victims were bludgeoned by starred mace heads or clubs. Most victims also had their throats cut, and some were completely decapitated. Their treatment is consistent with prisoners of war who are killed in action. Yet there was no solace in death for victims of Moche mass killings. Decomposing corpses were manipulated in jarring ways. Skeletons were covered in lacerations yet still fully articulated. Cut marks on the bones were located at muscle attachment sites and not at the joints. Bones would have still been held together by ligaments, which indicates that the purpose of mutilation was to deflesh the body, and not dismemberment. We do not know what happened to the flesh of captives, but the goal at this Moche abattoir was not cannibalisation. Patterns in butchery are unlike those observed in animals. There are no chop marks like those found in cuts of meat, nor are there bone breaks consistent with marrow extraction, and no signs that the bones were cooked or eaten. Rather, perpetrators were focused on flaying the bodies and using the defleshed skeletons as trophies for public display. Ropes still preserved around the necks, wrists and ankles of these articulated skeletons would have been used to suspend the stripped corpses like gruesome marionettes. Túcume Shrine of the Sacred Stone

Situated about 250 kilometres north of the Moche capital by way of the sinuous Pan-American Highway, and about 20 kilometres inland, is the monumental archaeological complex of Túcume,12 comprising twenty-six pyramidal platform mounds. Construction first began around 1000 C E, and as 12 J. M. Toyne, ‘Ritual violence and human offerings at the Temple of the Sacred Stone, Túcume, Peru’ in Living with the Dead in the Andes, ed. J. Fitzsimmons and I. Shimada (Tucson: University of Arizona Press, 2015), pp. 173–99.

581

danielle kurin

different states – first the Chimú around 1350 C E, then the Inca a century later – conquered the region, they built new pyramids and remodelled existing architecture. Within the Temple of the Sacred Stone, archaeologists uncovered the clustered remains of 116 individuals entombed in small chambers under the temple floor. Bodies were inhumed sequentially over the course of a hundred years. Nevertheless, the skeletons all exhibit the same complex pattern of execution and mutilation. Chest cavities were sliced open in uniform fashion, and extensive cut marks on neck bones indicate victims had their throats cut and were eventually decapitated. The victims also display similar patterns of post-mortem manipulation. The headless and mangled bodies were each deliberately posed in a seated fetal position and carefully deposited into the sub-surface tombs. Sometime well after death, the victim’s head was tossed in with the rest of the body. There was no effort to properly reposition the skull anatomically. Rather, heads literally rolled, settling uncomfortably on tomb floors. In the laboratory, bioarchaeologists found that all the victims were male. Furthermore, although the victims ranged in age from four years to their late forties, the majority of the dead were young adults who died between the ages of twenty and thirty-five. Bones of the very young and the very old were conspicuously absent from the assemblage. The century of sustained ritual activity and a uniform tradition of human sacrifice and treatment of the body after death characterise this example of massive serial murder. Extermination: Chimú Children at Huanchaquito-Las Llamas

The recently discovered Huanchaquito-Las Llamas site,13 near the modern coastal city of Trujillo, in northern Peru, is home to the largest single mass extermination of children known from the New World. The mass grave is just a kilometre west of Chan Chan, the capital city of the Chimú polity. The sprawling and labyrinthine twenty-kilometre-square mud-brick metropolis was built between the tenth and fifteenth centuries C E, and retains an imposing presence even today. Burial practices and radiocarbon dating from Huanchaquito-Las Llamas reveal a single mass killing event where almost 140 children were killed (along with three adults and over 200 juvenile llamas and alpacas), sometime around 1450 C E. The children range from approximately five to fourteen years of age, 13 G. Prieto et al., ‘A mass sacrifice of children and camelids at the Huanchaquito-Las Llamas site, Moche Valley, Peru’, PLoS One 14:3 (2019), e0211691, https://doi.org/10 .1371/journal.pone.0211691.

582

Mass Extermination in Prehistoric Andean South America

and DNA analysis indicates that both boys and girls are present in the assemblage. All showed a single transverse cut through the sternum. The breast bone cuts are uniform across the sample in their location, angle and direction. Around a dozen children also showed cut marks on the external surface of the third or fourth rib, near the armpit. Many of the children had ribs which were visibly displaced, indicating that the bones were manually spread apart as the chest was opened by force. This bisection of the thoracic cavity is unique among methods of sacrifice in the Andes. Ripping open the chests of the children by transverse sectioning proved an efficient way to remove the heart. The lack of hesitation cuts or chatter marks, indicative of false starts, points to an experienced perpetrator with a skilled hand. The programme of execution was uniform, and there were no other injuries that occurred at or around the time of death. On site, excavators took small bone samples from each child skeleton. In the lab, archaeo-chemists dissolved away all the inorganic material in the bones, leaving a collagenous soup of organic material, nearly ready for elemental and compositional testing. The prepped samples underwent stable isotope analyses of carbon and nitrogen, two elements that are used to infer and reconstruct the types and relative amounts of plant and animal food sources that a person ate during their lifetime. Results for the sacrificed child victims yielded a diversity of diets, which suggest that victims were selected from distinct geographic regions with unique ecosystems and food webs. Moreover, differences in head shape – a consequence of artificial cranial deformation and a marker of local group identity – are also consistent with a scenario whereby children from distinct ethnic groups were drawn from distinct territories rather than from a single local population. The sacrifice of hundreds of children (and animals) conscripted from the farthest reaches of the Chimú realm constituted a significant investment. The proximate motivation for the mass execution may have been a period of intense rainfall and widespread flooding caused by a major El Niño event. This is suggested by the presence of a mud flow covering on the burials as well as of footprints made in the muck while still wet. Normally, coastal Andean South America is a perpetually arid region, with negligible rainfall. Yet in this case, the mass execution of children during downpours and mudslides may have been a desperate attempt by Chimú priests to offer bodies of great value to the weather gods in order to mitigate the devastating sociopolitical, economic and ideological impacts of a major natural disaster. 583

danielle kurin

Genocidal Massacres and Killing Sprees Massacres or killing sprees in the Andes describe the slaughter of large groups of people (dozens to hundreds).14 These murderous sprees are often the result of surprise attacks, ambushes and raids by a small group of male perpetrators coming from outside or rival communities. Victims tend to be individuals of all ages and both sexes. Over half will exhibit lethal peri-mortem trauma on the skeleton – especially on the head and face. Often, corpses are haphazardly dumped in mass graves. Bodies may be contorted or piled one atop another and there may be signs of animal scavenging. Particular body parts may be conspicuously absent – the result of trophy-taking by assailants. Unlike regular funerary rituals, the massacred dead are not afforded the rites and offerings normally enacted by mournful descendants. Amato Village Slaughter

The end of an era known as the Formative Period (400 B C E – 100 C E) was a time of emerging social stratification, elite rivalry and competition for scarce resources. It is characterised by heightened violence in much of Andean South America. In the Acari valley, about 600 kilometres south of Lima, Peru, archaeologists discovered dozens of decapitated bodies discarded inside a centrally located windowless building within the ancient walled village of Amato.15 The bodies belonged to 47 men and women and 24 youths – including several infants, who were murdered en masse around 70 C E. In addition to the 71 missing heads, cervical bones of the uppermost neck were also absent. The vertebrae that remained displayed cut mark patterns consistent with mechanical removal of the head. Other bony signs of murderous intent are equally dramatic. For instance, many of the dead exhibit a bone-crushing injury on the lower arm bones known as a parry fracture. This particular fracture occurs when a victim raises their arms in a vain attempt to deflect an oncoming blow from a club, a nightstick or some other heavy blunt object. These defensive wounds are not accidental, but rather result from face-to-face confrontations that ended in death. Several Amato skeletons had fusing fractures enveloped by a thin bony callous – a type of scarring indicative of short-term healing. Yet most also displayed unhealed 14 M. D. Vega, ‘A history of violence: 3,200 years of interpersonal and intergroup conflicts from the Initial to the Early Colonial periods in the Peruvian central coast: a bioarchaeological perspective’ (PhD thesis, University of Western Ontario, 2016). 15 L. M. Valdez, ‘Walled settlements, buffer zones, and human decapitation in the Acari Valley, Peru’, Journal of Anthropological Research 65:3 (2009), 389–416.

584

Mass Extermination in Prehistoric Andean South America

peri-mortem wounds that occurred at or around the time of death. The timing of the bone breaks indicate that victims literally fought for their lives. In the moments before execution, the Amato villagers were stripped naked, corralled into a small, dark house, bound at the wrists and ankles, and summarily executed. Victims were buried where they fell, and lack any sort of orderly internment or sumptuary grave goods. These lines of skeletal and burial data all suggest that the village massacre at Amato was focused on annihilation; men, women and youngsters were all targeted for attack. Killing sprees of this sort were probably widespread. Indeed, every settlement in the Acari valley, including Amato village, is fortified and surrounded by a buffer zone and no-man’s-land. Village attacks were a constant threat; perpetrators were motivated to raid stores, abduct captives, viciously execute non-combatants and ultimately drive rivals from desirable lands. The extermination of rivals was intentional, and the procurement of victims’ heads as trophies served to memorialise the slaughter. Execution on Punta Lobos Beach

Punta Lobos is a beach cove on Peru’s central coast, about 300 kilometres north of Lima.16 There, archaeologists discovered the skeletal remains of over 180 men and adolescent boys. The youngest individual was about eight years old and the oldest was in his late forties; only the very old and the very young escaped an ignoble fate. Radiocarbon dating indicates the Punta Lobos massacre was a single mass extermination event that occurred around 1300 C E. Victims were blindfolded and physically restrained with bindings. We know this because the preservation at the site is excellent; archaeologists unearthed delicate cotton cloths covering the eyes and still tied around the head. Ropes had been wound about the skeletal remains of hands and feet. Sharp force trauma was observed on the upper neck bones of nearly all the Punta Lobos skeletons, indicating victims had their throats cut. Victims were also stabbed in the heart, as cut marks were found on the collarbones near the sternum, the first ribs, and on vertebrae low on the neck. Those few skeletons without cut marks etched across their bones were likely felled by mortal wounds to the soft tissue. In the frenzy to deeply lacerate individuals, multiple bones were cut. Far from surgical slices, tally mark-like cuts on 16 J. W. Verano and J. M. Toyne, ‘Estudio bioantropológico de los restos humanos del Sector II, Punta Lobos, Valle de Huarmey’ in Arqueología de la Costa de Ancash, ed. M. Giersz, Andes 8: Boletín del Centro de Estudios Precolombinos de la Universidad de Varsovia (Lima: Institut français d’études andines, 2011), pp. 421–46.

585

danielle kurin

bone are evidence of hesitation from a perpetrator’s unsteady hand. These chatter marks may also be the result of struggling by a desperate victim. Corpses were deposited across a large area of the beach and only superficially enshrouded by the beach’s wind-blown sand. The majority of the skeletons were found face down and positioned haphazardly with arms and legs akimbo. Human remains at Punta Lobos were not associated with any sort of mortuary architecture, grave markers or funerary goods. The lack of formal burial, and the fact that descendants did not come back and rebury the remains of loved ones, signals a symbolic indignation. It also suggests that women, infants and the elderly were no longer present in the area, having perhaps been dispersed as captives into slavery. Even if not, it is hard to imagine this community surviving the loss of its menfolk, likely an intended outcome. Leaving bodies to rot and the intentional abrogation of mortuary rituals aimed at steering the dead into the afterlife and ancestorhood was also a potent component of the perpetrators’ retribution. Based on the treatment of local men and boys before, during and after slaughter, scholars suspect that the Punta Lobos massacre was the result of a reprisal or retribution killing, or a political execution. The location, timing and profile of the victims, along with the brute force needed to corral and kill almost 200 men at once, fits the modus operandi of murder by the Chimú state. This major polity, based on Peru’s north coast, was in the midst of a militaristic expansion south that, for the people of Punta Lobos, would culminate in tragedy, seemingly a genocidal massacre. Reprisal Killing at Pacatnamú

Mighty as they were, things did not always end well for Chimú regiments who came upon new ethnic groups. Such may have been the case at the site called Pacatnamú,17 located at the mouth of the Jequetepeque river in northern Peru. There, archaeologists uncovered the skeletons of fourteen young men who were victims of a gruesome mass execution that took place during the fifteenth century. The condemned men were probably professional Chimú soldiers, given the multiplicity of healed war wounds they displayed from earlier conflicts. Against all odds, the local Pacatnamú villagers apparently captured and mercilessly slaughtered these invading enemies. The hapless victims were bound, lined up and murdered. The cause of death 17 J. W. Verano, ‘A mass burial of mutilated individuals at Pacatnamú’ in The Pacatnamú Papers, ed. C. B. Donnan and G. A. Cock (Los Angeles: Museum of Cultural History, University of California, Los Angeles, 1986), I , pp. 117–38.

586

Mass Extermination in Prehistoric Andean South America

was from lethal blunt force trauma and stab wounds to the chest. After death, their bodies were mutilated and tossed unceremoniously into one of the village’s many defensive trenches-cum-rubbish pits. Chachapoya Carnage at Kuelap

Travelling east over the Andean cordillera and rising magnificently out of the cloud forests is Kuelap, a large ridge-top settlement whose monumentally high walls protect 400 houses, administrative buildings and circular platforms that served as stages for public display. Kuelap was constructed and inhabited by the Chachapoyas people between 1000 and 1400 C E. When archaeologists excavated Kuelap,18 they uncovered a cache of over 2,500 sling stones. A plethora of broken stone axe-heads littered the site, a harbinger of the mass death that the researchers would soon find. At the site’s southernmost end, the remains of around 120 individuals were strewn about in grisly fashion. The skeletons were awkwardly splayed on floor surfaces. Corralled into the corners of rooms and corridors, dead bodies were piled one atop another. Victims were left unburied to slowly decay without dignity, lacking the grave goods and offerings an honourable death would merit. The only artefacts uncovered were everyday household items, located in situ inside the homes that had become impromptu tombs for the massacred. As a final insult, the walls of the buildings and corridors were forcefully pushed down on the bodies and the site was soon abandoned. Almost two-thirds of the Kuelap dead were juveniles, including many young children and several newborns. Few women were among the dead; they were vastly outnumbered relative to adult males by a ratio of more than six to one. The conspicuous absence of females is likely due to mass abductions, an act of sexual and gender violence that is often observed during genocidal moments. Such demographic data points to a bias focused on murdering male residents; those who remained at Kuelap perished during a single violent assault. Perpetrators followed a common pattern of murder by means of blunt force trauma. Bones of the dead show signs of weapon impacts and peri-mortem fractures. With few exceptions, every skull displayed multiple, deadly head wounds. Blows came from all directions. Among adults, wounds tended towards the left side of the head, consistent 18 J. M. Toyne and A. Narvaéz, ‘The fall of Kuelap: bioarchaeological analysis of death and destruction on the eastern slopes of the Andes’ in Embattled Bodies, Embattled Places: War in Pre-Columbian Mesoamerica and the Andes, ed. A. K. Scherer and J. W. Verano (Washington, DC: Dumbarton Oaks, 2013), pp. 341–64.

587

danielle kurin

with face-to-face assault by a right-handed opponent. The young, on the other hand, were more likely to be hit on the back of the head, a sign they were struck while fleeing an assailant. Together, the data reveals that Chachapoyan men, women and children at Kuelap experienced a ferocious assault resulting in kidnappings and carnage. This apparent genocidal massacre was intended to exterminate the entire resident population as well as to deprive the Chachapoyas people of descendants by abducting their women. After such outright annihilation, the city of Kuelap struggled to restore its earlier splendour.

Chronic Catastrophes and Casualties Episodes of genocide are predicated on the directed, attempted and intentional extermination of a group identified on the basis of perceived social, cultural or biological differences between in- and out-group members. This section reviews widespread and persistent genocidal campaigns which took the form of targeted assaults that often escalated into episodes of mass killings. This genre of prehistoric brutality occurred in polities much less complex than modern nation-states and often in the absence of empires. Indeed, evidence from bones and burials suggests that imperial leaders in the Andes rarely ordered the complete and total annihilation of despised enemies and unruly subjects. Rather than death outright, Andean empires favoured controlling living bodies. They compelled subordinate populations to adopt culturally distinguishing practices that would conspicuously mark their relative inferiority. The Inca and her imperial antecedents also regularly instituted policies of forced mass displacement and corveé labour, advocated for the capture and transfer of women and children, and engaged in damnatio memoriae – prolific acts of cultural erasure. Chanka Endemic Genocide

The Chanka people (1000–1400 C E ) of the south central Andean cordillera emerged in the absence of any sort of regional hierarchy and under conditions of prolonged drought. Their skeletons – some 300 from a half-dozen tombs – show evidence of extreme human suffering over generations. During this dark age, violence became ubiquitous: head trauma affected more than half of the entire Chanka population; blunt force trauma caused death for one out of every three individuals. However, violence was not experienced equally by all. Bloodshed was overwhelmingly directed towards people whose ethnicity was marked by cranial modification – the intentional

588

Mass Extermination in Prehistoric Andean South America

and permanent elongation of the skull achieved through the persistent use of turban-like bindings. Chanka people were divided into two groups: the rumpu uma, or people with natural, round heads, and the caytu uma, people with long, artificially deformed heads.19 Both groups lived side by side, but the physical attacks evident in the archaeological record were intentionally directed towards caytu uma men and women. A sobering suite of distinctive attributes reflect the demographics of modern genocidal moments.20 For instance, unlike traditional battlefield warfare involving adult male soldiers, wound frequency, lethality and location between long-headed males, females and juveniles are unnervingly similar (for example, over half of eighty-one men sustained wounds, as did less than two-thirds of seventy-four women). Death by homicide was the primary cause of excess mortality among caytu uma late adolescents and young adults of both sexes – arguably the most productive segment of society. Yet, the infants, elderly, sick and disabled from that group – all vulnerable non-combatants – were targeted for attack as well. The soaringly high ratio of murdered to maimed individuals – deduced from peri-mortem and ante-mortem head wounds – is consistent with scenarios where the perpetrators focused on kills and not casualties. Throughout the Chanka era, the caytu uma people experienced trauma at significantly higher rates, and their wounds were more numerous and more deadly; violent encounters were frequent, successive and ultimately fatal. One out of every six suffered multiple attacks, identified by the presence of both healed injuries and deadly wounds. Also, only the long heads have gaping wounds on the base of the skull – telltale execution blows. These ring-shaped fractures occur when a kneeling victim with head bowed is killed by closerange, high-impact blunt force trauma. Finally, only members of this targeted group had their faces physically obliterated by repeated, bone-crushing overkill trauma. Collectively, these corresponding lines of evidence show that during the Chanka era the caytu uma, indelibly and conspicuously marked by their intentionally remodelled heads, were targeted for violence, while the rumpu uma, whose members maintained a natural skull shape, led longer, healthier and more peaceful lives. All told, the case of the Chankas appears to have been one of endemic genocide, where marked social groups within a loosely organised chieftaincy sought to intentionally exterminate one another. 19 D. G. Holguín, Arte y Diccionario Quechua–Español (1608; Madrid: Imprenta del estado, 1901). 20 D. S. Kurin, The Bioarchaeology of Societal Collapse and Regeneration in Ancient Peru, Bioarchaeology and Social Theory (New York: Springer, 2016).

589

danielle kurin

Andean Empires as Anomalies The prehistoric Andes witnessed the genesis of three major empires. The Inca Empire is the most well known. Less well known are the other two roughly contemporaneous empires called Wari and Tiwanaku (500–1100 C E). Like the Inca, these neighbouring states controlled expansive, discontinuous territories through heavy investments in administration and infrastructure to facilitate the incorporation of scores of distinct ethnic polities. Strategic outposts controlled the movement of people and protected resources and transport corridors. The twin empires put considerable emphasis on consolidation through coercion and outright militarism. Through conquest and rebellions, villages were destroyed and adults and children were killed.21 Trophy heads and cranial trauma hint at raids by professional soldiers to procure body parts for display. So too, highly choreographed mass executions served as public spectacles and artistic celebrations that reinforced the ideals and power of the state and its leaders through the corporality of human sacrifice. Apart from the violent revelry, there is little archaeological evidence for large-scale interethnic warfare. We find no analogue among Andean empires for the type of systematic mass exterminations enacted by Rome in Carthago delenda est style. Like the eventual sacking of Rome, the capital cities of the Wari and Tiwanaku and their attendant holdings were ultimately abandoned and left mostly desolate and destroyed. But not entirely. Generations later, local people would hollow out parts of the formidable, still standing walls of the ruined Wari capital.22 Within these improvised niches, descendants entombed their dead. A plurality of the dead already had healed wounds, the scars of previous violent encounters. Victims – adults and children alike – were killed with multiple fatal blows to the skull, suggestive of killing sprees. But violent massacres of this type were relatively rare. Beginning in the late thirteenth century, the coalescing Inca state would follow and surpass the templates of infrastructure and governance set by the preceding twin empires. At its apogee, the Inca realm dwarfed other empires. Its boundaries encompassed an area equivalent to that between London and Tehran. With such vast territories, Inca success was predicated on the military prowess of a large, highly mobile professional army. 21 T. A. Tung, Violence, Ritual, and the Wari Empire: A Social Bioarchaeology of Imperialism in the Ancient Andes (Gainesville: University Press of Florida, 2012). 22 T. A. Tung, ‘Violence after imperial collapse: a study of cranial trauma among Late Intermediate period burials from the former Huari capital, Ayacucho, Peru’, Nawpa Pacha 29 (2008), 101–17.

590

Mass Extermination in Prehistoric Andean South America

Weigh stations and command posts built at regular intervals along the 40,000-kilometre Inca road system provided the supplies and manpower needed to support ambitious expeditionary campaigns. Suppressing revolts was another regular activity; documentary evidence suggests that rival groups were forcibly displaced and killed, but battlefields like those later described in colonial-era chronicles have yet to be found. The infamous Leyenda Negra – Spanish propaganda that highlighted the violent, inhuman savagery of indigenous Inca lords – is, to date, largely inconsistent with the extant skeletal record.23 Nevertheless, there is persuasive archaeological evidence that the Inca plundered and burned towns that resisted or revolted. Settlements were destroyed and unceremoniously abandoned as retribution for resistance. Another common tactic of the Inca was to raze indigenous local architecture or sacred sites and rebuild over them. As Inca expansion accelerated in the early fifteenth century, there were occasional clashes with local groups in the frontier zones of Ecuador, Argentina and Chile.24 Groups were pacified by manipulating local elites through intermarriage and by deploying a ritual called Capacocha, the sacrifice of ‘perfect’ children, by strangling and blunt force trauma. The internment of privileged children on the peaks of towering summits such as Mount Ampato and Mount Llullaillaco was a means by which associated elite lineages and entire ethnic groups were symbolically incorporated into the animistic and immemorial genealogy of the Inca world. As a result of this ethnic integration, mass violence in once hostile regions actually subsided.25 While expansive empires are thought to leave death and devastation in their wake, this does not appear to have been the case for prehistoric states in the Andes.

Genocidal Moments and Massacres in the Ancient Andes In this chapter, burial patterns and skeletal trauma have been approached as linked data sets marshalled to help detect instances of mass extermination. 23 M. S. Murphy, C. Gaither, E. Goycochea, J. W. Verano and G. A. Cock, ‘Violence and weapon-related trauma at Puruchuco-Huaquerones, Peru’, American Journal of Physical Anthropology 142 (2010), 636–49. 24 C. Torres-Rouff and M. A. Costa, ‘Interpersonal violence in prehistoric San Pedro de Atacama, Chile: behavioral implications of environmental stress’, American Journal of Physical Anthropology 130 (2006), 60–70. 25 V. A. Andrushko and E. C. Torres, ‘Skeletal evidence for Inca warfare from the Cuzco region of Peru’, American Journal of Physical Anthropology 146:3 (2011), 361–72.

591

danielle kurin

Careful study of this material record has revealed key attributes about the life history of the victims disproportionally singled out for attack and, sometimes, the motives of the perpetrators. Time and again, societies of the ancient Andes variably employed mass extermination as a strategy of forcing submission, a performance of religious and political power, a means of acquiring resources and territories, and as a form of reprisal against insubordinate groups. Perpetrators incorporated key tenets of genocidal violence including the intentional infliction of harm, aims to increase a victim’s suffering, the normalisation of torture and the physical destruction of the body. This dehumanisation extends to the (mis)treatment of the vanquished after death. Violence of this type was deeply symbolic, even ritualistic. Moments of mass murder in the Andes necessarily required particular, efficient means of executing large groups of people. That, combined with the cultural importance of certain body parts and forms of assault, resulted in distinctive but standardised killing methods that left indelible, patterned marks on bone. These material configurations help illuminate the cultural calculus of deadly combat and physical engagement with corpses by prehistoric people who left no written accounts of their intentions and motivations. The archaeological record bears evidence of violent circumstances consistent with genocidal moments and massacres, but the social contours of the kill varied by context. Groups of all sorts – based on age, sex, ethnicity, occupation and origin – were singled out for extermination. Genocide is made manifest in the indiscriminate slaughter of rival communities for material gain. It guides the taking and execution of captives, be they soldiers or non-combatants like women and children. It appears in reprisal killings and petty intra-group routs, but also through massive serial murder for symbolic or sacred purposes. Disparate as they may seem, differing modes of mass murder in the Andes may be best viewed through the lens of huañuchiyta atipani, the indigenous Quechua concept used by the Inca and her ancestors to denote the utter and absolute power of perpetrators to manufacture death.

Bibliographic Note The publications cited in the chapter above all represent key contributions to the study of violence, death and burial in the prehistoric Andes and are worth reading in their own right. For readers seeking additional information, there are several comprehensive and accessible monographs that merit consideration. A journey into the deep, and often violent, cultural history of the region is thoroughly chronicled in Jeffrey Quilter’s The Ancient Central Andes 592

Mass Extermination in Prehistoric Andean South America

(New York: Routledge, 2014). Rebecca Stone-Miller’s Art of the Andes: From Chavín to Inca (2005; 3rd ed., London: Thames & Hudson, 2012) traces the aesthetic history of the region over the past three millennia, with special attention to the art and iconography of power, domination and violence. Steve Bourget’s Sacrifice, Violence, and Ideology among the Moche: The Rise of Social Complexity in Ancient Peru (University of Texas Press, 2016) is equally compelling in its coverage of the role of murder in the rise of social complexity. The Inca Empire, in particular, continues to captivate scholars and the public alike: its capital city, Cusco, and the legendary ‘lost city’ of Macchu Picchu rank among the most visited tourist destinations in the western hemisphere. Johan Reinhard’s The Ice Maiden: Inca Mummies, Mountain Gods, and Sacred Sites in the Andes (Washington, DC: National Geographic Society, 2005) chronicles the breathtaking discovery and scientific analysis of the frozen bodies of young girls and boys, who were sacrificed to the mountain gods some 500 years ago. Among the most important collection of mummies ever found, their clothing and artefacts represent the richest trove of Inca treasures known to date. Likewise, cutting-edge research on their perfectly preserved bodies has yielded new insights on the mechanisms of widespread, sanctioned child murder within the expansive Inca Empire. Finally, those who yearn to better understand the methods bioarchaeologists use to read human bones would benefit from reading John Verano’s seminal work, ‘Advances in the paleopathology of Andean South America’ (Journal of World Prehistory 11:2 (1997), 237–68).

593

23

The Spanish Destruction of the Canary Islands A Template for the Caribbean Genocide i g o r p e´ r e z t o s t a d o ‘They go as naked as when their mothers bore them,’ Admiral Christopher Columbus famously wrote in his log book on Friday, 12 October 1492, of the people of Guanahani that greeted his fleet, the first time Europeans and Americans met. ‘They are the color of the Canarias, neither black nor white’, as were their eyes, he later added. ‘Nor should anything else be expected,’ he explained, ‘as this island is in a line east and west from the island of Hierro in the Canaries.’1 Columbus’ fleet had sailed from the island of Gomera, in the Canary Islands, on Thursday, 6 September 1492, having observed ‘a great fire issue from the mountain of the Island of Tenerife’.2 Tenerife and La Palma, the largest and more populous islands of the archipelago, remained unconquered. But only for a short time. The same Sevillian financiers who had made Columbus’ travel possible were pulling in resources to bankroll military campaigns against the two islands. By mid-1496, when Columbus returned from his second transatlantic travels, the almost century-long process of the conquest of the Canaries had been completed. The conquest of the Canary Islands was the learning ground and laboratory of European settler colonialism, later developed on a far larger scale in the Americas, starting in the Caribbean. The indigenous population of the Canaries would be Europeans’ readiest reference by which to interpret the new people they encountered in the western hemisphere. For profit 1 P. Hulme, ‘Tales of distinction: European ethnography and the Caribbean’ in Implicit Understandings: Observing, Reporting and Reflecting on the Encounters between Europeans and Other Peoples in the Early Modern Era, ed. S. Schwartz (Cambridge University Press, 1994), pp. 157–97. 2 Christopher Columbus, Journal of the First Voyage of Columbus, in The Northmen, Columbus and Cabot, 985–1503, ed. J. E. Olson and E. G. Bourne (New York: Scribner’s Sons, 1906), pp. 92–3, online facsimile, American Journeys digital library, www.americanjourneys.org/aj-062 (accessed 30 March 2022).

594

The Spanish Destruction of the Canary Islands

conquest, settler colonialism and plantation agriculture based on enslaved people forcibly brought from Africa were first successfully rehearsed in the Canaries before travelling west. It caused a population collapse due to a war of attrition, enslavement, deportation and new diseases. As a result, the indigenous population and its language, religion and culture were utterly destroyed, in a prelude to what later happened in the Antilles at an even greater pace and scale. These contemporary and connected genocides are thus historically relevant for their intersecting position between medieval and modern, European and colonial, and for the impact they had on the evolution of the early modern world.

Background The main environmental characteristics of the seven islands and many islets of the Canary archipelago is their isolation, from themselves and the African continent, and their volcanic nature (Map 23.1). This creates a unique ecological environment of rich coastal plains and rocky, impenetrable mountainous interiors. The islands were isolated, their inhabitants developing autonomously, both among themselves and regarding the nearby African continent. The Canary Islands population originated, according to the data, from central and west North Africa.3 They arrived from different parts of the continent in two major waves around the first millennium B C E and the first millennium C E. The first populated the whole archipelago and the second some islands. This resulted in different mixtures and evolution for each of the islands, arising from genetic isolation and reduced population size, especially in the less populated islands (Lanzarote, Fuerteventura, El Hierro and La Gomera).4 Pre-conquest demographic assessments derive from archaeological evidence. Minimal estimations reduce that number to around 30,000–50,000 and the maximal to around 100,000.5 A widely accepted figure is that of 80,000. Early European narratives suggest that the indigenous society was in a moment of increased demographic pressure over its natural resources. Their economy was one of subsistence, based on the raising of goats, farming 3 R. Fregel et al., ‘Mitogenomes illuminate the origin and migration patterns of the indigenous people of the Canary Islands’, PLoS ONE 14:3 (2019), 1–24. 4 Ibid., 9. 5 L. S. Owens, ‘The application of dental anthropology to population biology, diet and health in the prehispanic Canary Islands’ (PhD thesis, University of London, 2003), pp. 79–83.

595

Map 23.1 Spain, North Africa and the Canary Islands in 1544. (Portolan Map)

The Spanish Destruction of the Canary Islands

of cereals and vegetables, collection of wild fruits and herbs and fishing and collecting of seafood.6 On each island, several chieftaincies competed for the control of resources. Invaders depicted the indigenous people of the Canaries as redoubtable fighters, which could result from the social and political configuration of the islands.7 Chiefdoms formed alliances under the leadership of a dominant group in order to face rival confederations. Europeans later exploited these rivalries. The connection between the population of the Canary Islands and the rest of the world was thin. Some funerary and religious practices are connected with the Phoenician culture and some DNA traces have been traced to the eastern Mediterranean. The Carthaginians seem to have known of the islands. Roman sources and archaeological digs evidence their knowledge of and relations with the islands.8 After the Roman Empire, regular contact between the Mediterranean world and the Canary Islands was lost, although the works of Roman authors referring to them, such as Pliny, were still known. For the Christian point of view, they drifted beyond the limits of the ecumene into a fantastic landscape where the rules of nature and society were 6 M. Arnay-de-la-Rosa et al., ‘Bone trace element pattern in an 18th century population sample of Tenerife (Canary Islands): comparison with a prehistoric one’, Biological Trace Element Research 65 (1998), 45–51; M. Arnay-de-la-Rosa et al., ‘Paleodietary analysis of the prehistoric population of the Canary Islands inferred from stable isotopes (carbon, nitrogen and hydrogen) in bone collagen’, Journal of Archaeological Science 37 (2010), 1490–501; M. Arnay-de-la-Rosa et al., ‘Paleonutritional and paleodietary survey on prehistoric humans from Las Cañadas del Teide (Tenerife, Canary Islands) based on chemical and histological analysis of bone’, Journal of Archaeological Science 38 (2011), 884–95. 7 T. Delgado-Darias, V. Alberto-Barroso and J. Velasco-Vázquez, ‘Violence in paradise: cranial trauma in the pre-Hispanic population of Gran Canaria (Canary Islands)’, American Journal of Physical Anthropology 166 (2018), 70–83; T. J. Desch-Obi, ‘African ritual violence: close combat in Western Africa and the Diaspora’ in The Cambridge World History of Violence, ed. R. Antony, S. Carroll and C. Dodds Pennock, 3 vols. (Cambridge University Press, 2020), I I I , p. 573; J. Velasco-Vázquez, T. Delgado-Darias and V. Alberto-Barroso, ‘Violence targeting children or violent society? Craniofacial injuries among the pre-Hispanic subadult population of Gran Canaria (Canary Islands)’, International Journal of Osteoarchaeology 28:4 (2018), 388–96; T. Delgado-Darias, V. Alberto-Barroso and J. Velasco-Vázquez, ‘Isolation and violence on an oceanic island: lethal injuries in a pre-Hispanic burial in Gran Canaria (Canary Islands, Spain)’, Journal of Island and Coastal Archaeology (2020), 1–19, https://doi.org/10.1080/15564894 .2020.1783036. 8 A. Mederos Martín and G. Escribano Cobo, ‘Las Islas Afortunadas de Juba II. Púnico. Gaditanos y romano-amuretanos en Canarias’, Gerión 20:1 (2002), 315–58, at 345–6; M. A. Cháves Álvarez and A. Tejera Gaspar, ‘Evidencias arqueológicas de filiación romana en las Islas Canarias’ in XVIII Coloquio de Historia Canario-Americana, ed. F. Morales Padrón (Las Palmas de Gran Canaria: Cabildo de Gran Canaria, 2010), pp. 32–42; P. Atoche Peña, Evidencias arqueológicas del mundo romano en Lanzarote (Islas Canarias) (Arrecife: Servicio de Publicaciones del Cabildo Insular de Lanzarote, 1995).

597

igor pé rez tostado

antithetical to civilisation. The ‘Fortunate Islands’ were imagined as a peaceful haven amid the terrifying Atlantic Ocean. It was not until the fourteenth century that the Canary Islands began to leave this mythic landscape and their coastline became slowly incorporated into the portolans drawn by Mediterranean sailors.9

Historical Sources and Debates The conquest of the Canary Islands generated at least twenty-five contemporary chronicles that recorded these deeds and actions in order to legitimise the acts of conquest, to set legal claim to lands, booty and privileges.10 The number of actors implicated in the conquest, from private individuals to competing aristocracies and monarchies, explains the existence of so many contending contemporary narratives. Abundant travel descriptions of the first European visitors to the islands during the demise of the indigenous population have been preserved. Starting in the sixteenth century, narratives of long stays in the new colonial society were written.11 In contrast, the archaeology of the conquest period has not yet been fully studied.12 The impact of the conquest on the cultural awareness of Spanish society was remarkable. There are around fifty-two major sources related to the conquest of the Canary Islands written during the sixteenth and seventeenth centuries, including histories, treatises, religious histories and literary pieces.13 The histories differ from the chronicles of the conquest because they were written retrospectively. They took as a basis historical documents and sometimes witness accounts, in order to identify and make a synthesis of the most important events and processes that brought about the conquest. These histories fall usually in one of three categories. A group focuses on the Canary Islands themselves and thus revolves around the European conquest of the archipelago. These histories are the latest to develop, starting at the 9 K. R. Wittmann, Las islas del fin del mundo: representación de las afortunadas en los mapas del occidente medieval (Lleida, La Laguna: Edicions de la Universitat de Lleida, Universidad de La Laguna, 2016). 10 S. Baucells Mesa, ‘Las fuentes narrativas canarias y la construcción ideológica del indígena’, Tabona 18 (2009/10), 13. 11 Ibid., 16–18. 12 J. F. Navarro Mederos, ‘Arqueología de las Islas Canarias’, Espacio tiempo y forma. Serie I, Prehistoria y arqueología 10 (1997), 460; J. F. Navarro Mederos and J. C. Hernández Marrero, ‘Evidencias arqueológicas de los primeros asentamientos europeos en la Gomera (Islas Canarias)’ in XVI Coloquio de Historia Canario-Americana, ed. F. Morales Padrón (Las Palmas de Gran Canaria: Cabildo de Gran Canaria, 2004), pp. 388–407. 13 Baucells Mesa, ‘Fuentes narrativas’, 13.

598

The Spanish Destruction of the Canary Islands

end of the sixteenth century and developing through the seventeenth century. The second group are histories of the Castilian or Portuguese kingdoms which refer to the Canary Islands. These are the most abundant group and frame the Canary Islands (especially their conquest) within the historical narrative of the Iberian kingdoms. The last group are the histories of the Portuguese and Castilian overseas conquests in Africa, Asia and America, in which the Canaries get incorporated as a major stepping stone and point of comparison. It is worth stressing how the Canary Islands feature in the chronicles and literature traditionally associated with the conquest of America. Francisco López de Gómara dedicates one chapter of his General History of the Indies to the conquest of the Canary Islands.14 Spanish voices critical of the conquest and the treatment of indigenous people developed ethnohistorical accounts of the customs, population and culture of the collapsing Canarian and American societies. In fact, the two major Spanish intellectuals’ interpretative frameworks for Native American societies originated in the Canaries. The natural slavery approach, most famously defended by Fernandez de Oviedo, and the paternalistic defence of the rights of ‘natives’, championed by Bartolomé de Las Casas, both reflected on the conditions of the inhabitants of the Canary Islands.15 Religious writing, focused on the history of miracles or saintly people, also arose during the conquest of the islands. It tried to frame the pre-Christian religious world of the indigenous population in a narrative of conversion supported by supernatural intervention. It emphasised the miraculous deeds of certain sculptures, such as the Cristo de La Laguna and Our Lady of Candelaria. Religious works also feature in the biographies of godly people or martyrs of local origin, which are recognised as a sign of divine intervention in the world. In addition, there are some few philosophical, anthropological and linguistic works written at the time of the European conquest that refer to the Canary Islands.16 Finally, we find some literary works of the sixteenth and seventeenth centuries that employ the conquest of the Canary Islands as a background.17 The historiography of the Canary Islands is characterised by its peculiar position in between the regional and the insular, the Iberian and the Atlantic/ American, European/colonial, and even between the settler and the plantation colony. Note the absence (except for in ‘prehistory’) of a relation with 14 F. López de Gómara, Historia general de las Indias (Córdoba: El Cid, 2004), pp. 448–53. 15 B. de Las Casas, Historia de las Indias (Madrid: Atlas, 1957), pp. 64–84. 16 Baucells Mesa, ‘Fuentes narrativas’, 21. 17 Ibid., 23–4.

599

igor pé rez tostado

African history. This condition of in-betweenness and laboratory status applies to the historiography of settler colonialism. Felipe FernándezArmesto has argued that it was a colonial society in the narrow sense because it was a society formed through the process of settlement and mixed colonial and metropolitan elements.18 More recently, Patrick O’Flanagan has interpreted the conquest of the Canaries as the founding violence of a process of settler colonialism.19 That the European conquest was fatal for the indigenous population seems uncontested, both in contemporary sources and by later historians. However, an open debate about the genocidal nature of the conquest and colonisation is a recent addition to the historiography. Where genocide is mentioned, it is addressed in order to refute it.20 Survival of genetic material, or more precisely historians’ reading of geneticists’ work, serves as a refutation of genocide. There are individual but notable exceptions to this line of interpretation. Francisco Morales Padrón, one of the most eminent twentieth-century historians of the islands, was crystal clear about that, defining the conquest both as a genocide and an ethnocide.21 The study of the conquest of the Canary Islands from the perspective of Genocide Studies is, however, recent. Mark Levene has claimed that ‘the Castilians had not set out to exterminate the Guanches. Yet the result was this all the same’.22 More recently, Mohamed Adhikari has argued that the conquest of the Canary Islands constitutes the ‘Western world’s earliest overseas settler colonial genocide’, a turning point in the global history of genocide.23 More research is needed focused on the Canary Islands, but also in comparison with contemporary cases such as the Caribbean, combining unpublished archive sources, ethnohistorical evidence and archaeological data. 18 F. Fernández-Armesto, Las Islas Canarias después de la conquista (Las Palmas de Gran Canaria: Cabildo de Gran Canaria, 1997), p. 14. 19 P. O’Flanagan, ‘Mediterranean and Atlantic settler colonialism from the late fourteenth to the early seventeenth centuries’ in Routledge Handbook of the History of Settler Colonialism, ed. E. Cavanagh and R. Veracini (New York: Routledge, 2017), pp. 37–48. 20 B. Kiernan, ‘Is “genocide” an anachronistic concept for the study of early modern mass killing?’, History 99:336 (2014), 530–48; P. E. H. Hair, ‘How the south was won – and how Portuguese discovery began’ in Spain, Portugal and the Atlantic Frontier of Medieval Europe, ed. J. J. López Portillo (New York: Routledge, 2016), p. 356n9. 21 F. Morales Padrón, Como vivían los antiguos canarios (Las Palmas de Gran Canaria: Mancomunidad de Cabildos, Plan Cultural, y Museo Canario, 1978), p. 19. 22 M. Levene, Genocide in the Age of the Nation State, vol. I I: The Rise of the West and the Coming of Genocide (London: I. B. Tauris, 2005), pp. 34–7, at p. 37 (emphasis in original). 23 M. Adhikari, ‘Europe’s first settler colonial incursion into Africa: the genocide of the aboriginal Canary Islanders’, African Historical Review 49:1 (2017), 1–26; M. Adhikari, C. Carmichael, A. Jones, S. Kapila, N. Naimark and E. D. Weitz, ‘Genocide and global and/or world history: reflections’, Journal of Genocide Research 20:1 (2018), 136–7.

600

The Spanish Destruction of the Canary Islands

Perpetrators The first Europeans to navigate the Canarian coasts were merchants moving from Italian entrepôts such as Pisa and, especially, Genoa. Based in Portuguese and Andalusian ports and accompanied to a lesser extent by Mallorcan and Catalan merchants, they first encountered the islands when they started trying to circumnavigate Africa, hoping to reach India. These explorations, begun in the 1290s, gave concrete results by the 1330s with the outlining of the islands and their incorporation into European portolan charts. It was one of the first successful explorers, Lanzarotto Malocello, who renamed after himself the easternmost island of the archipelago, Lanzarote.24 The early arrivals were opportunists interested in slave raiding, complemented with trading local products such as hides, dyes and wax. They did not aim or dedicate resources towards the permanent conquest of the islands, nor their colonisation.25 Clergymen with their own agenda followed these Italian raiders and traders. By the 1350s the papacy had created the missionary bishopric of La Fortuna, based in Telde (Gran Canaria), aimed at the conversion of the indigenous people. These religious activities were part of a papal thrust to absorb peoples that had lived beyond the margins of European Christianity. Following the Mallorcan tradition of Ramon Llul, and with the backing of the king of Aragon, friars sought to convert the Canarian islanders by inserting missionaries within indigenous communities.26 In 1386, hermits established themselves in Gran Canaria.27 This religious endeavour did not stop the heightened slave raids from Andalusia and Vizcaya in the 1390s, particularly that of 1393, and their own abuses of the indigenous population. Islanders violently turned against the Christian missionaries, erasing their presence as a retaliation. In 1402, Norman troops arrived in the archipelago headed by Jean de Béthencourt, lord of Grainvile, in Normandy, and Gadifer la Salle, from Poitou. In two years they had conquered Lanzarote, Fuerteventura and Hierro, which became depopulated by famine, slave raiders and the diseases 24 M. A. Ladero Quesada, Los primeros europeos en Canarias: (siglos XIV–XV) (Las Palmas de Gran Canaria: Mancomunidad de Cabildos, 1979), p. 10. 25 C. Verlinden, ‘The Italians in the economy of the Canary Islands at the beginning of the Spanish colonization’ in The Beginnings of Modern Colonization: Eleven Essays with an Introduction (Ithaca, NY: Cornell University Press, 1970), pp. 132–57. 26 D. Abulafia, The Discovery of Mankind: Atlantic Encounters in the Age of Columbus (New Haven: Yale University Press, 2008), p. 71. 27 Ladero Quesada, Primeros europeos, p. 14.

601

igor pé rez tostado

brought by them. Their mentality diverged from that of the Italian traders and Aragonese-Mallorcan missionaries who had preceded them. They were seigneurial lords with experience in Mediterranean warfare who put their resources at risk and borrowed heavily in order to accomplish a military conquest. Their endeavour was blessed by Pope Benedict XIII as a crusade. According to Adhikari, the genocidal intent of the conquest is beyond doubt because the perpetrators were aware of the consequences of the compound effect of slavery and deportation.28 They established a permanent feudal dominion over the islands and their inhabitants under their formal vassalage to a distant monarch, the king of Castile. Lanzarote, Fuerteventura and Hierro formed a feudal lordship, to which La Gomera was added in the midfifteenth century. By then, the Norman component and leadership had been substituted by the Andalusian aristocratic Peraza family. The venture’s lack of profitability had forced them to sell their nominal rights. Despite the meagre return of the conquest, the islands were a strategic stepping stone for the opening up of maritime routes to West Africa. Thus, the dynamics and rivalry of Portugal and Castile dominated the subjugation of the unconquered islands, namely La Palma, Tenerife and Gran Canaria. The Peraza family was unable to make effective its nominal rights and extend its dominions to these islands. Meanwhile, Portugal was increasing its naval ambitions in the Atlantic at large under the leadership of Prince Henry ‘the navigator’. Portugal took possession of the uninhabited islands of Madeira and Azores between the 1420s and 1440s and later Cabo Verde in the 1460s. Over the same decades it tried to advance its interests in the Canaries. It made diplomatic moves claiming sovereignty in Castile and Rome, bought dubious legal rights from Maciot de Béthencourt (nephew and heir of Jean de Béthencourt) and built alliances with local chieftains. Portugal also sent large military expeditions and slave raiding parties to the Canaries. These moves forced the Castilian crown to reaffirm its alleged dominion over the archipelago. The treaty of Alcáçovas of 1479 between Portugal and Castile included Portugal’s recognition of the Canaries as belonging to Castile. On the ground, the kings stripped the Peraza family of its right of conquest, appointing instead, and closely monitoring, more aggressive and better resourced conquerors. In 1477, 100 foot soldiers and 30 cavalrymen under the command of Juan Rejón arrived on Gran Canaria.29 In 1480 the kings choose the more aggressive and savvier commander Pedro de Vera to complete the task.30 He 28 Adhikari, ‘Europe’s first settler’, 9, 21. 29 Abulafia, Discovery of Mankind, pp. 97–8. 30 F. Fernández-Armesto, 1492: The Year Our World Began (London: Bloomsbury, 2010), pp. 280–2.

602

The Spanish Destruction of the Canary Islands

accomplished it by stirring up division among the local rulers, utilising terror and scorched earth tactics and enlisting indigenous people from the island of Gomera into his army. A veteran of the conquest of Gran Canaria, Alonso de Lugo, obtained royal legal support (but no resources) to attempt the conquest of La Palma. His funding came from Genoese merchants in Seville interested in slaves, trade and plantations who had also financed Columbus’ voyage. He also enjoyed the support of the Duke of Medina Sidonia, the expansionist lord of the Andalusian Atlantic coast.31 Alonso de Lugo is presented as ‘unremittingly ruthless, unrestrainedly ambitious, unhesitatingly reckless, indefeasibly tough’ by Felipe Fernández-Armesto. This character profile, together with a pressing need to generate revenue, explain much of the rapacity and deceit of his campaigns, which resulted in the final conquests of La Palma in 1493 and Tenerife in 1496.32

Victims The conquest was accomplished thanks to fragmentation and political infighting among the Canaries’ indigenous population. Division derived in part from geography, with few if any communications between the islands by the late Middle Ages, and in part from each island’s inaccessible landscape and diverse ecosystems. Ritualised interpersonal combat that served to channel the group confrontation among polities within each island had been the key for regulating access to scarce resources. However, it proved fatal when threatened by external invasion. Conquerors exploited the divisions and shifted populations from one island to another to fight for them. The culture and language, although undoubtedly of a common origin, varied among the seven islands, adding to the fragmentation. Allencompassing terms such as ‘Canary Islanders’ (even worse, the synecdoche Guanche), or the slightly more nuanced terminology of ‘native population’, ‘aboriginal population’, ‘pre-conquest population’ or ‘pre-Hispanic population’ of the Canary Islands are imposed from outside. They do not do justice to the diversity and variety within and among the Guanches of Tenerife, the Canarians of Gran Canaria, the Auaritas of La Palma, the Majos of Lanzarote, the Gomeros of La Gomera, the Bimbachos of Hierro and the Majoreros of 31 L. Salas Almela, ‘“Melilla que es en las partes de África” and the House of Medina Sidonia: conquest, tenure and cession (1497–1556)’ in A Península Ibérica e o Norte de África (séculos XV a XVII): história e património, ed. J. Correia and A. Teixeira (Lisbon: Universidade Nova, Centro de História de Além-Mar, 2019), pp. 123–46. 32 Fernández-Armesto, 1492, pp. 283–4.

603

igor pé rez tostado

Fuerteventura.33 The differences deepened through the process of acculturation during the fourteenth and fifteenth centuries, most divisively in matters of religion.34 Fourteenth-century European descriptions claimed that the islanders lacked a common religion, in reference to a lack of institutional organisation and the simplicity of their cult structures.35 By the late fifteenth century the converts to Christianity among the indigenous population aided the Castilian conquest. Most famously, the former slave Francisca Gazmira was instrumental in the conquest of her native La Palma. She was able to bring it about that the leaders of the island both converted to Catholicism and became vassals of the Spanish monarchs. This move paved the way for the subsequent Spanish invasion of the island. In Tenerife, a local shrine and cult formed around an image of the Virgin and Child found on the shore at the beginning of the fifteenth century. The Spanish conquest launched by Alonso de Lugo in 1493 started with an alliance with the lord of Guimar, the owner of the image and the shrine. It involved his conversion and vassalage in exchange of military aid against his enemies.36 Another key element is the isolation in which the insular population had lived in the millennium prior to the Castilian conquest. They lacked effective maritime skills to leave the islands or travel among them, so they could not effectively retreat in the face of advancing Europeans. They faced them armed with lithic technology and their knowledge of an inaccessible and rugged landscape. However, they could not counter nor escape the new diseases that accompanied the conquerors, the famine that followed their destruction of crops, or the wrecking of family and community bonds brought by slavery and deportation.

The Unfolding of Genocide The conquistadores, missionaries, raiders and colonists acted in a diffuse and mostly uncoordinated manner, with little direction and even less control from the royal authority, which only timidly tried to rein in slavers. That local culture, language and especially religion must disappear went mostly unquestioned. The unfolding of genocide was the dynamic outcome of the 33 D. Abulafia, ‘Neolithic meets medieval: first encounters in the Canary Islands’ in Medieval Frontiers: Concepts and Practices, ed. D. Abulafia and N. Berend (Farnham: Ashgate, 2002), pp. 259–61. 34 E. Aznar Vallejo, ‘Exploración y colonización en la configuración de la Europa atlántica’, Historia, Instituciones, Documentos 35 (2008), 45–61. 35 Abulafia, Discovery of Mankind, p. 57. 36 Ibid., pp. 99–100.

604

The Spanish Destruction of the Canary Islands

forces and relationships at play in the sacking for slaves, deportation, conquest momentum, virgin soil diseases, land grabbing, exploitation of soil and labour and cash-crop farming – factors that, when associated and sustained in time, created the biological, environmental, economic and political pressure that brought about the genocide ‘by attrition’, following Sheri Rosenberg and Silina’s interpretation of the Genocide Convention.37 As these authors argue, rather than look for a dolus specialis, that is, a specifically stated and deliberated intent to destroy the victim group by an individual, which is what criminal prosecutors look for, the focus might rather be put, as attempted in this chapter, on the structures and agents that pursue genocidal plans. Connecting with this line of thought, Adhikari considers that enslavement and deportation constitute intentional (in the juridical sense of the term) ‘and irreparably damaging acts of collective destruction’.38 In addition, the insular environment of the Canaries means that the invasion of each island constitutes an individual case of genocide in which key elements combined differently.39 In the case of the Antilles (always with the caveat of the limited effect of central royal authority), orders to ‘destroy the Caribs as soon as possible’ through war and enslavement are plentiful.40 The key element of the destruction was slavery and deportation. Slaves were the most sought-after commodity of the islands from the European arrival on. Men and women were sold in the Iberian slave markets or brought to the uninhabited islands of Madeira and Azores, settled and exploited by the Portuguese. Early capture and enslavement of indigenous people was conducted through what the sources call saltos (literally, ‘jumpings’). The raiders would reach the coast at dawn, unnoticed, and capture anyone they could get their hands on before the alarm was raised, people fled towards the interior and resistance mounted. At that point, enslavers would retreat to their vessels with their human booty. The process could be restarted at another point along the coast, until the slavers were satisfied and departed. Slave raiding was common in the Mediterranean world and on occasion brought about the depopulation or 37 H. Fein, ‘Genocide by attrition 1939–1993: the Warsaw ghetto, Cambodia and Sudan: links between human rights, health and mass death’, Health and Human Rights 2:2 (1997), 10–45; S. P. Rosenberg and E. Silina, ‘Genocide by attrition: silent and efficient’ in Genocide Matters: Ongoing Issues and Emerging Perspectives, ed. J. Apsel and E. Verdeja (New York: Routledge, 2013), pp. 106–26. 38 Adhikari, ‘Europe’s first settler’, 21. 39 Ibid., 20–3. 40 P. Castañeda Delgado, ‘La política española con los caribes durante el siglo XVI’, Revista de Indias 30 (1970), 73–130. Later European powers in the region, such as the English, tried to co-opt them as ethnic soldiers for their own advantage: N. L. Whitehead, ‘Carib ethnic soldiering in Venezuela, the Guianas, and the Antilles’, Ethnohistory 37:4 (1990), 357–85.

605

igor pé rez tostado

abandonment of villages. In the Canary Islands, its destructiveness multiplied. Purposefully organised expeditions, such as the 1393 raids, had especially dramatic effects. When the conquest began in the fifteenth century, slaves were the only quick return that conquerors could offer to their financial backers.41 The ravages of conquest warfare were added to the already high toll of enslavement in the least populated islands. Lanzarote (1402), Fuerteventura (1402–5) and El Hierro (1405) easily fell to Jean de Béthencourt and Gadifer de La Salle. European overlordship extended to La Gomera (1420) under the leadership of the PerazaHerrera family. European troops employed new deadly weapons. Swords afforded an enormous advantage in close-range confrontation. Horses allowed the Europeans to move quickly through the coastal plains and provided an advantageous, elevated fighting position.42 Finally, the establishment of fortified camps around a square stone tower, supplemented by forward posts, allowed for a permanent control of the territory and exerted pressure over surrounding areas (Figure 23.1).43 The direct involvement of the Crown of Castile from the 1470s onwards made military resources available on a larger scale, pulling together private investors and conquerors. This renewed push brought about the subjugation of the major islands: Gran Canaria (1478–83), La Palma (1492–3) and Tenerife (1494–6). The warfare developed at this stage followed the patterns already established in the conquest of Muslim territory in the Iberian peninsula and later also applied in America. It consisted mainly of a war of attrition carried out through sorties or razzias, also called jornadas or cabalgadas. These aimed at the destruction of livelihood and food supplies, the taking of prisoners (for ransom and slavery) and the division of rival groups through negotiation. 41 M. Gambín García, ‘Los procesos judiciales de liberación de esclavos palmeses y guanches en Sevilla (1496–1512): nuevos datos para su estudio’ in XXI Coloquio de Historia Canario-Americana, ed. E. Acosta Guerrero (Las Palmas de Gran Canaria: Cabildo de Gran Canaria, 2016), pp. 1–10. 42 J. Santana-Cabrera et al., ‘The paths of the European conquest of the Atlantic: osteological evidence of warfare and violence in Gran Canaria (XVth century)’, International Journal of Osteoarchaeology 26 (2016), 776. 43 J. Onrubia Pintado and M. C. González Marrero, ‘The archaeology of the early Castilian colonialism in Atlantic Africa: the Canary Islands and western Barbary (1478–1526)’ in Archaeologies of Early Modern Spanish Colonialism, ed. S. Montón-Subías, M. Cruz Berrocal and A. Ruiz Martínez (Cham: Springer, 2016), pp. 119–51; M. C. González and A. Larraz, ‘Fortificaciones y expansión atlántica en el Mar de Canarias’ in Mil Anos de Fortificações na Península Ibérica e no Magreb (500–1500): Actas do Simpósio Internacional sobre Castelos, ed. I. C. Ferreira Fernandes (Lisbon: Edições Colibri, Câmara Municipal de Palmela, 2002), pp. 781–8.

606

The Spanish Destruction of the Canary Islands

Figure 23.1 Torre del conde, built in Gomera c.1450 by the Peraza family, is one of the best surviving examples of a military tower built during the conquest of the Canary Islands. Photograph, 1890–5. (Colección José A. Pérez Cruz, Fundación para la Etnografía y el Desarrollo de la Artesanía Canaria (FEDAC) / Cabildo de Gran Canaria/Biblioteca Virtual del Patrimonio Bibliográfico – España)

Conquerors made widespread use of terror through the staging of atrocious executions: dragging, quartering, hanging, cutting of hands and feet, impaling and throwing people into the sea tied and weighted with rocks.44 Pedro de Vera’s campaigns in Las Palmas in the 1480s resorted more than usual to terror tactics. In 1488 Vera launched an indiscriminate retaliation against the Gomeros, for plotting to the murder of the Spanish overlord of the island, Hernán Peraza. His actions were so extreme that the bishop of Lanzarote 44 A. Espino López, ‘Granada, Canarias, América. El uso de prácticas aterrorizantes en la praxis de tres conquistas’, Historia 45 (2012), 382–3.

607

igor pé rez tostado

protested, forcing the monarchs to reconsider the righteousness of this policy. In practical terms, it brought only the discharge of Vera.45 The final element was the spread of virgin soil diseases. The relative isolation of the islands until the fourteenth century had reduced the flux of pathogens across the archipelago. However, plagues accompanied the conqueror (brought from both Europe and the coast of Guinea), decimating the indigenous population. Epidemic outbreaks helped Europeans, starting probably in the fourteenth century, and continued long after the conquest.46 Great outbreaks are closely connected with key moments of the conquest. A great pestilence in Gran Canaria took place soon before the conquest began in 1478. Contemporary sources estimated it killed two-thirds or threequarters of the island’s populace.47 The so-called modorra (literally ‘drowsiness’) epidemic of 1494–5 in Tenerife (and Isabella in the Caribbean) was key to the European success in breaking indigenous resistance.48 The description of the symptoms connects modorra to influenza,49 though new research suggests that it might be a kind of encephalitis.50 Its impact was multiplied by the virgin soil effect described by Crosby: as nearly all the population falls ill, few are available to take care of the sick, plant, harvest, maintain flocks or even cook.51 To this situation, the scorched earth and terror tactics of the conquerors were added. Yet, the people fought back defiantly.

Resistance Although indigenous infighting and collaboration were keys to the Castilian onslaught, it is also necessary to stress the strength of the resistance. What is remarkable about the conquest of the Canary Islands is how long it took to 45 Fernández-Armesto, 1492, pp. 280–2. 46 M. J. Betancor Gómez and L. A. Anaya Hernández, ‘Las epidemias en Gran Canaria hasta la tercera década del XVI’ in X Coloquio de Historia Canario-Americana, ed. F. Morales Padrón (Las Palmas de Gran Canaria: Cabildo de Gran Canaria, 1994), I I , pp. 829–58. 47 A. Crosby, Ecological Imperialism: The Biological Expansion of Europe, 2nd ed. (Cambridge University Press, 2004), p. 92. 48 J. P. Hernández González, ‘En torno a una biografía global del primer médico de América Diego Álvarez Chanca (circa 1450 – post 1515)’, Anuario de Estudios Atlánticos 58 (2012), 40. 49 C. Rodríguez-Martín, ‘Una epidemia histórica en las Islas Canarias: la modorra de los guanches (1494–1495)’, Revista de Antropología y Arqueología 10:1 (1998), 69–87. 50 J. Hernández, ‘Epidemiología histórica de una enfermedad atlántica: la modorra’, Anuario de Estudios Canarios 54 (2010), 95–111. 51 A. Crosby, ‘Virgin soil epidemics as a factor in the aboriginal depopulation in America’, William and Mary Quarterly 33:2 (1976), 289–99.

608

The Spanish Destruction of the Canary Islands

accomplish, taking in mind the imbalance in weaponry and the swiftness of the subsequent American conquest.52 The first contacts in the fourteenth century were relatively tame, focused mainly on trade and evangelisation. Slave raiding however gave rise to increased resistance. The indigenous population could not protect themselves effectively against these hit-and-run razzia attacks. The result was deep resentment and distrust. By the time of the big raiding expedition of 1393, the indigenous population turned completely against all newcomers, destroying the Mallorcan settlement in Gran Canaria and exterminating its inhabitants, missionaries included.53 At the turn of the century, the Europeans’ goals moved from trade to conquest. Jean de Béthencourt and Godifer de La Salle did not encounter strong opposition to their taking of Lanzarote, Fuerteventura and Hierro (1402–5). These first conquered islands lacked enough population and inaccessible highlands, both necessary for building up an effective resistance. Early slave raiders had been advertising that the islands were easy to conquer and at low cost.54 However, the situation of the major islands of Gran Canaria, La Palma and Tenerife proved them wrong. Here the preconquest populace was plentiful and the interior extremely rough. Lack of armour made inhabitants more vulnerable in face-to-face combat, but provided them with greater speed in the mountains, where European warhorses were useless. This manoeuvring advantage was increased by whistle language, which allowed for the quick gathering and later dispersal of thousands of fighters for co-ordinated lightning attacks. Their traditional weaponry, manufactured of wood and stone, was weaker but still deadly. At range and from high ground, rock launching and avalanches effectively parried the crossbows and occasional rudimentary fire weapons of the Europeans.55 In addition, the people were quick to learn from their enemies: they used captured utensils and metal weapons, and even carts to attack and burn down towers.56 52 Fernández-Armesto, 1492, p. 275. 53 J. Abreu Galindo, Historia de la conquista de las siete islas de Gran Canaria (Santa Cruz de Tenerife: Valentín Sanz, 1940), pp. 27–9; A. Rumeu de Armas, El obispado del Telde: misioneros mallorquines y catalanes en el Atlántico (Telde: Ayuntamiento, 1986), pp. 112–14. 54 C. Rosell, ed., Crónicas de los reyes de Castilla, desde don Alfonso el sabio hasta los católicos don Fernando y doña Isabel (Madrid: M. Rivadeneyra, 1877), I I , p. 214. 55 Crosby, Ecological Imperialism, pp. 84–6. 56 E. Aznar Vallejo, ‘The conquest of the Canary Islands’ in Implicit Understandings: Observing, Reporting and Reflecting on the Encounters between Europeans and Other Peoples in the Early Modern Era, ed. S. Schwartz (Cambridge University Press, 1994), p. 143.

609

igor pé rez tostado

The island of Tenerife was the most populous and its inhabitants reputedly the fiercest fighters of the archipelago. This was where the most resounding European defeats took place. The islanders repelled several landing forces in the 1460s. In the spring of 1494, the first invasion by Alonso de Lugo was defeat at the ‘slaughter’ of Acentejo. His army of around 1,000 foot and 120 horse and artillery was destroyed (contemporary sources speak of 800 or 900 Spanish deaths) and survivors forced to leave the island by June.57 Religious critics of the conquest, such as Las Casas and Espinosa, used this defeat as proof of its unjust and ungodly nature.58 Battles were a rare event in the islands, Acentejo, Arucas or Ajodar being exceptional – people were not used to that kind of fighting and learned the hard way to avoid confrontation in open fields. The conquest was carried out through raids, skirmishes, punishments and counter-attacks. In that setting, indigenous defenders were capable of reciprocating violence, yet retribution did not serve as a deterrent but rather as an intensifier of violence, and is one of the key factors explaining the unfolding of genocide.59 Indigenous resistance and retaliations increased bitterness and revenge as a cause and justification for further excesses, feeding a spiral of violence. Spanish sources on the conquest abound with rash actions and disobedience to military command resulting from a desire for revenge.60 Comparison with the subjugation of the Antilles, begun in the 1490s, shows that effective resistance did not avert genocide but made it much more difficult for the conquerors to achieve. Indigenous resistance successfully slowed down the process. In fact, the Canary Islands were conquered largely because the Castilians could pit one group against another, use warriors from one island as troops in others, along with the impact of diseases.61 Even then, after the official end of the conquest, rebel bands continued to resist incorporation into the new colonial society, especially in Gran Canaria and Tenerife.62 In the Antilles, the dominant Taino societies were not able to put up an effective defence against the conquerors and succumbed much more rapiddly to the same tactics used in the Canaries. By 57 A. Macías Hernández, ‘Expansión europea y demografía aborigen: el ejemplo de Canarias’, Boletín de la Asociación de Demografía Histórica 10:2 (1992), 20–1. 58 E. M. Merediz, ‘Bartolomé de Las Casas in the Canary Islands’ in Bartolomé de Las Casas, O.P.: History, Philosophy and Theology in the Age of European Expansion, ed. D. Thomas Orique and R. Roldán Figueroa (Leiden: Brill, 2019), pp. 342–5. 59 Rosenberg and Silina, ‘Genocide by attrition’, 110. 60 See e.g. Abreu Galindo, Historia de la conquista, p. 167. 61 L. Rosa Olivera, Canarios en la conquista y repoblación de Tenerife (Las Palmas de Gran Canaria: Mancomunidad de Cabildos, Plan Cultural y Museo Canario, 1980). 62 Onrubia Pintado and González Marrero, ‘Archaeology of the early Castilian’, 123.

610

The Spanish Destruction of the Canary Islands

contrast, the Carib societies, with the same lithic technology as the Tainos but formidable warriors to European eyes, were able to avoid the worst effects of slave raiding and outright conquest and survive the impact of virgin soil diseases. Geography and navigation skills also help to explain this outcome. The Canarians’ best chance to parry the conquest was to fortify themselves in the natural highland strongholds of a few islands, from where there was no way of retreating nor protecting their food production. The Caribbean, in contrast, is a vast region of seven thousand islands and islets whose inhabitants had mastered inter-island navigation. Thus the Caribs, long accustomed to being the ones to carry out naval raid warfare, were able to effectively attack the newcomers and disperse when needed, either in the inaccessible mountain forests or by sea. This successful resistance initiated a spiral of violence with the Europeans that lasted centuries, in which the Caribs stepped up their aggressiveness and Europeans attempted a complete extermination when subjugation proved impossible.63

Long-term Impact From the onset of European contact with the Canary Islands in the fourteenth century, slave raiding disjointed and weakened the indigenous population. The patchy numbers of the chronicles and from chattel markets in Europe lead us to think that the process affected thousands. Alonso de Lugo alone was accused of unlawfully capturing 1,200 people in La Palma and 1,000 in Tenerife.64 Some slaves remained on their island of origin, but many were brought to other islands and the Continent, especially at the peak of the war in the late fifteenth century. Deportation magnified the disruptive effect of slavery. The death toll of the conquest of the Canary Islands is very difficult to calculate. It has to be inferred through partial (and often exaggerated) contemporary estimates. The accounts of the early conquests of Lanzarote, Fuerteventura and Hierro offer very imprecise numbers. Archaeological records of abandoned settlements and toponyms such as the Aldea de la Gente Muerta, give an impressionistic glimpse over an undoubted population 63 Abulafia, Discovery of Mankind, pp. 128–9. See note 40 above and, for the French case in which warfare alternated with gift giving and evangelisation efforts, P. P. Boucher, Cannibal Encounters: Europeans and Island Caribs, 1492–1763 (Baltimore: Johns Hopkins University Press, 1993). 64 E. Aznar Vallejo, La integración de las Islas Canarias en la corona de Castilla (1478–1526): aspectos administrativos, sociales y económicos (Seville and La Laguna: Universidad de Sevilla and Universidad de la Laguna, 1983), p. 151.

611

igor pé rez tostado

collapse following the first wave of conquest.65 Open field battles cost many indigenous deaths, even when their side ended victorious, as in the ‘massacre’ of Acentejo.66 The outbreaks of plague in Gran Canaria and modorra in Tenerife in the key moments of the conquest were devastating. References to bodily weakness and practices of female infanticide among the indigenous population of Gran Canaria hint at the heightened insecurity of the local people just before the epidemic hit the island. Gran Canaria lost between 65 per cent and 85 per cent of its inhabitants when the conquest was completed.67 The big modorra outbreak of Tenerife is estimated to have taken between 3,000 and 5,000 indigenous lives, out of a population in the range of between 15,000 and 20,000.68 By the early sixteenth century almost the entire indigenous population of the islands had disappeared. Estimations vary between 2,000 and 6,000 survivors.69 To the number of deaths the near total enslavement, dispossession, destruction of ways of life and deportation of survivors has to be added. Simultaneously, indigenous culture had disappeared and the population’s existence as an ethnicity had been erased. Contemporary chronicles and traveller accounts are unanimous: the indigenous population was almost extinct, but for a few, and its culture had already disappeared.70 In the Caribbean, the demographic evolution was very similar. Pre-contact Taino population estimations vary wildly, but a conservative and very much respected calculation puts it in the several hundred thousand. By 1550, contemporary sources and later historians agree that the Taino population of the Greater Antilles was almost extinct.71 Rabbits, sheep, chicken, asses, camels and other animals brought in European ships became the phalanx of the ecological transformation of the 65 Macías Hernández, ‘Expansión europea’, 30. 66 Aznar Vallejo, La integración, p. 151n2. 67 J. Onrubia Pintado, La Isla de los Guanartemes: territorio, sociedad y poder en la Gran Canaria indígena (siglos XIV–XV) (Las Palmas de Gran Canaria: Cabildo, 2003), p. 263; Crosby, Ecological Imperialism, pp. 92–3. 68 C. Rodríguez-Martín, ‘Estudio demográfico de la población Guanche de Tenerife’, Chungara 32:1 (2000), 31; Macías Hernández, ‘Expansión europea’, 22. 69 L. A. Anaya Hernández, ‘Los aborígenes canarios y los estatutos de limpieza’, Museo Canario 49 (1992–4), 127–33. 70 A. Quartapelle, ed., Quatrocientos años de crónicas de las Islas Canarias (La Orotava, Tenerife: Veredalibros, 2015), pp. 163, 218, 225. 71 M. Livi-Bacci, ‘Return to Hispaniola: reassessing a demographic catastrophe’, Hispanic American Historical Review 83:1 (2003), 3–51; M. Livi-Bacci, ‘The depopulation of Hispanic America after the conquest’, Population and Development Review 32:2 (2006), 199–232.

612

The Spanish Destruction of the Canary Islands

archipelago.72 Fire was the standard method by which the new settlers cleared forests for agricultural use. The regrowth of trees was impeded by newly imported fauna, especially rabbits, and this, together with tree cutting for construction and other uses, meant that by the sixteenth century the large forests of the westernmost islands had disappeared. Erosion followed, while timber became a scarce commodity.73 The Canaries became major food exporters for Europe, its colonies and the long-distance travellers among them.74 Cereal, sugar and wine served as the agricultural foundations of the new economy, making the colonial society possible, and in the process the natural environment was transformed by the demand for land, water and wood.75 Subsistence herding, one of the primary activities of the pre-conquest economy, continued to be practised in the margins of the new society. However, the changes in the ecosystem brought about by the Europeans disastrously affected indigenous nutrition and health, and consequently population size.76 The ecological transformation of the Caribbean followed the same lines, but to an even greater effect, due to its previous complete biological isolation from Eurasia and Africa. After the conquest, the Canary Islands attracted European migrants and forcedly transplanted people from Africa and America. They also became a transit point of migration between three continents. The dominant group were the Iberian soldiers and sailors, their descendants and successive generations of Genoese, Portuguese, English and Flemish landowners, traders and investors.77 They ruled the islands with the backing of the Castilian crown 72 S. Halikowski Smith, ‘The mid-Atlantic islands: a theatre of early modern ecocide?’, International Review of Social History 55:18, supp. (2010), 61. 73 Crosby, Ecological Imperialism, p. 96; Halikowski Smith, ‘Mid-Atlantic islands’, 57–8; A. Polónia and J. M. Pacheco, ‘Environmental impacts of colonial dynamics, 1400–1800: the first global age and the anthropocene’ in Economic Development and Environmental History in the Anthropocene: Perspectives on Asia and Africa, ed. G. Austin (London: Bloomsbury, 2017), pp. 26–9. 74 J. Bouchard, ‘Shetland sheep and Azorean wheat: Atlantic islands and provisioning centers, 1400–1550’, Global Food History 6:3 (2020), 169–93. 75 A. Macías Hernández, ‘Población, producción y precios del trigo, 1498–1560’, Anuario de Estudios Atlánticos 57 (2010), 327–84; A. Vieira, ‘Sugar islands: the sugar economy of Madeira and the Canaries, 1450–1650’ in Tropical Babylons: Sugar and the Making of the Atlantic World, ed. S. Schwartz (Chapel Hill: University of North Carolina Press, 2004), pp. 42–84; M. Lobo Cabrera, ‘Formas de poder y economía canaria entre los siglos XV– XVII’, Investigaciones históricas: Época moderna y contemporánea 18 (1998), 13–28. 76 Crosby, Ecological imperialism, pp. 91–2. 77 F. Bruquetas de Castro, ‘Los genoveses en Canarias. Siglos XVI y XVII’ in Comercio y cultura en la edad moderna: Comunicaciones de la XIII reunión científica de la Fundación Española de Historia Moderna, ed. J. J. Iglesias Rodríguez, R. M. Pérez García and M. F. Fernández Chaves (Editorial Universidad de Sevilla, 2015), pp. 57–68; M. Lobo Cabrera, ‘Los flamencos en Canarias en el siglo XVI’ in El greco . . . y los otros: La

613

igor pé rez tostado

and dominated the economy and land ownership.78 To make this new society possible, slavery became an essential institution of the colonial era. Indigenous inhabitants were forced into slavery both during the conquest and afterwards. Soon after, a captive Berber population from northern Africa, sub-Saharan Africans brought from Cabo Verde by the Portuguese or brought directly in Guinea, and even a few Tainos from Hispaniola were subjugated into slavery.79 Women were also forcibly imported as slaves to the islands as agricultural workers, house servants and breeders of new generations of slaves and sexual objects.80 In the Caribbean, this same process unfolded at a greater extent both in relative and absolute terms, due to the greater success of intensive cash crop agriculture.

Territorial Ambition, Ethnicity, Race, Religion and Gender The territorial expansion of Iberian aristocrats and competition among the respective kingdoms in the late Middle Ages were what propelled Europeans beyond the boundaries of the peninsula. They expanded towards the southern side of the Strait of Gibraltar and down the Atlantic coast of Africa, thus reaching the Canaries. The conquest was a private enterprise in which the state sanction was loose or came ex facto, as both Castilian and Portuguese monarchs competed to control the region. Quick return on investment was

contribución de los extranjeros a la monarquía hispánica, 1500–1700, ed. L. Ruiz Molina, J. J. Ruiz Ibáñez and B. Vincent (Editorial Universidad de Murcia, 2015), pp. 413–25; M. Lobo Cabrera and B. Rivero Suárez, ‘Los primeros pobladores europeos en Las Palmas de Gran Canaria’, Anuario de Estudios Atlánticos 37 (1991), 17–115; M. Marrero Rodríguez, ‘Los genoveses en la colonización de Tenerife’, Revista de Historia 89 (1950), 52–65. 78 E. Aznar Vallejo, ‘Expansión atlántica y construcción del estado moderno: el caso de Canarias’ in XXIII Coloquio de Historia Canario-Americana, ed. E. Acosta Guerrero (Las Palmas de Gran Canaria: Cabildo de Gran Canaria, 2020), pp. 1–25. 79 M. Marrero Rodríguez, La esclavitud en Tenerife a raíz de la conquista (La Laguna: Instituto de Estudios Canarios, 1966); J. Santana et al., ‘The early colonial Atlantic world: new insights on the African diaspora from isotopic and ancient DNA analyses of a multiethnic 15th–17th century burial population from the Canary Islands, Spain’, American Journal of Physical Anthropology 159:2 (2016), 300–12; M. Lobo Cabrera, ‘La esclavitud del indígena canario’, Museo Canario 55 (2000), 125–38; E. W. Stone, ‘Indian harvest: the rise of the Indigenous slave trade and diaspora from Española to the Circum-Caribbean, 1492–1542’ (PhD thesis, Vanderbilt University, 2004), pp. 3–6. 80 M. Lobo Cabrera, ‘La mujer esclava en España a comienzos de la edad moderna’, Baetica: Estudios de arte, geografía e historia 15 (1993), 295–316; M. Lobo Cabrera, ‘La población esclava de Las Palmas en la primera mitad del siglo XVI’ in XXI Coloquio, ed. Acosta Guerrero, pp. 9–10.

614

The Spanish Destruction of the Canary Islands

Figure 23.2 Luis Arencibia Betancort, Hautacuperche, 2007, bronze, Valle Gran Rey, Gomera. This four-metre-high contemporary rendering of the indigenous leader who led the insurrection against the Peraza family in 1488 aims at representing, as its own caption states, ‘the nobility, courage, dignity and love for freedom of the people of La Gomera’. It was erected as part of the ‘recovery of the historical memory’ by the regional government of the Canary Islands. (Photo by Ángel Guzmán Ramírez)

essential, and this brought about the enslavement and selling of the indigenous population, and massive usurpation of land.81 From the first point of contact on, islanders came to be (with few exceptions) homogenised, negativised and simplified as the savage antithesis of the 81 J. E. Kicza, ‘Patterns in early Spanish overseas empire’, William and Mary Quarterly 49:2 (1993), 229–59; A. Guimerá Ravina, ‘¿Canarias, ensayo de colonización americana?: el repartimiento de la Tierra en la Española’ in America y la España del siglo XVI, ed. F. P. Solano Pérez Lila and F. del Pino Díaz (Madrid: Consejo Superior de Investigaciones Científicas, 1983), I I , 175–90; A. M. Macías Hernández, ‘La construcción de las sociedades insulares, el caso de las Islas Canarias’, Anuario de Estudios Atlánticos 45 (2001), 131–60; A. M. Macías Hernández, ‘Una revisión necesaria. El diezmo de la primera agroindustria azucarera del atlántico, 1483–1543’, Anuario de Estudios Atlánticos 56 (2010), 245–88.

615

igor pé rez tostado

civilised and Christian European in the eyes of Europeans: they were ‘Canarians’.82 This new ethnic identity, invented and imposed by the conquerors, was later absorbed and adapted by the surviving indigenous population through acculturation, integration and miscegenation.83 As Columbus’ log book shows, the concept of race was loosely connected to appearance, and thought to result from several malleable factors, including latitude. However, a sense of hierarchy based on pigmentation mirrored the social stratification in which Europeans and their descendants (many of indigenous mothers) occupied the higher echelons of society while indigenous and enslaved Africans remained at the bottom. This blueprint was replicated repeatedly in the Americas.84 The difference in the Caribbean was that the Spaniards distinguished (or adopted the existing distinction; contemporary, historical and archaeological data here is inconclusive) between the Taino, who for them were those more peaceful natives whose surrender they obtained with relative ease, and the Caribs, the fearsome raiders who ate human flesh and thus against whom any excessive violence became legitimate. As in the New World, an antiquarian interest in the archaeological remains of ‘vanished’ indigenous civilisations developed in the eighteenth century. It provided the creole elite of the islands with a prestigious antiquity.85 New ideas on race, evolution, anthropology and archaeology reached the islands in the second half of the nineteenth century. As a result, twisted theories connected the origin of the population of the islands with the more distinguished ancient Celtic or Iberian cultures in a kind of generic Atlantic civilisation.86 This shift in thinking, aided by new scientific societies that had arisen at the end of the nineteenth century, was part of the creation of a regional political identity in the archipelago.87 Religion justified the conquest but was never its cause. Indigenous spirituality was considered idolatry, not a real religion in the sense of an organised 82 S. Baucells Mesa, ‘La antítesis entre aborígenes canarios y europeos: el distanciamiento como criterio en la representación historiográfica del indígena’, Museo Canario 58 (2003), 35–58. 83 Aznar Vallejo, ‘Exploración y colonización’, 48–52. 84 E. M. Merediz, ‘Más allá de América: Las Casas en una crónica sobre las Islas Canarias’, Museo Canario 51 (1996), 245–56. 85 C. Ortiz García, ‘“Antigüedades guanchinescas”: comercio y coleccionismo de restos arqueológicos canarios’, Culture & History 5:2 (online) (2016), 1–23. 86 A. R. Farrujia De la Rosa, W. Pichler, A. Rodrigue and S. García Marín, ‘Las escrituras líbico-bereber y latino-canaria en la secuenciación del poblamiento de las Islas Canarias’, Museo Canario 54 (2009), 18–19. 87 C. Ortiz García, ‘Guanchismo y nacionalismo en las sociedades científicas canarias de fines del siglo XIX’, Revista de Estudios Generales de la Isla de La Palma 2 (2006), 379–93.

616

The Spanish Destruction of the Canary Islands

institution, dedicated personal, sacred books or temples.88 Peaceful attempts at evangelisation were carried out by Mallorcan friars who counted on the backing of Roman popes and were helped by natives, ex-slave converts to Christianity.89 The conquest held, from the very beginning, the status of a crusade, sanctioned as it was by Benedict XIII in 1403, in the papal bull Apostolatus officium. The military invasion was accompanied by conversion, often forced and en masse. Converted natives and their leaders (generally after capture or slavery) played a key role in fragmenting the resistance to the conquerors.90 The same ideas were transferred to America, where a papal donation legitimated Iberian rule on the basis of conversion. The cult of Our Lady of Candelaria developed in this framework of ongoing conquest and acculturation. This was a local cult that started in Tenerife around the 1400s, so well before the conquest of the island by Alonso de Lugo in the 1490s. At its heart was a washed-off image of the Virgin Mary, which was believed to be miraculous. The cult of Candelaria, which incorporated many of the elements of the maritime virgins and superposed Christianity over pagan beliefs, served in the long term to put the islanders within the frame of Christian history. It also became a transatlantic tool for identity formation, both in the Canaries and the Americas.91 After the conquest, a religious administration was established, spearheaded by the efforts of missionary orders to convert the indigenous population.92 As part of the establishment of royal institutions, the Inquisition arrived to the islands. A special commissary arrived in 1493, with a tribunal in Las Palmas set up in 1505, which later in 1569 became independent from Seville. The primary targets of the institution were ‘false converts’ from Judaism.93 As for the indigenous people, the Inquisition mostly researched if slave owners took care of their evangelisation and if, after receiving baptism, they educated converts in the doctrine, prayers and faith.94 The Inquisition did not discriminate against the scions of the 88 Abulafia, Discovery of Mankind, pp. 86–7, 101. 89 A. Rumeu de Armas, La política indigenista de Isabel la Católica (Valladolid: Instituto ‘Isabel la Católica’ de historia eclesiástica, 1969), pp. 21–2. 90 Fernandez-Armesto, 1492, pp. 280–5. 91 E. M. Merediz, ‘Traveling icons: the virgin of Candelaria’s transatlantic journeys’, Arizona Journal of Hispanic Cultural Studies 5 (2001), 115–32; M. Dall’Agnola and E. Navarro Marrero, ‘Nuestra señora de Candelaria y otras vírgenes Del Mar: ¿un arquetipo precristiano?’, Museo Canario 54 (2009), 131–56. 92 Rumeu de Armas, Política indigenista, pp. 29–31. 93 A. Acosta González, ‘La inquisición en Canarias durante el siglo XVI (una aproximación estadística)’, Anuario de Estudios Atlánticos 32 (1986), 129–93. 94 Lobo Cabrera, ‘Población esclava’, 3.

617

igor pé rez tostado

indigenous population, despite their pagan ancestry, contrary to what it did to offspring of Muslim or Jewish converts.95 Post-conquest indigenous descendants came almost solely from the maternal side, since most of the mortal victims were male. According to genetic studies of the present-day population, 50 per cent of the inhabitants of the islands descend from the indigenous population on the maternal side, but only 10 per cent on the paternal side.96 European conquerors were mostly male who killed or displaced the males of the islands and mated with the indigenous females by force, persuasion or desperation.97 The post-conquest Amerindian populace also shows similar sexual asymmetry.98 The conquest of the Canary Islands brought about the destruction of the indigenous populations of all of its islands, namely the erasure of their names, language, customs, economy, land ownership systems, beliefs, culture, social structure and political organisation, and also the natural ecosystems they had created. Enslavement, deportation, disease and the strategic use of terror were the means by which it was achieved. Acculturation, miscegenation and the building of a colonial society based on plantation agriculture and longdistance trade did the rest to erase any trace of the indigenous culture. Genocide in the Canary Islands can be understood both as a process of attrition, following its definition by Fein and Rosenberg, and as an outcome, the aggregate result of thousands of specific instances of violence.99 Was it a genocide as defined in international law by the 1948 Genocide Convention? Morales Padrón in the 1970s and Adhikari more recently agree on this while coming from different angles.100 With those same arguments we can no doubt also include the case of the Caribbean. However, that legal definition 95 M. Lobo Cabrera, ‘Los indígenas canarios y la inquisición’, Anuario de Estudios Atlántico 29 (1983), 63–84; L. A. Anaya Hernández, ‘La limpieza de Sangre en Canarias y su relación con América’, Museo Canario 64 (2009), 101–30; J. M. Santana, ‘The formation of north African otherness in the Canary Islands from the 16th to the 18th centuries’, Culture and History 9:2 (online) (2020), 8. 96 R. Fregel et al., ‘Demographic history of Canary Islands male gene-pool: replacement of native lineages by European’, BMC Evolutionary Biology 9:1 (2009), 1–14. 97 A. Millares Torres, Historia general de las Islas Canarias (La Habana: Selecta, 1945), p. 312; C. Flores et al., ‘A predominant European ancestry of paternal lineages from Canary islanders’, Annals of Human Genetics 67 (2003), 146. 98 S. Wang et al., ‘Geographic patterns of genome admixture in Latin American Mestizos’, PLoS Genetics 4:3 (2008), 1–9. 99 See note 37 above. 100 Adhikari, ‘Europe’s first settler’, 21–3. It also fits the more restricted definition of genocide used by the author elsewhere; see M. Adhikari, ‘“We are determined to exterminate them”: the genocidal impetus behind commercial stock farmer invasions of hunter-gatherer territories’ in Genocide on Settler Frontiers: When Hunter-Gatherers and Commercial Stock Farmers Clash, ed. M. Adhikari (New York: Berghahn, 2014), p. 2.

618

The Spanish Destruction of the Canary Islands

did not exist at the time of the events, and so a suppler interpretation of the Genocide Convention, such as the one discussed by Rosenberg and Silina, would be needed in that case.

Conclusion The conquest of the Canary Islands was a necessary prerequisite to the American conquest. First, because of its location in the wind routes, and second, because it served as the learning ground for the conquest of the continent and how to build a colonial society and economy on the spoils. In both cases, however, conquest unfolded as a genocide and developed following the same pattern. In comparison, the Caribbean proved much faster, over a considerably bigger population and encompassing a larger territory. It made use, however, of the same techniques: a private enterprise driven by economic profit, the use of terror and divide-and-conquer tactics, decisively aided by the spread of virgin soil diseases and the devastating effects of slavery and dislocation. The relevance of these two cases to an understanding of the present state of colonialism, global ethics and human rights is made more visible if we tackle the perspective at the time of the events. The European experiences in the Canaries and America led to the expansion and rethinking of ius gentium, the forefather of what we today call international law.101 Rather than making it a crime, ius gentium offered a rationale and justification for what we today call genocide. The resulting consensus, which lasted centuries, explains why the crime of genocide is such a latecomer to international law. Consensus did not equate with unanimity. The dissenting voices at the time, such as Bartolomé de Las Casas, considered what happened in the Canarias as in the Caribbean as a horrendous crime from which ‘the obligation to make restitution was born’.102 That restitution never materialised. Historians today thus have the deontological responsibility to establish the facts as accurately as possible, so that jurists can assess the rights to justice and reparation.103

101 David M. Lantigua, Infidels and Empires in a New World Order: Early Modern Spanish Contributions to International Legal Thought (Cambridge University Press, 2020). 102 Las Casas, Historia, p. 77. 103 B. Clavero, ‘Memory’ in A Cultural History of Genocide in the Early Modern World, ed. I. Pérez Tostado (London: Bloomsbury, 2021), pp. 177–91.

619

igor pé rez tostado

Bibliographic Note The conquest of the Canary Islands created an important number of sources and accounts, many of which have been edited. Some of the most important ones became available in English early on. The first one was the chronicle of J. Abreu de Galindo, The History of the Discovery and Conquest of the Canary Islands, translated by George Glas and first published in London by R. and J. Dodsley in 1764 (London: Adamantine Media Corporation, 2006). Later, two important translating endeavours were carried out by the Hakluyt Society. The first in 1872 was The Canarian, or Book of the Conquest and Conversion of the Canarians in the Year 1402, by J. de Béthencourt, translated by R. H. Major (New York: Burt Franklin, 1970), which covers the Norman Conquest at the beginning of the fifteenth century. The second is the chronicle of Antonio the Espinosa, The Guanches of Tenerife: The Holy Image of Our Lady of Candelaria and the Spanish Conquest and Settlement, translated by C. Markham and published by the society in 1907. In the Spanish language, the finest edition on sources on the conquest is Francisco Morales Padrón, Canarias, crónica de su conquista: transcripción, estudio y notas (Las Palmas de Gran Canaria: Museo Canario, Ayuntamiento, 1978; Gran Canaria: Cabildo Insular, 1993). The Instituto de Estudios Canarios, El Museo Canario and the University of Las Palmas de Gran Canaria have edited and made available online the chronicles of the conquest, further historical sources, classical writings and the main specialised historical journals. The best starting point is the Viera y Clavijo virtual library maintained by the Institute of Canarian Studies, http:// iecanvieravirtual.org/ (accessed 27 February 2021), and the Memoria digital de Canarias hosted by the University Library of Universidad la Las Palmas de Gran Canaria, https://mdc.ulpgc.es (accessed 27 February 2021). For the Anuario de Estudios Atlánticos see http://anuariosatlanticos.casadecolon.com/index.php/ aea (accessed 27 February 2021). Although the archipelago has produced a wide historiography (see notes to the chapter), only a little of it is accessible in English translation. Eduardo Aznar Vallejo’s chapter ‘The conquest of the Canary Islands’ is a summary of the subject by one of the major specialists on its sources. The authors available in English best versed with Canarian sources are Charles Verlinden, ‘Italians in the economy of the Canary Islands’; FernándezArmesto, 1492; Abulafia, ‘Neolithic meets medieval’; and Abulafia, Discovery of Mankind.

620

The Spanish Destruction of the Canary Islands

Finally, on the place of the Canary Islands in wider debates on biological imperialism, settler colonialism and genocide, see Alfred Crosby, ‘Virgin soil epidemics’ and Ecological Imperialism; O’Flanagan, ‘Mediterranean and Atlantic settler colonialism’; and Mohamed Adhikari’s ‘Europe’s first settler colonial incursion in to Africa’ and ‘“We are determined to exterminate them”’.

621

24

Genocidal Massacres in the Spanish Conquest of the Americas Xaragua, Cholula and Toxcatl, 1503–1519 harald e. braun

This chapter will examine and compare three genocidal events in the early modern Caribbean and Mesoamerica: the massacres of Xaragua (Hispaniola, 1503), Cholula (Mexico, 1519) and Toxcatl (Mexico, 1520; also known as the massacre of Templo Mayor). These mass killings represent turning points in the history of the Spanish Atlantic conquest and share important characteristics. Each targeted Amerindian communities. Each was entirely or partially planned and executed by European actors, namely Spanish military entrepreneurs under the leadership of friar Nicolás de Ovando, Hernán Cortés and Pedro de Alvarado respectively. Each event can be described as a ‘genocidal massacre’ targeting a specific community because of its membership of a larger group.1 The European perpetrators intended to provide an object lesson for the surviving members of that group. These object lessons took the form of unexpected mass killing of civilians and combatants, and were executed – often combined with other forms of excessive violence – for the purpose of intimidating indigenous populations in order to facilitate their subjection to Spanish rule. Each of these acts of violence also met the cognate objective of destroying or significantly diminishing the political, religious and military leadership of the targeted community. Altogether, these three mass 1 I follow Leo Kuper’s definition of the term; see his Genocide: Its Political Use in the Twentieth Century (New Haven: Yale University Press, 1981), esp. ch. 4. For the state of research on violence in colonial Mesoamerica and Latin America more widely, see the chapters by J. G. McCurdy, W. Gabbert, H. Langfur and M. Restall in the Cambridge World History of Violence, vol. I I I : AD 1500 – AD 1800, ed. R. Antony, S. Carroll and C. Dodds Pennock (Cambridge University Press, 2020); an introductory, comparative overview of incidents of mass killings in colonial Latin America is E. A. Johnson et al., ‘Murder and mass-murder in pre-modern Latin America: from pre-colonial sacrifices to the end of colonial rule, and introductory comparison with European societies’, Historical Social Research 37 (2012), 233–53.

622

Genocidal Massacres in the Spanish Conquest of the Americas

killings emerged from a ‘dynamic of submission’: they were carried out in order to impose political, economic and cultural domination.2 There are differences, too. A complicating factor – in the cases of Xaragua and Cholula – is the role of indigenous actors as initiators, facilitators and perpetrators of massacre, a factor evident yet frequently distorted or ignored in European records.3 In the case of the Toxcatl massacre, it is possible that the intention to serve an object lesson and destroy the leadership of the targeted community mingled with or was even outweighed by a combination of greed and the ‘weight of fear and of the imaginary’ on the minds of the perpetrators.4 The Spanish adventurers perpetrating those massacres hailed from a violent place. The mass killing of civilians and prisoners of war was a depressing feature of European siege warfare, for instance during the Italian Wars of the late Renaissance.5 The Atlantic arena of early modern European colonial expansion, however, witnessed the escalation of specific forms of violence and shock tactics, such as the regular use of dogs in combat and regular, organised mass killings.6 Massacre in the Canary Islands, the Caribbean and the early Spanish main did not simply follow the – rather 2 As opposed to a ‘dynamic of eradication’, indicative of genocide and elimination of a group. See the brief discussion in J. Sémelin, ‘In consideration of massacres’, Journal of Genocide Research 3:3 (2001), 377–89, at 381. 3 Historical scholarship now acknowledges that even the description of indigenous actors as ‘allies’ providing logistical and military support to European invaders during the conquest of the Americas is prone to obscure or distort their agency and self-perception. Florine Asselbergs, Conquered Conquistadors. The Lienzo de Quauhquechollan: A Nahua Vision of the Conquest of Guatemala (Boulder: University Press of Colorado, 2004); and the contributions in L. E. Matthew and M. R. Oudijk, eds., Indian Conquistadors: Indigenous Allies in the Conquest of Mesoamerica (Norman: University of Oklahoma Press, 2007) are examples of works bringing indigenous perspectives back to life and into the study of early modern Mesoamerican and colonial history. 4 Sémelin, ‘Consideration’, 384. 5 On practices, perceptions and normative frameworks of violence in European theatres of war, see G. Parker, ‘The etiquette of atrocity: the laws of war in early modern Europe’ in Success is Never Final: Empire, War and Faith in Early Modern Europe (New York: Basic Books, 2002), pp. 143–68; and H. E. Braun, ‘Killing innocents? Massacre, war, and boundaries of violence in early modern Europe’ in A Cultural History of Genocide in the Early Modern World, ed. I. Pérez-Tostado (London: Bloomsbury, 2021), pp. 41–61. S. D. Dowd, Renaissance Mass Murder: Civilians and Soldiers during the Italian Wars (Oxford University Press, 2018) discusses a period of escalating violence in war in contemporary Europe. 6 J-F. Schaub, ‘Violence in the Atlantic: sixteenth and seventeenth centuries’ in The Oxford Handbook of the Atlantic World, 1450–1850, ed. N. Canny and P. D. Morgan (Oxford University Press, 2011), pp. 113–29, at p. 114 observes that ‘the threshold of tolerance for violence seems to have been lowered in Western Societies due to [the] Atlantic experience’. See also B. Sandberg, ‘Beyond encounters: religion, ethnicity, and violence in the early modern Atlantic world, 1492–1700’, Journal of World History 17 (2006), 1–25.

623

harald e. braun

malleable – rules of conduct observed in European theatres of war and conquest. Massacres in the form of cunningly conceived and carefully planned mass killings of large, unprepared communities became integral to the Spanish strategy of conquest and colonisation. Arguably, the experience of conquest in the Atlantic tested and stretched existing boundaries of violence and established a distinct yet increasingly controversial ‘etiquette of atrocity’.7 Genocidal massacres as a distinct form of excessive violence, then, marked the shift from initial exploration, slave raiding and trading to territorial conquest and subjection of populations to colonial rule and systemic exploitation of indigenous labour. Early massacres on the American main followed the loose pattern of treacherous hostage taking combined with genocidal atrocity established during the Spanish conquests of the Atlantic archipelagos and the Caribbean. The situation in the Canaries as well as the Caribbean differed from that on the American main in that the original number of indigenes was small and their technological capability limited. Strategies deployed during a protracted and costly war of attrition waged by technologically far superior European invaders against Neolithic populations in difficult territory and along fragile lines of supply, however, were successfully transmitted and adjusted to the technologically more advanced and more complex polities and vastly larger populations of the tierra firme. This escalation of violence in the Atlantic was driven by a number of factors. One was the mentalité of the so-called conquistadores, mostly younger men from the middle ranks of society and with the means to buy their own weapons and food to secure a place on exploratory expeditions and slave raids.8 They saw themselves as entrepreneurs whose legitimate business it was to subjugate and exploit Amerindian resources and the labour of Amerindian populations at breakneck speed. They justified their actions on the grounds that their ventures were self-funded and high-risk, often marked by extremely high casualty rates among European participants.9 They also strongly identified with the ideals and the rhetoric of the Reconquista. The 7 The term is borrowed from G. Parker; see ‘Etiquette of atrocity’. 8 For the social and economic profile of the conquistadores, see H. Thomas, Who’s Who of the Conquistadors (London: Cassell, 2000) and R. Himmerich y Valencia, The Encomenderos of New Spain, 1521–1555 (Austin: University of Texas Press, 1991); also J. Lockhart, Men of Cajamarca: A Social and Biographical Study of the First Conquistadors of Peru (Austin: University of Texas Press, 1972) and J. I. Avellaneda, The Conquerors of the New Kingdom of Granada (Albuquerque: University of New Mexico Press, 1996). 9 Fatality rates were significantly higher among African and indigenous slaves, carriers and fellow combatants, though conquistador narratives do not tend to acknowledge this fact.

624

Genocidal Massacres in the Spanish Conquest of the Americas

subjugation and exploitation of non-Christian peoples was understood as the reward for individual valour as well as the expression of divine providence. Victory against the odds was common in the Americas, Hernán Cortés proclaimed on the eve of the siege of Tenochtitlán in May 1521, because the Spanish ‘dare face the greatest peril, consider fighting their glory, and have the habit of winning’.10 He also reminded his companions that the ‘principal reason for our coming to these parts is to glorify and preach the faith of Jesus Christ, even though at the same time it brings us honour and profit’. The Spanish, Cortés concluded, were instruments of divine justice, because the Mexica were cannibals and deserving of a ‘great punishment’. Papal bulls authorising conquest and colonisation as well as the legal instruments by which the Castilian crown aimed to regulate its progress, like the Requerimiento, sanctioned and nourished this belief in superiority over the indigenous peoples of the Americas.11 The relative absence of royal and ecclesiastical authority during the first few decades following Columbus’ landfall further emboldened this sense of entitlement. On the one hand, Atlantic conquest operated within a framework of legitimacy and probity unilaterally defined by the Crown. Conquistadors depended upon royal licensing, recognition and endorsement of their actions and achievements while at the same time entertaining a profound disdain for secular and ecclesiastical authority as parasitical beneficiaries of individual valour and sacrifice. Royal officials and missionary 10 Francisco López de Gómara, Historia de la conquista de México (Zaragoza: Agustín Millán, 1552), trans. L. B. Simpson, Cortés: The Life of the Conqueror by His Secretary (Berkeley: University of California Press, 1964), p. 240. In his Historia, which is effectively a biography and defence of Cortés, Gómara claims to render a speech delivered before the Spanish left Tlaxcala to commence the siege of Tenochtitlán. Historians now trace and dissect the glorifying ‘mythstory’ of conquest and conquistador constructed within the frame of European culture, consciousness and desires. See M. Restall, Seven Myths of the Conquest (Oxford University Press, 2003) and When Moctezuma met Cortés: The True Story of the Meeting That Changed History (New York: Ecco, 2018). 11 English translations of the papal bulls and treaties providing much of the legal framework for Iberian expansion in the Americas – especially Inter caetera, issued by Pope Alexander VI to the Catholic Monarchs on 4 May 1493, and the Treaty of Tordesillas concluded between Castile and Portugal in 1494 – are in F. G. Davenport, European Treaties Bearing on the United States to 1648 (Washington, DC: Carnegie Institution of Washington, 1917): Inter caetera, pp. 71–8; Tordesillas, p. 95. For the English text of the Requerimiento – drafted by the Castilian lawyer Juan López de Palacios Rubios on behalf of the Castilian crown in 1513 in order to channel papal doctrine – see, for instance, J. Cowans, Early Modern Spain: A Documentary History (Philadelphia: University of Pennsylvania, 2003), pp. 34–6. A recent discussion of papal doctrine framing European conquest is D. M. Lantigua, Infidels and Empires in a New World Order (Cambridge University Press, 2020).

625

harald e. braun

friars at the coalface of conquest and colonisation, on the other hand, often lacked the will and support or the resources to restrain conquistador violence. Some shared in the disregard for the lives (and souls) of indigenous peoples, as appears to have been the case with the Hieronymite friars on Hispaniola. Their attitude reflected the fact that the theological–legal vernacular of violence available during the first decades of conquest tended to blur the boundaries between Amerindian pagans, Muslim enemies of the Christian faith and heretics.12 The conquistadors’ ruthless pursuit of personal gain, though, increased tensions with secular and ecclesiastical authorities’ intent on preserving, converting and establishing political and economic dominion over Amerindian peoples. The Crown and Church of Castile had felt the need to protect indigenous populations early on.13 With the dramatic demographic decline of indigenous Caribbean populations – partly the work of European diseases14 – the violence of conquest became problematic. The impassioned sermon friar Antonio de Montesinos delivered to Spanish settlers on Hispaniola (now Haiti and the Dominican Republic) on 21 December 1511 marked the first known public denouncement of settler violence. The question of whether violent conquest and excessive violence facilitated or obstructed the legitimising objective of Catholic mission became a pressing issue during the decades leading up to the mid-sixteenth century.

Xaragua 1503 The massacre of Xaragua took place on Hispaniola in 1503. It represents the paradigm for shock mass killing and the extermination of indigenous leaders as integral to the subjugation of Caribbean and subsequently Amerindian peoples to Spanish colonial rule. It was ordered by friar Nicolás de Ovando (1460–1511), a knight and commander of the military order of Alcántara and 12 For juridical-theological negotiations of the violence of conquest and empire, see, for instance, Lantigua, Infidels; and the relevant chapters in J. Tellkamp, ed., A Companion to Early Modern Spanish Imperial and Political and Social Thought (Leiden: Brill, 2020). 13 A good example are the attempts to regulate the slave trade in the Caribbean during the late fifteenth and early sixteenth centuries; see Erin Woodruff Stone, Captives of Conquest: Slavery in the Early Modern Spanish Caribbean (Philadelphia: University of Pennsylvania Press, 2021). 14 For the complex, at times heated debate on the relative impact of European pathogens on Amerindian populations, see N. D. Cook, Born to Die: Disease and New World Conquest, 1492–1650 (Cambridge University Press, 1998); M. L. Bacci, Conquest: The Destruction of the American Indios, trans. C. Ipsen (Cambridge: Polity Press, 2008); and M. L. Bacci, ‘The depopulation of Hispanic America after the Conquest’, Population and Development Review 32:2 (2006), 199–232.

626

Genocidal Massacres in the Spanish Conquest of the Americas

royal Gobernador general de Las Indias (1502–9). Ovando established royal control over Spanish settlers by settling them in townships, adapting and imposing the encomienda system successfully practised on Muslim populations in Castile as a means to organise, instruct and exploit Amerindian populations.15 He pursued and completed the conquest of Hispaniola with a ruthlessness and a degree of treacherous brutality already noted by contemporaries, including the Dominican friar Bartolomé de Las Casas (1484– 1566), the most prominent critic of the conquest.16 At some point in 1503, Ovando called a meeting with Anacaona, the chief cacique or ‘Queen’ of Xaragua, one of the main chiefdoms of the Taino people of Hispaniola. Xaragua paid tribute to the Spanish at the time, but was able to resist more pervasive forms of Spanish rule. The fact that Xaragua hosted a sizeable number of Spanish deserters and rebels married to Taino women caused further concern and suspicion. From Ovando’s point of view, Xaragua represented a latent threat to Spanish rule on Hispaniola. The meeting, allegedly arranged to consult and celebrate peace and friendship, was a pretext for extinguishing this real, imagined or purported menace. More than 300 Spanish soldiers accompanied by an indeterminate number of Taino warriors from the neighbouring chiefdom of Marién, rivals of the Xaragua at the time, arrived at the agreed place. Once the festivities had begun, on a signal from Ovando, the Spanish and their allies apprehended the caciques and slaughtered their entourage, with many women and children among the victims. It is likely that Spanish rebeldes with their Taino wives and mestizo children were among the victims.17 Anacaona was captured and soon afterwards executed by hanging. Many Xaragua caciques taken prisoner were 15 The encomienda system granted indigenous labour and tribute to a settler (encomendero), in return, theoretically, for spiritual instruction and legal protection within nascent colonial society. 16 Our main sources are Bartolomé de Las Casas, Brevísima relación de la destrucción de las Indias (manuscript completed in 1542 and circulated widely thereafter; first printed Sevilla, 1552); I reference Bartolomé de Las Casas, A Short Account of the Destruction of the Indies, ed. and trans. N. Griffin, with an introduction by A. Pagden (London: Penguin, 1992), esp. ‘Events at Cholula’, pp. 21–2. Gonzalo Fernández de Oviedo de Valdés, Historia general y natural de las Indias (first version, Toledo, 1526); I reference Gonzalo Fernández de Oviedo de Valdés, Historia general y natural de las Indias, ed. J. Pérez de Tudela Bueso, vol. I I , Biblioteca de Autores Españoles 118 (Madrid: Ediciones Atlas, 1959), esp. ‘Events at Cholula’, p. 103. We have no testimony from either indigenous or Spanish survivors. 17 Las Casas reports that some Spanish horsemen attempted to save the lives of some of the indigenous children while others cut them down methodically. It is possible that members of Ovando’s company took pity or wanted to take slaves or sought to preserve the lives of mestizo children.

627

harald e. braun

burned alive.18 Some of the surviving caciques fled to Cuba, where they would soon confront one of Ovando’s captain’s in charge of the killing on the day, Diego Velázquez de Cuéllar (1465–1524). The massacre eliminated the Xaragua as a political and military force and as a rival for both the Spanish and other Taino chiefdoms, and marked the beginning of the end of the process of Spanish subjugation of Hispaniola. Ovando appears to have found it necessary to take statements from those involved in the massacre in order to prove that the betrayal and slaughter of an ally had been justified – an indication that he expected criticism – though these have not been found.19 He was recalled in 1509 and replaced by Diego Colón, who would seek to rule more benignly over Spanish settlers and indigenous alike. At the time of Ovando’s recall, it was evident to the Castilian crown that European diseases, settler violence and existing strategies of economic exploitation wreaked havoc among the indigenous peoples. The justification of genocidal massacres became more tenuous if the perpetrator did not deliver on the moral, political and economic objectives of crown and church: conversion of a demographically stable indigenous population and delivery of stable revenue. Hernán Cortés learned the ropes of conquest under the patronage of Ovando and rose under Velázquez, but would prove remarkably adept at presenting his actions in ways aligned with changing expectations on the part of secular and ecclesiastical authorities. The massacre of Xaragua displays core characteristics of early sixteenthcentury Spanish massacres of conquest in the Atlantic. Invariably, these massacres happened within the context of indigenous political structures, conflict and competition. They show the Spanish as generally competent in reading, manipulating and exploiting indigenous political dynamics to their advantage. They usually involve indigenous individuals and groups not only as victims or bystanders but as pursuing their own agenda while providing intelligence as well as logistical and active military support for the European invaders. The aim and outcome of the massacres was the extermination of indigenous leaders and the imparting of terror. The use of terror – including other stock measures such as public executions by hanging or burning and 18 Las Casas claims that more than 300 caciques were burnt alive and ‘countless’ other natives slaughtered; Oviedo speaks of sixty executed caciques. Some years after the massacre, Dominican and Franciscan friars reported the burning of forty caciques, Colección de documentos inéditos relativos al descubrimiento, conquista y organización de las antiguas posesiones españolas de América y Oceanía (Madrid: M. Bernaldo de Quirós, 1867), V I I , p. 410. 19 T. S. Floyd, The Columbus Dynasty in the Caribbean, 1492–1526 (Albuquerque: University of New Mexico Press, 1973), p. 63.

628

Genocidal Massacres in the Spanish Conquest of the Americas

the feeding of dismembered bodies to dogs – was a workable strategy of conquest, and even a psychological necessity for tiny, beleaguered bands of violent entrepreneurs operating in difficult, unfamiliar territory, surrounded by enemies and unreliable allies, and depending on extremely stretched lines of supply.

Cholula 1519 and Toxcatl 1520 Frustrated Spanish settlers and adventurers eventually extended their search for precious minerals, cheap labour and social status from the exhausted Caribbean archipelago to the American main. Many of them veterans of the Caribbean campaigns, they imported their ambitions, rivalries and tactics of conquest, including torture, hostage taking of indigenous leaders under diplomatic pretext and genocidal massacre. Two exploratory expeditions (1517, 1518) – supported and partly funded by Diego Velázquez, now adelantado of Cuba – established contact with highly developed coastal Mayan communities.20 The second expedition under Juan de Grijalva was met by envoys of Moctezuma II, the leader (tlatoani) of the Mexica, the dominating member of the Triple Alliance of Nahuatl-speaking communities ruling over much of what is today central Mexico. Communication was limited. The envoys deployed the customary tactics of Mesoamerican diplomacy, seeking to impress and possibly intimidate the strange foreigners with gifts and references to Moctezuma’s power. Contrary to likely intention, the splendid attire and especially the rich gold ornaments worn by the envoys made the Mexica the target of Spanish conquest. Velázquez ordered and partly financed a further expedition under his client and kinsman Hernán Cortés. Undoubtedly intent on being in charge of future conquest, he would find himself outmanoeuvred by the latter. On reaching the point of contact between Grijalva and the ambassadors of Moctezuma II, the expedition founded the town of Veracruz and elected Cortés alcalde. This symbolic act aimed to establish a direct relationship and right to 20 For a detailed account of events, though beholden to the European hero-leader figure, see H. Thomas, Conquest: Moctezuma, Cortés, and the Fall of Old Mexico (London: Hutchinson, 1993); revising Eurocentric perspectives and acknowledging the complexity of the sources, for instance, are Restall, When Moctezuma Met Cortés, and D. M. Carballo, Collision of Worlds: A Deep History of the Fall of Aztec Mexico and the Forging of New Spain (Oxford University Press, 2021). For the politics of conquest on the Spanish side, see the excellent essay by J. H. Elliott, ‘Cortés, Velázquez, and Charles V’ in Hernán Cortés, Letters from Mexico, trans., ed. and with a new introduction by A. Pagden (New Haven: Yale University Press, 1986), pp. xi–xxxvii.

629

harald e. braun

communication between the group and the emperor over the head of Velázquez. As a result, subsequent events were characterised not only by intercultural conflict but also by substantial acts of violence among the conquistadors fuelled by the enmity between Cortés, Velázquez and their followers and supporters. The situation also fostered tensions and a widespread attitude of ‘all or nothing’ among members of the expedition. They had taken unfathomable risks by placing themselves between indigenous communities of unknown strength and the most powerful Spanish leader in the Caribbean. They were condemned to succeed in their objective to subject a sizeable part of the indigenous population and territory on the American main on terms that would impress the Crown and secure its protection. In his letters (cartas de relación) to the emperor, Cortés maintains that the expedition consistently followed Spanish protocol of conquest, for instance by referencing the Requerimiento.21 Read out to uncomprehending indigenous populations, the document demanded voluntary subjection to the rule of the Catholic rulers of Spain. If they chose to resist, they would suffer punishment, even enslavement. By the time Cortés dropped anchor at Veracruz, however, the Requerimiento had been subject to criticism, especially from ecclesiastical quarters, and the threat of slavery had been partially disowned by the Crown. This, and probably even more so the need to receive recognition for their actions from the Crown and over the head of Velázquez, convinced Cortés and his followers that their conquest had to be presented as a series of acts of ‘voluntary’ submission to royal authority. Correspondingly, acts of genocidal violence, whether prompted by indigenous resistance or not, were described as pre-emptive strikes provoked by indigenous ‘treason’ and ‘rebellion’ mingled with insufferable violations of European norms (such as human sacrifice). Cortés’ correspondence with Charles V is carefully curated to profile himself and his expedition – in deliberate contrast to Caribbean conquest and conquistadors like Velázquez – as the beginning of a process of sustainable longterm European settlement and rule over indigenous peoples. The victims of the massacres of Cholula and Toxcatl were the Mexica (aka Aztecs) and other ethnically, culturally and politically aligned Nahuaspeaking communities.22 The Mexica had been late arrivals to the fertile 21 Cortés, Letters, ‘First Letter’, pp. 10, 21, and ‘Second Letter’, p. 59. The fact that the Requerimiento was not read out at Cholula led to a charge against Cortés in his residencia. 22 The term ‘Aztec’ is a nineteenth-century invention used to describe Nahua-speaking, ethnically and culturally related communities in the basin of central Mexico still

630

Genocidal Massacres in the Spanish Conquest of the Americas

landscape of the Valley of Mexico, but rapidly expanded from their centre, the twin city of Tenochtitlán-Tlatelolco situated in the marshlands of Lake Texcoco. In 1440, they formed a triple alliance with the altepetls (communities with urban centres and surrounding areas headed by a tlatoani) of Texcoco and Tlacopan, and set out on annual campaigns of aggressive expansion, establishing tributary overlordship over most of central Mexico. Despite its aggressive stance, military prowess and wealth, this was a comparatively fragile political structure characterised by rivalry among the members of the alliance and under persistent pressure to suppress resistance and continue expansion. Tensions between the Mexica, their allies and subject peoples might have been exacerbated by the excessive practice of ritual violence. Like most Mesoamerican peoples, the Mexica practised human sacrifice, sacrificing enemy warriors taken prisoner on the battlefield as well as victims delivered as tribute.23 Under tlatoani Ahuitzotl (r. 1486–1502), the taking of sacrificial victims – by means of so-called ritual Flower Wars traditionally arranged between communities for that purpose – proliferated and escalated. Cortés and other Spanish sources claim that the coastal communities allying themselves to the Spanish had pointed to excessive demand for sacrificial victims as a particular pressure point. In short, there were significant holes in the tapestry of Mexica dominance. Most importantly, some key regional powers had maintained their independence or were in a position to reclaim their independence from the Triple Alliance, notably the altepetls of Cempoala and Tlaxcala.

Cholula The massacre of Cholula took place in mid- to late October 1519, when the Spanish and a large number of warriors from Cempoala and Tlaxcala advanced from the coast on to the rich and expansive agricultural community

popular with historians for reasons of convenience. The ‘Aztecs’ referred to themselves as Mexica-Colhua. Still one of the most comprehensive and sensitive approaches to precolonial Mexica or Aztec society is I. Clendinnen, The Aztecs: An Interpretation (Cambridge University Press, 2014). F. F. Berdan and M. E. Smith, Everyday Life in the Aztec World (Cambridge University Press, 2020) gives a panoramic view of Mexica life. 23 On human sacrifice, warfare and empire in precolonial Mesoamerica, see, for instance, R. Hassig, Aztec Warfare: Imperial Expansion and Political Control (Norman: University of Oklahoma Press, 1988); D. Carrasco, City of Sacrifice: The Aztec Empire and the Role of Violence in Civilization (Boston: Beacon Press, 1999); and C. Dodds Pennock, Bonds of Blood: Gender, Life-Cycle and Sacrifice in Aztec Culture (Basingstoke: Palgrave Macmillan, 2008).

631

harald e. braun

of Cholula. Cholula was situated in the Puebla–Tlaxcala valley in central Mexico, halfway between Tenochtitlán and the Spanish anchor point at the coast. The city was an important trade hub and religious centre hosting a major temple of the god Quetzalcoatl, one of the most important deities in the shared Mesoamerican pantheon. Its relation to the Triple Alliance is no longer altogether clear. Historiography tended to describe Central American indigenous communities as politically unified city-states, ethnically and politically more homogeneous even than the likes of Renaissance Florence. This certainly was the understanding of the Spanish. Cholula, though, might rather have been a cluster of communities with differing histories and fluid, sometimes conflicting political allegiances. Though difficult to verify in the sources, this structure might have played into the massacre. The complex make-up of indigenous urban and political conglomerates certainly added a layer of complication and potential peril in situations of political crisis. This played out powerfully during the final months of Tenóchtitlan-Tlatelolco, when many previous allies of the Triple Alliance and even political groupings within Tlatelolco, resentful of Tenóchtitlan’s hegemony, decided to side with the Spanish. In the case of the massacre of Cholula, extreme hostility between the altepetls of Cholula, Tlaxcala and Cempoala appears to have been a major, albeit not fully determinable factor. Our information about the massacre comes almost exclusively from Spanish sources.24 The main sources are Cortés’ second letter to emperor Charles V, testimony from Spanish eyewitnesses and other contemporary accounts, Tlaxcalan accounts and narratives provided by the Mexica allies of the Cholulans. We have no Cholulan testimonies. The sources betray politically motivated and ultimately irreconcilable disagreement over what happened: whether a pitched battle involving armed combatants or an unprovoked attack on an unprepared community took place, whether or not women and children were among the victims, and who ultimately instigated the violence. Estimates concerning the number of victims vary considerably, ranging from some 2,000 to around 27,000 (out of an estimated

24 A survey of the main textual sources, including telling archaeological evidence, is G. G. McCafferty, ‘The Cholula massacre: factional histories and archaeology of the Spanish conquest’ in The Entangled Past: Integrating History and Archaeology, ed. M. Boyd, J. C. Erwin and M. Hendrickson (University of Calgary Press, 2000), pp. 347–59.

632

Genocidal Massacres in the Spanish Conquest of the Americas

urban population of approximately 30–50,000).25 Population estimates, analysis of historical accounts and archaeological evidence suggest a number between 2,000 and 6,000, with many women and children among the victims. Spanish sources stress that the community subsequently received settlers from other altepetls and continued to exist. Cortés embedded Cholula in a carefully construed sequence of events leading up to the alleged submission of Moctezuma II to Charles V on 8 November 1519, less than a month after the massacre.26 He presents Spanish actions as a pre-emptive strike punishing a treacherous Cholulan– Mexica plot to overwhelm his forces. Cortés’ main objective is to portray himself as a capable commander and selfless servant of Charles V. He is at pains to stress that he and his men observed Spanish law, custom and honour in dealing with indigenous peoples. The Spanish, he posits, exerted legitimate pressure on Moctezuma and received an invitation to proceed to Tenochtitlán passing through Cholula en route. Integral to his justification of subsequent actions is the claim that Mexica and Cholulan promises of safe passage were part of a treacherous design to lure his company into a deathtrap, just as his Tlaxcalan and Cempoalan allies had warned would be the case. The Spanish and their allies, Cortés claims, discovered and avoided a first ambush laid by a large Mexica army along the main road to Cholula by taking a different route. After their arrival in Cholula and a seemingly friendly welcome, though, clear signs of hostile intent piled up. The Spanish were not offered sufficient provisions. Women and children were nowhere to be seen and appeared to have abandoned the city. Cortés reports this as evidence that the city was preparing for battle. Intelligence from different sources – locals tortured and interrogated, corroborating information gathered by Cortés’ indigenous interpreter Malinche (whom he does not name) – provided further evidence that the Cholulans intended to strike against the Spanish. With all escape routes blocked and under threat of imminent attack, the Spanish decided that a pre-emptive strike was the only solution. Consequently, the Cholulan chieftains were invited to the Spanish compound under a pretext and taken prisoner, though not harmed, according to Cortés. The Spanish then fell upon the surprised Cholulan warriors and, with the help of allies rallying to their plight, defeated them after several hours of intense combat. Other Spanish eyewitness accounts, recorded well after the event, offer partially overlapping yet in some important aspects strikingly different accounts. The narratives of Andrés de Tapia, Francisco de Aguilar, 25 Ibid., 355.

26 Cortés, Letters, ‘Second letter’, pp. 69–75, esp. pp. 72–3.

633

harald e. braun

Bernadino Vázquez de Tapia, Bernal Díaz de Castilla, Bartolomé de Las Casas and Francisco López de Gómara discussed below, as well as other testimonies and accounts, have to be read in the light of personal rivalries among conquistadors and rising tensions between conquistadors and the authorities framed by the evolving early sixteenth-century debate concerning the legitimacy of Spanish conquest and conduct in the Americas. Indigenous sources composed after the conquest, too, are prone to political bias, and reflect entrenched ethnic rivalries as well as the fact that many communities were keen to negotiate their position within the emerging Spanish hegemony. They tend to negotiate the choppy waters of memory during the years immediately after the conquest by foregrounding superstition and treachery on the part of their indigenous rivals and carefully avoid apportioning blame or primary agency to the Spanish. Andrés de Tapia (1498?–1561) was one of Cortés’ captains and a lifelong friend and supporter. His gave testimony during the residencia into Cortés’ actions and penned his own narrative during the late 1540s.27 Tapia claims that Cholulan and Mexica dignitaries intended to send the Spanish on their way to Tenochtitlán accompanied by a large escort of Cholulan warriors with orders to ambush them once they reached a particularly difficult section of the road. He also stresses the crucial role of Malinche in discovering the plot and informing the Spanish (though, like Cortés, he does not give her name). On receiving evidence of the plot, Cortés decided to strike first, and on the day of departure apprehended most of the local leaders, extracted confessions without torture, and killed most of them for their treachery. The Spanish and the Amerindians in their company then fell upon the warriors disguised as carriers who had gathered in front of the Spanish quarters. After killing the warriors assembled in the courtyard, the Spanish and their allies went out to burn and ravage the city and temples for two more days, soon supported by numerous Tlaxcalan reinforcements. Tapia 27 A residencia was the routine examination of the conduct of an official after the end of their term of office, an integral part of the Spanish administrative system. Cortés’ residencia was conducted from 1526 to 1545, but never concluded; see the documentation in L. Martínez, ed., Documentos cortesianos II, 1526–1545. Sección IV: Juicio de Residencia (Mexico City: Universidad Autónoma de México / Fondo de Cultura Económica, 1991), including Tapia’s testimony concerning Cholula, pp. 599–602; Andrés de Tapia, Relación de algunas cosas de las que acaecieron al muy ilustre señor don Hernando Cortés, marqués del Valle . . . in Collección de documentos para la historia de México, ed. J. G. Icasbalceta (Mexico City: J. M. Andiade, 1866), I I , pp. 554–94; and for an English translation of the relevant passages see S. B. Schwartz, Victors and Vanquished: Spanish and Nahua Views of the Conquest of Mexico (Boston: Bedford/St Martin’s Press, 2000), pp. 115–19.

634

Genocidal Massacres in the Spanish Conquest of the Americas

claims that the Spanish, at the express order of Cortés, did not kill any women or children, but that many were killed by indigenous allies who did not feel bound by his order. Though he does not give numbers, his narrative implies a very high casualty count. His account only partly overlaps with that of Francisco (formerly Alonso) de Aguilar (1479–1571), a conquistador and eyewitness who wrote his Relación breve de la conquista de la Nueva España in the 1560s, after he had joined the Dominican order.28 Aguilar confirms a growing sense of hostility after the Spanish had arrived in the city, though no plot or provocation. After several days during which the local people did not bring food and water, ‘the captains’ acted against the advice of Cortés and decided to kill the Amerindians who had eventually arrived to deliver foodstuffs and water. He gives the number of victims as approximately 2,000. The captains took this decision, according to Aguilar, in order to intimidate the Mexica and secure a safe passage to Tenóchtitlán and quick submission. Bernadino Vásquez de Tapia (no relation to Andrés), who developed a notoriously inimical relationship with Cortés as well as his second-incommand Pedro de Alvarado in the years after the conquest, is the eyewitness deviating most blatantly from Cortés’ account.29 He claims that there was no plot to ambush the Spaniards, that the Cholulans provided everything the Spaniards requested, and that he, though a captain, had received no notice or orders when Cortés suddenly ordered the killing of around 4,000 Amerindian porters assembled in the courtyard before their quarters. The reason for Cortés’ decision to order the massacre, Vásquez de Tapia states – here in line with Aguilar’s account – was the hope that indigenous peoples would be deterred from further resistance and the Mexica and Moctezuma would be subdued more easily. Another co-perpetrator, Bernal Díaz del Castillo, offers a more comprehensive and rhetorically enhanced account of the massacre in his Historia 28 There are several editions, including Francisco de Aguilar, Relación breve de la conquista de la Nueva España, ed. J. Gurría Lacroix (Universidad Nacional Autónoma de México, 1977). For an English translation of the relevant passages see P. de Fuentes, ed., The Conquistadors: First-Person Accounts of the Conquest of Mexico (Norman: University of Oklahoma Press, 1993), pp. 143–4. 29 For his testimony for the residencia concerning events at Cholula see Martínez, ed., Documentos cortesianos II, pp. 56–7; Bernadino Vázquez de Tapia, Relación de merítos y servicios del conquistador Bernadino Vázquez de Tapia, vecino y regidor de esta gran ciudad de Tenustitlan, ed. J. Gurría Lacroix (Universidad Nacional Autónoma de México, 1972); translation of extracts in D. A. Peterson and Z. D. Green, eds., ‘The Spanish arrival and the massacre at Cholula’, Notas Mesoamericanas 10 (1987), 203–22, esp. 209–10.

635

harald e. braun

verdadera de la conquista de Nueva España.30 He describes the negotiations between Cortés, Moctezuma’s ambassadors and the Cholulans as well as the failed ambush on the road to Cholula. He provides detail on Cholulan preparations for a surprise attack on the Spanish while still housed in the city: holes had been dug in the streets and furnished with stakes to kill the horses; side streets had been blocked with barricades; missiles had been collected and placed on the roofs of houses. He also names and gives much credit for Spanish survival to Malinche. Díaz deviates from other defensive accounts in crucial points. He agrees with Andrés de Tapia that women and children had not been evacuated, but claims they actively participated in the preparations for war, gathering stones and other projectiles, and preparing pots with tomato and chilli to cook human flesh. According to Díaz, the entire community – warriors and civilians, including women and even children – was actively involved in the treacherous plot, and therefore deserved punishment. The actual fighting and killing, though, he says, only involved the warriors. Like many of the most outspoken critics of Spanish conduct of war in the Americas, Bartolomé de Las Casas was a member of a religious order. He joined the Dominican order after witnessing the brutal conquest of Hispaniola and Cuba. Las Casas did not query the legitimacy of Spanish rule in the Americas, sanctioned by papal bulls, but focused firmly on the violence of conquest as a human tragedy, contrary to Christian ethics and dogma, and a fateful obstruction to peaceful mission and conversion of the Amerindians. Up to the later 1550s, such criticism circulated freely in the Americas and in Spain, not least at the royal court. The description of events at Cholula in his Brevísima relación is representative of his critique of Spanish strategy and use of violence.31 The friar denies the existence of a plot. The Cholulan dignitaries were deceived, tortured and later burned at the stake for no fault of their own. The victims of the subsequent mass slaughter were helpless slaves willingly sent as carriers. The massacre was staged simply in order to conquer by means of intimidation. In fact, Las Casas defines Cholula and similar events during the conquest in Leo Kuper’s terms as an object lesson for other members of a group. Once they had arrived at Cholula, he says, the Spanish 30 Bernal Díaz’s text is available in a number of English editions and translations. I reference Bernal Díaz del Castillo, Historia verdadera de la conquista de Nueva España, ed. David Carrasco (Albuquerque: University of New Mexico Press, 2008), ‘The massacre at Cholula’, pp. 106–15. 31 Las Casas, Destruction, pp. 45–8.

636

Genocidal Massacres in the Spanish Conquest of the Americas

decided that the moment had come to organize a massacre (or ‘punishment’ as they call it) in order to inspire fear and terror in all the people of the territory. This was, indeed, the pattern they followed in all the lands they [the conquistadors generally] invaded: to stage a bloody massacre of the most public possible kind in order to terrorize those meek and gentle peoples.32

Las Casas’ fervent, at times polemical and altogether efficient critique made him the bête noire of conquistadors like Bernal Díaz and apologists like Ginés de Sepúlveda or the Sevillian humanist Francisco López de Gómara (1511–66). The latter, in his Historia general de las Indias, set out to provide a narrative legitimising Spanish conquest in the Americas.33 His version of events at Cholula reflects that intention. Faced with the need to reconcile divine providence with the startling violence, greed, and destruction wreaked by the conquistadors, Gómara resolved to heroise Cortés.34 Concerning Cholula, he alleges that Mexican–Cholulan treachery was the sole cause of violence.35 According to Gómara, a plot to slaughter the Spanish on arrival in Cholula contrived by Moctezuma, the Cholulans and a high-ranking Tlaxcalan noble was uncovered and reported by an unnamed woman given to Pedro de Alvarado while the Spanish still rested in Tlaxcala. The commander was strangled on Cortés’ orders, apparently without a Tlaxcalan backlash. Like Cortés, he mentions a Mexica force – he speaks of more than 30,000 men – waiting to ambush the Spaniards. Cortés, though, does not again mention it as in any way involved in the actual confrontation. Gómara explains that the Cholulans did not trust Moctezuma enough to admit his troops into their town. He then follows Cortés’ second letter, but offers more detail. He mentions the entrenched enmity between Tlaxcalans and Cholulans, initially generous provision of food soon replaced by hostile behaviour, the removal of women and children from the city, and a Cholulan plot to overwhelm the Spanish on their way to Tenochtitlán uncovered by Malinche (he mentions her by name). He also records that Cortés extracted confessions from Cholulan leaders through intimidation rather than torture, and ordered several of them to be killed as punishment. The Spanish, supported by 32 Ibid., p. 45. 33 See note 10 above. 34 On Gómara’s efforts to reconcile violent imperialism and divine providence, see C. A. Roa-de-la-Carrera, Histories of Infamy: Francisco López de Gómara and the Ethics of Spanish Imperialism, trans. S. Sessions (Boulder: University of Colorado Press, 2005), and G. E. Carman, Rhetorical Conquests: Cortés, Gómara, and Renaissance Imperialism (West Lafayette: Purdue University Press, 2006). 35 Gómara, Life of the Conquerer, trans. Simpson, esp. pp. 123–30.

637

harald e. braun

indigenous reinforcements that had been secretly summoned, then attacked and defeated the Cholulan warriors in fierce battle, burned the temple and sacked the city. Gómara emphasises Cholulan fear of Moctezuma, treachery, and the unnatural desire to gorge on human flesh. His account of events ends with a lasting reconciliation between Tlaxcala and Cholula negotiated by Cortés. Accounts from Tlaxcala tend to stress its early alliance with the Spanish and conversion to Christianity. The Tlaxcalan chronicler Diego Muñoz Camargo (1529–99) in his History of Tlaxcala (Historia de Tlaxcala; completed before 1585) claims that the Cholulan leaders dismissed the Tlaxcalan offer of peace and alliance with the Spanish.36 Convinced that Quetzalcoatl would assist their warriors and consume the Spanish invaders with fire and water, they even broke Mesoamerican diplomatic protocol by mutilating the Tlaxcalan envoy. When defeat became obvious, many Cholulans became victims of their heathen superstition and committed suicide by throwing themselves from the walls of their burning temples (the same claim is made by Andrés de Tapia). Muñoz Camargo’s account is not corroborated by other than Tlaxcalan sources and can be read as an attempt to put the blame for the destruction of Cholula squarely on the Cholulans and erase memory of the massacre from the historical record. The perceived need to do so points at Tlaxcalan agency at Cholula as well as awareness of Tlaxcalan responsibility for what happened at Cholula among other Nahua-speaking communities. The Lienzo de Tlaxcala – a series of cloth paintings mixing indigenous and European visual traditions created during the mid-1500s and one of Muñoz Camargo’s sources – adds another facet to a complex sequence of events.37 Though a rather condensed visual representation of events, the Lienzo points to Malinche as an important indigenous actor, conquistadora, and a crucial conduit for Tlaxcalan–Spanish collaboration, especially in the run-up to the massacre of Cholula (Figure 24.1).38 36 Modern editions include Diego Muñoz Camargo, Historia de Tlaxcala, ed. G. Vázquez, Crónicas de América 26 (Madrid: Librería Rodréguez, 1986); relevant extracts in English in M. León-Portilla, ed., Broken Spears: The Aztec Account of the Conquest of Mexico, expanded and updated edition (Boston: Beacon Press, 1992), pp. 41–9. 37 The original Lienzo de Tlaxcala is now lost, but for a digital reconstruction see the Universidad Autónoma de México’s well-advanced project based on the Glasgow manuscript, https://lienzodetlaxcala.com. 38 On representations of Malintzin/Malinche/Doña Marina in indigenous painted sources, see C. Townsend, Malintzin’s Choices: An Indian Woman in the Conquest of Mexico (Albuquerque: University of New Mexico Press, 2006), pp. 67–79; also C. J. Rogers, ‘Malintzin as a conquistadora and warrior woman in the Lienzo de

638

Genocidal Massacres in the Spanish Conquest of the Americas

Figure 24.1 La Matanza de Cholula, Lienzo de Tlaxcala.

Indigenous agency is also confirmed and highlighted by sources closer to the victims. This is what Mexica sources (mainly from Tlatelolco) relay through the mediation of friar Bernadino de Sahagún (1499–1590): The Spaniards asked: ‘Where is [Tenochtitlan]? Is it far from here?’ The Tlaxcaltecas said: ‘No, it is not far, it is only a three-day march. And it is a great city. The Aztecs are very brave.’ . . . At this time the Tlaxcaltecas were enemies of Cholula. They feared the Cholultecas; they envied and cursed them; their souls burned with hatred for the people of Cholula. This is why they brought certain rumours to Cortés, so that he would destroy them. They said to him ‘Cholula is our enemy. It is an evil city. The people are as brave as the Aztecs and they are the Aztecs’ friends.’ When the Spaniards

Tlaxcala (c.1552)’, Historical Journal 64:5 (2021), 1173–97; briefly McCafferty, ‘Cholula massacre’, 356–8.

639

harald e. braun

heard this, they marched against Cholula. They were guided by the Tlaxcaltecas and the chiefs from Cempoala, and they all marched in battle array. . . . When they arrived, the Tlaxcaltecas and the men of Cholula called to each other and shouted greetings. An assembly was held in the courtyard of the God, but when they all gathered together, the entrances were closed, so that there was no way of escaping. Then the sudden slaughter began: knife strokes, and sword strokes, and death. The people of Cholula had not foreseen it, had not suspected it. They faced the Spaniards without weapons, without their swords or their shields. The cause of the slaughter was treachery. They died blindly, without knowing why, because of the lies of the Tlaxcaltecas.39

The sources collected and curated by Sahagún stress the psychological impact of the massacre – shock and terror among the communities associated with the victims, and despair on the part of Moctezuma – along lines similar to what we read in Las Casas. These are colonial sources, though, and it is likely that retrospect presentation of events and identification of agency was shaped by the intention to shift some of the agency and moral responsibility from the new Spanish overlords to indigenous rivals (Tlaxcala) and scapegoats (the role allocated to a ‘weak’ Moctezuma). Spanish accounts of events at Cholula focus on the legitimacy of violence and hinge on two main points: the existence of an indigenous plot that provoked and justified a pre-emptive strike and due punishment of treachery, and the status of the victims (warriors, unarmed temames (porters), women and children). While an indigenous plot remains a possibility, existing archaeological and textual evidence strongly points to a surprise attack on an unprepared community and the large-scale killing of civilians at the hands of the Spanish, Tlaxcalans and Cempoalans. On the part of the Spanish, we are likely to encounter what Cortés and his men knew as a tested and proven tactic: diplomatic subterfuge combined with excessive violence and the intimidation or extermination of indigenous leadership. Tlaxcalan and Cempoalan motivation cannot be reconstructed with a satisfying degree of certainty, though it is highly probable that vengeance, the taking of prisoners

39 Book X I I of the Florentine Codex offers a detailed narrative of the conquest, including the massacres of Cholula and Toxcatl. The quotation is from León-Portilla, Broken Spears, pp. 39–41. This is not the place to discuss the complex gestation of Sahagún’s bilingual (Spanish and Nahua) compendium of Mexica culture and history, and its many layers of meaning and interpretation. A starting point is C. E. Dibble, ‘Sahagún’s historia’ in Bernadino de Sahagún, Florentine Codex: General History of the Things of New Spain, trans. A. J. O. Anderson and C. E. Dibble, 13 parts (Salt Lake City: University of Utah Press and Santa Fe: School of American Research, 1950–82), part I, pp. 9–23.

640

Genocidal Massacres in the Spanish Conquest of the Americas

and the weakening of Mexica hegemony and destruction of a powerful Mexica ally were driving factors.

Toxcatl Turning to the Toxcatl massacre (aka the massacre of Templo Mayor) (Figure 24.2), historians are again forced to acknowledge that we will probably never be able to establish the facts of the matter with anything approaching certainty. What is clear is that in May 1520 the Spanish garrison in Tenochtitlán, under the command of Pedro de Alvarado, killed a great number of Mexica nobles and warriors – leading members of the Mexica elite, in fact – who had gathered at the Templo Mayor to honour their god Huitzilopochtli at the Feast of Toxcatl. Cortés and a sizeable number of his company were absent at the time, confronting and defeating a Spanish force under the command of Pánfilo Narváez sent by Diego Velázquez to apprehend and return Cortés to Cuba for trial. Again, the sources reflect the contentious politics of the conquest and the ways in which its history was written and rewritten over time.40 Indigenous

Figure 24.2 La Matanza de Templo Mayor – Toxcatl, Codex Dúran.

40 For a source-based narrative of the massacre see Thomas, Conquest, pp. 382–7, 389–93; for a brief discussion of the relationship between eyewitness testimonies and other accounts see P. García Loaeza, ‘Telling violence: the Toxcatl massacre at the Templo Mayor in sixteenth-century sources’, Journal of Iberian and Latin American Studies 22 (2016), 109–23.

641

harald e. braun

accounts relay a sense of shock and outrage. The Florentine Codex is both representative and the most exhaustive in its treatment of the massacre. After describing joyful preparations and the beginning of the most important religious feast in the Mexica calendar, the codex continues: At this moment, when the dance was loveliest and when song was linked to song, the Spaniards were seized with an urge to kill the celebrants. They all ran forward, armed as if for battle. They closed the entrances and passageways, all the gates of the patio . . . They posted guards so that no one could escape, and then rushed into the Sacred Patio to slaughter the celebrants . . . They ran in among the dancers, forcing their way to the place where the drums were played. They attacked the man who was drumming and cut off his arms. Then they cut off his head, and it rolled across the floor. They attacked all the celebrants, spearing them, striking them with their swords. They attacked some of them from behind . . . Others they beheaded . . . They slashed others in the abdomen and their entrails all spilled to the ground. Some attempted to run away, but their intestines dragged as they ran; they seemed to tangle their feet in their own entrails. No matter how they tried to save themselves, they could find no escape.41

Cortés, on the other hand, omitted any reference to the event in his second letter, presumably because he wanted to keep the focus on him as a heroic leader taking swift and decisive action and uphold his carefully curated claim that he was successful in avoiding the ‘scandal’ of excessive violence even when confronted with treacherous behaviour and unpalatable violence on the part of the indigenous population. Soon, though, Cortés would be compelled to offer his version of events. The abortive información that Velázquez initiated in 1521 in order to put Cortés on trial pressed witnesses on a great number of issues, question 47 asking ‘whether they knew that Pedro de Alvarado and his men had slain more than 600 Mexica nobles and 5– 6,000 commoners who peacefully celebrated their idol, thus causing the uprising that put the conquest into jeopardy’.42 One exemplary witness,

41 León-Portilla, Broken Spears, pp. 74–6. The shock at the violence and treachery is palpable in other indigenous accounts, such as the Codex Aubin, which stresses that permission to celebrate was given by both Cortés and Alvarado; ibid., pp. 80–2, 80, 81. 42 An información represented the first step in a legal process of indictment before a court of laws, namely the gathering of witness statements justifying the need for a charge. L. Martínez, ed., Documentos cortesianos I: 1518–1528. Secciones I a III (Mexico City: Universidad Nacional Autónoma de México / Fondo de Cultura Económica, 1990), pp. 280–349, question 47, at pp. 308–9.

642

Genocidal Massacres in the Spanish Conquest of the Americas

Juan Alvárez, reported a Mexica plot foiled by pre-emptive action on the part of Alvarado and his men.43 He claimed that the Spanish had noticed preparations for war, seized and interrogated Mexica under torture, and had thus been able to confirm their suspicion. As in the case of Cholula, the defence was that the Spanish had to kill unarmed warriors and non-combatants to prevent an attack. With conflicts and lawsuits among the conquistadors snowballing, it is no surprise that Pedro de Alvarado, too, became subject to a residencia a few years later, in 1529.44 Toxcatl loomed large among the charges brought against him, not least because of the loss of pillaged gold and the fact that it put the whole conquest in grave peril. Bernadino Vázquez de Tapia makes another appearance as a key witness.45 He states that interrogation and torture of arbitrarily apprehended indios did not yield any evidence of a plot, both because testimony from torture was notoriously unreliable and because the indigenous translator did not convey what the prisoners were saying but merely confirmed what Alvarado claimed they were saying.46 Vázquez also states that the victims – he speaks of 200–300 Mexica lords and 2,000–3,000 spectators, though witnesses gave different numbers in the course of the inquiry – peacefully gathered in the temple and simply ignored the armed Spanish taking position at the exits. The brutality of the Spanish assault was execrable. The greed and folly of Alvarado and his men, so Vázquez de Tapia says, resulted in the loss of treasure, the death of many Spaniards and indigenous allies, and the near collapse of the Spanish enterprise.47 Alvarado and his men in turn denied that the Spanish struck out and struck first at unarmed, unsuspecting people.48 The interrogation of commoners, so Alvarado says, confirmed suspicions that Mexica nobles were planning to replace the statue of the Virgin Mary in the Templo Mayor with that of Huitzilopochtli, their god of war, and then go on to destroy the Spanish.49 Allegedly, the intention was to impale captured Spaniards as a punishment. Crucially, Alvarado and his witnesses contend that they went to the temple in order to prevent sacrilege even though their quarters were already 43 Ibid., pp. 346–9. 44 Procesos de residencia instruidos contra Pedro de Alvarado y Nuño de Guzmán, ed. J. F. Ramirez (Mexico City: Valdés y Redondas, 1847). Toxcatl is raised in question 7, p. 31. 45 His testimony ibid., pp. 34–42. Testimony during Cortés’ residencia: Martínez, Documentos cortesianos I, pp. 47–8. 46 Procesos . . . contra Pedro de Alvarado, ibid., p. 37. 47 Ibid., p. 38. 48 Ibid., pp. 65–8. 49 Ibid., pp. 66–7.

643

harald e. braun

besieged.50 In a nutshell: the Mexica attacked first; the Spanish did not act preemptively. Once the Spanish arrived at the main temple courtyard, they were confronted by increasing numbers of Mexica warriors. Forced to retreat and defend their quarters, they held the city for forty days until the return of Cortés and the main force. The party of Alvarado offers a familiar tale of conquistador heroism laced with points that were politically and legally relevant: they stress that an indigenous uprising had been only a matter of time; they are at pains to assert that nothing was done to provoke violence, that everything had been done to corroborate their suspicion of imminent attack, and that the Mexica had begun their assault by the time the Spanish decided to retrieve the statue of the Virgin Mary from the Templo Mayor. Later accounts rearranged the core elements of these confused narratives, selecting, emphasising, sometimes adding details.51 Indigenous sources are unequivocal in their condemnation. Most Spanish sources are either ambivalent or follow Vázquez de Tapia in being downright dismissive of the explanation given by Alvarado and some of his men. Francisco López de Gómara, for instance, indirectly corroborates Vázquez de Tapia’s version of events. After describing the dance, he states that some say that he [Alvarado] was warned that those Indians, as city leaders, had gathered there to coordinate their subsequent mutiny and rebellion; [whereas] others say he went there in the first instance to see their much praised and famous dancing, but seeing them so richly adorned, that they bent under the gold on their heads and shoulders . . . without sorrow nor Christian piety knifed and killed them, and took what was on their heads.52

Gómara makes clear that Cortés, though horrified by events and anything but persuaded by Alvarado’s jumbled explanations, could not pursue the issue any further given the desperate situation Alvarado had brought about. Bernal Díaz, not present at the massacre, bluntly rejects Alvarado’s claims.53 He does not describe the assault as such, but recounts that four of Moctezuma’s envoys intercepted the Spanish on their way back to Tenóchtitlán, and bitterly and credibly complained that Alvarado had killed many of their ‘principales y caciques’ even though they had received permission from both Cortés and Alvarado himself to celebrate the feast of Toxcatl. 50 This is the slant of the question Alvarado put to his chosen witnesses (question 19), ibid., pp. 93–4. 51 See García Loaeza, ‘Telling violence’, 113–19. 52 Gómara, Life, pp. 197–8. 53 Díaz, Historia, ‘Spanish massacre of the dancers’, pp. 155–62.

644

Genocidal Massacres in the Spanish Conquest of the Americas

He also goes into detail about Cortés’ interrogation of Alvarado. According to Díaz, Alvarado gave confused and conflicting accounts of an alleged Mexica conspiracy which left Cortés and everyone else in little doubt that Moctezuma’s envoys had told the truth. Alvarado escaped punishment only, Díaz suggests, echoing Gómara, because the Spanish needed to maintain unity in the face of Mexica retaliation. Las Casas – again prefiguring Leo Kuper – decries Toxcatl as yet another example of a massacre conceived as a means to instil fear and forestall indigenous resistance. Echoing the sense of loss felt by the Mexica, he briefly states: ‘The garrison decided to stage a show of strength and thereby boost the fear they inspired in the people of this kingdom, a classic Spanish tactic in these campaigns, as we have had occasion to remark before . . . the sad story of a massacre which wiped out their entire nobility, beloved and respected by them for generations and generations.’54 Though relations between the Spanish and the Mexica were deteriorating rapidly and a rebellion would probably have broken out at some point, the evidence, on balance, does not stack up to proof of an indigenous plot or in fact a Mexica surprise attack. It is possible that Alvarado and his men, weakened in numbers and holed up in the midst of an increasingly hostile population, succumbed to fear and paranoia. It is equally possible, though, that the opportunity to strike a decisive blow against increasingly uncooperative Mexica leaders and to enrich themselves were weightier factors and reason enough for killing ‘without sorrow or Christian piety’.

Conclusion The communities involved in the massacres of Xaragua, Cholula and Toxcatl represented vastly different cultures, including distinct cultures of violence, but shared a sense of the fragility of political alliances and body politics nonetheless. The Spanish, through a mixture of luck, bold ignorance and intuitive albeit limited understanding of indigenous political structures, were able to exploit the tensions and rivalries that characterised the relationships among Caribbean and Mesoamerican peoples. They were dependent on indigenous collaboration more than they were prepared to concede and possibly realised, and narrowly survived their own divisions. The use of genocidal massacre against indigenous communities was a factor in their success. In the case of Xaragua and Cholula, indigenous 54 Las Casas, Destruction, pp. 50–1.

645

harald e. braun

actors co-operated with the European strategy and practice of excessive violence. In the case of Cholula, in fact, it is more than likely that indigenous actors (the altepetls of Tlaxcala and Cempoala, Malinche) directed strategy and outcomes. Though it is unlikely that we will ever be able to establish whether the Toxcatl massacre was conceived as least in part as an object lesson, it does follow the pattern and illustrates the impact of genocidal massacre and excessive violence on indigenous communities. Most Spanish and indigenous perpetrators either denied that a massacre took place in the first instance or defended their actions as the necessary and justified response to indigenous transgression of European norms they assumed to be universally binding. The indigenous victim of genocidal massacre was denounced as a rebel, a traitor and a cruel cannibal intent on slaughtering the Spanish under the pretext of friendship and hospitality. The excessive violence of massacre was obscured or altogether omitted from testimonies. We have indigenous accounts, however, which challenge the perpetrator narratives. We also have contemporary Spanish testimonies and Spanish accounts which demonstrate that many Spanish felt that the excessive violence of genocidal massacre violated the very norms the conquistadors claimed to represent and uphold.

Bibliographic Note The text and notes of the chapter reference contemporary sources concerning the massacres of Xaragua, Cholula and Toxcatl available in partial or full English translation. These include the collections of extracts translated from Spanish and Nahua into English edited by P. de Fuentes, M. León-Portilla and S. B. Schwartz. León-Portilla and Schwartz include helpful introductions and illustrations. Another very useful collection is J. Lockhart’s We People Here: Nahuatl Accounts of the Conquest of Mexico (Eugene, OR: Wipf & Stock, 2004). Lockhart presents bilingual translations (Nahua–English and Spanish–English) from a number of contemporary indigenous sources – some of them directly relevant to this chapter (Florentine Codex, Annals of Tlatelolco and Codex Aubin) – as well as contextualisation, commentary and ample illustrations. Readers interested in full translations of Book X I I of the Florentine Codex are served by A. J. O. Anderson and C. E. Dibble’s The War of Conquest: How It Was Waged Here in Mexico (Salt Lake City: University of Utah Press, 1978) and Anderson and Dibble’s Florentine Codex: General History of the Things of New Spain. The volume by H. F. Cline (trans.) and S. L. Cline (ed.), Conquest of New Spain, 1585 Revision (Salt Lake City: University of Utah Press, 1989), offers 646

Genocidal Massacres in the Spanish Conquest of the Americas

a translation of a late redaction of Book X I I that indicates the shift in Franciscan missionary ideology from the mid- to the late sixteenth century. The idea that Moctezuma mistook Cortés for the returning god Quetzalcoatl propagated in Book X I I of the Florentine Codex and by Francisco López de Gómara can be regarded as a construct illustrating the complexity of politics of memory in colonial Mexico. See, for instance, J. H. Elliott, ‘The mental world of Hernán Cortés’ in Spain and Its World, 1500–1700 (New Haven and London: Yale University Press, 1992), pp. 27–41; and C. Townsend, ‘Burying the white gods: new perspectives on the conquest of Mexico’, American Historical Review 108:3 (2003), 659–87. L. B. Simpson’s translation of Gómara’s Historia de la conquista de Mexico is only available in the hardback originally published in 1964, whereas Pagden’s English translation of Cortés’ Cartas de relación is now available as an ebook. So are D. Carrasco’s English edition of Bernal Díaz de Castillo’s Historia verdadera de la conquista de Nueva España referenced above and Bernal Díaz, The Conquest of New Spain, translated and with an introduction by J. M. Cohen (Baltimore: Penguin Books, 1963). A fresh, more rugged translation is The True History of the Conquest of New Spain, translated and with an introduction and notes by J. Burke and T. Humphrey (Indianapolis: Hackett, 2012). The most reliable translation of Bartolomé de Las Casas’ Brevísima relación de la destrucción de las Indias remains the one by N. Griffin with an introduction by A. Pagden.

647

Index

Aardakh genocide, 96 Abbasids, 512, 515, 554 abnormal graves, 59, 62–6, 69–77, 80–1, 82–4 disordered, 73 ordered, 72, 77–9 see also burials, graves aboriginal religious traditions, 95, 98 absolute power, 51, 381, 463, 572, 592 Abu’l-Faraj Ibn al-‘Ibrı¯, 455 abuse, 65, 106, 108, 222, 344, 549, 575, 601 Abydeans, 263 Abydus, 263–4 Acari valley, 584–5 acculturation, 604, 616–18 Acentejo, 610, 612 Achaea, 299, 300 Achaean League, 267, 288, 295, 303 Achaeans, 261, 267–8, 270, 273, 288, 303 Achan, 188 Acragas, 249 administration, 335, 338, 340, 394, 434–5, 562–3, 568, 590 Roman, 330, 333–5, 338, 340–2, 345, 348, 363 administrators, 370, 533, 555–8 British, 547 admixture biocultural, 81 genetic, 81, 84 adolescents, 68, 77, 81, 84, 228, 470, 585, 589 Aduatuci, 323–5, 326 adult females, see adult women adult males, 61, 68, 73, 75, 77–8, 112–13, 189–90, 205, 236, 247, 249, 383, 389–90, 464, 578, 581, 587, 589, see also males adult women, 61, 68, 71, 73–4, 77, 336 see also females, women adultery, 562, 568 Aedui, 320, 325 Aegina, 242–3, 251 Aeginetans, 242–3

Aelia Capitolina, 350 Æthelwine, Bishop, 413 Aetolians, 261, 270–2, 273 Afghanistan, 11–12, 508, 511 Afghans, 11, 548 Africa, 69, 97–8, 261, 363–4, 371, 595, 613–14 See also Carthage, German Southwest Africa, Rwanda, South Africa, Sudan North, 157, 248, 278, 340, 595, 596, 614 agency, 61, 76, 104, 108, 152, 348, 496, 634, 638–40 human, 138, 148, 168–9, 180, 467 agricultural crises, 48 see also famine, harvests, starvation agricultural lands, 199, 527 see also farms, harvests agriculture, 23, 141, 147, 162, 166, 169, 175, 297–8, 336–7, 613–4 see also farms, harvests plantation, 595, 618 Agrippa, King, 337–8 Aguilar, Francisco de, 633–5 Akkad, 212–19, 220–1, 224, 231 kings, 209, 212 Akkadian Empire, 216, 217 Alamu¯t, 511, 514–15 Ala-ud-Din, Sultan, 550, 558–64, 569 Alaul Mulk, 558–9 Albert of Aachen, 46, 429, 430–1, 451–2, 454, 458–9, 462, 465–6 Albi, 439, 482, 485, 488, 491 Albigenses, 472, 474, 479, 485 see also Cathars Albigensian Crusade, 3, 37, 95, 470–96 Alesia, 325 Alexander the Great, 169, 201, 235, 253–5, 261, 271, 559 Alexandria, 169, 337–8, 340, 371 ‘Ali ibn Tahir al-Sulami, 467–8 ‘All under Heaven’, 393, 397–8

648

Index allies, 61, 72, 98, 242, 243, 253, 267, 285, 294, 301, 303, 501–2, 627–9, 631–5, 643 Allies, the (World War 2), 133, 135 Mexica, 632, 641 Alsace, 17, 74, 437 altepetls, 631–3, 646 Alvarado, Pedro de, 622, 635, 637, 641–5 Amalek, 22–4 Amalekites, 22, 188 Amalric, Arnau, 470–1, 474, 488–9, 491–2 Amato, 573, 584–5 Amazon, 579 Ambaghai Khan, 500, 501 ambassadors, 269, 316, 512, 629, 636 Roman, 288, 323–4 Ambiorix, 317–18 Americas, 599, 613, 616–17, 619, 622, 625, 634, 636–7 see also Andean South America Amerindians, 618, 624–7, 634–6 see also individual entries for Amerindian peoples Amielh, Bernart, 483 Ammonites, 189, 195, 197–8 Ammonius, Aurelius, 364 Amorite, 187, 197, 229 Amounderness, 412, 420 Ampato, 573, 591 Amu Darya river, 508, 513 Anacaona, 627 anachronisms, 31, 42, 45, 51, 239, 280, 495, 498 anarchy, 170, 443, 571 Anatolia, 174–5, 353 ancestry, 59, 65, 77, 81, 83, 84 mixed, 78, 82 steppe, 78, 81, 84 Andahuaylas, 573, 578–9 Andean Empires, 578–9, 588, 590–1 Andean South America, 572–92 Andes, 572–3, 578–9, 583–4, 588, 591–2 anger, 13, 46–7, 257, 268, 272, 274, 317, 320 Anglo-Saxon Chronicle, 408–9, 422 animal bones, 72, 75–6 animal imagery, 189, 535 animals, 239, 244, 581, 583 wild, 1, 222 annihilation, 60, 62, 67, 252, 255, 572, 585, 588 ante-mortem trauma, 576, 577 anthropocidal texts, 192, 203 anthropocidal violence, 186, 192–3, 206 anthropocide, 192, 224, 233 anti-Jewish ideologies, 54, 428 anti-Jewish urban riots in Castile and Aragon, 426, 443–5

anti-Jewish violence, 426, 428, 430, 433, 434, 436, 438, 444–5 Antilles, 595, 605, 610–12 anxiety, 32, 46, 286–7, 307, 565 existential, 293–5, 307–8 apocalyptic texts, 186, 193, 206 Appian, 44, 282, 284–5, 293, 300, 304, 307, 347 Arab sources, 458, 460 arable land, 259, 272, 419 see also agricultural land, farms Arabs, 27–8, 158–9, 163, 377 Aragon, 426, 438–9, 440, 443, 445, 488 Arameans, 161, 167, 195, 198, 229 Arcadians, 266, 268–70, see also Cynathea archaeological evidence/sources, 55, 60, 83, 84, 317–18, 590–1, 595, 600 archaeologists, 62, 198, 252, 574–5, 582, 584–5, 586–7 Arevaci, 289–91, 300, 303–4 see also Celtiberians, Numantia Argentina, 10, 574, 579, 591 Argos, 244–5, 249, 288 Arisba, 237–9 Armenians, 3, 13, 26–7, 95, 97, 129, 450, 451 Armleder, 425, 435–6, 437–8 rebellion, 437–8 arrests, 21, 367, 440 arrows, 74, 76, 78, 457, 503, 557, 576 Ashurbanipal, King, 155–7, 163, 165 Asparn-Schletz, 70, 74 assaults, 7, 66, 427, 428–9, 433–4, 576, 578, 644 Assyria, 41, 95, 155, 157, 162, 166, 178–9, 198–9 culture, 166, 231, see cultures inscriptions, 226–9, 230–2 kings, 161, 175, 179, 226, 228 Assyrians, 158, 163–6, 174, 201, 209, 226, 228, 231–2 Athenians, 235, 241–2, 244–8, 252, 254 Athens, 99, 235, 240–4, 245, 246–8, 249, 253 atrocities, 17, 132–3, 146, 155, 264, 270–1, 406, 411 mass, 127, 131–2, 143, 145–6, 147 Vietnamese, 537–9 Auschwitz, 24–6, 127 Australia, 14, 24 Austria, 26, 71, 437, 441–2 Lower, 74, 75, 77, 437 Upper, 437 authoritarianism, 29, 195, 199, 206 authority, 36–7, 283, 432–3, 437–8, 505, 510, 555, 570 ecclesiastical, 425, 625–6, 628 Roman, 287, 289–91, 294, 301–3, 368 Avaricum, 314, 318–19, 320–1, 324

649

Index Averni, 319, 320, 325 Aztecs, 630, 639

bones, 72, 76–7, 209, 575–7, 581–4, 585, 587–8, 592 animal, 72, 75–6 broken, 577, 581 booty, 228–9, 271, 272, 317, 322, 324, 557, 559 Borjigid Mongols, 498, 501, 504 Bosnia, 7, 145 Bosnia-Herzegovina, 144 Bosnian Serbs, 144–5 bows, 397, 483–4 compound, 388 boys, 246, 470, 477, 483–4, 498, 501, 583, 586 see also ribauds Bronze Age, 59, 63, 67, 80, 81, 83 Early, 71, 75, 217 Late, 71–2, 75–6, 81, 168, 195 Middle, 75, 78, 84 brutality, 185, 199, 222, 224, 236, 245, 343, 377 Buddhists, 95–6, 561 burial mounds, 64, 216 burials, 61, 63–4, 72–3, 78–80, 82, 572, 583, 588 see also funerary rituals, grave goods, graves Burma, 24, 400 burning, 24, 47, 222, 228, 231, 434, 440 Burundian genocide of Hutus and Tutsis, 96 Byzantines, 450, 453

Ba’athist genocide, 96 Babylon, 155, 160, 188, 194, 196, 199, 201, 227 Babylonian exile, 201–3 Baghdad, 453, 515, 553–4 Balban, Sultan, 549, 551, 554–6, 557, 569 Baldric of Dol, 458, 460–1, 468 Bambuti genocide, 96 Ban-la Tra-toan, 532, 540 baptism, 434, 437, 439, 445, 617 Bar Kochba, 339, 347–9 Barani, Ziya-ud-Din, 550–2, 554–5, 557–63, 566–71 barbarians, 44, 219, 241, 246, 269–70, 297, 380–1, 398–9 Barcelona, 440, 445, 488 Batavians, 342 Bede, Venerable, 4 beggar boys, see ribauds beheadings, 524, 536–7, 557, 566 Belarusian genocide, 96 belief, 13, 16, 464, 474, 479, 487, 618, 625 religious, 7, 334, 548 systems, 159, 479 Bell Beaker culture, 81, 84 Benedict of York, 433–4 Benedict XIII, Pope, 602 Bengal, 133, 550, 555–6, 566 Bengali/Bangladeshi genocide in East Pakistan, 96 Bergheim, 70, 74 Bethencourt, Jean de, 601–2, 609 Beverley, 413 Béziers, 470–4, 477, 486, 488–91, 495 Bible, Hebrew, 22, 186, 189, 192–4, 203, 372, 461, 463 biblical texts, 23, 99, 185, 189–92, 194, 196–201, 205, 229 biological reproduction, 60, 305, 322, 324 bishops, 94, 429–30, 432, 433, 437, 461, 481, 607 Bituriges, 318–19, 320–1 Black Death, 425, 428, 438, 440–2, 444 Black War genocide, 95 blockades, 133–5, 143 Blucˇina Cerzavy, 70, 75 blunt force trauma, 73, 74, 79, 575, 576, 587–9, 591 see trauma, blunt force Bo Qi, 390 Boeotia, 240, 242–3, 245–6, 250–1 bone beds, 60–1, 72

cadavers, 455–8 Caedwalla, King, 4 Caesar, Julius, 23, 42, 45, 309–29, 358 actions, 309–11, 325, 326, 329 mass killings, 314–21 Caesarea, 334, 353, 356, 367 Caesarius of Heisterbach, 471–2, 486 California genocide, 95 Calixtus II, Pope, 429 Cambodia, 1–2, 5, 11, 18, 21, 24, 96, 140–1 Canaanites, 101, 187–9, 192, 204–5, 229 Canada, 27 Canary Islands, 41, 594–619, 623–4 cannibalism, 64, 75, 291, 409–10, 421, 460, 575 Canopus Decree, 171, 173 capital punishment, 218, 361, 567–8 captives, 68, 80–1, 214, 227–8, 537–8, 580–1 see also enslavement, mass enslavement, prisoners, slavery, slaves execution, 45, 592 Carcassonne, 439, 488, 491, 493 Caribbean, 594, 608, 611–12, 614–16, 618–19, 623–4, 626, 629–30 Caribs, 605, 611, 616 Carnutes, 318, 320–1, 325

650

Index Carthage, 19, 40–1, 248, 278, 281–9, 294–8, 301–2, 304–8 destruction, 19, 40, 248, 278, 282–7, 297–300, 301 Carthaginians, 41, 235, 262, 283–7, 297, 301–6, 308, 597 caste, 171 Castile, 426, 443, 602–4, 606, 626–7 castles, 403–5, 433–4, 439, 441, 445 royal, 433–4, 439, 445 Catalonia, 440, 445 Catharism, 95, 476–81, 487, 489 Cathars, 95, 474–5, 477–80, 482 Catholicism, Roman, 95–8 Cato, 19, 127, 284–7, 296–7, 303, 316, 329 cattle, 23, 188, 364–5, 396, 408 causes of genocide, 11–15, 48–54, see also motivations Caux, Bernart de, 483 cavalrymen, 556, 559, 562, 602 Celtiberians, 289–91, 296, 303–4, 306 see also Arevaci, Numantia Cempoala, 631–3, 640, 646 Cenabum, 318–21, 324–5, 326 Cerro Sechín, 573, 578 Chachapoyas, 579, 587–8 Chahan, 508–9, 521 Chalcidians, 251–2 Chalk, Frank, 12, 36, 474–5, 487, 490, 518 Champa, 2, 523–43 Chams, 54, 523–30, 531–43 captive, 537–9 leaders, 527, 541 rulers, 528, 531–2, 535, 536–7 territory, 523, 530, 533, 536, 541–3 Chan Chan, 573, 582 Chankas, 579, 588–9 chariots, 228, 388 Charny, Israel, 27–8, 88 Cheshire, 412, 416, 419 children, 20–1, 74–80, 263–5, 313, 315–17, 560–1, 582–3, 632–5 forcible transfer, 6, 8, 210, 332, 373, 379, 465, 579 young, 137, 412, 503, 587 Chimú, 579, 582–3, 586 China, 140–1, 143, 376–99 court, 532, 534, 540–1 First Emperor, 391–4 historical sources, 384–6 northern, 499, 506 southern, 395 Spring and Autumn Period, 388–9

Warring States Period, 378, 384–6, 389, 397, see also individual dynasties Chinchorro, 573, 578 Chinggis/Chengiz Khan, 498–9, 503–9, 510–11, 517, 520–1 Cholula, 622–3, 629, 630–40, 643, 645–6 Christendom, 445, 464, 466, 486, 489, 494, 512 Latin, 477, 484, 486, 490 Christian communities, 357, 360, 367, 370, 373, 462 Christian sources, 46, 371, 452, 455 Christianity, 97, 99, 353–5, 358–63, 369–71, 426–7, 431, 479 Christians, 95–6, 353, 360–71, 372–4, 425–7, 435–6, 439–41, 443–5 see also persecutions Latin, 450, 467, 472, 481, 487 ordinary, 446, 472, 488 Orthodox, 95, 450 Christmas, 403 chroniclers, 127, 436, 448–9, 451, 453, 462, 552 Chu, 389, 390 churches, 356, 363–7, 412–13, 480–1, 494, 496, 626, 628 Cicero, 282, 289, 294, 296, 302, 312, 325 Circassian genocide, 95 city-states, 38–40, 68, 128, 213, 220, 232, 278, 303 ciuitates, 320–1 civic communities, 278–81, 304, 305–6 civic identity, 257, 262, 271, 273 civil wars, 142–3, 175, 179, 443–4 civilian populations, 8, 13, 16, 136–7, 143, 145, 332, 378 civilians, 130–1, 133, 137, 145–6, 390, 466, 540, 622–3 Clement VI, Pope, 442 Cleomenes, King, 267, 273 clergy, 89, 364, 367, 373, 442 clerics, 100, 356, 428, 434, 438–9, 443, 489 Cliffs End Farm, 70–1 climate, 48, 150–80, 269 change, 14, 62, 147, 151–3, 180 role in genocide, 150–82, see also weather climate–conflict pathways, 151, 167, 175, 178, 180 clothes, 364–5, 397, 415, 493, 556 Cnossus, 264–5 coins, 334, 342, 344, 554, 555, 559, 565 collective punishment, 318, 321, 323, 325, 328 collectivities, 9, 37–9, 258–60, 266, 333, 486, 490, 495 Cologne, 429, 441–2 colonial societies, 600, 613, 618–19 colonists, 73, 248, 604

651

Index commanders, 339, 341, 509, 513, 557, 559, 633, 637 see also generals, kings, leaders and individual entries Mongol, 512, 514 Communist Party of China, 18, 376 Communist Party of Kampuchea, 18 communities, 39, 73–4, 91–2, 97–9, 305–7, 330–3, 437, 631–3 civic, 278–81, 304, 305–6 Greek, 338, 340 Jewish, 336, 338, 347, 427–30, 432, 435, 438, 442–4 local, 161, 435 Nahua-speaking, 630, 638 of faith, 93 political, 215, 219 regional, 10, 212, 231, 311, 518 religious, 92–3, 374 unprepared, 624, 632, 640 competition, 48, 50, 195–8, 206, 383, 389 international, 389, 398 concentration camps, 17, 24–5, 140 Confucian school, 387–8 Confucius, 387 conquerors, 390, 406, 549, 551–2, 603–4, 606–8, 610, 616 conquistadors, 625–6, 630, 634–5, 637 conscription, 382, 389 consensus, 150, 178, 194, 215, 406, 436, 548, 619 conspicuous destruction, 294, 306, 307, 320–1, 328, 333 conspiracy, 6, 440, 560–1, 645 Constantine I, Emperor, 355–6, 366, 369 Constantine V, Emperor, 463 consuls, 285, 288, 289–91, 299, 304, 346 contemporary sources, 405, 549, 600, 608, 610, 612 contempt, 63, 74, 80, 83, 98, 513 conversion, 428–9, 549, 599, 601, 617, 636 forced, 429, 431–3, 437, 445 mass, 438 to Islam, 551, 571 Copres, 368 copycat violence, 433, 444 Corded Ware culture, 77–8, 81, 84, see cultures Corinth, 278–9, 287–9, 293, 295–6, 297–8, 299, 302, 305–6 Corinthian War, 249–50 corpses, 128, 221–2, 226–8, 455–7, 463, 466, 584, 586 Cortés, Hernán, 622, 625, 628, 629–30, 631–44 cortezia, 483–4 corvée, 179, 380, 386, 393–4, 397

countryside, 232, 261, 270, 300, 413–14, 415, 423, 434 Crete, 264 crises, 144, 155, 438, 530, 565 agricultural, 48 economic, 143, 146 food, 142–3 humanitarian, 131, 138, 145 political, 449, 571, 632 revelatory, 153, 160, 173 crops, 143, 175, 415, 416–17, 572, 604 Crow Creek, 60–1, 70 cruelty, 27–8, 286, 296, 303, 391, 406, 411, 460 holy, 464 crusaders, 46, 429–30, 448–68, 475–8, 488–94 slaughter of, 464, 466 violence, 431, 448 crusades, 438, 448–51, 458–60, 461–4, 466–7, 470–7, 485–6, 487–92 see also Albigensian crusade, First Crusade, Second Crusade, Third Crusade, Fourth Crusade, holy wars exceptionality of violence, 463–4 extermination policies, 450 mythistory, 467–8 unreliability of sources, 451–62 cult, 53, 218, 338, 435–6, 604, 617 cultural diversity, 62 cultural genocide, 5, 7–8, 27, 33, 41, 164, 170, 355, 373–4 cultural institutions, 8, 321, 373 cultural practices, 27, 304, 307, 374, 542, 551 cultural reproduction, 81, 223 cultural superiority, 538, 543 cultural traditions, 60, 528 cultures, 40–1, 53, 61–2, 232, 236, 595, 599, 612 Assyrian, 166, 231 Bell Beaker, 81, 84 Corded Ware, 81, 84 Hellenistic, 202, 273 Linear Pottery, 73 Mesopotamian, 232–3 Urnfield, 75, 77 Cusco, 573 Cynaetha, 257, 268–70, 272–4 see also Arcadians Cyprus, 346–7 Danes, 403–5, 408, 413, 417 Darfur, 96, 131, 132 Darius I, King, 255 Darius III, King, 254 Darré, Richard Walther, 134–5 debt slavery, 196–7

652

Index diversity, 50, 548–9, 583, 603 cultural, 62 ethnic, 380 religious, 450, 548 divine vengeance, 471, 474 doctrines, 89, 91, 425, 445, 479, 481, 617 Dodona, 257, 270–1, 274 dogs, 1, 222, 397, 408, 470, 623, 629 wild, 380 Domesday Book, 405, 417–18, 420 Domitian, 344, 356 Dönitz, Karl, 26 Drerus, 257, 264–5, 270, 272 drought, 139, 143, 151–3, 155, 157, 160, 171, 175 prolonged, 579, 588 Duan Jiong, 396 Durham, 403, 405, 410–11, 413, 417, 420–3

Deccan, 550, 558, 566 Deeds of Abbot Æthelwig, 411 defiance, 520, 543, 566, 570 definitions, 2–11 generic, 486, 490 of warfare, 65, 66 definitions of genocide, 6, 7, 9, 11, 32–4, 38, 90–1, 204 Convention, 6–11, 33–6, 40–3, 53, 118–19, 136, 141, 205, 210–11, 222, 230–1, 259, 280–1, 304–6, 322, 326, 332–3, 373, 378, 465, 474, 486–7, 517–18, 605, 618–9, see United Nations Genocide Convention scholarly, 33, 37, 311 sociological, 8–10, 258, 274, 422 dehumanisation, 44, 60, 62, 83, 219, 225, 268, 273, 296 deities, 50–1, 53, 89, 192, 193–4, 542, 632 Delhi, 549, 553–6, 557–60, 563–4 power elite, 551, 565 Sultanate, 549, 553, 561, 564, 570 sultans, 550, 552, 556, 558, 569, 570 deliberate intent, 166, 521 democracy, 247, 249, 253, 381–2 depopulation, 419–21, 605 deportation, 40, 43, 595, 602, 604–5, 611–12, 618 deportees, 166–7 depravity, heretical, 470, 476, 478, 486, 492, 494–6 Derbyshire, 412 deserts, 22, 129, 130, 158, 202, 222, 578 ‘Diary of a Madman’, 376 diasporas, 336, 344, 347, 349–50 Carthaginian, 305 revolts, 345–7, 348 Díaz de Castilla, Bernal, 634, 635–6, 644 dictators, 11, 156, 487 Diocletian, 356, 358–63, 366, 368 Diodorus Siculus, 248–50, 252, 260, 282, 289, 293 direct evidence, 575, 578, 581 direct rule, 383, 394, 398 direct violence, 130, 132, 137, 328 disasters, 132, 143, 171, 246, 409 natural, 143, 152–3, 420, 583 disease, 127–8, 131, 155–6, 169, 439–40, 601, 610, 618 see also illness, plague epidemic, 74, 82, 612 virgin soil, 605, 608, 611, 619 disobedience, 230, 294, 301, 566, 568, 610 dispersal, 41, 244, 280, 304–5, 306 displaced camps, 131 Dium, 257, 270–1, 274

Early Bronze Age, 71, 75, 217 Early Iron Age, 73, 81, 84, 195 earthquakes, 159–60 Eburones, 42, 314, 317–18, 320, 326, 328–9 ecclesiastical authorities, 425, 625–6, 628 ecclesiastical elites, 428, 435 ecclesiastical leaders, 427, 432 edicts, 43–4, 338, 353, 357, 363–9, 370–1, 372, 534–5 Edom, 189, 203 Edomites, 189–90 education, 88, 100, 269, 367 civic, 261 Egypt, 162, 169–72, 174, 201, 344, 345–7, 357, 368 Middle, 364 Ptolemaic, 162, 169–74, 180 Elam, 155–7, 163, 209, 212 elephants, 523, 532, 556–7, 559, 561, 566 eliminationist violence, 227–8 elites, 152, 164, 170, 196–7, 298–300, 504, 520–1, 524 creole, 616 ecclesiastical, 428, 435 local, 328–9, 443, 591 Eljigidei, 513 emotions, 255, 257, 263, 268, 274, 454, 466 emperors, 334–8, 348–9, 358–62, 370–2, 392–5, 532–3, 535–6, 630 empires, 216–17, 293–4, 327–8, 337, 341–5, 349–50, 358–60, 367–70 enemies of Christ, 426, 430 enemies of God, 47, 451 England, 414–15, 420, 421, 425, 428, 432, 433, 434–5, 512 1189–90 massacres, 432–5

653

Index England (cont.) northern, 4, 403, 405, see also William the Conqueror enslavement, 39, 298, 304–5, 317, 322–4, 328–9, 426–7, 605–6 see also captives, prisoners, slavery, slaves mass, 40–1, 257, 264, 266, 272, 298, 305–6, 321–8 envoys, 519, 523, 554, 629, 644–5 see also ambassadors Ephraim of Bonn, 432–3 epidemic disease, 74, 82, 612 Epp, Franz Ritter von, 25 equity, 260, 268–9, 271–2, 274 eruptions, 170, 173–4, 176–9, 339, 568 explosive, 169, 174 tropical, 170, 173–4, 176, 179 escalation, 12, 320, 367, 623–4 eschatological violence, 193, 233 ethics, 89–90, 92, 137, 259, 267, 270, 274–5, 296, 301–2, 308, 463, 478, 489, 619, 636 see also justification, morality Ethiopia, 143, 169 ethnic diversity, 380 ethnic groups, 2–3, 10–11, 32, 40, 194–5, 203–4, 206, 229–30 ethnic identities, 51, 95–6, 166, 202, 206, 259, 344, 616 see also collectivities, identities ethnic violence, see interethnic violence ethnicity, 53–4, 60, 61, 164, 169–71, 588, 612, 614 ethnocide, 41, 203–4, 210, 230, 232–3, 580, 600 ethnohistorical sources, 62, 67 Eugenius III, Pope, 432 Eulau, 70, 78 Eusebius of Caesarea, 348, 353–7, 360–1, 363–4, 366–9 evangelisation, 609, 617 Evesham, 411–12 evidence, 81–3, 172–4, 205–6, 209–12, 224–6, 272–3, 588–9, 633–4 direct, 575, 578, 581 Mesopotamian, 210, 225, 232 papyrological, 346–7 religious, 94–8 skeletal, 572, 578 surviving, 35, 466, 607, see also sources excessive violence, 42, 572, 616, 622–4, 626, 640, 642, 646 exclusivism, 91, 99 exemplary punishment, 246–7, 324 existential anxiety, 293, 307–8 expansion, 43–4, 59, 62, 68, 74, 392, 396 imperial, 49, 161–3, 167

territorial, 62, 69, 77, 83, 84, 180, 530, 614 expropriation, land, 135 expulsions, 21, 255, 369, 428, 438, 443, 445, 465 see also deportation extermination, 4–5, 8–10, 332, 377–9, 381–2, 386–8, 390–2, 395–6 deliberate, 32, 73 mass, 572–5, 578, 579, 582, 590, 591–2 of indigenous leaders, 626, 628 exterminationist violence, 47, 50, 52, 189, 194, 203–4 extinction, 13–14, 25, 386, 387 extreme violence, 185, 190–2, 199, 257, 259–60, 263, 268–71, 272–4 extreme weather, 152–3, 156, 159–60, 174, 179–80 see also climate eyewitnesses, 312, 409–10, 411, 459–60, 633–5 see also evidence, sources faith, 93, 99–100, 357, 361, 449, 457, 617, 625 families, 27, 61, 63, 223, 227, 412–13, 501, 559–60 famine, 129–30, 131–3, 137–44, 145–7, 155–7, 408–10, 411–12, 420–3 see also food crises, Great famine genocide, great famine in England, Great Soviet Famine/genocide, harvests, hunger, mass starvation, starvation crimes, 141–3 recurrent, 139, 143 faminogenesis, 139–44 farmers, 81, 170–2 Neolithic, 81 farms, 70–1, 137, 397 see also agriculture, agricultural lands, arable land, harvests fatality sites, 63, 69–71, 78–80 fear, 13, 198, 260, 285–6, 444–6, 555, 637–8, 645 Fein, Helen, 6, 9, 36, 258, 333, 477, 486–7, 490 females, 50, 81–2, 128, 535, 587, 589 adult, 71, 77 indigenous, 61, 81, 84 see also women fiction, 406, 449, 459–60, 467 fifteenth century, 530, 540, 542, 579, 582, 586, 604–6, 611 ‘final solution’, 326, 350, 379, 390 see also Holocaust First Crusade, 425, 428–32, 448, 451, 460, 461–2, 474, 488–9 First Emperor, 391–4 Fischer, Eugen, 25 Flavian dynasty, 344 floods/flooding, 169, 172, 175, 192–3, 203, 221, 224, 343

654

Index genocidal intent, 34–6, 127, 130, 136, 139, 143 genocidal massacres, 2, 9–10, 28, 40, 48, 61, 67, 83–4, 89, 212, 304, 311, 316, 318, 321, 425, 446, 466, 468, 490, 492, 498–9, 518–19, 521, 550, 559, 571, 584, 586, 588, 622, 624, 628–9, 645–6 genocidal moments, 61–2, 83–4, 490–2, 494–5, 573, 575, 587, 591–2 genocidal regimes, 24, 130 connections between, 24–6 genocide, see individual entries Genocide Studies, 4, 27–8, 49, 88, 131, 141, 371, 374 German Southwest Africa, 25–6, 129 see also Herero Germani, 23, 327–9 Germanic tribes, 42, 69, 315, 317 Germans, 15–17, 94, 136, 315–17, 327–8, 478 Germany, 15–16, 17, 73–4, 78, 425, 433, 435–7, 441–2 Nazi, 16, 20, 25–6, 94 SS, 16–18, 93–4 Gesta Francorum, 455–8, 459, 461–2, 464, 468 Ghaznavids, 552 Ghiyas-ud-Din, Sultan, 549, 551, 564, 566 Ghurid, 552 Glick, Leonard, 86, 88 Godfrey of Bouillon, 455, 458 Gomera, 594–5, 602–3, 606–8, 615 good faith, 285, 302, 316, 514 good men, 483–4 good women, 483–5, 495–6 Göring, Heinrich, 25 Göring, Hermann, 25 governance, 151, 152, 328, 549, 552, 558–9, 562, 566 governors, 158–9, 167–8, 300, 335, 337–8, 347–8, 555 Gran Canaria, 601–3, 606–9, 610–12 grave goods, 72, 74, 76, 585, 587 see also burials, funerary rituals, graves graves, 17, 69, 72–4, 75, 78, 81, 83, 223 abnormal, 69–73, 77, 80 mass, 27, 59, 61, 64, 83–4, 580, 582, 584 multiple, 61, 63, 78 Great Famine genocide, 95 great famine in England, 95, 138–9, 420, 422–3 Great Soviet famine/genocide, 95 Great Theban Revolt, 170, 173 see also Thebans, Thebes Greece, 34, 99, 133, 238, 244, 278, 282, 305 central, 253

Florentine Codex, 642 followers, 29, 91–3, 437, 474, 504, 549, 630 food, 134–5, 147, 222, 411–12, 415, 635 availability, 138, 147 crises, 142–3 see also famine, Great famine genocide, great famine in England, Great Soviet Famine/genocide, harvests, hunger, starvation markets, 142, 146 production, 147, 157, 611 foot soldiers, 388, 430, 470, 557, 602 foraging, 137, 416, 421 forced conversion, 429, 431–3, 437, 445 forced labour, 135 forced marriage, 22 forced migrations, 175, 196, 394 forces, natural, 220, 221, 224 forcible transfer of children, 6, 8, 210, 332, 373, 379, 465 Fourth Crusade, 459–60 Fourth Lateran Council, 492 France, 425, 428, 432–3, 435, 436–40, 475, 478, 494–5 Southern, 237, 439, 440, 473–4, 479 freed slaves, 44, 318 Fuerteventura, 595, 601–2, 604, 606, 609, 611 Fulcher of Chartres, 451, 457–8, 459, 462 funerary rituals, 575, 584 see also burials, grave goods, graves Gadara, 339–41 Galerius, Emperor, 357–8, 362–3, 366, 369 Gallus, Cestius, 338, 343 Garonne, 472, 482–3, 485–6, 488, 492 Gascony, 439, 482 Gauls, 262, 309–10, 312, 314–15, 316–21, 327–8, 329, 330 geld accounts, 418–19 gender, 21–2, 50, 155, 313, 519–20, 614 bias, 91 prejudice, 19–22 gendercide, 20, 50, 111–12, 117, 245, 247, 313, 503, 520 gendered violence, 20, 39, 103–26, 587 General Plan Ost, 96 generals, 254, 331, 346–7, 349, 381, 394 see also commanders, kings and individual entries genetic admixture, 81, 84 genetics, 59, 81–2, 84 génocidaires, 91–2, 93, 130, 478, 487, 491, 493–4, 495 see also perpetrators genocidal ideologies, 19, 54, 350

655

Index Hindus, 96, 377, 547–8, 559, 564, 567, 570 Hindustan, 554, 558 Hirimmu, 226–7 ˘ Hispaniola, 41, 128, 614, 622, 626–8, 636 historical factors, 11–12, 185, 195, 199 historical materialism, 18 historical memory, 219, 615 historical sources, 281, 377, 598 see also evidence, sources and individual entries for authors and works historiography, 86, 261, 281, 330, 355, 464, 478, 599–600 Hitler, Adolf, 15–16, 20, 23, 25, 29, 91, 376 Hittites, 187, 229 Hivites, 187, 229 Holocaust, 3, 13, 15, 18, 26, 28, 32–3, 36, 45, 88, 96, 103–4, 108, 111, 116, 120, 205, 236, 275, 281, 306, 311, 373, 446, 480, 517 see also ‘final solution’ Holodomor, 95, 140 holy cruelty, 464 Holy Land, 429, 430, 432, 474, 482, 487–9 holy wars, 430, 472, 475–7, 480, 485, 487–8, 491–2, 494–5 see also Albigensian crusade, First Crusade, Second Crusade, Third Crusade, Fourth Crusade, jihad Hò ̂ ng Đu´̛ c, Emperor, 533, 534–6 honour, 273, 483–4, 557, 625, 633, 641 horses, 413, 458–9, 556–7, 606 hostages, 159, 164, 228, 245, 285, 319, 323, 508 Huaca de La Luna, 573 Huanchaquito-Las Llamas, 573, 582 Huari, 573 Hugh the Chanter, 410 Huitzilopochtli, 641, 643 Hula¯iia, 228 ˘ Hülegü, 513–15 human agency, 138, 168–9, 180, 467 human sacrifice, 66, 575, 580, 582, 590, 630–1 see also sacrifice humanitarian crises, 131, 138, 145 humanitarian supplies, 130, 135, 145 Humber estuary, 403 humiliation, 13, 60, 289, 293, 295, 307–8, 395, 463 hunger, 127, 128–33, 134, 138, 142, 146, 408–9, 412–13; see also famine, food crises, Great famine genocide, great famine in England, Great Soviet Famine/ genocide, mass starvation, starvation Hungerpolitik, 132–3 hunter-gatherers, 73, 81 Hutus, 96, 200

Greece (cont.) mainland, 241, 251, 253 urbicide, 235–256, 257–77 Greek cities, 240, 248–9, 270 see also polis and individual entries for Greek cities Greeks, 95, 170–1, 235–41, 248–9, 254, 268–70, 295–6, 303–4 Gregory Bar Hebraeus, 455 Gregory IX, Pope, 494 group identity, 50, 324, 373, 490, 583 see also collectivities, identities Guatemalan genocide, 96, 98 Guibert of Nogent, 457, 461–2 Gujarat, 558–9, 566, 569 Güyük, 512–13 Hadrian, Emperor, 267, 348–50 Haifa, 28 Halberstadt, 70, 74 Han, 377, 379–80, 385–6, 390, 394–9, 506 Harrying of the North, 403–9, 411, 414, 416–17, 419–23 harvests, 139, 157, 160, 167, 409, 608 Hasan, Amir, 551, 555 hate, 13, 89, 99, 466 heads severed, 380, 554, 578 trophy, 580, 590 health infrastructure, 143 heaven, 19, 22–3, 393, 398, 493–4, 505, 520 Hebrew Bible, 22, 186, 189, 192–4, 203, 372, 461, 463 Heinrich von Diessenhoven, 441 Hellenistic culture, 202, 273, see cultures Hellenistic period, 202, 206, 257–75 Hellenistic violence, 257–8 Hercules Victor, 299–300, 302 hereditary enemies, 498, 501, 519 h¯erem, 43, 48, 50, 54, 185–90, 200, 205, 229 ˙Herero, 25, 26, 95, 129 heresy, 472–4, 476–7, 478, 480–1, 483–4, 485–92, 494–5 heretical depravity, 470, 476, 478, 486, 492, 494–6 heretics, 463, 470–5, 479–80, 481–5, 487, 489–92, 495–6, 626 Provençal, 470–4, 478, 485, 488, 489, 495 Herodotus, 157, 237–9, 241 Heshbon, 186, 197 Hierro, 594–5, 601–2, 603, 606, 609, 611 hillforts, 74–5, 77, 80 Himera, 248–9 Himmler, Heinrich, 18, 23–4

656

Index hydroclimatic shocks, 172–5, 178 Hysiae, 244–5, 247–8

indigenous populations, 67, 128, 594, 608–9, 617, 622, 626, 628 indigenous resistance, 608, 610, 630, 645 Indonesia, 12, 24, 28, 133, 173, 176 infanticide, female, 612 infidels, 430, 449, 454, 462–3, 466, 520, 553 infrastructure, 166, 325, 590 health, 143 inhabitants, 171, 226–7, 237, 288, 320–1, 340, 419–20, 609–12 inheritance, 187, 197, 229 Innocent III, Pope, 470–4, 482, 485–9, 491–2 Inquisition, 470, 475–6, 478, 483, 486, 494–6, 617 inscriptions, 42–3, 191, 209, 214, 215–16, 218–19, 226, 227–8, 299–300 see also sources Assyrian, 226–9, 230–2 Mesha, 189–90, 197, 200–1, 205 royal, 209, 213, 216, 218, 225 institutions, 12, 80, 87, 89, 385, 394, 614, 617 cultural, 8, 321, 373 military, 59, 66–7, 521 religious, 99, 272 intent, 6–7, 9–10, 33–6, 42, 83, 120–1, 127–8, 130, 132, 136, 139, 141–5, 166, 193, 210–13, 226–7, 230, 244, 280, 303–7, 310–11, 316–18, 326, 350, 370–1, 374, 378–9, 423, 446, 498–9, 521–2, 548, 574, 578–9, 602, 605 see also justification, motivations deliberate, 166, 521 genocidal, nature of, 34–5, 127, 130, 136, 139, 143 intentional destruction, 227, 304, 310, 314, 370 intentional killing, 131, 350, 381, 580 interethnic violence, 185, 195, 203–4, 205, 206 actual, 199, 204, 206 internal migration, 151–2 international law, 10–11, 132, 133–4, 135, 465, 517, 618–19 invasions, 19, 155, 249, 524, 529, 534 inventions, 52, 59, 67, 311, 382, 486–7, 489 Iran, 499, 508, 511, 512–13, 515–18, 549, 565 Iraq, 129, 143, 161, 163–5, 174, 565, 570 Ireland, 14, 20, 95, 139, 162 Iron Age, 69, 76, 186, 194–5, 196, 209 Early, 73, 81, 84, 195 Late, 80 Iron Age II, 178, 185, 195–6, 199, 202 Isaaq genocide, 96–7 ISIS genocide, 96 Islam, 94–5, 97, 99, 449, 551–3, 559, 561–2, 570–1 see also Muslim conversion to, 551, 571

Ibn al-‘Arabı¯, 453, 462 Ibn al-Athı¯r, 453–4, 458, 464, 468 Ibn al-Jawzı¯, 453, 468 Ibn al-Qala¯nisı¯, 453 Ibn Batuta, 550, 566 iconography, 53, 342, 526 identities, 50, 231, 260, 320–1, 379–80, 480–1, 486–7, 504 see collectivities civic, 257, 262, 271, 273 ethnic, 51, 95–6, 166, 202, 206, 259, 344, 616 group, 50, 324, 373, 490, 583 tribal, 503–4, 505 identity formation, 50, 206, 549, 617 ideological factors, 12–13, 18–22, 35, 43–4, 53–4, 55, 61 ideologies, 10, 43, 48, 52, 53–6, 97, 108, 159–67, 180, 259, 272, 297, 431, 460, 505, 548 anti-Jewish, 54, 428–9 genocidal, 19, 52–6, 306–7, 311, 350, 474, 487, 510, 568 imperial, 159–67, 216–18, 282, 298–301, 325, 358–9, 395 jihad, 451 racialist, 43–5, 55 see also racism ideologues, 547, 570, 571 Iliad, 236 illness, 77, 368–9, 439, 509, 512, 569 see also disease, plague impacts, 48, 94, 144, 151–3, 172, 295, 577, 610–11 paradigmatic, 281, 311 psychological, 307, 435, 640 imperial courts, 360, 366, 394, 398 imperial expansion, 49, 161–3, 167 imperial ideology, 282, 358–9 imperial power, 44, 360, 398, 498 imperialism, 49, 54, 129, 281, 307, 325, 334 see also empire Roman, 282–3, 285, 325, 328–9, 334–5 imperium, 283, 294, 304, 306–7, 328–9 impiety, 220, 267, 271, 301, 567 Inca, 572, 579, 582, 588, 590–2 incitati, 315–16, 319–20 incumbents, 158, 167 independence, 27–8, 240, 267, 511, 530, 555, 631 India, 42, 139–40, 527, 547–71, 601 upper north, 552, 559, 567 indigenes, 624 indigenous peoples, 625–6, 628, 630, 633 see also individual indigenous peoples entries female, 61, 81, 84

657

Index Ismailis/Isma¯ʿı¯lı¯s, 499, 511–15, 519–21 Nizari/Niza¯rı¯, 511–17, 518–21 Israel ancient, 22–3, 27–8, 178, 188, 190, 196–8, 201–2, 203 Israelite armies, 22, 200–1 Israelite sources, 185–207 Israelite texts, 185, 189–92, 200, 207 Israelites, 185–8, 190–3, 194–6, 198–9, 200–1, 203–5, 230, 232–3 Italians, 44, 318 Italy, 19, 157, 286, 298, 303, 307, 436 Izz-ud-Din Isami, 550, 566

Joshua, 22, 185, 186–9, 193–5, 199–200, 204–5, 230, 233 Judah, 189, 196–9, 201–2, 237 Judaism, 86, 94, 95, 99, 427, 431, 617 Judea, 333–8, 341, 343, 345–8, 350 Judeans, 193–4, 199, 201–4, 206 Jurilovca-Orgame, 70 justice, 243, 257–9, 263, 267–9, 271–2, 274, 392, 571 traditions of, 257, 268 justification, 22, 32, 46, 93, 200, 205, 230, 246, 267–8, 270–5, 284–5, 293–7, 301–2, 306, 308, 316, 318, 322, 324–5, 327–8, 391–4, 436–8, 440, 450–1, 465, 476, 489, 544–5, 547–50, 571, 610, 616, 619, 624, 628, 633, 640, 646 see also ethics, intent, morals, motivations justified retribution, 322, 324 see also punishment, revenge, vengeance Juwaynı¯, 515 jüyin, 500

Jajnagar, 555–6, 558 Jamshed, 567–8 Jamuqa, 501–2, 504 Jebel Sahaba, 67, 70 Jebusites, 187, 229 Jerusalem, 46–7, 336, 341, 343–5, 349–50, 448–68 massacres, 460, 466–7 politicising fall, 464–5 siege, 340, 451, 466 Jesus Christ, 427, 430, 461, 464, 482–3, 490, 493–4, 625 Jewish communities, 336, 338, 347, 427–30, 432, 435, 438, 442–4 Jewish genocide by Polish collaborators with Nazis, 96 Jewish moneylenders, 428, 437 Jewish population, 330, 334, 335, 337, 339, 341, 441, 443 Jewish quarters, 433, 440–1, 444–5 Jewish revolts, 338, 341–3 Jewish sources, early, 185, 204, 206 Jews, 17, 330, 333–4, 335–6, 337–8, 343–8, 349–50, 425–46 1189–90 massacres in England, 432–5 and Rome, 330–50 anti-Jewish urban riots in Castile and Aragon, 426, 443–5 anti-Jewish violence, 426, 428, 430, 433, 434, 436, 438, 444–5 Babylonian exile, 201–3 Bavarian, 25 medieval western Europe, 425–46 York, 434–5 jihad, 451, 467, 570 ideology, 451 Jin, 389, 499–500, 501–2, 506–8, 510, 521 Johan de Sant Peire, 483 John of Worcester, 408–11 Jonassohn, Kurt, 12, 36, 474, 487, 490, 518 Josephus, 202, 330, 334, 339, 343

Kalahari desert, 129 Kapelle, W. E., 406 Kazakh genocide, 95 Kazakhstan, 140 Kedar, Benjamin Z., 46–7, 462 Kemosh, 190 Kenya, 67, 79–80 Kereit, 498, 501, 504–6 Ket-Buqa Noyan, 514, 520 Kfar Kassem, 28 Khmer Rouge, 5, 11, 21–2, 129, 140–1 Khusrau, Amir, 550–1, 555, 559, 563 Khwa¯razmians, 499, 508, 510–12, 514 Khwa¯razmsha¯hs, 508, 511–12 Khwursha¯h, Rukn al-Dı¯n, 514–15, 520 kingship, 218, 262, 555 kinship, 65, 68, 77, 246 knights, 413–14, 430, 437, 470–2, 477, 485, 488–9, 491 Koszyce, 70, 77, 80, 84 Kuelap, 573, 587–8 Kuper, Leo, 40, 446, 490, 518 Kush, Tughril, 557 La Palma, 594, 602–4, 606, 609, 611 La Paloma, 573, 578 labour, 319, 527, 624 forced, 135 labourers, 68, 84 enserfed, 41 Lactantius, 355–7, 360–4, 366, 369

658

Index Lakhnauti, 555–6, 557–8 Lammasar, 515 land, 156–7, 187–90, 196–9, 240–2, 417–21, 487–9, 510 arable, 259, 272, 419 see also agricultural lands, farms expropriation, 135 ownership, 614, 618 shortages, 48, 197–8, 297 tax, 389, 393 Lanzarote, 595, 601–2, 603, 606–7, 609, 611 Las Casas, Bartolomé de, 128, 599, 619, 627, 634, 636–7, 640, 645 Late Bronze Age, 71–2, 75–6, 81, 168, 195 Late Iron Age, 80 Latin Christendom, 477, 484, 486, 490 Latin Christians, 450, 467, 472, 481, 487 Latin sources, 452–3, 458–9, 464 see also sources and individual author and work entries Latvian genocide by NKVD, 96 Lauterpacht, Hersch, 4 law, 19, 132, 133, 263, 271, 368–9, 463–4, 570 international, 10–11, 132, 133–4, 135, 465, 517, 618–19 Roman, 365, 372 shari’a, 561, 570 leaders, 28, 49, 91–3, 315, 317, 556–7, 559 see also commanders, generals, kings and individual entries Cham, 527, 541 ecclesiastical, 427, 432 Mongol, 498, 500, 554 political, 131, 143, 547 Tatar, 500, 503 Vietnamese, 533, 536 leadership, 92, 94, 504–5, 597, 602, 606, 622–3 religious, 91, 98, 372 Legalists, 384, 387, 394 Lemkin, Raphael, 2–4, 32–3, 43, 99–100, 136, 278–80, 309–10, 373 Lesbos, 235, 237, 247 Leuctra, 251–2, 266 Levant, 194, 195–9, 201, 206, 449, 451, 454, 458, 459 before the Crusades, 449–50 Iron Age, 194–200 Southern, 337, 339 Levantine groups, 189–92, 197–8, 200, 203, 205 Libyan genocide, 95 Licinius, Emperor, 369 Lienzo de Tlaxcala, 638 Lima, 573, 584–5

Lincoln, England, 433 Linear Pottery culture, 73, see cultures literary sources, 232, 281, 357 see also sources and individual author and work entries livestock, 14, 187–8, 418, 420, 510 Livy, 19, 156–7, 282, 293 Llullaillaco, 573, 591 local elites, 328–9, 443, 591 London riot, 433–4 looting, 131, 259, 302, 324, 454, 529 López de Gómara, Francisco, 599, 634, 637, 644 Louis IX, King, 494 Louis VIII, King, 474, 478, 494 Lu Xun, 376 Lugo, Alonso de, 603, 604, 610–11, 617 Lyttus, 257, 264–5, 270 Macedon, 201, 251–3, 271, 282 Macedonians, 253, 254, 261, 266, 293 Mainz, 46, 429–32 males, 50, 81–2, 84, 128, 189, 503, 535 adult, 77, 82, 247, 249, 389–90, 581, 587, 589 Malik-ul-Umara, 556 malnutrition, 76–7, 131, 138, 412; see also famine; hunger Manchuria, southern, 395, 398–9 Manichaeism, 371–4, 479 manors, 418–19, 420 northern English, 417, 420 manpower, 217, 328, 529, 591 Mantinea, 244, 249–51, 257, 265–8, 271 Manual Redeye, 29 Mao Zedong, 141, 376–7, 394 Maori genocide, 95 Marathon, 240, 243 maritime powers, 133, 286, 306 Marmande, 493–5 marriage, 22, 68, 500 alliances, 395, 500, 526 forced, 22 Martínez, Ferrán, 444 martyrdom, 429, 431, 520 mass atrocities, 127, 131–2, 143, 145–6, 147 mass conversions, 438 mass enslavement, 40–1, 257, 264, 266, 272, 298, 305–6, 321–8 mass extermination, 572–5, 578, 579, 582, 590, 591–2 mass graves, 27, 59, 61, 64, 83–4, 580, 582, 584 mass killings, 310–11, 321–2, 327–9, 376–9, 381–2, 390–3, 466, 622–3 Caesar, 314–21

659

Index mass murder, 22, 24, 27, 472, 474, 494–5, 574, 592 mass starvation, 127, 129–30, 132, 134, 138, 140, 142–4, 145–6 see also famine, Great famine genocide, great famine in England, Great Soviet Famine/genocide, harvests, hunger, mass starvation, starvation mass violence, 35–6, 39–43, 44–8, 185–6, 230–2, 295–6, 312–13, 326–9 massacres see individual entries material culture, 65 material factors, 48–9 materialism, 18 Maximin, Emperor, 368–9 meat, 503, 559, 581 Megalopolis, 257, 265–7, 272 Melos, 246–8, 251 Mencius, 387, 393 mental harm, 6, 40, 60, 210, 222, 332, 373, 379 Meos, 554–5 mercenaries, 245–6, 261–3, 472, 482–3, 485, 492, 559 Mesha inscription, 189–90, 197, 200–1, 205 Mesoamerican peoples, 631, 645 Mesopotamia, 199, 201, 209–33, 346, 349 Mesopotamian culture, 232–3, see cultures Mesopotamian evidence, 210, 225, 232 see also sources Mesopotamians, 49, 186, 198, 210, 219, 224, 231–3 Mexica, 476, 625, 629, 630–1, 633–5, 643–5 see also Amerindians allies, 632, 641 nobles, 641–3 Mexico, 622, 631 central, 629, 631–2 Middle Ages, 422, 425, 428, 445–6, 449, 463, 465, 474 Middle Bronze Age, 75, 78 migration, 62, 151, 158, 168, 613 forced, 175, 196, 394 internal, 151–2 Miletus, 240, 363 military campaigns, 46, 66, 131, 214, 403, 408, 419, 421 see also individual entries Vietnamese, 523, 535 military institutions, 59, 66–7, 521 military technologies, 261–2, 530 militias, 35, 92, 130–1 Milley, General Mark, 29 Minhaj-us-Siraj, 554 Moab, 190, 198, 200–1, 203

Moabite Stone, 191 Moabites, 51, 185, 189–92, 194–5, 197–8, 230, 233 Moche Temple of the Moon, 580 Moctezuma, Emperor, 633, 635, 637–8, 640 Moctezuma II, Emperor, 629, 633 Moctezuma, envoys, 644–5 money, 1, 300, 364–5, 418, 430, 565 moneylenders, 444 Jewish, 428, 437 Möngke Khan, 513, 515 Mongolia, 398–9, 498, 499, 502, 505, 508–9, 514–15, 520 central, 498, 501 eastern, 498, 502, 520 western, 502, 515 Mongols, 377, 390, 391, 498–522, 553–4, 558, 560 Borjigid, 498, 501, 504 commanders, 512, 514 leaders, 498, 500, 554 reprisals, 508, 517 monks, 405, 436, 459, 461, 468, 470, 481, 489 Montfort, Simon de, 491–3 Moore James Frazer, 94 R. T., 476 morals, 31–2, 45–6, 246, 280, 282, 297, 302, 310–11, 431, 439, 468, 472, 474, 483–4, 489, 517, 531, 533–6 see also ethics, justifications mosques, 99, 452–9, 462, 466 mothers, 21, 221, 223, 411, 442, 535, 560, 594 motivations, 6–7, 9, 32, 35–6, 41, 46, 61, 72, 80, 87–8, 108, 140, 151, 230–1, 241, 244, 259–60, 272–5, 285–7, 289, 293–301, 306–8, 311, 316, 319–20, 325–6, 331, 350, 370–1, 374, 430, 435, 467–8, 487, 519–22, 548, 552, 578–9, 583, 592 see also intent, justifcations religious, 370, 435, 467 motives, see motivations Mozi, 387–8 Mughis, Qazi, 562–3 Mu‘iz-ud-Din bin Sam, 552 Multan, 567 Mummius, Lucius, 288, 299–301, 302–3 municipal authorities, 439, 441 Muqaddar, Malik, 556–7 murder, 9, 90, 94, 572–4, 577, 579, 586, 587 mass, 22, 24, 27, 472, 474, 494–5, 574, 592 ritual, 30, 435–6, 439 Muslim law, 562, 570 Muslim sources, 452–3 Muslims, 46–7, 95–6, 432, 450, 452–5, 457–8, 463–4, 466–7; see also Islam

660

Index mutilation, 463, 476, 478, 575–7, 580–2 Myanmar, 96 Mycalessus, 244–6 myths, 92, 127, 219, 236, 464, 479, 578 Mytilene, 247

Nuon Chea, 21–2, 24 Nuremberg, 20, 437 Nuremberg Laws, 25 Nuremberg trials, 3, 25 Nusrat Khan, 559–60

Nahua-speaking communities, 630, 638 Nama, 25, 26, 129 Namibia, 26 Naram-Sin, King, 209, 212–17, 218–20 Nataruk, 67, 70, 79–80 nation-states, 38, 86, 92, 97–8, 100, 588 natural archives, 154, 168–79, 180 natural disasters, 143, 152–3, 420, 583 natural forces, 220, 221, 224 natural resources, 595 Nazi Germany, 16, 20, 25–6, 94 Neanderthals, 14, 66 Nebo, 190, 205 negotiations, 315, 316, 334, 508, 514, 606, 636 Neo-Assyrian Empire, 154–5, 164–8, 174–5, 197–8, 224, 227 Neo-Assyrian sources, 178, 224–32 Neolithic farmers, 81 Neolithic period, 73–4, 81, 83–4 neologisms, 31, 33–4, 236 neo-Muslims, 559–61, 570 Nerva, 345 Netherlands, 78–9 New Zealand, 95 Nicomedia, 356, 359–60, 362–3, 366, 369 Nineveh, 158–61, 163–5, 175 Nippur, 155, 220–1 Nizari Ismailis, 511–17, 518–21 Nizaris, 511, 517, 520 Nìžná Myšl’a, 70 nomads, 23, 158, 499–500, 503, 506, 510, 519, 521 pastoral, 499, 519 non-combatants, 60, 164, 257, 266, 274, 314–15, 572, 579 non-state genocide, 59–84 Norman Conquest, 406, 413, 421, 432 Normandy, 405, 415, 438, 601 Normans, 408, 410, 414, 417, 420–2 North Africa, 157, 248, 278, 340, 595, 596, 614 northern Vietnam, 395, 398 Northumbria, 403–5, 408, 410, 416–17, 420 Northumbrians, 405–6, 411, 415, 421–2 Noviodunum, 319, 320 Numantia, 34, 278, 282–4, 289–300, 302–4, 306–8 Numantines, 289–91, 295, 297, 300, 302–4, 306–7

object lessons, 212, 218, 311, 320, 518, 622–3, 636, 646 see also conspicuous destruction offerings, 160, 170, 190, 584, 587 votive, 271 Ofnet, 70, 73 Olynthus, 251–2, 253 omens, 159–60 Orchomenus, 250–1 Orderic Vitalis, 405, 414–17, 422, 460 Orthodox Christians, 95, 450 otherness, 60, 87, 327–8 see also outsiders, prejudice, racism Ottoman Armenian, 26 Ottoman Empire, 26, 97 outsiders, 61, 67, 72, 73–4, 84, 90, 93, 162 see also otherness, prejudice, racism Ovando, Nicolás de, 622, 626–8 Pacatnamú, 573, 586 Palestine, 195–6, 202, 348, 349, 356, 366–8 western, 196, 202 papacy, 474, 487, 490, 601 papyrus, 357, 360, 364 pastoral nomads, 499, 519 pastoralists, 162, 167 Pastoureaux, 438–40 Patarenes, 482 peace, 244, 246, 300, 392–3, 395, 547, 548–9, 571 peaceful routine, 334, 339, 341, 346, 350 peasant farming, 140–1 peasants, 139, 140, 406–7, 409, 418–19, 428–30, 481, 482–4 Pennines, 403, 412, 416, 417, 421 Pequots, 23 Peraza family, 602, 607, 615 perennial themes of genocide, 22–4 perfidia, 315–16, 318 peri-mortem trauma, 64, 71, 74–5, 77–8, 576, 577, 584 Perizzites, 187, 229 perpetrators, 9–10, 12–13, 35–7, 87–8, 144–6, 258–9, 311–12, 381–2 see also génocidaires persecutions, 21, 353–7, 360–1, 363, 369–70, 372, 440–2, 479–80 Tetrarchic, 353–75 Persepolis, 235, 254–5 Persians, 201, 235, 240–1, 243, 254–5, 377

661

Index Peru, 176, 572–4, 578–9, 580, 584–6 northern, 580, 582, 586 Philip I, King, 271 Philip II Augustus, King, 487–8, 493–4 Philip IV, King, 436 Philip V, King, 263, 271–3, 438, 440 Phrygia, 353, 368, 370 Phú Yên, 541 Phylarchus, 266, 268 physical destruction, 6, 7, 32, 35, 235, 240, 373, 379 Pierre des Vaux-de-Cernay, 484–5, 489, 492 pigs, 408, 535 pilgrimage, 461, 464, 489, 490, 566 armed, 429, 488 martial, 474, 486, 488, 490 pilgrims, 429, 489, 492 pits, 71–3, 74–8, 83, 272, 441, 562, 610 plague, 48, 71, 438, 441–2, 486, 490, 608, 612 see also epidemic disease, illness plantation agriculture, 595, 618 Plataea, 240–1, 243–4, 251 plunder, 135, 158, 269, 272, 286, 288, 298–300, 433–4 Plutarch, 19, 282, 287, 309, 315, 322 Po Binasor, 528–9, 532 poems, 155, 220, 222–3, 433 see also sources and individual author and work entries poisoning accusations, 438–40 Pol Pot, 1, 11, 18, 21–2, 141 Poland, 15, 17, 30, 96, 133, 135 polis, 235–6, 250, 258–9, 261–2, 272–3 see also Greek cities, urbicide and individual entries Polish genocide by NKVD, 96 political communities, 215, 219 political crises, 449, 571, 632 political factors, 49–50 political groups, 8, 10–12, 33, 332 political leaders, 131, 143, 547 political power, 98, 257, 261, 376, 395, 543, 549, 592 Polybius, 260–1, 263–4, 266–73, 281–4, 289, 294, 295–7, 305 Pontius Pilate, 335 popes, 430, 440, 458, 472–5, 481, 485–6, 487–9, 492; see also individual names Porajmos, 96 Portugal, 291, 602 Portuguese, 599, 601, 605, 613–14 post-exilic sources, 200–4 post-mortem trauma, 577, see trauma, postmortem

power, 11–12, 80, 92–3, 259–61, 344, 381, 527–9, 565 absolute, 51, 381, 463, 572, 592 imperial, 44, 360, 398, 498 political, 98, 257, 261, 376, 395, 543, 549, 592 privileged access to, 91 social, 91, 250 pregnant women, 189–90, 198, 205 prehistory, 14–15, 47, 54, 59, 62–3, 69, 71–2, 574–5 prejudice, 15, 30, 44, 98, 273, 286–7, 296–7, 327–8, 466 see also racism gender, 19–22 premodern genocide, 31–56 primary sources, 330, 371, 465, 537 see also inscriptions and individual author and work entries prisoners, 215–16, 452, 454, 462–3, 631, 633 see also captives, enslavement, mass enslavement, slavery, slaves of war, 133, 580–1, 623 propaganda, 28, 200, 213, 217, 342, 349, 467 property, 267, 269, 272, 363–4, 366, 372, 428, 488 Prophecy of Neferti, 162, 171–2 prophets, 53, 562, 568, 570 protected groups, 37, 53, 136–7, 141 Protestants, 96, 479 Provençal heretics, 470–4, 478, 485, 488, 489, 495 provincial populations, 335, 345–6, 348, 350 provincial rebellions, 340, 341 Psarras, Sophia-Karin, 386, 395 psychological impact, 307, 435, 640 Ptolemaic Egypt, 162, 169–74, 180 Publicani, 482 Punic genocide, 95; see also Carthage punishment, 272, 274, 365–7, 559, 562–3, 566–7, 569–70, 637 see also reprisals, retribution, retributive measures, revenge, vengeance capital, 218, 361, 567–8 collective, 318, 321, 323, 325, 328 exemplary, 246–7, 324 punitive campaigns, 379, 381 Punjab, 552–4, 556 Punta Lobos, 573, 585–6 Qalandars, 557–8, 568 Qi, 378, 389 Qin, 379, 384–5, 389–93, 394, 398 dynasty, 377, 380, 391, 394 mass killing under unification, 393

662

Index resistance, 393–4 state, 383, 391 Qin Shihuang, Emperor, 377, 383, 394 Quetzalcoatl, 632, 638, 640 Quhista¯n, 511, 514–15, 520 Quiders, Bertran de, 483 Quietus, Lusius, 346, 349 Quintilian, 313 Qumran, 193, 202 Quran, 91, 99, 452–3, 563, 570 Qutb Minar, 552–3

religious rituals, 284, 302 religious traditions, aboriginal, 95, 98 repetitive behaviours, 92–3 reprisals, 17, 156, 212–14, 347, 406, 586, 592 see also punishment, retribution, retributive measures, revenge, vengeance Mongol, 508, 517 reproduction, 223, 259, 373 biological, 60, 305, 322, 324 cultural, 81, 223 social, 9, 258, 333, 373 reputation, 232, 235, 286, 289, 294, 296, 298, 302 resentment, 11, 48, 170, 427, 432, 437, 443, 609 residential schools, 27 resistance, 164, 346, 348, 350, 365–6, 383, 393–4, 608 effective, 609–10 indigenous, 608, 610, 630, 645 resource challenges, 195, 203 resources, 195–7, 200–1, 206, 419, 602–3 natural, 595 scarce, 48, 584, 603 water, 168 restitution, 254, 619 retribution, 41, 46, 250–1, 255, 324–6, 514, 519, 521 see also punishment, reprisals, retributive measures, revenge, vengeance justified, 322, 324 retributive measures, 340, 342 see also punishment, reprisals, retribution, revenge, vengeance revelatory crises, 153, 160, 173 revenge, 13, 240, 242, 250, 255, 556, 560, 610 see also punishment, reprisals, retribution, retributive measures, vengeance personal, 427 see also punishment, reprisals, retributive measures, retribution, vengeance revenues, 171, 427, 485, 489, 562, 603 revolts, 170–1, 212–14, 318–19, 324–5, 335–7, 338–40, 342–9, 566–7 see also rebellions Great Theban, 170, 173 internal, 155, 179–80 Jewish, 338, 341–3 Rhine, 315–17 Rhineland, 425, 429, 432, 435, 441 Rhône, 472, 482–3, 485–6, 488 ribauds, 470–1, 488–9, 491, 495 see also boys Riggsby, A., 312, 324, 327–8 Rimuš, 209, 212–13, 216–17, 219 Rintfleisch, 425, 435–8

race, 5, 16, 21, 53–4, 468, 476, 519, 616 racialist ideology, 43–5 racism, 6–7, 21, 23, 25, 30, 36, 44–5, 54, 55, 103, 110, 136, 219, 296–7, 327–8, 380–1, 446, 466 see also dehumanisation, prejudice nature and limits, 15–19 scientific, 38 raids, 527, 529, 532, 536, 580, 584, 606, 610 slave, 601, 624 Raimon VI, count of Toulouse, 472, 478, 482, 487–8, 489, 492, 494 Raimon VII, count of Toulouse, 478, 494 rainfall, 156, 169, 174, 583 rape, 22, 104, 107, 110, 114–17, 122–23, 126, 414 see also sexual or sexualised violence Raymond of Toulouse, 455–9, 466 rebellions, 212, 390, 394, 437–8, 560, 562–3, 568, 644–5 see also revolts provincial, 340, 341 rebellious cities, 209, 212, 226 rebels, 228, 230, 348, 390, 394, 567–8, 569, 571 recurrent famines, 139, 143 refugees, 130–1, 409, 412–13, 417, 421, 485, 491 regional communities, 10, 212, 231, 311, 518 regulations, 530, 552, 563, 568, 570 relevance, 67, 92, 162, 341, 619 religion, 52–3, 86, 87, 88, 89, 91, 92, 93, 94, 97, 272–3, 360–3, 519–20, 548–9, 559–60 as participating factor in genocide, 95–6 definition, 89–90 evidence, 94–8 religion–genocide nexus, 86–100 religious affiliation, 374, 465 religious beliefs, 7, 334, 548 religious communities, 92–3, 371, 374 religious diversity, 450, 548 religious groups, 6, 7–10, 33, 37, 52–3, 210, 212, 378–9 religious institutions, 99, 272 religious leaderships, 91, 98, 372 religious motivations, 370, 435, 467

663

Index ritual murder, 30, 435–6, 439 ritual violence, 231, 631 rituals, 66, 71, 75–6, 92, 302–3, 361, 365, 368 funerary, 575, 584 religious, 284, 302 rivals, 294–5, 302, 501, 537, 543, 579, 585, 627–8 Robert the Monk, 457, 461, 468 Rohingya genocide, 96 Roman administration, 330, 333, 338, 341–2, 345 Roman ambassadors, 288, 323 Roman army, 285, 297, 324, 340, 362 Roman authority, 287, 289–91, 294, 301–3, 368 Roman Catholicism, 95–8 Roman law, 365, 372 Roman rule, 204, 318, 330, 340 Roman sources, 69, 283, 300–1, 597 Roman traders, 319, 324 Roman world, 51, 280, 295, 301, 321–2, 343, 372–3 Romani, 3, 15, 25, 96 Romania, 77, 322–3 Romans, 282–5, 287–91, 297–8, 301–6, 318–19, 324–5, 327–9, 348–9 massacre of, 44, 318, 321 Rome, 278–308, 318–19, 327–9, 334–7, 349–50, 358–60 and Carthage, see Carthage and Corinth, see Corinth and Jews, 330–50 and Numantia, see Numantia Caesar, 22, see Caesar, Julius empire, land, agriculture and mass violence, 297–8 greed and glory, 298–300 imperialism, 282–3, 285, 325, 328–9, 334–5 imperium, 283, 294, 304, 306–7, 328–9 prejudice and justification of violence, 296–7, 327–8 provincial conflict, 338–49 provincial routine, 333–8 reactions and ethical constraints on mass violence, 301–2 religious context, 302–7 religious violence in the late empire, 353–74 senate, 19, 284–6, 291, 301, 314–5, 331, 348–9 sources, 69, 281–2, 283, 300–1, 355–8, 597 spectrum of violence, 325–6 Tetrarchic persecutions, 34, 353–6, 358–61, 370–1, 374 see also individual emperors and leaders Rome Statute of the International Criminal Court, 8, 136–7, 332, 378

Röttingen, 437 routine, 160, 334–5, 337, 345 peaceful, 334, 339, 341, 346, 350 royal authorities, 48, 436, 443, 445, 604–5, 630 royal castles, 433–4, 439, 445 royal inscriptions, 209, 213, 216, 218, 225 see also inscriptions RPF (Rwandese Patriotic Front), 130 Ru¯dba¯r, 514 Rummel, R. J., 376–7, 381, 386, 398 Russia, southern, 133, 140 Rwanda, 1–2, 91, 94, 96, 98, 130, 199–200, 205 Rwandese Patriotic Front (RPF), 130 sacrality, 93, 99, 495 sacred books/texts, 91, 98–9, 363–4, 366, 372–3, 617 sacred sites, 254, 365, 542, 579, 591 sacrifice, 71, 75, 189–90, 361, 367–8, 578, 580, 583 human, 66, 575, 580, 582, 590, 630–1 sacrificial victims, 580, 631 sacrilege, 241, 270–2, 274, 643 sailors, 433, 443, 445, 613 Saint-Sernin, 483 Salamis, 240, 241–2, 249 Samaria, 178, 203 Samuel, 22–3, 188, 192 sanctuary, 47, 262–3, 413, 421, 462, 579 Sandby Borg, 70, 80 Saracens, 455–8, 466, 474 Sargon II, King, 158, 161, 166–7, 178–9 Saul, 22–3 Savoy, 441 scale of genocide, 39–40 Schleinbach, 70, 75 Schöneck-Kilianstädten, 70, 73 schools, 1, 5, 7–8, 27, 246, 262, 387 residential, 27 Scipio Aemilianus, 281–3, 291–4, 297–8, 299–302, 304, 306 Scipio Nasica, 284, 287 scorched earth policy, 291, 407 Second Crusade, 432 Secret History of the Mongols, 502, 503 Secularists, 95 sedentary societies, 510, 519, 521 Segesta, 248 Seleucids, 201–3 Selinus, 248 Seljuqs, 511, 514 Selk’nam genocide, 95

664

Index senate, Roman, 19, 284–6, 291, 301, 314–15, 331, 348–9 Sennacherib, King, 226–8 Serbs, 96 Bosnian, 144–5 servants, 412, 470 settlements, 71, 75, 78, 298, 340–1, 349, 350, 579–80 settler violence, 626, 628 settler-colonialism, 595, 600 settlers, 61, 242, 251, 599 severed heads, 380, 554, 578 Seville, 443–4, 453, 603, 617 sexual or sexualised violence, 50, 69, 104, 113–17, 121–23, 126, 587 see also rape Shamash-shumu-ukin, 155–6 shame, 13, 257, 273, 289, 555 Shams al-Dı¯n, 513 shared victimology, 574–5 sharp force trauma, 63, 71, 76, 585 sheep, 188, 222–3, 232, 396, 408 shepherds, 425, 438, 492 Shiha¯b al-Dı¯n, 513 Shiji, 378, 385 ships, 263, 403, 527 Shoah, 96, 236 shock, 147, 170, 174, 218, 454, 464, 640–2 environmental, 420 hydroclimatic, 174 volcanic-related, 176 Shropshire, 412 Sicily, 157, 238, 248, 260, 283, 474 siege engines, 261, 434, 457 sieges, 128, 133–4, 283–4, 293, 319, 456 see also Alesia, Avaricum, Carthage, Numantia Jerusalem, 340, 451, 466 long, 46, 252, 305–7, 470 silver, 188, 364–5, 510, 559, 565 Sindh, 553–4, 569, 570 sites, 64, 70–1, 74–80, 251–2, 578, 583, 585, 586–7 fatality, 63, 69–71, 78–80 sixteenth century, 391, 422, 540, 548, 598–9, 612–13, 626, 628 skeletons, 64, 76, 77–9, 577–8, 580–2, 584, 585–7, 588 skulls, 73, 75–7, 518, 582, 587–90 slave raiding, 601, 602, 605, 611, 624 slavery, 236, 239, 313, 602, 604–6, 614, 617, 619 debt, 196–7 see also enslavement, mass enslavement slaves, 239, 322–4, 503, 504, 555, 557, 605–6, 614 see also captives, prisoners

freed, 44, 318 social groups, 9–10, 33, 51, 52–3, 210–12, 231–2, 574, 580 social order, 220, 394, 398 social power, 91, 250 social reproduction, 9, 258, 333, 373 social status, 363, 629 sociocultural factors, 50–3 socio-economic factors, 151, 152 soldiers, 315–17, 320, 321–2, 341, 390, 470, 531–3, 555 German, 16, 26 Zhao, 386, 390 soldiers of Christ, 488–9 solitude, 410–11, 417 Somalia, 96–7, 143, 147 songs, 30, 93, 247, 255, 451, 461, 642 souls, 140, 439, 485, 493, 612, 626, 639 sources, 46–7, 54–5, 330–2, 355, 408–9, 451, 452, 632–3, 639–41 see also inscriptions and individual author and work entries Arab, 458, 460 biblical, 23, 185, 189, 199–200, 205 contemporary, 405, 549, 600, 608, 610, 612 early Jewish, 185, 192, 204, 206 ethnohistorical, 62, 67 historical, 281, 377, 384, 598 Latin, 452–3, 458–9, 464 literary, 232, 281, 357 primary, 330, 371, 465, 537 Roman, 69, 281–2, 283, 300–1, 355–8, 597 secondary, 384 Spanish, 610, 631–3, 644 Syrian, 454, 458 unreliability, 451–62 written, 14, 55, 350 South Africa, 98, 205 South America, Andean, 572–3, 583–4 South Sudan, 143, 146–7 southern France, 237, 439, 440, 473–4, 479 southern Levant, 337, 339 southern Manchuria, 395, 398–9 southern Russia, 133, 140 Soviet ethnic cleansing/genocide, 95 Spain, 289–91, 295, 298–300, 340, 427–8, 438, 630, 636 Spaniards, 616, 635, 637–42 Spanish conquests, 41, 45, 604 of the Americas, 622–46 Spanish sources, 610, 631–3, 644 Sparta, 240, 242–3, 244, 247, 249–52, 288 allies, 242, 244, 249 Spartans, 242–5, 246, 251, 265, 267, 288

665

Index Srebrenica, 39, 145 SS/Schutzstaffel, 16–18, 93–4 St Cuthbert, body of, 410, 413 Staffordshire, 412 Stalin, 18, 140, 376 starvation, 127–37, 139–42, 143–7, 196, 199 see also famine, Great famine genocide, great famine in England, Great Soviet Famine/genocide, harvests, hunger crimes, 131, 133, 136, 137, 143, 145, 147 mass, 127, 129–30, 132, 134, 138, 140, 142–4, 145–6 policies, 132, 140, 145 state capacity, 235, 377, 381, 383, 394 statistics, 377, 384, 390, 524, 539–40 statues, 270–1, 299, 336, 347, 643–4 steppe ancestry, 78, 81, 84 steppe influxes, 64, 68, 83, 84 steppes, 81, 222, 395, 397, 499–502, 503 Stillfried, 70, 77 subjugation, 225–6, 338–9, 343, 345, 347–9, 610–11, 624, 626 Sudan, 67, 130, 143 South, 143, 146–7 Sui, 380, 398 suicides, 94, 134, 291, 429, 431, 638 Sulpicius Severus, 348 Sumer, 155, 209, 212, 220, 222–4 Sund, 70, 76 superiority, 91–2, 97, 625 cultural, 538, 543 moral, 534 supplies, 128, 168, 315, 319, 420, 423, 624, 629 humanitarian, 130, 135, 145 surprise attacks, 536, 584, 636, 640, 645 surrender, 258–9, 285, 291–4, 323, 364–5, 390, 395–6, 514–15 unconditional, 263, 541 survival, 18, 90, 132, 137, 145, 199, 304, 358 surviving evidence, 35, 466, 607 see also archaelogical evidence/sources, inscriptions, sources survivors, 187, 188, 227, 228–9, 283, 431, 455, 612 Susa, 156, 254–5 suspicion, 331, 364, 409, 451, 568, 627, 643–4 Swabia, 437 symbolic violence, 293–4 see also conspicuous destruction, object lessons symbolisation, 93 symbols, 93, 222, 247, 479, 552 Symeon of Durham, 405, 410–11, 413, 417, 421–2 synagogues, 47, 99, 441, 452–3 Syracuse, 248

Syria, 64, 129, 142, 146, 152, 335, 338–9, 347–8 Syrian sources, 454, 458 Tacitus, 69, 335, 342, 344 Talheim, 70, 73 Tang, 380, 398 Tangut, 377, 499, 506–10, 515–17, 519–21 Tapia, Andrés de, 633–6, 638, 643–4 Tarain, 552 Tatars, 498–506, 509, 515–17, 518–21 jüyin, 500 leaders, 500, 503 taxes, 170, 172, 179, 342–4, 382–3, 397–8, 417–19, 421 Tayichi’ud, 502, 504 techniques of genocide, 40–1 technologies, 18–19, 55, 146, 179, 261, 382, 389, 547 military, 261–2, 530 Tees river, 403–5, 410–11, 416 Tel Shalem, 349 Tell Brak, 64, 70 Temoney, Kate, 86–7 Templars, 464 temple floors, 581–2 temples, 170–1, 218–21, 299–300, 302, 336, 343–5, 638, 643 Templo Mayor, 622, 641–4 Temüjin, 498–505, 520; see also Chinggis Khan Temüjin Üje, 500 Tencteri, 314–15, 317–18, 320, 329 Tenerife, 594, 603–4, 606–12, 617 territorial ambitions, 519, 520, 614 see also imperialism territorial expansion, 62, 69, 77, 83, 84, 180, 530, 614 see also imperialism terror, 260, 264, 293–4, 603, 618–19, 628, 637, 640 terrorism, 147, 293, 466 Tetrarchic persecutions, 353–6, 358–61, 370–4 texts, 91–2, 188–92, 193–4, 199–202, 213, 217–22, 228–30, 355–8 see also inscriptions, sources and individual author and work entries Thebans, 237, 243, 250–1, 253–4 Thebes, 240, 243, 249–54 theologians, 426, 445, 551, 561 Thermopylae, 240, 243, 253 Thermum, 257, 270–4 Thespiae, 240, 251 Third Crusade, 433, 435 Third Lateran Council, 482

666

Index thirteenth century, 454, 459, 487, 498–522, 528, 550, 554 Thucydides, 239, 241, 244–8 Thyrea, 242–3 Tiberius Alexander, 340 Tigris, 159, 174–5, 346 Tineus Rufus, 348 Titus, 340, 427, 489 Tiwanaku, 573, 578, 590 Tlaxcala, 631–2, 637–40, 646 Tlaxcaltecas, 639–40 Toghril Ong-Khan, 498, 501–2, 504–5 toleration, 164, 360, 369 torture, 363, 366–7, 559, 563, 629, 634, 637, 643 total destruction, 40, 223–4, 230, 244, 255, 260, 328, 377 totalising violence, 50, 55, 190, 230–2 totalistic killing, 205, 227, 229–30, 233 Toulouse, 439–40, 457–8, 472, 478, 481–3, 484–6, 488, 492–4 Toxcatl, 622, 629, 630, 641–3, 645 feast of, 641, 644 trade, 83, 297, 526, 527, 565, 603, 609 traders, 297, 565, 601–2, 613 Roman, 319, 324 traditions, 98, 213, 274, 380, 413, 562–3, 568, 570 cultural, 60, 528 transhistorical phenomenon, 31, 42–7, 386 trauma, 59, 66, 78, 82, 199, 575–7 ante-mortem, 576, 577 blunt force, 73, 74, 79, 575, 576, 587–9, 591 patterns, 64, 577 peri-mortem, 64, 71, 74–5, 77–8, 576, 577, 584 post-mortem, 577 sharp force, 63, 71, 76, 585 war-related, 63, 67 see also wounds travelogues, 550, 566 treachery, 267, 300–1, 315–16, 318, 320, 321, 634, 640 treason, 568, 630 treasures, 470, 557, 559–60, 643 Trencavel, Raimon-Roger, 488, 491 tribal identity, 503–4, 505 tribalism, 91 tribes, 68–9, 315, 317, 318–19, 320, 501–2, 504–5, 519 see also individual entries Germanic, 42, 69, 315, 317 neighbouring, 42, 69, 317 tribute, 179, 510, 519, 520–1, 523, 528, 627, 631 Triple Alliance, 629, 631–2 trophy heads, 580, 590 tropical eruptions, 170, 173–4, 176, 179 Troy, 236, 255

Trump, Donald, 28 Túcume, 573, 581 Tudebode, Peter, 455–7, 459, 462 Tughlaqabad fort, 564 Tughluq, Sultan Firuz Shah, 551, 566 Tughluq, Sultan Ghiyas-ud-Din, 564 Tughluq, Sultan Muhammad, 550–1, 564–5, 566, 569, 571 Tughril Khan, 555–8 Turkish Committee of Union and Progress, 97 Tutsis, 96, 130, 200 twelfth century, 405, 411, 418, 427–9, 431–2, 435, 478, 481 Tyne river, 403, 411 Typhonians, 172 Tyre, 248, 340, 369, 448–9, 452, 458, 462 Ukraine, 133, 140, 407 unconditional surrender, 263, 541 uninhabited islands, 602, 605 United Nations, 2, 6, 99, 137, 146, 205, 465, 474 United Nations Genocide Convention (1948), 5–7, 9–11, 33, 40, 135–6, 310–11, 517, 618–19 definition of genocide, 6, 8, 33–4, 35, 39, 40–3, 305, 310–11 United States of America, 27, 30 unity, 54, 272, 384, 387, 393, 399, 450, 645 Ur, 155, 219–21, 223 urbicide, 33–4, 38–9, 51–2, 55, 210, 212–13, 216, 218–20, 224, 227, 231–3, 235–6, 255–6, 297, 301, 303, 306, 341, 349–51 see also individual entries ancient Greece, 235–56, 257–77 Urnfield culture, 75, 77 Uruk, 220–1 Ustasˆe genocide, 96 Vaccaei, 291, 300 values see ethics, justifications, morals vassalage, 198, 602, 604 Vázquez, Bernadino, 633, 643–4 Velázquez, Diego, 628–30, 641–2 Velim-Skalka, 70, 75 Vellaunodunum, 319, 320 Veneti, 314–15, 323–5, 326 vengeance, 268, 294, 325, 415, 448, 451, 500, 504 see also punishment, reprisals, retribution, retributive measures, revenge divine, 471, 474 Vera, Pedro de, 602, 607

667

Index Veracruz, 629–30 Vercingetorix, 313, 318–20, 324–5 Vespasian, Emperor, 339–43, 345, 427, 489 victim groups, 37–9, 90, 132, 230, 303, 333, 355, 605 victimology, shared, 574–5 victims, 35–7, 44–5, 142–4, 575–82, 583–7, 627–8, 630–3, 636–40 victors, 61, 249, 254, 448, 463, 578 victory, 42, 192, 216, 241, 349, 541, 557, 560 Vietnam, 21, 399, 523–43 northern, 395, 398 Vietnamese, 21, 523–43 actions, 535, 537, 540 aggression, 524, 532, 540 armies, 523, 529, 531, 536 atrocities, 537–9 attacks, 526, 530–3, 537, 543 emperor, 534, 536, 538 genocide, 2, 534, 540, 542–3 invasion, 523, 535 leaders, 533, 536 military campaigns, 523, 535 rulers, 524, 534, 536, 538 territory, 523, 524, 528, 531–2, 537 Vijaya, 527, 530, 536, 538–9 villages, 24, 250, 406, 410, 483–4, 488, 492, 587 violence, 45–52, 179–88, 194–5, 257–62, 429–33, 434–41, 443–5, 547–51 anthropocidal, 186, 192–3, 206 anti-Jewish, 426, 428, 430, 433, 434, 436, 438, 444–5 copycat, 433, 444 crusader, 431, 448 dehumanising, 226, 232 direct, 130, 132, 137, 328 eliminationist, 227–8 eschatological, 193, 233 excessive, 42, 572, 616, 622–4, 626, 640, 642, 646 exterminationist, 47, 50, 52, 189, 194, 203–4 extreme, 185, 190–2, 199, 257, 259–60, 263, 268–71, 272–4 gendered, 20, 39, 587 Hellenistic, 257–8 interethnic, see interethnic violence mass, 35–6, 39–43, 44–8, 185–6, 230–2, 295–6, 312–13, 326–9 ritual, 231, 631 settler, 626, 628 symbolic, 293–4 totalising, 50, 55, 190, 230–2 Virgin Mary, 617, 643–4

virgin soil diseases, 605, 608, 611, 619 Vitellius, 335 volcanic eruptions, see eruptions Völkermord, 4 voluntarism, 18–19 von Papen, Franz, 26 votive offerings, 271 vulnerability, 146, 152–3, 156, 178, 295 Waldensians, 479, 483–5 war crimes, 135, 137, 263, 407, 499 war wounds, 581, 586 warfare, 59, 63–8, 78–9, 80, 82–3, 146–7, 261, 274–5 definitions, 65, 66 without the state, 65–9 war-related trauma, 63, 67, see trauma, war-related Warring States Period, 378, 384–6, 389, 397 warriors, 59, 66, 78, 631, 634, 636, 638, 640–1 Wassenaar, 70, 78–9 waste, 72, 128, 318, 411, 417–20, 422, 529 values, 419–20 water, 129, 130, 221–3, 635, 638 resources, 168 wealth, 156, 263, 284, 285–7, 300, 510, 552, 631 weapons, 63, 66–7, 76, 82–3, 132–3, 135, 143–6, 382 lethal, 80, 83, 382 weather, 143, 155–6, 158–9, 179 see also climate extreme, 152–3, 156, 159–60, 174, 179–80 Wei, 387, 389–90 wells, 264, 439–40, 441, 561 western Europe, 426, 440, 446, 476 Wiederstedt, 70, 74 William of Poitiers, 414–16 William the Conqueror, 403–23 Williams, Stephen, 365, 367 wine, 561–2, 613 sellers, 561–2 wives, 68, 172, 431, 437, 503, 557, 559, 562 potential, 82 women, 19–22, 78–80, 187–8, 455–7, 470–2, 483–4, 492–3, 579–80 see also females adult, 61, 74, 190 and children, 263–4, 313, 316–17, 520, 556–7, 560–1, 632–3 good, 483–5, 495–6 pregnant, 19, 189–90, 198, 205 young, 68, 71, 73, 77, 81, 84 Worms, 46, 429–31 wounds, 577, 585, 587, 589 war, 581, 586 see also trauma

668

Index Yemen, 146–7 Yesügei, 500–1 York, 403, 408, 410, 412, 414, 416, 419–21, 433–5 Jews, 434–5 Yorkshire, 403–5, 412–13, 417, 419, 420, 423, 433, 434 young women, 68, 71, 73, 77, 81, 84 Yue, 378, 385

Wu Guang, 394 Wu Hu genocide, 95, 97 Xaragua, 622–3, 626–8, 645 Xenophon, 245, 250, 463 Xerxes, 240–1, 255 Xi Xia, 506–10, 517, 518, 521 Xianbei Murong, 380 Xianbei Tuoba, 380 Xiongnu, 380, 386, 393, 395, 504 Yahweh, 50, 185, 187–90, 192–3, 197, 200, 229–30 Yan, 378 Y-chromosomes, 65, 81 Yeke Monggol Ulus, 499, 505–6, 521

Zhanguo ce, 378, 385 Zhao soldiers, 386, 390 Zhongxing, 508–9, 517, 520, 521 Zunghar (Dzungar) genocide, 95 Zuochuan, 378

669