War and Collective Identities in the Middle Ages: East, West, and Beyond [New ed.] 1641893621, 9781641893626, 9781802701067

This book uses sociological perspectives to bring together work on war and identity in the Middle Ages relating to a ran

239 121 2MB

English Pages 246 Year 2023

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

War and Collective Identities in the Middle Ages: East, West, and Beyond [New ed.]
 1641893621, 9781641893626, 9781802701067

Table of contents :
Contents
Notes on Contributors
Chapter 1. War and Peoplehood in the Middle Ages: An Introduction
Chapter 2. War and Peoplehood through Time: A Sociological Longue Durée Perspective
Chapter 3. Making War Ethnic: Arab–Persian Identities and Conflict on the Euphrates Frontier
Chapter 4. Captive Identities: Inscribing Armenianness from Sebēos to Ayrivanec’i
Chapter 5. War and Identity in Early Medieval Bulgaria
Chapter 6. Collective Identifications in Byzantine Civil Wars
Chapter 7. Warfare and Peoplehood: The Vikings and the English
Chapter 8. Medieval European Civil Wars: Local and Proto- national Identities of Toulousains, Parisians, and Prague Czechs
Chapter 9. The Crusades and French Political Identity in the Thirteenth- Century Mediterranean
Chapter 10. The Song– Jurchen Conflict in Chinese Intellectual History
Chapter 11. Faithful to a Vanishing Past: Narrating Warfare and Peoplehood in Yuan China
Chapter 12. War and Collective Identifications in Medieval Societies: Drawing Comparisons
Selected Bibliography
Index

Citation preview

i

WAR AND COLLECTIVE IDENTITIES IN THE MIDDLE AGES

ii

WAR AND CONFLICT IN PREMODERN SOCIETIES Further Information and Publications www.arc-​hum​anit​ies.org/​our-​ser​ies/​arc/​wcp/​

iii

WAR AND COLLECTIVE IDENTITIES IN THE MIDDLE AGES EAST, WEST, AND BEYOND Edited by

YANNIS STOURAITIS

iv

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. © 2023, Arc Humanities Press, Leeds

The author asserts their moral right to be identified as the author of this work.

Permission to use brief excerpts from this work in scholarly and educational works is hereby granted provided that the source is acknowledged. Any use of material in this work that is an exception or limitation covered by Article 5 of the European Union’s Copyright Directive (2001/​29/​EC) or would be determined to be “fair use” under Section 107 of the U.S. Copyright Act September 2010 Page 2 or that satisfies the conditions specified in Section 108 of the U.S. Copyright Act (17 USC §108, as revised by P.L. 94–​553) does not require the Publisher’s permission.

ISBN (HB): 9781641893626 eISBN (PDF): 9781802701067 www.arc-​hum​anit​ies.org

Printed and bound in the UK (by CPI Group [UK] Ltd), USA (by Bookmasters), and elsewhere using print-​on-​demand technology.

v

CONTENTS

Notes on Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Chapter 1. War and Peoplehood in the Middle Ages: An Introduction YANNIS STOURAITIS�����������������������������������������������������������������������������������������������������������������1 Chapter 2. W  ar and Peoplehood through Time: A Sociological Longue Durée Perspective SINIŠA MALEŠEVIĆ. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Chapter 3. M  aking War Ethnic: Arab–​Persian Identities and Conflict on the Euphrates Frontier PETER WEBB. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Chapter 4. Captive Identities: Inscribing Armenianness from Sebēos to Ayrivanec‘i SERGIO LA PORTA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Chapter 5. War and Identity in Early Medieval Bulgaria PANOS SOPHOULIS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Chapter 6. Collective Identifications in Byzantine Civil Wars YANNIS STOURAITIS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

Chapter 7. Warfare and Peoplehood: The Vikings and the English CLARE DOWNHAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 Chapter 8. M  edieval European Civil Wars: Local and Proto-​national Identities of Toulousains, Parisians, and Prague Czechs PHILIPPE BUC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

vi

vi

Contents

Chapter 9. T  he Crusades and French Political Identity in the Thirteenth-​Century Mediterranean GREGORY LIPPIATT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 Chapter 10. The Song–​Jurchen Conflict in Chinese Intellectual History SHAO-​YUN YANG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 Chapter 11. F  aithful to a Vanishing Past: Narrating Warfare and Peoplehood in Yuan China FRANCESCA FIASCHETTI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 Chapter 12. W  ar and Collective Identifications in Medieval Societies: Drawing Comparisons YANNIS STOURAITIS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203

Selected Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231

vii

NOTES ON CONTRIBUTORS

Philippe Buc is Professor of Medieval History at the Universität Wien and at the Leiden Universiteit. He works on religion and violence in Western Europe and in Japan. His most recent article on the topic is “Civil War and Religion in Medieval Japan and Medieval Europe: War for the Gods, Emotions at Death, and Treason,” The Indian Economic and Social History Review 57 (2020): 1–​27. Clare Downham is Professor of Medieval History at the University of Liverpool. She specializes in Viking Age connections between Ireland and Britain. Her most recent major publication is the monograph Medieval Ireland (Cambridge: Cambridge University Press, 2017).

Francesca Fiaschetti is University Assistant and director of the “Mongolia Cluster” at the University of Vienna. She specializes in the history of the Mongol empire and nonHan Chinese dynasties. She has published extensively on Mongol diplomacy, and on the cultural and political history of the Yuan dynasty, and she is the co-author of the Historical Dictionary of the Mongol World Empire, Second Edition (Scarecrow Press, 2018).

Sergio La Porta is the Haig and Isabel Berberian Professor of Armenian Studies, and the Associate Dean of the College of Arts and Humanities at California State University, Fresno. He specializes in Armenian commentaries on the works of Dionysius the Areopagite, and medieval Armenian cultural interactions with the Islamicate, Byzantine, and Latinate worlds. His most important recent project with Dr. Alison Vacca is a new translation of and commentary on the eighth-​century History of Łewond (LAMINE series, University of Chicago Press, forthcoming). Gregory Lippiatt is Lecturer in Medieval History at the University of Exeter. He researches aristocratic government in the High Middle Ages and its intersection with Christian reform movements, particularly the Crusades. His most recent major publication is Simon V of Montfort and Baronial Government, 1195–​1218 (Oxford: Oxford University Press, 2017). Siniša Malešević is Professor of Sociology at the University College, Dublin and Senior Fellow at the Conservatoire national des arts et métiers (CNAM), Paris. His main research interests include the sociology of war and violence, ethnicity and nationalism

newgenprepdf

viii

viii Notes on Contributors and comparative historical sociology. His most recent publication is the book Why Humans Fight: The Social Dynamics of Close-​Range Violence (Cambridge: Cambridge University Press, 2022).

Panos Sophoulis is Associate Professor of South-​ Eastern European History at the University of Athens. He specializes in the history and culture of southeastern Europe, with a particular interest in the Balkans during the Middle Ages. His most important recent publication is the book Banditry in the Medieval Balkans, 800–​1500 (Cham: Palgrave Macmillan, 2020). Yannis Stouraitis is Senior Lecturer in Byzantine History at the University of Edinburgh. His research focuses on collective identifications, war ethics and the social aspects of warfare, migration, and mnemohistory. His most important recent publication is the edited volume Identities and Ideologies in the Medieval East Roman World (Edinburgh: Edinburgh University Press, 2022).

Shao-​yun Yang is Associate Professor of History at Denison University. He specializes in the intellectual history of medieval China, focusing on ideas relating to empire, ethnicity, and foreign relations. His first monograph is the book The Way of the Barbarians: Redrawing Ethnic Boundaries in Tang and Song China (Seattle: University of Washington Press, 2019). Peter Webb is Lecturer in Arabic Literature and Culture at the University of Leiden. He specializes in the literature and culture of premodern Islam with particular focus on community, ethnicity, and social identities in the early Islamic world. His most recent major publication is the monograph Imagining the Arabs: Arab Identity and the Rise of Islam (Edinburgh: Edinburgh University Press, 2016).

1

Chapter 1

WAR AND PEOPLEHOOD IN THE MIDDLE AGES: AN INTRODUCTION Yannis Stouraitis The role of

war in shaping and reshaping various types of polities and visions of community in the Middle Ages is a topic that deserves more attention than it has hitherto received. In historical sociology, it is commonplace to claim that collective identifications, group solidarity, and homogeneity are not a cause but rather a product of warfare and intergroup violence.1 Historical evidence, on the other hand, demonstrates that the impact of war on the stability, cohesion, and/​or cultural homogenization of various kinds of groups such as ethnic or national communities, as well as kingdoms, empires, and nation-​states, can be both constructive and destructive. This means that, even if warfare is a situation that often enhances discourses of collective demarcation and othering, thus sharpening group boundaries and hardening stereotypes, collective identities can be considered as neither a natural nor an automatic result of war. When I conceived the idea of a workshop on the relationship between war and collective identities in the Middle Ages that would result in a comparative volume, one of the main incentives, besides cross-​cultural comparison, was to encourage a more direct dialogue between the fields of historical sociology and social history—​in particular, the history of collective identities in the Middle Ages.2 While I delved into modern sociological theories on ethnicity and nationhood for the purposes of my own research on collective identity in the so-​called Byzantine Empire some years ago,

1 For a comprehensive overview of the various sociological approaches to the topic, see Siniša Malešević, The Sociology of War and Violence (Cambridge: Cambridge University Press, 2010), 17–​88.

2 The workshop was organized at the University of Vienna in June 2018 in the context of a research project on civil wars in the Byzantine Empire, funded by the Austrian Science Fund (FWF P 30089-​G25: Contested Empire: Civil War in the Medieval East-​Roman Imperial State, c. 500–​1204; Department of Byzantine and Modern Greek Studies, University of Vienna). I am indebted to Philippe Buc, Clare Downham, Francesca Fiaschetti, Sergio La Porta, Gregory Lippiatt, Siniša Malešević, Panos Sophoulis, and Peter Webb for their contributions both to the workshop and the volume. I am also grateful to Shao-​yun Yang who accepted my invitation to step in at a later stage and write a chapter on Song China, which enriched our comparative approach considerably. Last but not least, I would like to thank those that contributed to the workshop with their participation and ideas but for various reasons could not take part in the making of the volume: Ilya Afanasyef (also for his important contribution to the organization of the workshop), Clemens Gantner, John Haldon, Elena Kalmykova, Nick Matheou, and Naomi van Stapele.

2

2

Yannis Stouraitis

it became increasingly apparent to me that historical sociologists often avoid serious engagement with the debates of medieval specialists on identities.3 Historians, on the other hand, can be prone sometimes to base their methodology on a single sociological approach, usually the one that has established itself as the current orthodoxy in the field. The result of that is, often, books or articles based on evidence from the sources that is cherry-​picked to suit and confirm the theory. Both these strategies testify to a lack of proper communication and meaningful dialogue between the two fields, with notable exceptions of course; a dialogue which would help both sides refine their arguments and address evident problems which they are often allowed to overlook in the safe environment of their respective field bubbles. In this regard, inviting a sociologist to participate in a project by historians of the Middle Ages was a conscious attempt to achieve a minimum level of direct dialogue between the two disciplines. I was particularly glad that Siniša Malešević accepted the challenge to contribute with a chapter to this interdisciplinary dialogue because his work straddles the topics of collective identity and warfare from the longue durée perspective of historical sociology, thus making him a rather ideal discussion partner for historians working on premodern cultures.4 More importantly, Malešević’s work has provided the most sustained criticism of the ethno-​symbolist approach which is currently enjoying the position of a new theoretical orthodoxy in the humanities. The founding father of the theory of ethno-​symbolism, Anthony D. Smith, began back in the early 1980s with a rehabilitation of ethnicity as a sociopolitical factor in the premodern world and used that as a point of departure to argue about the role of persistent premodern ethnic cultures in providing cultural cores for the emergence of modern nations.5 In this context, Smith advertised ethno-​symbolism as a critical approach to modernism, which posits that in some cases the nation as an ethnic group predates the modern phenomenon of nationalism which has transformed premodern ethnies into nations and nation-​states. While Smith was very cautious to distinguish modern nationhood from premodern ethnic identity in the largest part of his work, in his latest phase he endorsed the work of Azar Gat, who has argued that nations and 3 As John Breuilly has put it, they usually choose their own “pet medievalist” to rely on, see Atsuko Ichijo and Gordana Uzelac, eds., When is the Nation? Towards an Understanding of Theories of Nationalism (London: Routledge, 2005), 47. 4 His three major works on the topic are: Sinisa Malešević, Identity as Ideology: Understanding Ethnicity and Nationalism (Basingstoke: Palgrave Macmillan, 2006); Malešević, The Sociology of War and Violence; Malešević, Nation-​States and Nationalisms: Organization, Ideology and Solidarity (Cambridge: Cambridge University Press, 2013).

5 Anthony D. Smith, The Ethnic Origins of Nations (Oxford: Blackwell, 1986); Smith, National Identity (Reno: University of Nevada Press, 1991); Smith, Myths and Memories of the Nation (Oxford: Oxford University Press, 1999); Smith, Chosen Peoples: Sacred Sources of National Identity (Oxford: Oxford University Press, 2003); Smith, The Cultural Foundations of Nations: Hierarchy, Covenant, and Republic (Oxford: Blackwell, 2008); Smith, Ethno-​ Symbolism and Nationalism: A Cultural Approach (London: Routledge, 2009).

3



War and Peoplehood in the Middle Ages

3

nation-​states can be traced in all periods of history in those social orders where a dominant ethnic discourse converged with state formation.6 Ethno-​symbolism has been very successful in highlighting the evident problems and shortcomings of modernism, targeting with its criticism the latter’s authoritative claim of an absolute dividing line between a premodern and a modern world that made ethnicity and nationhood impossible in traditional agrarian societies. That criticism was conducive for the emergence of a new research trend, which in regard to Medieval Studies has definitely made a significant contribution to relegitimizing ethnicity as a research topic; especially against its quasi “demonization” by vulgar modernism and postmodernism toward the end of the previous century.7 Within that framework, the ethno-​symbolist approach has offered considerable insights into the relationship between war and collective identity formation, in particular ethnicity, arguing about how war has been a central factor that shaped crucial aspects of ethnic identities through mobilization, propaganda, and myth-​ making.8 Even though Smith accepts that warfare may also at times weaken a polity and endanger the survival of its dominant ethnie, he has argued that, more often than not, prolonged warfare between two or more communities has been a main reason of reinforcement of the dominant ethnicity within the state. In particular, he has stressed that, more important than mobilization and propaganda during the conflict, is the historic consciousness of ethnic identity that emerges as an outcome of prolonged warfare due to the mythologization of war deeds in epics, ballads, dramas, or hymns.9 Another representative of the ethno-​symbolist approach, John Hutchinson has recently reiterated the argument about premodern nationalism from the angle of the relationship between war and “peoplehood,” positing that warfare has been crucial for the formation of national communities since the High Middle Ages.10 When it comes to the debate about the history and perenniality of the phenomena of ethnicity and nationhood, ethno-​symbolism, besides offering a new perspective regarding the question of when the nation as a phenomenon came into being, has also—​and most importantly—​given the modernist camp a necessary push in order 6 On Gat’s sociobiological approach to perennial ethnicity, see Azar Gat (with Alexander Yakobson), Nations: The Long History and Deep Roots of Political Ethnicity and Nationalism (Cambridge: Cambridge University Press, 2013); cf. Anthony D. Smith, “Review of Azar Gat with Alexander Yakobson, Nations: The Long History of Deep Roots of Political Ethnicity and Nationhood,” Nations and Nationalism 19 (2013): 819–​34. See also John Hutchinson, Chris Wickham, Bo Stråth, and Azar Gat, “Debate on Azar Gat’s Nations: The Long History and Deep Roots of Political Ethnicity and Nationalism,” Nations and Nationalism 21 (2015): 383–​402. 7 Anthony D. Smith, “The Poverty of Anti-​Nationalist Modernism,” Nations and Nationalism 9 (2003): 357–​70.

8 Smith, “War and Ethnicity: The Role of Warfare in the Formation, Self-​images, and Cohesion of Ethnic Communities,” Ethnic and Racial Studies 4 (1981): 375–​97; Smith, Ethnic Origins, 38–​41. 9 Smith, Ethnic Origins, 38–​39.

10 John Hutchinson, Nationalism and War (Oxford: Oxford University Press, 2017), esp. chap. 1.

4

4

Yannis Stouraitis

to rethink and refine their arguments, nuance their approach, and respond with their own criticism11—​a development which historians should welcome if they want to avoid the tyranny of a theoretical orthodoxy that often fosters a kind of methodological myopia. Within this framework, Malešević’s work represents a thorough criticism of the ethno-​symbolist approach, while claiming middle ground between modernists and ethno-​ symbolists when it comes to the historical background of the phenomena that led to the emergence of national communities. While he maintains that nations and nation-​states should be seen as a distinct phenomenon of roughly the last 300 years, he considers them as the outcome of a longue durée process that long preceded that period.12 However, he confutes the ethno-​symbolist claim that at the epicentre of this longue durée process is the persistence of ethnic cores, consisting of similar perceptions and sentiments generated by, and encoded in, specific beliefs, values, and practices.13 Instead, he argues for a central role of the political and economic forces which brought about the cumulative bureaucratization of coercion and the intensification of centrifugal ideologization. The culmination of the evolution of these phenomena in the period after the seventeenth century provided the conditions for the emergence of nations and nation-​states. The concepts of “ideology” and “attachment” are central in Malešević’s critical approach to the concept of “identity” in general, and “national identity” in particular.14 Departing from a sociological approach to “ideology” free of any epistemological concerns of mystification,15 he suggests that nationhood should be understood as a form of ideological attachment to a particular vision of political community, namely the culturally and territorially delimited nation of morally equal citizens, which is generated and reproduced on a daily basis through nationalism. Anthony D. Smith attributed to nationalism the role of the ideological trigger that gave birth to modern nations, defining it as “an ideological movement for attaining and maintaining identity, unity and autonomy of a social group some of whose members deem it to constitute an actual 11 See the debate between modernists, perennialists, and ethno-​symbolists in Ichijo and Uzelac, eds., When is the Nation?

12 Malešević, Identity as Ideology; Malešević, Nation-​ States and Nationalisms; Malešević, “Nationalism and the Longue Durée,” Nations and Nationalism 24 (2018): 292–​99.

13 Sinisa Malešević, “ ‘Divine Ethnies’ and ‘Sacred Nations’: Anthony D. Smith and the Neo-​ Durkhemian Theory of Nationalism,” Nationalism and Ethnic Politics 10 (2004): 561–​93; Malešević, Identity as Ideology, 110–​35. 14 Malešević, Identity as Ideology, 13–​80; cf. Rogers Brubaker and Frank Cooper, “Beyond ‘Identity’,” Theory and Society 29 (2000): 1–​47. For an extensive overview of the history of identity as a concept and a deconstruction of its uses, see Lutz Niethammer, Kollektive Identität: Heimliche Quellen einer unheimlichen Konjunktur (Reinbek: Rowohlt, 2000).

15 Sinisa Malešević and Iain McKenzie, eds., Ideology after Poststructuralism (London: Pluto Press, 2002), especially the chapter by Malešević on “Rehabilitating Ideology after Poststructuralism.” One of the most useful critical overviews of the main epistemological and sociological strands of thought on ideology research remains Terry Eagleton, Ideology: An Introduction (London: Verso, 1991).

5



War and Peoplehood in the Middle Ages

5

or potential nation.”16 Malešević goes beyond such definitions and views nationalism as the dominant operative ideology of modern societies and, therefore, as the main determinant of nationhood. Nationalism shapes and reproduces nationhood by being systematically sustained and disseminated through mass media, the educational system, the public sphere, government decrees and proclamations, and civil society actions. This is what makes it equally important not only for the emergence but also for the cohesion and perpetuity of national communities and nation-​states.17 In this context, Malešević adopts a critical stance toward the ethno-​symbolist approach to war as a catalyst of ethnic formation, that is, as one of the main historical generators of ethnicity and nationhood.18 He acknowledges that warfare plays a crucial role in the creation and destruction of states and other organizational structures, and that war commemorations can help reinforce group attachments. However, he argues that group attachments that are enhanced through the memory of war cannot be taken a priori to be of ethnic or national nature. Warfare does not necessarily and mechanically trigger and disseminate an attachment to the nation or the ethnic group as distinct visions of community, especially when a social order lacks the interest and/​or the organizational capacity to generate and disseminate nationalism or ethnocentrism as its dominant operative ideology. Malešević’s criticism of ethno-​symbolism offers new insights and poses important questions. By highlighting the central role of centrifugal ideologization and organizational capacity in the emergence of nations and nationhood, instead of longstanding shared cultural markers and myths, he confutes the ethno-​symbolist claim that we can detect different forms of the nation and nationhood in premodern and modern times, respectively. For Malešević, nations and nationhood cannot come into being without robust organizational mechanisms in place, which foster, enforce, preserve, and popularize the idea that the nation, as a culturally and/​or territorially delimited group of morally equal citizens, is the principal unit of attachment and political loyalty, as well as the principal source of political legitimacy.19 Given that ethno-​ symbolists also accept that what they view as premodern nations lacked many of the elements that characterize modern nations,20 it becomes evident that a fruitful way to move forward by gaining insights from both positions is to accept that the problem refers to how we make use of terms as categories of analysis;21 that is, to which extent we 16 Smith, Myths and Memories of the Nation, 18–​19.

17 Malešević, Identity as Ideology, 83–​108; cf. Michael Billig, Banal Nationalism (London: Sage, 1995).

18 Malešević, The Sociology of War and Violence, 65–​70.

19 Malešević, “Nationalism and the Longue Durée,” 293; Malešević, Nation-​States and Nationalisms, 8–​13, 20–​43. 20 See, for instance, John Hutchinson, “Bringing the Study of Warfare into Theories of Nationalism,” Nations and Nationalism 24 (2017): 1–​16.

21 On the need to distinguish categories of analysis from similar categories of practice in the study of collective identities, see Brubaker and Cooper, “Beyond ‘Identity’,” 4–​6.

6

6

Yannis Stouraitis

use concepts such as nation, ethnie, nation-​state, city-​state, kingdom, and empire with the intention to distinguish heuristically between qualitatively different phenomena. That said, the part of Malešević’s work that touches upon the question of collective identities in medieval times would have profited a lot had he engaged critically with the arguments of medievalists such as Susan Reynolds, who has claimed that, at least in parts of high-​medieval Europe national communities had taken shape whose members perceived them as natural, objective communities of descent, possessing distinctive cultural attributes and with rights to be self-​governing.22 Equally interesting and fruitful would have been an engagement with the debate among medieval historians regarding the role of the disintegration of the Western Roman Empire in giving birth to new kingdoms shaped around dominant ethnic discourses in medieval Europe.23 In particular, Walter Pohl and his research collaborators in Vienna have produced a great number of important studies on collective identities in medieval Europe and 22 Susan Reynolds, Kingdoms and Communities in Western Europe 900–​1300 (Oxford: Oxford University Press 1997, rev. ed.), especially chap. 8; cf. Reynolds, “Nations, Tribes, Peoples, and States,” Medieval Worlds 2 (2015): 79–​88.

23 For the various approaches to collective identities in the post-​Roman western European kingdoms and the development of the historians’ debate, see Patrick Geary, “Ethnicity as a Situational Construct in the Early Middle Ages,” Mitteilungen der anthropologischen Gesellschaft in Wien 113 (1983): 15–​26; Geary, The Myth of Nations: The Medieval Origins of Europe (Princeton: Princeton University Press, 2002); Andrew Gillett, ed., On Barbarian Identity: Critical Approaches to Ethnogenesis Theory (Turnhout: Brepols, 2002); Gillett, “Ethnogenesis: A Contested Model of Early Medieval Europe,” History Compass 4 (2006): 241–​60; Walter Goffart, Barbarians and Romans, A.D. 418–​584: The Techniques of Accommodation (Princeton: Princeton University Press, 1980); Goffart, Barbarian Tides: The Migration Age and the Later Roman Empire (Philadelphia: University of Pennsylvania Press, 2006); Goffart, “Rome’s Final Conquest: The Barbarians,” History Compass 6 (2008): 855–​83; Guy Halsall, Barbarian Migrations and the Roman West, 376–​568 (Cambridge: Cambridge University Press, 2008); Peter Heather, “Disappearing and Reappearing Tribes,” in Strategies of Distinction: The Construction of Ethnic Communities, 300–​800, ed. Walter Pohl and Helmut Reimitz (Leiden: Brill, 1998), 95–​111; Heather, “Ethnicity, Group Identity, and Social Status in the Migration Period,” in Franks, Northmen, and Slavs, ed. Ildhar H. Garipzanov, Patrick J. Geary, and Przemysław Urbańczyk (Turnhout: Brepols, 2008), 17–​49; Heather, Empires and Barbarians: The Fall of Rome and the Birth of Europe (Oxford: Oxford University Press, 2010); Ian Wood, “Barbarians, Historians, and the Construction of National Identities,” Journal of Late Antiquity 1 (2008): 61–​81; Walter Pohl, “Conceptions of Ethnicity in Early Medieval Studies,” in Debating the Middle Ages, ed. Lester K. Little and Barbara H. Rosenwein (Oxford: Blackwell, 1998), 15–​24; Pohl, “Ethnicity, Theory, and Tradition: A Response,” in On Barbarian Identity: Critical Approaches to Ethnogenesis Theory, ed. Andrew Gillett (Turnhout: Brepols, 2002), 221–​40; Pohl, “Telling the Difference: Signs of Ethnic Identity,” in From Roman Provinces to Medieval Kingdoms, ed. Thomas F. X. Noble (New York: Routledge, 2006), 120–​67; Pohl, “Introduction—​Strategies of Identification: A Methodological Profile,” in Strategies of Identification: Ethnicity and Religion in Early Medieval Europe, ed. Walter Pohl and Gerda Heydemann (Turnhout: Brepols, 2013), 1–​64; Pohl, “Von der Ethnogenese zur Identitätsforschung,” in Neue Wege der Frühmittelalterforschung: Bilanz und Perspektiven, ed. Walter Pohl, Martin Diesenberger, and Bernhard Zeller (Vienna, 2018), 9–​34; Helmut Reimitz, History, Frankish Identity, and the Framing of Western Ethnicity, 550–​ 850 (Cambridge: Cambridge University Press, 2015).

7



War and Peoplehood in the Middle Ages

7

beyond,24 departing from an understanding of “identity” as relational and categorical modes of identification that can be ethnic, religious, political, regional, linguistic, and so on.25 Especially with regard to ethnicity, the methodology of the contemporary “Vienna school”—​from which I have drawn extensively for my own research—​has moved away from rigid definitions based on lists of criteria, which in Pohl’s own words simplify and de-​historicize matters.26 Other good examples of applying similar methodologies to the research of collective identity in a non-​western-​European context are Florin Curta’s work on how Slavic ethnicity came into being or Gregory Areshian’s recent take on Armenian identity through time.27 Such approaches work against essentializing and reifying approaches to ethnicity, as well as against hard notions of “identity,” thus paving the ground for a conceptual dialogue with Malešević’s theoretical deconstruction of the heuristic value of identity as a concept. Moreover, a closer look at developments beyond medieval western Europe shows that the role of ethnicity as an ideological factor for the cohesion and solidarity of premodern political orders needs to be addressed critically.28 The study of—​by the standards of the 24 See, indicatively, Walter Pohl and Helmut Reimitz, eds., Strategies of Distinction: The Construction of the Ethnic Communities, 300–​800 (Leiden: Brill, 1998); Walter Pohl and Mathias Mehofer, eds., Archaeology of Identity—​Archäologie der Identität (Vienna: ÖAW Verlag, 2010); Walter Pohl and Gerda Heydemann, eds., Post-​Roman Transitions. Christian and Barbarian Identities in the Early Medieval West (Turnhout: Brepols, 2013); Walter Pohl and Gerda Heydemann, eds., Strategies of Identification: Ethnicity and Religion in Early Medieval Europe (Turnhout: Brepols, 2013). For more of the group’s recent publications see also below.

25 For a recent discussion of the usefulness of such a methodological approach to historical research on collective identities, see John Haldon and Yannis Stouraitis, “The Ideology of Identities and the Identity of Ideologies,” in Identities and Ideologies in the Medieval East Roman World, ed. Yannis Stouraitis (Edinburgh: Edinburgh University Press, 2022), 1–​16. 26 Pohl, “Introduction—​Strategies of Identification,” 6.

27 Florin Curta, The Making of the Slavs: History and Archaeology of the Lower Danube Region, c. 500–​ 700 (Cambridge: Cambridge University Press, 2001); Gregory Areshian, “Historical Dynamics of the Endogenous Armenian, i.e. Hayots, Identity: Some General Observations,” in Reflections of Armenian Identity in History and Historiography, ed. Houri Berberian and Touraj Daryaee (Irvine: UCI Jordan Center for Persian Studies, 2018), 15–​35.

28 There are a number of recent publications with a strong comparative aspect that bring cultures beyond western Europe into perspective: Ildhar H. Garipzanov, Patrick J. Geary, and Przemysław Urbańczyk, eds, Franks, Northmen, and Slavs (Turnhout: Brepols, 2008); Walter Pohl, Clemens Gantner, and Richard Payne, eds, Visions of Community in the Post-​Roman World: The West, Byzantium and the Islamic world, 300–​1100 (Farnham: Ashgate, 2012); Eirik Hovden, Christina Lutter, and Walter Pohl, eds, Meanings of Community across Medieval Eurasia: Comparative Approaches (Leiden: Brill, 2016); Walter Pohl, Clemens Gantner, and Richard Payne, eds, Transformations of Romanness in the Early Middle Ages: Early Medieval Regions and Identities (Berlin: De Gruyter, 2018); Walter Pohl and Daniel Mahoney, eds, Historiography and Identity IV: Writing History across Medieval Eurasia (Turnhout: Brepols, 2021); Walter Pohl and Rutger Kramer, eds, Empires and Communities in the Post-​Roman and Islamic World, c. 400–​1000 (Oxford: Oxford University Press, 2021). See also the papers of the thematic cluster “Revisiting Pre-​modern Ethnicity and Nationhood,” in Medieval Worlds 5 (2017): 54–​169, put together by Ilya Afanasyef and Nicholas Matheou.

8

8

Yannis Stouraitis

time—​highly centralized imperial states such as Byzantium or the Chinese Empire can profit a lot from the concept of ideologization and the exploration of multiple modes of collective identification, of which ethnicity was only one and not the most important for reproducing ideological attachment to the vision of a unitary political community. As I have tried to show in my own studies on collective identity in the Byzantine Empire, the formation of a dominant Roman ethnic category within that empire, which gradually took shape after the fall of western Rome, can be explained as a product of the unbroken continuation of a geopolitically, administratively, and culturally transformed Roman imperial state under the centralized rule of the emperor in Constantinople. However, it can hardly be viewed as the most important mode of collective identification or, for that matter, as the ideological factor that kept that state together up to 1204. It was rather the dominant operative ideology of divinely ordained Roman imperial monocracy which prioritized religious over ethnic affiliation in order to promote loyalty to centralized imperial rule, thus underpinning the unity of the East Roman political order up to the late-​twelfth century, especially among the members of the ruling elite as the main agents of groupness. The regression of that ideology in the aftermath of the Crusader sack of Constantinople in 1204 sealed the fragmentation of the East Roman political community in the Late Middle Ages.29 Post-​1204 Byzantine historiography testifies to a perception of the events of the Fourth Crusade as a collective trauma and a major turning point in the history of the so-​ called Byzantines.30 If we look at that from an ethno-​symbolist sociological viewpoint, the memory of catastrophe caused to the community by war should be considered as conducive for endowing the ethnic vision of the Greek-​speaking Romans with a sense of meaning and unique destiny.31 If this were so, however, one cannot help noticing that the memory of 1204 hardly underpinned the feeling of belonging together among the Greek-​speaking Roman populations. On the contrary, the regional ruling elites and their subject populations that are considered to have shared memories of a common past alongside common cultural markers were keen to fight against each other in the name of their regional political autonomy rather than reuniting in the name and for the sake of a common ethnic destiny. Thus, the Byzantine case study invites us to revisit critically the notion of an ethnic group in the absence of groupness.32 The Greek-​ speaking Christian 29 Yannis Stouraitis, “Reinventing Roman Ethnicity in High and Late Medieval Byzantium,” Medieval Worlds 5 (2017): 70–​94.

30 On the reception of the events of 1204, see the chapters by Ruth Macrides, Chrya Maltezou, and Dimiter Angelov in the collected volume Angeliki Laiou, ed., Urbs Capta: The Fourth Crusade and its Consequences (Paris: Éditions Lethielleux, 2005). On Roman ethnic discourse in the period after 1204, see Gill Page, Being Byzantine: Greek Identity before the Ottomans, 1204–​1420 (Cambridge: Cambridge University Press, 2008). 31 Hutchinson, “Bringing the Study of Warfare into Theories of Nationalism,” 10–​11.

32 Rogers Brubaker, “Ethnicity without Groups,” European Journal of Sociology 43 (2002): 163–​89, especially at 167–​70.

9



War and Peoplehood in the Middle Ages

9

populations of diverse regions may have used the self-​ designation Rhōmaios (Roman) in order to culturally demarcate themselves from the Latin conquerors in the Balkans or the Turkish conquerors in Asia Minor, but they did not employ it as a means of identification with each other, that is, a sign of attachment to the vision of a unitary Roman community. This means that, even though the impact of war may have sharpened the boundaries of cultural demarcation between populations that were circumscribed and categorized through different shared cultural markers (Greek-​speaking Romans, Bulgarians, Serbs, Franks, Turks) in the core territories of the disintegrated Byzantine Empire, these shared cultural markers did not enhance groupness and the need to be loyal to one and the same community in the case of the Greek-​speaking Roman populations. The questions and problems raised by my research on the Byzantine case study formed the background against which the idea of putting together a cross-​cultural study on war and peoplehood in the Middle Ages took shape. The current study aspires to complement recent research which has addressed the role of different factors such as language,33 religion,34 historiography,35 and ethnic terminology36 in the configuration of collective identity discourses in the medieval period from a cross-​ cultural perspective. The individual chapters explore the relationship between war and collective identification in various medieval societies, posing some basic questions as a point of departure: What was the role of warfare in generating, enhancing, or weakening collective identifications in various medieval cultures at different times? What type of identifications were those? Moreover, what similarities or differences can we trace in the role of war in group formation between western European cultures and those of the eastern Mediterranean and beyond? It goes without saying that the current publication does not claim to offer a wholistic coverage of the topic but rather an introduction that will problematize it, hopefully leading to more focused research on the issues and questions that will be raised. In light of this, Chapter 2 has been ceded to a sociologist. It offers a comprehensive overview of the relationship between the phenomena of war and the formation of 33 Walter Pohl and Bernhard Zeller, eds., Sprache und Identität im frühen Mittelalter (Vienna: ÖAW Verlag, 2012).

34 Bas ter Haar Romeny, ed., Religious Origins of Nations? The Christian Communities of the Middle East (Leiden: Brill, 2010); Walter Pohl, Ingrid Hartl, and Gerda Heydemann, eds., Ethnicity and Religion in the Middle Ages (Leiden: Brill, forthcoming).

35 Pohl and Mahoney, eds, Historiography and Identity IV; Pohl and Kramer, eds., Empires and Communities in the Post-​Roman and Islamic World.

36 Walter Pohl, Cinzia Grifoni, and Sophie Gruber, eds., Ethnic Terminologies in the Early Middle Ages (Leiden: Brill, forthcoming).

10

10

Yannis Stouraitis

high-​level collective identities from the longue durée perspective of historical sociology. In the first part of his chapter, Siniša Malešević presents the classical sociological theses regarding the relationship between war and group formation. Beginning with a short overview of older biologically determinist perspectives that saw in the biological and cultural difference of groups a main cause for the emergence of wars, he moves on to the criticisms of the sociobiological and “hard culturalist” approach, which in contemporary historical sociology have gone beyond the question of the causal relationship between war and collective identity, seeking to analyse and understand when and how war, ethnicity, and nationhood converged through time.37 Malešević highlights the differences between the modernist and the ethno-​symbolist approaches and presents his own criticisms of the shortcomings of both positions in the second part of his chapter. His thesis is that the longue durée historical relationship between war and collective identity building has been determined by three long-​term historical processes: the cumulative bureaucratization of coercion, centrifugal ideologization, and the envelopment of micro-​solidarity. The sociological overview of the longue durée relationship between war and peoplehood, besides offering an important insight into the different methodological approach of historical sociology, sets the background against which the historical chapters of the volume should be read. These chapters have been arranged chronologically, the presented case studies spanning the period between the sixth and the fifteenth centuries, and ranging geographically from western and northern Europe to the Near East and East Asia. There are two chapters dealing with identifications in the context of civil wars, three chapters focusing on identity formation in the context of interstate conflicts, and four chapters whose point of departure are the identities that emerged through conquest. In Chapter 3, Peter Webb provides an in-​depth analysis of the strategies of memorialization of the pre-​Islamic battle of Dhū Qār, focusing on the renegotiation of the identities of the battle’s opponents in Islamic sources. The battle took place on Iraq’s Euphrates frontier around the year 610 CE as a result of the Sasanian king’s aim to stop Arabian nomadic incursions. The fighting parties consisted of a combined force of Sasanian heavily armed cataphract cavalrymen and allied Arabian groups on the one side, and several lineage groups of the Arabian macro-​tribal unit Bakr ibn Wāʾil on the other side.38 Webb undertakes a meticulous study of how the representation of the identities of the two sides in the battle developed in Islamic historiographical sources that emerged from late eighth century onwards, and in Arabic poems composed from 37 Charles Tilly, “War Making and State Making as Organized Crime,” in Bringing the State Back In, ed. Peter B. Evans, Dietrich Rueschemeyer, and Theda Skocpol (Cambridge: Cambridge University Press, 1992), 169–​91; Michael Mann, The Sources of Social Power II: The Rise of Classes and Nation-​ States, 1760–​1914 (Cambridge: Cambridge University Press, 1993); Andreas Wimmer, Waves of War: Nationalism, State Formation, and Ethnic Exclusion in the Modern World (Cambridge: Cambridge University Press, 2012); Hutchinson, Nationalism and War. 38 On the battle, see Ella Landau-​Tesseron, “Ḏū Qār,” Encyclopaedia Iranica 7 (1996): 574–​75.

11



War and Peoplehood in the Middle Ages

11

the time of the battle until the ninth century. His point of departure for comparing these accounts is the established image of the battle in Abbasid historiography in which it was depicted as an ethnic conflict between the Arabs and the Persians. His analysis of Arabic poetry written very close to the event deconstructs the image of two homogenous ethnic groups fighting each other. Chapter 4, written by Sergio La Porta, deals with war and ethnicity from the viewpoint of mnemohistory in a neighbouring culture. La Porta explores the reconfiguration of Armenian identity discourse through the lens of a story about deported prisoners of war, first reported in the mid-​seventh-​century History of Armenia attributed to Sebēos. The latter’s narrative of the captives’ re-​Armenization, as a product of the new geopolitical context of the second half of the seventh century, sets the background against which the retelling of the story by Armenian authors in subsequent centuries is explored. Yovhannēs V Drasxanakertc‘i in the tenth century, Step‘anos Tarōnec‘i in the eleventh century, the priests of the Ani Cathedral, Samuēl and Mxit‘ar Anec‘i, in the twelfth century, and Vardan Arewelc‘i and Mxit‘ar Ayrivanec‘i, all provided different versions of the same story, each one stressing different visions of Armenianness intended to serve current political and cultural needs, as well as the authors’ personal ideologies of identity. In Chapter 5, Panos Sophoulis traces the formation of a Bulgar identity in roughly the first two centuries after the conquest of parts of the northern Balkans by a group of nomadic warriors of Türkic-​Oğuric origin. The first Bulgar state emerged in the late seventh century in the northeastern Balkans, close to the Danube Delta, as a result of the defeat of the armies of the East Roman Emperor Constantine IV that forced him to accept a peace treaty in 680.39 This was an entity dominated by a Türkic-​Oğuric elite that ruled over a majority of Slavic groups that had also gradually settled in these areas during the seventh century, and the remaining indigenous Christian populations. Sophoulis sees the political and social development of that entity between the late seventh and the early ninth century as a result of high levels of militarization. The latter was the outcome of the awareness of the threat posed to the new polity by its neighbouring rival, the Byzantine Empire, which never abandoned the aim of reconquest. Chapter 6, written by Yannis Stouraitis, explores identity discourses in the context of Byzantine civil wars. The first part of the chapter zooms in on the terminology of internal armed conflict in Byzantine sources, which testifies to the image of community that set the framework of internal armed conflict, namely the imperial-​political community circumscribed by the limits of enforceable imperial authority. The conceptualization of eastern Roman civil war points to the apolitical nature of ethno-​cultural categorizations of imperial subjects within the medieval empire.40 In the second part of the chapter, the identity discourse of the accounts of civil wars is analyzed in order to scrutinize the 39 On the migration and settlement of the Bulgars, see Daniel Ziemann, Vom Wandervolk zur Großmacht: Die Entstehung Bulgariens im frühen Mittelalter (7. bis 9. Jh.) (Cologne: Böhlau, 2007); Jonathan Shepard, “Slavs and Bulgars,” in The New Cambridge Medieval History, II: c. 700–​c. 900, ed. Rosamond McKitterick (Cambridge: Cambridge University Press, 1995), 228–​48. 40 On different approaches to the relationship between ethnicity and political community in Byzantium, see Stouraitis, “Reinventing Roman Ethnicity”; Anthony Kaldellis, Romanland: Ethnicity and Empire in Byzantium (Cambridge, MA: The Belknap Press of Harvard University Press, 2019).

12

12

Yannis Stouraitis

role of a shared Christian faith in providing a shared religious-​cultural background that underpinned political attachment well into the High Middle Ages. The developments of the period after 1204, when the Romans are attested as a politically fragmented ethnic category in the written sources, demonstrate that the image of a Roman ethnic community was not perceived as the framework for civil war by contemporaries. In Chapter 7, Clare Downham focuses on the vikings and the English in the transition from the Early to the High Middle Ages. She begins with the problems surrounding the controversial term “viking” as a definition for a militarized trading diaspora, not exclusively Scandinavian, and she argues that the regional polities that emerged in England under viking rulers had an integrative character. In this context, she points to the established notion of a shared English identity since the eighth century, but cautions that the concept of Englishness remained open enough to accommodate those that were willing to submit or cooperate irrespective of ethnic background.41 Her central thesis maintains that medieval English identity was not the result of a war between two cultures, that of Alfred the Great and his descendants and that of the Scandinavian settlers. It was rather an outcome of integration and negotiation. Chapter 8, written by Philippe Buc, takes us back to the role of civil war in shaping attachments. Buc undertakes a scrutiny of three case-​studies of civil war in late-​ medieval western Europe: the Albigensian Crusade (1208/​9–​1229), the Franco-​French civil war between the Armagnacs and the Bourguignons (1407–​1435), and the Crusades of Emperor Sigismund against the dissident Bohemian Hussites who resisted papal Catholicism (1419–​1436). The thread that connects the three cases is religion. Buc takes his lead from current research that has stressed the role of religion in the formation of collective identities in the Middle Ages in order to explore how civil war contributed to shaping or accentuating different modes of collective identification.42 From the three civil wars, the role of religious identification in configuring the opposite parties in the internal armed conflict is far less explicit in the Armagnac–​Bourguignon conflict, whereas the conceptualization of the other two cases as Crusades highlights religion’s central role in setting the political framework. Chapter 9 explores the development of French identity during the High Middle Ages into a diaspora identity in the areas occupied by the Crusaders across the Mediterranean. Gregory Lippiatt’s point of departure is the historical background of the relationship between Frenchness and Crusading that attributed to the French a distinct role in the 41 On the debated issue of English nationhood and its medieval or early modern beginnings, either in the eighth, fourteenth, or sixteenth century, see: Adrian Hastings, The Construction of Nationhood (Cambridge: Cambridge University Press, 1997), 35; Geoffrey Elton, The English (Oxford: Blackwell 1992), 69–​70; Liah Greenfeld, Nationalism: Five Roads to Modernity (Cambridge, MA: Harvard University Press, 1992), 14. For a refutation of these arguments see Krishan Kumar, The Making of English National Identity (Cambridge: Cambridge University Press, 2003), esp. chap. 3 and 5.

42 Pohl and Heydemann, eds., Strategies of Identification: Ethnicity and Religion in Early Medieval Europe.

13



War and Peoplehood in the Middle Ages

13

defence of Christendom already before the Crusades. “Frankishness” or French identity had become inclusive categories in the aftermath of the First Crusade, while linguistic Frenchness was the main cultural marker that connected the ruling elites of various polities that emerged all over the Mediterranean between the late eleventh and the thirteenth centuries.43 These polities formed part of a broader Francophone community that was culturally linked with France through language but also through various notions of a shared heritage pertaining to common attitudes toward marriage and inheritance. In Chapter 10, Shao-​yun Yang scrutinizes the impact of the Song–​Jurchen conflict (1125–​1234) on Song Chinese identity.44 His argument runs against modern approaches that see the claims of leading Southern Song thinkers and literati for a reconquest of north China as evidence of a Song Chinese elite nationalism that was nurtured by the notion that the Song should unite all Chinese people under their single rule.45 According to Yang, the irredentist views of Song literati about the north were shaped around the notions of “filial revanchism” and Chinese imperial supremacism. He traces how the concept related to and was influenced by classical Confucian and Neo-​Confucian thought. The latter entertained a notion of ethnocentric moralism according to which becoming immoral turned someone into a barbarian, thus compromising the demarcation of the Chinese from the barbarians in purely ethnic terms. Chapter 11 deals with the spread of Mongol identity through conquest. Francesca Fiaschetti argues that Mongol expansion in the Inner Asian and Chinese geographical, political, and military context in the thirteenth century brought about the renegotiation of both the identity of the conquerors and that of the subjugated populations.46 “Mongolization” at the initial stage of expansion referred to a process of acquiring full-​ blown political rights and privileges, dependent on participation in military activity on behalf of the Mongol imperial rule. Moreover, the Mongol conquest had a profound impact on the renegotiation of loyalties and collective attachments in South China and Southeast 43 Peter Lock, The Franks in the Aegean, 1204–​1500 (London: Routledge, 1995); Jean Dunbabin, The French in the Kingdom of Sicily, 1266–​1305 (Cambridge: Cambridge University Press, 2011); Malcolm Barber, The Crusader States (New Haven: Yale University Press, 2012).

44 On the conflict see Peter Lorge, War, Politics and Society in Early Modern China, 900–​1795 (London: Routledge, 2005), 50–​57; Jing-​shen Tao, “The Move to the South and the Reign of Kao-​ tsung,” in The Cambridge History of China: Volume 5, The Sung Dynasty and Its Precursors, 907–​1279, ed. Paul J. Smith and Dennis C. Twitchett (Cambridge: Cambridge University Press, 2009), 556–​643.

45 On the nationalist–​irredentist approach, see Nicholas Tackett, The Origins of the Chinese Nation: Song China and the Forging of an East Asian World Order (Cambridge: Cambridge University Press, 2017).

46 On Mongol identity, see Pavel Rykin, “The Social Group and Its Designation in Middle Mongolian: The Concepts Irgen and Oboq,” trans. Sarah Turner, Antropologicheskii Forum /​Forum for Anthropology and Culture 1 (2004): 183–​21; Igor de Rachewiltz, “The Genesis of the Name ‘Yeke Mongγol Ulus’,” East Asian History 31 (2006): 53–​56; Lhamsuren Munkh-​Erdene, “Where Did the Mongol Empire Come From? Medieval Mongol Ideas of People, State and Empire,” Inner Asia 13 (2011): 211–​37; Michal Biran, “The Mongols and Nomadic Identity: The Case of the Kitans of China,” in Eurasian Nomads as Agents of Cultural Change: The Mongols and Their Eurasian Predecessors, ed. Reuven Amitai, Michal Biran, and Anand A. Yang (Honolulu: Hawaii University Press, 2015), 152–​81.

14

14

Yannis Stouraitis

Asia, since the subjugation of the Southern Song led to legitimization of the Yuan empire and the integration of Han and non-​Han groups that participated in the process. Chapter 12 provides a comparative summary of the main conclusions of each of the individual chapters, seeking to highlight similarities and differences regarding the impact of military conflicts and/​or the memorialization of war on the formation or renegotiation of collective identities. The insights offered by the authors go beyond the narrow horizon of the debate about premodern ethnicity’s relationship to modern nationhood, seeking to demonstrate how and when war converged with various high-​ level identifications (ethnic, religious, regnal, imperial, regional) in different cultures at different times in the era we schematically call the Middle Ages.

15

Chapter 2

WAR AND PEOPLEHOOD THROUGH TIME: A SOCIOLOGICAL LONGUE DURÉE PERSPECTIVE Siniša Malešević Warfare has been

one of the most significant forces to shape social relations throughout history. Although scholars still disagree about how old the institution of war is, there is a consensus that once it emerged on the historical scene warfare radically transformed the patterns of social stratification, political institutions and polity formation, economic relations and forms of production, religious structures, gender divisions, and citizenship rights. Wars have also framed the dynamics of collective attachments. In this chapter, I briefly explore the relationship between war and peoplehood through time. The chapter focuses on the debates within historical sociology, exploring the following questions: How have wars shaped collective identities over long periods of time? When do cultural attachments attain political meaning and what role do violent conflicts play in this process? What are the key mechanisms that make the relationship between war and peoplehood possible and sociologically meaningful? The first part of the chapter offers a critical review of the debate on the relationship between war and nationalism in modern and premodern contexts. I identify the strengths and weaknesses of the two leading perspectives (the modernist and the ethno-​symbolist) and then make a case for an alternative approach. In the second part of the chapter, I provide the contours of the sociological longue durée perspective which attempts to move beyond the modernist versus ethno-​symbolist debate. In addition to articulating the key propositions of this perspective, I also illustrate its heuristic relevance by briefly exploring several examples from the Middle Ages.

War and Cultural Difference: The Classical and Contemporary Debates

One of the key questions that preoccupied late nineteenth and early twentieth-​century social science was: Are cultural differences the cause or the consequence of violent conflicts? The popularity of Darwinian and Lamarckian biology at the time deeply influenced social thought and many scholars were adamant that shared cultural and biological mores are the root cause of organized violence. Hence, the leading fin de siècle sociologists and social psychologists such as Ludwig Gumplowitz, Gustav Ratzenhofer, William McDougall, William James, and Benjamin Kidd argued that the cultural and biological differences between groups ultimately lead to conflict. The social Darwinists

16

16 Siniša Malešević Kidd and James wrote about the unavoidable “race struggle” as a product of evolution.1 As James puts it in a highly influential essay “The Moral Equivalent of War,” war is “the gory nurse that trained societies to cohesiveness.”2 A more Lamarckian-​influenced McDougall, Ratzenhofer, and Gumplowitz also invoked biological categories to explain the origins of group conflicts. For McDoughall, warfare stems from the “pugnacious instincts” that regularly find their expression in wars between culturally and biologically different groups.3 Similarly, Gumplowitz and Ratzenhofer identified ethnocentrism, a term coined by Gumplowitz, as the primary source of violent conflicts. Nevertheless, these biologically determinist perspectives had already been challenged by their contemporaries such as Georg Simmel, William Graham Sumner, Max Weber, Marcel Mauss, and later also by Lewis Coser and others. Both Simmel and Sumner reversed the biological interpretations, arguing that rather than causing violence cultural difference is often a direct consequence of protracted conflict. In Simmel’s view, external conflicts regularly enhance in-​group solidarity as they sharpen the social boundaries between groups and intensify polarity between the two foes.4 In this process the group structures also become more centralized. Sumner makes a similar point: in the “competition for life,” it is “the exigencies of war with outsiders … what makes peace inside.”5 Coser later developed this idea further and attempted to show how conflicts help solidify groups that are loosely structured.6 The protracted wars are likely to generate strong in-​group cohesion and out-​group animosity. Furthermore, Coser argues that intra-​group conflicts can also bring about social change as they can mobilize isolated individuals to join specific groups. Weber and Mauss have also questioned the biological and hard culturalist interpretations by emphasizing the dynamic quality of collective violence. For Weber, collective violence is not an innate, biological, propensity but something that entails discipline, division of labour, and a coordinated social action.7 Hence instead of being an automatic and natural response, war requires a great deal of organizational work. It is through the rationalization of social action that warfare becomes routinized, instrumentalized, and bureaucratized, thus bringing about violent conflicts over social status. Thus, under modern conditions “every victorious war enhances the [state’s] cultural prestige.”8 Mauss also highlights the social dynamics of violent conflicts. 1 Benjamin Kidd, Social Evolution (London: Macmillan, 1893); William James, The Moral Equivalent of War (New York: Read, 2015). 2 William James, “The Moral Equivalent of War,” Lecture 11 in Memories and Studies (New York: Longman Green, 1911), 267–​96, at 272. 3 William McDougall, An Introduction to Social Psychology (London: Methuen, 1915).

4 Georg Simmel, Conflict and the Web of Group Affiliations (Glencoe: Free Press, 1955). 5 William G. Sumner, Folkways (Boston: Ginn, 1906), 12.

6 Lewis Coser, The Functions of Social Conflict (New York: Free Press, 1956). 7 Max Weber, Economy and Society (New York: Bedminster, 1968), 926. 8 Weber, Economy and Society, 926.

17



War and Peoplehood through Time

17

To avoid insecurity and distrust, individuals and groups engage in gift exchange rituals. The exchanging of gifts is less of an economic transaction and much more a mechanism for forging strong group bonds, thus avoiding war. In his own words: to trade, the first condition was to be able to lay aside the spear … Only then did people learn how to create mutual interests, giving mutual satisfaction, and, in the end, to defend them without having to resort to arms.9

These early debates have been revisited and augmented in the context of recent debates on the relationship between ethnicity, nationhood, and war. Contemporary historical sociologists have pushed this debate further by zooming in on the temporal dimension in order to identify not only the causal direction between these phenomena but also to find out when and how ethnicity, nationhood, and war converge through time. There are two leading general perspectives that have analyzed this relationship: the modernist and the ethno-​symbolist approaches. The modernist school of thought argues that since nations and nationalisms are distinctly modern phenomena, they could not be the primary cause of war. In contrast, the ethno-​symbolists, who see ethnicity and nationhood in more historically continuous terms, insist that warfare has played a crucial role in the formation of ethnic and national identities. The modernist perspective, as developed by Charles Tilly, Michael Mann, and Andreas Wimmer among others, recognizes the centrality of war in the nation-​formation processes. However, for modernists rather than being a cause of warfare strong national attachments are the long-​term consequence of violent conflicts. Both Tilly and Mann argue that national identities are the byproduct of state formation, which itself originated through the intensification of warfare. In Tilly’s highly influential account, nationalism developed gradually through war and state-​making processes.10 Zooming in on early modern Europe, he found that protracted warfare was a key catalyst of state formation as it forced rulers to develop the administrative structures, transport and communication networks, and fiscal capacity in order to increase their organizational capacity to extract taxes and other resources for war. This long-​term process, involving intensified coercion but also negotiations with their citizens (mostly bankers, merchants, and traders who provided money and loans for war), enhanced the coercive and economic powers of states that won these protracted conflicts. As Mann explains, warfare played a crucial role in the increase of fiscal capacities, capital accumulation, and legal systems.11 Ultimately, the states could obtain more resources, taxes, and conscripts for war. However, this process was often premised on the expansion of 9 Marcel Mauss, The Gift (London: Routledge, 1990), 82.

10 Charles Tilly, “War Making and State Making as Organized Crime,” in Bringing the State Back In, ed. Peter B. Evans, Dietrich Rueschemeyer, and Theda Skocpol (Cambridge: Cambridge University Press, 1992); Tilly, Coercion, Capital and European States (Oxford: Blackwell, 1992).

11 Michael Mann, The Sources of Social Power II: The Rise of Classes and Nation-​States, 1760–​1914 (Cambridge: Cambridge University Press, 1993); Mann, The Sources of Social Power III: Global Empires and Revolution, 1890–​1945 (Cambridge: Cambridge University Press, 2012).

18

18 Siniša Malešević citizenship rights, which eventually gave rise to constitutional and parliamentary orders as expanding political and social rights were the price rulers paid for more taxation and conscription. Tilly and Mann perceive the rise of nationalism in the late eighteenth and nineteenth centuries as a direct consequence of state building. For Tilly, European nationalism developed mostly in two dominant forms—​the state-​enforcing and state-​ seeking nationalism. State-​enforcing nationalism is associated with France, Spain, and Portugal and it was a top-​down project of centralization and assimilation coordinated by the state authorities. In contrast, state-​seeking nationalism comprised the bottom-​ up undertakings headed by the cultural and political elites representing the social movements of cultural minorities as exemplified in Greece, Serbia, or Ireland, among many others. Mann also emphasizes that the rise of nationalism was dependent on the interplay between the rising state power and the expansion of commercial capitalism.12 Both processes contributed toward proliferation of the infrastructural capacities of nation-​states. Wimmer extends this analysis further by demonstrating empirically that an overwhelming majority of wars that took place in the world in the last two hundred years were driven by nationalist ambitions.13 In his own words: “over three-​quarters of all full-​scale wars … were fought either by nationalists who seek to establish a separate nation-​state or over the ethnic balance of power within existing state.”14 Furthermore, modernists share the view that nationalism could only be a consequence of warfare in modernity. Since nationalism has been around for less than 300 years and wars have been fought for thousands of years, this indicates that nationalism could not play any role in premodern warfare. All modernist interpretations emphasize that the politicization of cultural difference can happen only under specific structural conditions. They differentiate sharply between the modern nation-​states and the premodern forms of polity such as empires, patrimonial kingdoms, city-​states, chiefdoms, or tribal confederacies. The modernist argument is built on the idea that there were no structural conditions for development and society-​wide expansion of nationalism before the modern era. For example, the premodern imperial orders were economically and politically highly stratified to allow for the emergence of the notion of popular sovereignty. Moreover, such polities were also culturally stratified as they deployed cultural differences to mark the status and class boundaries between the aristocracy and the commoners. Hence it is only with the development of greater infrastructural capacities, centralized authorities, advanced division of labour, developed communication and transport networks, standardization of vernaculars and expansion of discursive literacy that nation-​states replaced empires and patrimonial kingdoms and nationalism became a society-​wide model for political legitimacy. Thus, for modernists 12 Michael Mann, “A Political Theory of Nationalism and its Excesses”, in Notions of Nationalism, ed. Sukumar Periwal (Budapest: Central European University Press, 1995), 44–​64.

13 Andreas Wimmer, Waves of War: Nationalism, State Formation, and Ethnic Exclusion in the Modern World (Cambridge: Cambridge University Press, 2012). His analysis is based on the database of 464 wars waged between 1816–​2001. 14 Wimmer, Waves of War, 2–​3.

19



War and Peoplehood through Time

19

wars play a central role in the development of peoplehood but this can only happen under modern conditions and through the rise of powerful state structures. The ethno-​symbolist perspective, as articulated by Anthony D. Smith and John Hutchinson, questions this modernist account. Their critique identifies three main weaknesses of modernist theories: a) the overly tight link between nationalism and modernization; b) the top-​down explanations; and c) the too rationalist account of nation-​formation. Smith characterizes modernist accounts as suffering from “blocking presentism,” which means that their theories overemphasize more recent developments such as industrialization, secularization, or urbanization, while largely ignoring long-​ term social change.15 To rectify this, ethno-​symbolists have developed an approach which traces historical continuities between premodern ethnic groups and modern nations. For Smith, nation-​formation is a longue durée process that in most instances involves gradual transformation whereby modern-​ day nationhood is built on the existing cultural cores of premodern ethnies. Much of this continuity is rooted in the cultural domain—​shared collective myths, memories, and symbols. Smith emphasizes the cultural significance of the myth of common descent which underpins many nationalist ideologies and is just as present in the premodern accounts of ethnic origins. This potent myth makes sense of one’s collective identity as it locates unique origins in time and space and in this way provides meaning to one’s collective existence. Smith identifies several ethnic collectivities that have preserved their name and some shared symbols and memories for thousands of years including Jews, Armenians, Greeks, Persians, Chinese, and Japanese. He singles out the shared religious traditions as playing a central role in the reproduction of ethnic and national communities.16 Furthermore, although ethno-​symbolists recognize the significance of cultural and political elites in the nation-​formation processes, they are critical of the top-​ down interpretations and insist that ethnic and nationalist symbols attract popular appeal regardless of the behaviour or motivation of the elites. In other words, for Hutchinson and Smith political and cultural entrepreneurs cannot manipulate ethnic and national symbols at will. While elites can instrumentalize shared cultural markers for selfish political ends, they still have “to remain within the cultural traditions of the populations they wish to rouse.”17 This point is linked with the modernists’ overemphasis on the rationality of ethnic and nationalist identifications. The ethno-​ symbolists insist that rather than being driven by economic, political, or other more instrumental goals, ethnicity and nationhood are primarily motivated by emotional responses. Nations are not contract-​based associations of self-​interested individuals 15 Anthony D. Smith, “History and National Destiny: Responses and Clarifications,” Nations and Nationalism 10 (2004): 195–​209, at 197.

16 Anthony D. Smith, The Ethnic Origins of Nations (Oxford: Blackwell, 1986); Smith, Nationalism and Ethno-​Symbolism: A Cultural Approach (London: Routledge, 2009).

17 Anthony D. Smith, Chosen Peoples: Sacred Sources of National Identity (Oxford: Oxford University Press, 2003), 362.

20

20 Siniša Malešević but are, as Smith calls them, “sacred communions of citizens,” while nationalism represents a “political religion.”18 In this context, for ethno-​symbolists war and peoplehood have historically been deeply interwoven. In their view, this link has less to do with state formation and much more to do with the sense of moral, emotional, and cultural obligations that individuals have toward their ethnic and national collectivities. Hutchinson rejects Tilly’s interpretation and argues that rather than developing only in the modern context national communities have been shaped by, and have in turn shaped, warfare since the Middle Ages.19 In this view, war has historically contributed toward “the sacralization of national communities” as past military victories and sacrifices are regularly commemorated and remembered. These ritualistic commemorations of the “glorious dead,” often reinforced through monuments full of symbolism such as cenotaphs, mass graves, and war memorials, glorify the sacrifices of one’s ancestors and in this way set the boundaries of normative obligation for their dependents—​to keep their nation alive as their forefathers have made the ultimate sacrifice for this noble goal. Hence, unlike the modernists who see nationalism and war as being indirectly related via modern state formation, the ethno-​symbolists identify a direct and transhistorical link between war and peoplehood. In Smith’s neo-​Durkhemian interpretation, national commemorations such as Armistice Day in the UK or the Fourth of July in the United States constitute “a reflexive act of self-​worship” through which “the nation is revealed as a sacred communion of the people, a union of the prematurely dead, the living and the yet unborn, its ‘true self’ lodged in the innate virtue of the Unknown Warrior and symbolised by the empty tomb.”20 While ethno-​symbolism successfully challenges some modernist claims, it is not immune to criticism. As I have argued before, the ethno-​symbolist approach is explanatory limited in three main respects.21 First, Smith and Hutchinson overemphasize the integrative quality of myths and memories. By focusing almost exclusively on cultural variables such as shared ideas, values, and perceptions of group membership, this functionalist perspective tends to overlook the centrality of political and economic forces that make these cultural links possible. Hence rather than being a spontaneous and automatic process, the commemorations of past sacrifices in war entail the presence of organizational structures and ideological projects which are realized through such ritualistic events staged by the state elites or civil society groups. In other words, there are no collective acts of remembrance without the organizations and groups interested in creating such events. The Fourth of July parades and the Armistice Day 18 Smith, Chosen Peoples. 19 John Hutchinson, Nationalism and War (Oxford: Oxford University Press, 2017).

20 Smith, Chosen Peoples, 249.

21 Siniša Malešević, Identity as Ideology: Understanding Ethnicity and Nationalism (New York: Palgrave Macmillan, 2006); Malešević, Nation-​States and Nationalisms: Organisation, Ideology and Solidarity (Cambridge: Polity, 2013).

21



War and Peoplehood through Time

21

commemorations involve great deal of organizational work and ideological framing, and this can only happen if some groups are politically or ideologically motivated to stage such events. Furthermore, even when such agents put in a great deal of work to make this ritualism widely accepted there is no guarantee that ordinary individuals will respond to such rituals of commemoration. The rituals can be contested, or one could see attempts to organize alternative events of remembrance. A good example is Ireland, where one can find the competing commemorations of the 1916 uprising held annually by the government and Sinn Féin and the other republican groups that do not recognize the official version of events. Second, the ethno-​symbolists regularly exaggerate the sacrificial dimension of ethnic and national solidarity. Smith often invokes the notion of martyrdom for the nation as a powerful indicator of how strong attachments of individuals to their ethnies and nations are. However, years of meticulous research on the battlefield behaviour and motivations of individuals who join clandestine political organizations show that most people tend to die not for abstract ideological principles but for the micro-​level groups to which they have strong emotional and moral ties.22 Hence even when individuals loudly proclaim that they are willing to die for their motherlands and fatherlands, one must dissect such claims and not take them at face value. In addition, commemorations of previous wars are shaped by specific historical contexts and as such contexts change so does the character of a commemoration. As Brubaker and Feischmidt show, the 1848 nationalist uprisings in Hungary, Romania, and Slovakia were traditionally commemorated in a sacralized sense through the prism of national liberation struggle.23 However, once these three countries joined the European Union, the commemorations attained a “desacralized” form as 1848 was publicly remembered as a liberal pan-​European resistance against the feudal ancien régime. The more recent commemorations of the same event have again acquired a more nationalist dimension. The changing feature of war commemorations also indicates that some violent historical episodes are remembered, and others ignored, while on other occasions the “forgotten” wars can be commemorated. All of this shows that the discourse of national martyrdom often has very little to do with the actual self-​ sacrifice and much more to do with the contemporary ideological framing of the past. Finally, the ethno-​symbolists often conflate the macro with the micro-​level phenomena and in this way tend to reify and essentialize group dynamics. Smith’s understanding of nations as the modern equivalent of traditional ethnies follows in the

22 Siniša Malešević, The Rise of Organised Brutality: A Historical Sociology of Violence (Cambridge: Cambridge University Press, 2017); Malešević, Grounded Nationalisms: A Sociological Analysis (Cambridge: Cambridge University Press, 2019); Randall Collins, Violence: Micro-​ Sociological Theory (Princeton: Princeton University Press, 2008); Donatella della Porta, Clandestine Political Violence (Cambridge: Cambridge University Press, 2013). 23 Rogers Brubaker and Margit Feischmidt, “1848 in 1998: The Politics of Commemoration in Hungary, Romania, and Slovakia,” Comparative Studies in Society and History 44 (2002): 700–​44.

22

22 Siniša Malešević footsteps of Durkheim’s distinction between mechanical and organic solidarity.24 As Smith argues: a transition from Gemeinschaft to Gesellschaft finds its confirmation in the … vital sphere of ethnicity: in the modern era, ethnie must become politicised … and must begin to move towards nationhood … [and] take on some of the attributes of Gesellschaft, with its features of rational political centralisation, mass literacy, and social mobilisation.25

This evolutionary view is simply too one-​dimensional to account for the complexity and variety of collective experience. For one thing, Smith does not differentiate the interpersonal, face-​to-​face, small group dynamics from the large-​scale bureaucratic construct that is a nation-​state, which is sustained through organizational work and framed around abstract ideological principles. As Simmel and Collins show nicely, group size impacts on different patterns of solidarity as face-​to-​face interaction is very different from the mediated action that characterized nation-​states.26 This has profound consequences for how individuals perceive their group membership. Hence, instead of reifying and essentializing large-​scale entities such as nations and states and treating them as being the same as ethnies and small group associations, it is crucial to recognize that any macro-​level cohesion requires intense organizational and ideological work. To sum up the two dominant perspectives in nationalism studies, modernism and ethno-​symbolism provide some useful guides for understanding the relationship between war and peoplehood through time. However, they both also exhibit several explanatory weaknesses. Thus, to address these weaknesses it is necessary to develop a different perspective which could help us understand better the war and peoplehood dynamics in the Middle Ages.

War and Peoplehood: The Sociological Longue Durée

To understand the changing dynamics of war and peoplehood through time, it is necessary to articulate a theoretical framework which clearly differentiates between the organizational and ideological structures in the modern and premodern world, yet which also recognizes elements of continuity that are present throughout history. In other words, one can reconcile the modernist argument that there is no nationalism before modernity while also insisting that nationalism developed gradually out of the premodern interactions between warfare and peoplehood. In a similar vein, one can accept the ethno-​symbolist view that it is necessary to track historical continuities in the formation of large collectivities while arguing that such continuities are rarely cultural but primarily shaped by the structural dynamics of organized violence. Hence, the sociological longue durée view is built around the idea that the historical dynamics of 24 Malešević, Identity as Ideology, 113–​18.

25 Smith, The Ethnic Origins of Nations, 157.

26 Simmel, Conflict and the Web of Group Affiliations; Randall Collins, Interaction Ritual Chains (Princeton: Princeton University Press, 2004); Collins, Violence.

23



War and Peoplehood through Time

23

war and peoplehood is moulded by three long-​term historical processes: the cumulative bureaucratization of coercion, centrifugal ideologization, and the envelopment of micro-​ solidarity.27 Let us explore how these processes operate. War and Organizational Power

The cumulative bureaucratization of coercion stands for an ongoing historical process through which different social organizations attain the capacity to accumulate coercive power and to use this power to control their social environment. Historically, the state has been the most significant “bordered power container” as it was capable of dominating and eventually monopolizing the use of coercive power over the territories under its control.28 However, this process has affected a variety of non-​state organizations too—​private corporations, merchant guilds, religious institutions, social movements, political parties, professional associations, and so on. In all these cases, increased organizational might was used to successfully dominate individuals under the control of such organizations. Although coercive organizational power has historically waxed and waned as some states, religious organizations, private enterprises, and other entities have been dominated, amalgamated, or destroyed by other social organizations, coercive organizational power as such has been on the increase for the past 10,000 to 12,000 years. Furthermore, while some parts of the world have experienced periodic decline in coercive organizational capacities, other regions have often taken their mantel and continued this process. For example, the Persian Empire under the Achaemenids possessed the greatest coercive organizational powers and as such dominated much of the known world in its heyday—​in 480 BCE it controlled close to 50 million people and it had influence not only in the Middle East but also in North Africa, Central Asia, Europe, the Mediterranean world, and India. As the Persian Empire declined, others took its place—​the Roman and Chinese Empires developed potent coercive-​organizational capacities and were able to dominate huge populations. At its peak in 117 CE, the Roman Empire controlled over five million square kilometres and over 60 million people. Its military was composed of ca. 450,000 soldiers. Similarly, the Chinese Empire ruled over more than 60 million people and could mobilize up to 500,000 soldiers under the Han dynasty.29 The decline of the Roman Empire in the West and later the Chinese Empire was followed by a rise in coercive-​organizational power in the Middle East, North Africa, and Mediterranean Europe with the expansion of the Rashidun, Umayyad, and Abbasid Caliphates from the seventh till the thirteenth centuries. 27 For an extensive elaboration of this theoretical model see my books: Sinisa Malešević, The Sociology of War and Violence (Cambridge: Cambridge University Press, 2010); Why Humans Fight: The Social Dynamics of Close-​Range Violence (Cambridge: Cambridge University Press, 2022); Nation-​States and Nationalisms; The Rise of Organised Brutality; Grounded Nationalisms.

28 Anthony Giddens, The Nation-​State and Violence (Cambridge: Polity, 1985).

29 Jane Burbank and Frank Cooper, Empires in World History: Power and Politics of Difference (Princeton: Princeton University Press, 2010).

24

24 Siniša Malešević Although the power dynamics have shifted through time and place, the coercive-​ organizational capacity continued to increase, mostly through warfare. To wage wars successfully, it was necessary to develop more centralized structures of governance and administration but also to build more effective systems of transport and communication. Hence the increase in coercive-​organizational capacity is mostly visible in the rise in infrastructural capacities (including more efficient models of tax collection, census taking and information collection, standardization of weights and measures, etc.), the development of complex divisions of labour with elaborate hierarchies or rule, the expansion of coercive apparatuses including military, police, courts, and private security structures, and the development of administrative structures. The cumulative bureaucratization of coercion has also been characterized by deeper societal penetration, wider territorial scope, and greater infrastructural reach. For example, the territorial expansions of the Sasanian Empire (224 to 651 CE) and the Umayyad Caliphate (661–​750 CE) largely transpired on the back of developed organizational capacities in the military and administrative spheres. Hence, the Sasanian rulers built on the organizational legacy of the Parthian Empire and developed a new military and administrative structure capable of ruling vast territories (stretching from today’s Egypt, Turkey, and Yemen to Pakistan and India) and containing their powerful Roman and later Byzantine neighbours for over 400 years. At the core of the Sasanian state was an efficient bureaucracy—​a centralized state apparatus that allocated substantial resources to urban planning, technological development, and the expansion of agriculture. The civil service was headed by the vizier and its top echelon was staffed by the Zoroastrian priests. However, the ruling elite also accommodated other elite groupings including the heads of the traders and merchants syndicate and the heads of the farming communities. The empire had a large and efficient military consisting of the cavalry, infantry, and the navy. The cavalry, initially modelled on the Parthian Empire, were the main military force consisting of nobility who were involved in lifelong military service and were very well trained. The cavalry also utilized war elephants and Sasanian strategists devised advanced siege engines which were successfully deployed against the Roman and Byzantine cities. The infantry, Paygan, were mostly peasant conscripts who were lightly armed and usually used for physical labour (i.e., mine excavations, digging ditches and canals, and fortification of walls). The navy controlled the Persian Gulf and was highly instrumental in keeping the trading routes open and safe from piracy and sudden invasions. The empire possessed a substantial organizational capacity with very good road networks and bridges that were regularly patrolled, thus enabling the relatively smooth mobility of trading caravans, state postal services, and people and keeping open the trade routes from Europe to India and China. The Sasanian traders dominated and at times monopolized international trade in luxury goods.30 30 Zeev Rubin, “Persia and the Sasanian Monarchy (224–​651),” in The Cambridge History of the Byzantine Empire, c. 500–​1492, ed. Jonathan Shepard (Cambridge: Cambridge University Press, 2009), 130–​55.

25



War and Peoplehood through Time

25

Similarly, the Umayyad Caliphate was built on the empire’s military might, which itself entailed the presence of sophisticated coercive organizational structures. At the apex of its power, the empire stretched from today’s Spain, Portugal, and Morocco to Caucasus, Arabia, Pakistan, and India, covering over 15 million square kilometres and including 62 million people. The organizational backbone of the state was a centralized and for its time efficient bureaucracy with an elaborate division of labour. The governance was divided into four principal areas: political affairs, military affairs, religious administration, and economic administration focused on tax collection. These centralized branches were further divided into many lower-​level departments and offices. Although the Caliph had the ultimate power, he was aided by six well-​organized Diwans (for revenue, correspondence, posts, justice, military, and the imperial seal). The huge territories of the expanding empire were ruled though centrally appointed and controlled governors who were responsible for tax collection and military recruitment. The empire operated an efficient and regular post service and introduced its own currency and coinage. The Umayyads inherited a powerful army from the Rashidun Califate (632–​661) that initially had over 100,000 soldiers and continued to expand until the collapse of the empire. The military was characterized by high discipline, good organizational structure, and strategic innovativeness. The Rashiduns introduced the new system, which the Umayyads developed further, with the army organized as a state department. This organizational novelty included the establishment of a register of all Muslim adult males who could be recruited for war and also introduced fixed salary rates for different military ranks.31 Since the army was paid in cash, it was dependent on state administration.32 Both of these examples indicate that the state’s power has historically been determined by its coercive organizational capacity, which was often created for warfare and through warfare. Furthermore, this bureaucratization of coercive power was historically cumulative as in both of these cases empires were built on the coercive organizational scaffolds of the previous imperial structures—​from Parthians to Sasanians and from Rashiduns to Umayyads. In other words, there was a significant element of continuity here, but this was not a cultural continuity favoured by ethno-​ symbolists, but coercive-​organizational continuity. The possession of a high degree of coercive organizational power was a precondition for waging successful wars and expanding one’s territory. However, while the experience of war was important in forging state institutions, in itself it was not sufficient to mould state subjects into culturally and politically homogenous entities. For this to happen, the coercive organizational capacities had to link well with centrifugal ideologization.

31 Non-​Muslims could not join the military unless they converted and as such were obliged to pay a poll tax to the state.

32 Gerald R. Hawting, The First Dynasty of Islam: The Umayyad Caliphate AD 661–​750 (London: Routledge, 2002).

26

26 Siniša Malešević Ideology, Peoplehood, and War There is no doubt that wars play a crucial role in the creation and destruction of states and other organizational structures. Nevertheless, the shared experience of protracted violent conflicts also impacts on group formation. For Smith, “the historical consciousness that is so essential a part of the definition of what we mean by the term ‘ethnic community’ is very often a product of warfare.”33 In the ethno-​symbolist view, war is a catalyst of ethnic formation as it provides sacrificial and traumatic memories that establish the principal moral parameters of group membership. While one can agree that war commemorations help reinforce group attachments, there is no sociological reason that such attachments must be ethnic or national. In fact, modernists are right that before modernity there were no structural preconditions for the development of society-​wide political identities and individuals tended to identify more in terms of shared religion, kinship, clans, and local residential attachments. However, the modernist arguments seem less persuasive when they simply assume that the ties of nationhood developed suddenly at the end of the eighteenth and throughout the nineteenth centuries. I argue that in explanatory terms we can gain more by recognizing that the formation of collective attachments has a much longer history while also moving away from the ethno-​symbolist obsession with cultural persistence through time. This can be achieved by exploring the historical dynamics of centrifugal ideologization. This concept refers to the open-​ended process through which different ideological and proto-​ideological doctrines penetrate and permeate different social groups within a specific society. They provide normative justification for the rulers and can also be utilized to mobilize different social groups. They are centrifugal in a sense that they are created by the cultural and political elites and are often reinforced through existing institutions including religious and political bodies. Similarly to the centrifuge in the washing machine, ideologization operates through sedimentation where acceleration pushes particles to move outward in a circular fashion—​they radiate from the centre to the periphery. Rather than being a static doctrine, ideologization is an ongoing historical process through which specific ideas, values, and practices are produced and disseminated to justify the existence of a specific organization. Hence ideologization is not culture; instead, it is a process based on the selection of specific cultural markers that can facilitate justification of particular action and in modernity can also aid popular mobilization. In the premodern contexts, elements of different belief systems and practices were utilized to legitimize the existing social orders—​from animism, totemism, the mythologies of common descent, imperial creeds, beliefs in the divine origins of monarchs to a variety of religious doctrines and civilizing missions. In modernity, one encounters fully fledged ideological projects—​liberalism, conservatism, socialism, religious fundamentalism and, most of all, nationalism. The key issue here is that modern ideologies such as nationalism could not appear abruptly and out of nothing. Instead, 33 Anthony D. Smith, “War and Ethnicity: The Role of Warfare in the Formation, Self-​Images, and Cohesion of Ethnic Communities,” Ethnic and Racial Studies 4 (1981): 375–​97, at 379.

27



War and Peoplehood through Time

27

they gradually developed on the premodern structures—​the slowly increasing coercive organizational capacities and the intensified centrifugal ideologization. There is no doubt that the traditional social orders were deeply stratified and culturally polarized with the aristocracy developing horizontal instead of vertical networks of solidarity. In medieval Europe, cultural markers were used not to establish cross-​class national identities as in the modern world but they were primarily deployed to differentiate aristocrats from their, mostly peasant, subjects. Hence the social world of aristocratic Europe was confined largely to fellow aristocrats across the borders of different polities. The nobility developed a rich cultural repertoire to dissociate themselves from the commoners—​from aristocratic endogamy, elaborate courtly protocols, and etiquette to use of different linguistic practices and rampant diglossia. Nevertheless, for nationalism to eventually emerge in modernity it had to be developed out of some organizational and ideological ingredients that existed in the premodern world. In this context, shared religious beliefs and practices have often acted as a stepping stone toward the development of peoplehood narratives that would later be transformed into ethnonationalist histories. For example, national flags and coats of arms without which one cannot envisage contemporary nation-​states have been mostly built on medieval aristocratic heraldry. Hence the symbols that were traditionally associated with specific aristocratic lineages and families have gradually been “nationalized,” thus becoming associated with specific nations. Hence the coat of arms of Poland, godło of a white crowned eagle, was originally a heraldic symbol of the Piast nobility, who later become a ruling dynasty. In this context, the eagle appeared on the ensigns, seals, and shields of the Piast dukes in the twelfth century and has eventually become transformed into a modern-​day national symbol for all Poles. Thus, there is an important element of continuity here, but this is not a spontaneous celebration of myths and memories, as Smith would argue, but an ideological project spearheaded by specific organizations that were ultimately successful in transforming an aristocratic symbol used to differentiate nobility from the peasants into a national, and thus a cross-​class symbol, that expresses a society-​wide unity. It is through centrifugal ideologization that warfare has contributed to the transformation of peoplehood. For example, the concept of the nation is much older than the sociological meaning it has today. Initially the term referred to “place of origin” and it was deployed in this sense in medieval Europe.34 Fourteenth-​century universities in Paris, Prague, and Uppsala categorized their students as belonging to different “nationes” (i.e., Bohemian, Bavarian, Saxon, and Silesian in the Prague case). This designation did not have any political meaning and it was utilized as a practical measure to keep track of students’ region of birth and to rule them by using the laws they were familiar with. This term was also used in a similar way by the Order of Knights of the Hospital of Saint John of Jerusalem in the

34 The Latin term nātĭō means birth and the later French use naissance has the same meaning indicating one’s place of birth or origin.

28

28 Siniša Malešević fifteenth century, who operated inns at Rhodes where members of different “nationes” were accommodated in the same inn. However, it was the shared war experience that played a significant role in the politicization of cultural difference. The religious wars against the infidels abroad fostered a degree of unity among the aristocrats, which was later used to build stronger political alliances at home. For example, participation in the Crusades enhanced the social status of aristocrats who fought in the Holy Land and also fostered lasting bonds between some of them. Later on, these aristocrats could call upon these ties to fight the wars and uprisings at home. Furthermore, the crusading experience and imagery was later used to justify participation in various European wars among Christian populations. For example, the Hundred Years’ War (1337–​1453) started as a conflict over the competing dynastic claims between the House of Plantagenet and the House of Valois. Nevertheless, as the conflict became protracted it expanded to other aristocratic groups where both sides deployed proto-​ideological discourses to justify their political claims over sovereignty. Drawing on the Crusading rhetoric of religious sacrifice but also on cultural difference between the two royal households, the Hundred Years’ War fostered greater politicization of cultural difference between the “French” and “English” nobilities.35 Hence the prolonged and destructive war stimulated the gradual codification of distinct ideas of peoplehood. The legacy of this war was the decline of French as the dominant language of aristocracy in England and the establishment of English as the official language of the country in 1362. The centrifugal ideologization went hand in hand with the cumulative bureaucratization of coercion as the war fostered state centralization and the introduction of new systems of taxation. The direct legacy of this war was the establishment of a regular standing army in France which was much more disciplined and professional than the unreliable and undisciplined mercenary units that fought during the war. The war also gave impetus to military inventions with significant breakthroughs in military tactics, technology, and weaponry, and the organization of the military. This war and many others waged in the next several centuries were highly instrumental in forging distinct forms of peoplehood although mostly at the elite, aristocratic, level. However, it is this ideological and organizational kernel that eventually gave birth to mass-​scale nationalisms in the late nineteenth and twentieth centuries. Micro-​Solidarity and Warfare

There is no doubt that military success regularly depends on one’s organizational strength. As Collins emphasizes, “armies fight, not in order to kill soldiers, incapacitate weapons, and take ground, but to destroy the ability to resist. Organization is both the weapon and 35 Denise Baker, Inscribing the Hundred Years’ War in French and English Cultures (New York: State University of New York Press, 2000).

29



War and Peoplehood through Time

29

the target of war.”36 Ideological power is also indispensable for the mobilization for, and justification of, war. However, wars are not only political and military projects; they also involve extreme forms of physical interaction between human beings who are expected to injure and kill other human beings. Hence in addition to this wider structural context, wars also generate new forms of micro-​dynamics. When people fight together and share the everyday experience of danger, sacrifice, and loss, they often develop unique and strong bonds of attachment. In many instances, the shared battlefield experience moulds intense emotional and moral ties that resemble and could even be stronger than that of very close family members. The soldiers often invoke kinship terms to describe these unique bonds. As a World War II US marine puts it: “those men on the line were my family, my home. They were closer to me than I can say, closer than any friends had been or ever would be. They never let me down, and I couldn’t do it to them. Any man in combat who lacks comrades who will die for him, or for whom he is willing to die, is not a man at all.”37 In this context, warfare can significantly enhance group identifications. As humans are first and foremost emotional creatures, they are prone to developing strong emotional ties with other individuals involved in the same situation. Micro-​sociological research has shown on numerous occasions that external threats and prolonged shared exposure to danger foster greater internal social cohesion.38 Nevertheless, this is not an automatic process that simply and inevitably unfolds as the two collectivities engage in a violent conflict. The size and complexity of the group impacts on the response toward external threats, and small-​scale groups are more likely to develop stronger and lasting bonds than large-​scale and complex societies. Furthermore, unlike the small face-​to-​face groups which can act instantly in response to external stimuli, large-​scale social orders require much more coordination, division of labour, and deliberation. Whereas small groups can directly perceive or experience an external attack, in complex organizations any information about an outside danger is likely to be mediated by intermediaries. This means that when states are at war with each other, their populations do not instantly hate each other or show a willingness to join the war effort. Instead, the transition from peaceful citizens to war-​mongering enthusiasts entails a great deal of organizational and ideological work. Hence wars do not necessarily and mechanically trigger strong nationalist feelings; such feelings have to be developed over time and then generated, prompted, and sustained through a variety of organizational and ideological channels including mass media, the educational system, the public sphere, government decrees and proclamations, civil society actions, and so on. As these organizational and ideological mechanisms are much more developed and institutionalized in modernity, it is no surprise that most interstate wars are now 36 Randall Collins, “Sociological Theory, Disaster Research and War,” in Social Structure and Disaster, ed. Gary Krebs (Newark: University of Delaware Press, 1989), 365–​86, at 366. 37 Richard Holmes, Acts of War: Behaviour of Men in Battle (New York: Free Press, 1985), 300.

38 Malešević, The Sociology of War and Violence; Collins, Violence; Coser, The Functions of Social Conflict.

30

30 Siniša Malešević accompanied by intense nationalist rhetoric and practice. In this context, modernists are right that wars can trigger mass nationalist upsurges only under modern conditions. However, this strong link between nationalism and war in modernity is built on organizational, ideological, and micro-​interactional processes that have been in place for much longer periods of time. In other words, warfare has played a significant role in transforming micro-​level solidarities into the future sense of peoplehood. Both nationalism and war depend on the ability of social organizations to penetrate the microcosm of small-​ group attachments. As human beings develop their strongest emotional ties with family members, close friends, lovers, peers, and other face-​to-​face intimate groups, all successful large-​scale organizations tend to tap into this micro world in order to project these emotions of micro-​solidarity onto their larger organizational canvas.39 Hence the representatives of nation-​states and nationalist movements regularly appropriate the language of close kinship by depicting their organizations as motherlands, fatherlands, brotherhoods, and sisterhoods. Although this process is rampant in modernity, its origins are much older. Many premodern organizations also utilized the discourse of micro-​solidarity in order to enhance collective bonds. For example, the many military religious orders that emerged in the Middle Ages such as the Knights Hospitaller, the Knights Templar, or the Teutonic Knights were built around strong group ties among the noblemen. The Teutonic Knights (officially called “Order of Brothers of the German House of Saint Mary in Jerusalem”) were largely shaped by military experience including the Crusades, the wars with the Cumans in Transylvania and later in the Baltics and Poland. The Order established a hierarchical organization with several layers of authority, starting from the Grandmaster at the top heading the general chapter (Generalkapitel), the “national” and regional leaderships in the middle, and the numerous local chapters (Kommende) at the bottom of the hierarchy. This hierarchical system successfully incorporated effective bureaucratic organization underpinned by a religious proto-​ideology that permeated the entire order with networks of micro-​solidarity rooted in the local chapters. The local chapters were the real hubs of micro-​solidarity as they were composed of individuals who lived together and also fought together in various wars. These local units consisted of small groups of religious brothers who lived in an ascetic monastic way and were organized into two main groups—​the knight brothers who fought in wars and the priest brothers involved in religious ceremonies. The local units also included ordinary armed soldiers (Sariant brothers) and half-​brothers responsible for economic and health-​ related issues. These brothers, noblemen, developed strong emotional attachments which were then successfully utilized by the Order, making the Teutonic Knights a formidable military machine of its time. Following a series of military successes, the Order eventually established its own independent state—​the Monastic State of the Teutonic Knights—​in 1230.40 39 Malešević, The Rise of Organised Brutality; Malešević, Grounded Nationalisms.

40 Nicholas Morton, The Teutonic Knights in the Holy Land, 1190–​1291 (Woodbridge: Boydell & Brewer, 2017).

31



War and Peoplehood through Time

31

While this polity was still a highly hierarchical feudal state that had no resemblance to a nation-​state in any way, it ultimately paved the way for the future development of the German nation-​state in the nineteenth century. This is not to say that there was some inevitable and automatic cultural continuity between the Teutonic Order and the future German nation. On the contrary, the Teutonic Knights were a rather typical medieval order that sharply differentiated between the cultures of elites and the ordinary peasantry and as such had no interest in binding these two profoundly different cultural worlds. Instead, the continuity that truly matters and that played a crucial role in the development and transformation of different polities on this territory was not cultural but was built from organizational, ideological, and micro-​interactional ingredients. Hence the coercive bureaucratic organizational form of the Teutonic Order together with their Christian proto-​ideology and the well-​embedded networks of micro-​solidarity provided a stepping stone for the emergence of other polities that ultimately brought about nineteenth-​and twentieth-​century Germany. The Teutonic Order unwittingly initiated long-​term structural developments ranging from the formation of Brandenburg Prussia to the Holy Roman Empire to the modern-​day German nation-​state. Although a premodern polity such as the Monastic State of the Teutonic Knights did not and could not generate society-​wide national attachments, it did initiate the organizational, ideological, and micro-​interactional processes that would eventually forge a sense of peoplehood in the modern era. In addition, as the Order was developed and expanded through warfare, it was this military experience that contributed substantially toward the growth of structures which would eventually generate a sense of peoplehood.

Conclusion

Heraclitus’s famous fragment states that “war is father of all, and king of all. He renders some gods, others men; he makes some slaves, others free.” Looking at the last five thousand years of human history, it is difficult to disagree with Heraclitus as warfare has shaped nearly all human institutions. However, although war impacts on all social relations, this is far from being a mechanical and preordained reality. Instead, different historical contexts matter as wars and societies change through time. Hence, although wars influence group identities they do not instantly create a sense of peoplehood. Rather, this is a long-​term and complex process that entails social change across different structures. In this chapter, I have advanced a sociological longue durée view which analyzes the relationships between war and peoplehood through the prism of three large-​scale historical processes—​the increasing coercive organizational capacity, the changing dynamics of ideological penetration, and the mechanisms that trigger the envelopment of micro-​solidarity. I have argued that the society-​wide sense of peoplehood, mostly articulated as nationalism, develops only under modern conditions but the structural ingredients that make this development possible have a much longer history. Zooming in on the examples from the Middle Ages, I have shown how organizational, ideological, and micro-​interactional processes contribute toward the gradual politicization of cultural differences over time.

32

33

Chapter 3

MAKING WAR ETHNIC: ARAB–​PERSIAN IDENTITIES AND CONFLICT ON THE EUPHRATES FRONTIER Peter Webb Battles and ethnic identities seem natural companions in theory. Both engage

strong emotions, both are conceptualized in binary oppositional terms of “us” versus “them,” and via memories of battles, ethnic groups can plot their history in heroic terms explaining how “we” as a people emerged from the crucible of conflict, how strong “we” are, and how “we” have always fought “them.” But the well-​known vicissitudes of collective memory readily recraft past events to harmonize them with a community’s present circumstances. In practice, therefore, battles may not always be as ethnically charged in the thick of the fight as they come to be memorialized afterwards. This chapter studies the memorialization of the Battle of Dhū Qār, a pre-​Islamic clash on Iraq’s Euphrates frontier ca. 610 CE, which acquired a prominent and evolving significance in Muslim historiography.1 Dhū Qār presents itself as a natural case study for war and peoplehood since its story has been told in explicitly ethnic terms for over 1,000 years. Arabic literature from at least the third/​ninth century regularly references Dhū Qār as the “the Arabs’ first victory over the Persians,”2 or “the first time the Arabs showed themselves equal to the Persians,”3 and the pre-​Islamic battle is depicted as foreshadowing the imminence of the Muslim conquest of Iraq which extinguished the Sasanian Empire just a few decades later.4 Muslim-​era histories invoke an ethnic binary of “Arab” versus “Persian” to project 1 An outline case study on Dhū Qār was published in Peter Webb, Imagining the Arabs: Arab Identity and the Rise of Islam (Edinburgh: Edinburgh University Press, 2016), 88–​95; this chapter undertakes a fuller survey of the textual evidence and develops the conclusions. 2 Aḥmad ibn Abī Yaʿqūb al-​Yaʿqūbī, al-​Tārīkh, ed. Michael Jan de Goeje, vol. 1 (Beirut: Dār Ṣādir, n.d.), 215, 225; Al-​Ṭabarī, Muḥammad ibn Jarīr, Tārīkh al-​rusul wa-​l-​mulūk, ed. Muḥammad Abū al-​Faḍl Ibrāhīm, vol. 2 (Cairo: al-​Maʿārif, 1960–​69), 188–​93.

3 Abū ʿAlī Aḥmad Miskawayh, Tajārib al-​umam wa-​taʿāqub al-​himam, ed. Sayyid Kisrawī Ḥasan, vol. 1 (Beirut: Dār al-​Kutub al-​ʿIlmiyya, 2003), 158; al-​Andalusī, Ibn Saʿīd, Nashwat al-​ṭarab fī tārīkh jāhiliyyat al-​ʿArab, ed. Nuṣrat ʿAbd al-​Raḥmān, vol. 1 (Amman: al-​Aqṣā, 1982), 286; Ibn al-​Athīr, ʿIzz al-​Dīn ʿAlī ibn Abī al-​Karam, al-​Kāmil fī al-​tārīkh, ed. C. J. Tornberg, vol. 1 (Beirut: Dār Ṣādir, 1979), 482–​83.

4 Ella Landau-​Tesseron, “Ḏū Qār,” 574–​75, at 575; Michael Morony, Iraq After the Muslim Conquest (Princeton: Princeton University Press, 1984), 152–​53, 220; Peter Heath, “Some Facets of Poetry in Pre-​modern Historical and Pseudohistorical Texts,” in Poetry and History: The Value of Poetry in Reconstructing Arab History, ed. Ramzi Baalbaki, Saleh Said Agha, and Tarif Khalidi

34

34

Peter Webb

Dhū Qār as a crucial turning point in world history: under this worldview, Persian kings had dominated the Middle East for centuries before Muhammad, but Muhammad’s Arab people were to end Persian supremacy once and for all via the Muslim Conquests. Hence Dhū Qār was the “beginning of the end,” the turning of the tide, the moment when the Arabs first displayed their potential. According to these Muslim-​era narratives, once the Arabs embraced Islam shortly after Dhū Qār, they could commence their irresistible conquest of the Middle East.5 The Arab–​Persian divide is salient in Middle Eastern social history, and competition between partisans of both identities has persisted in varying degrees since early Islam to the present with Dhū Qār as the pivot point of the rhetoric. Iranian nationalists celebrate memory of an enlightened millennium of ancient Persian civilisation pre-​Dhū Qār, lamenting the destruction by “Arab conquest,”6 whereas Arab nationalists champion history from Dhū Qār onwards, expressly mobilizing memories of Arab victories over Persians in contemporary competition against Iran.7 But if we leave nationalist discourses aside, a very different picture of the battle emerges. This chapter analyzes the preserved poetry and early prose narratives of Dhū Qār, ranging from poems composed at the time of the battle, poetry from the Umayyad era (ca. 660–​750), and the first prose accounts written ca. 800–​1000. The textual layers reveal varied memories as Dhū Qār’s history was transmitted over time and across different communities, alongside a gradual emergence of its now-​familiar Arab–​Persian ethnic binary. (Beirut: American University of Beirut, 2011), 39–​60, at 48, 50–​52; Aziz al-​Azmeh, The Emergence of Islam in Late Antiquity (Cambridge, Cambridge University Press, 2014), 118, refers to Dhū Qār as an “Arab” victory, but notes difficulties in the sources too (127).

5 The presentation of Dhū Qār as the “tipping point” is particularly explicit in Miskawayh, 1: 157–​58; Ibn al-​Athīr, 1:480–​82. Not all premodern historians indulge the “tipping point” narrative to the full, for example Abū al-​Fidāʾ, al-​Mukhtaṣar fī akhbār al-​bashar, vol. 1 (Cairo: Dār al-​Maʿārif nd.), 56 does not narrate the portentous material, though he does nonetheless define the battle in binary ethnic terms of “Arab” versus “Persian” (1:72).

6 Mirza Fatḥ-​ʿAlī Akhundzade, Maktubāt: Nāmehā-​ye Kamāl al-​Dawleh beh Shāhzādeh Jamāl al-​Dawleh (Frankfurt: Alborz, 2006) and Mirza Aqā Khān Kermānī, Seh Maktūb (Frankfurt: Alborz, 2005) state the case explicitly. The rise of Persian nationalism in a form that casts the Arabs as binary enemies is detailed in Reza Zia-​Ebrahimi, The Emergence of Iranian Nationalism: Race and the Politics of Dislocation (New York: Columbia University Press, 2016). The 1979 Iranian Revolution rehabilitated the esteem for Islam, but without eradicating the Persian ethnic symbolism, and much of the Pahlavi-​era rhetoric in this regard has persisted.

7 See Talal Atrissi, “The Image of Iranians in Arab Schoolbooks,” in Arab–​Iranian Relations, ed. Khair El-​Di Haseeb (Beirut: Centre for Arab Unity Studies, 1998); D. Gershon Lewental, “ ‘Saddam’s Qadisiyyah’: Religion and History in the Service of State Ideology in Baʿthi Iraq,” Journal of Near Eastern Studies 50 (2014): 891–​910 and see the essays in Writing the Modern History of Iraq: Historiographical and Political Challenges, ed. Jordi Tejel, Riccardo Bocco, and Peter Sluglett (Singapore: World Scientific, 2012). The majority of twentieth-​century politicised attention was focused on the Battle of Qādisiyya where Muslim armies defeated Sasanian imperial forces and opened the conquest of Iraq, but Dhū Qār featured too: Iraq’s Baathist regime named their 7th Army Corps the “Dhū Qār Corps,” Pesach Maloveny, Wars of Modern Babylon: A History of the Iraqi Army from 1921 to 2003 (Lexington: University of Kentucky Press, 2017), 892.

35



Making War Ethnic

35

Dhū Qār: The Battle and Its Sources Dhū Qār is enveloped within the general cloud of historiographical uncertainty that covers much of Late Antique Arabia’s history. The earliest written Arabic histories emerged between the later second/​eighth and mid-​third/​ninth centuries, that is, 150–​200 years after Dhū Qār was fought. The Sasanians had written records, but following the Muslim conquest in the 640s, much was lost, and no pre-​Islamic Persian memories of Dhū Qār are extant.8 It is therefore impossible to confidently reconstruct the action with maps and lines of troop movements as military historians like to draw, but from the perspective of identity, matters are not so bleak. Dhū Qār was mentioned in Arabic poems composed from the time of battle until the period of recording in the late second/​eighth and third/​ninth centuries, and while embellishment and manipulation reshaped old poems over time, it seems at least that much of the preserved poetry does reflect genuine survivals from earlier periods.9 The poems furnish ethnonyms and sentiments that together offer a window into the kinds of communities which poets in late pre-​Islam and Islam’s first two centuries associated with the battle, and this chapter diachronically analyzes the complete gamut. Alongside the poetry, we also possess prose accounts of the battle which accrued over successive generations of narration which historians of the third/​ninth and fourth/​ tenth centuries compiled into contiguous battle accounts. These Muslim-​era histories are composites of multiple mnemonic layers, and they can be read alongside the poetry to permit critical diachronic analysis of the associations drawn between the pre-​Islamic battle and Arab, Persian, and Muslim identities. In the round, the prose accounts explain Dhū Qār as follows.10 The Sasanian Empire had long appointed agents to guard its Iraqi frontier against Arabian nomadic incursions. These agents were powerful men with ties to groups within Arabia, and 8 A tenth-​century Persian view on the battle is preserved in Muḥammad Balʿamī, Tārīkhnāme-​ ye Ṭabarī, ed. Muḥammad Rawshan, vol. 2 (Tehran: Alborz, 1366–​73/​1987–​94), 812–​24; it is a Muslim-​era Persianized memory, not a Sasanian one; it is considered below. 9 For a discussion of Arabic poetry authenticity, see Walid Arafat, “The Historical Significance of Later Anṣārī Poetry—​I,” Bulletin of the School of Oriental and African Studies 29 (1966): 1–​11; Arafat, “The Historical Background to the Elegies on ʿUthmān b. ʿAffān attributed to Ḥassān b. Thābit,” Bulletin of the School of Oriental and African Studies 33 (1970): 276–​82; Webb, Imagining the Arabs, 69. Alan Jones, “The Oral and the Written: Some Thoughts about the Quranic Text,” in Proceedings of the Colloquium on Logos, Ethnos, Mythos in the Middle East and North Africa Part One: Linguistics and Literature, ed. Kinga Dévényi and Tamás Iványi (Budapest, 1996), 57–​66, at 58, and Suzanne Pinckne Stetkevych, The Mute Immortals Speak: Pre-​Islamic Poetry and the Poetics of Ritual (Ithaca: Cornell University Press, 1993), 122, allude to the impact of “Abbasid guises” in shaping preserved poetry, but consider the bulk to be authentic; Said Saleh Agha, “Of Verse, Poetry, Great Poetry and History,” in Poetry and History: The Value of Poetry in Reconstructing Arab History, ed. Ramzi Baalbaki, Saleh Said Agha, and Tarif Khalidi Khalidi (Beirut: American University of Beirut, 2011), 1–​35, at 8 describes the retreat from earlier “vigorous” doubts about authenticity. 10 The following is derived from the most detailed sources on Dhū Qār: Muḥammad Ibn Ḥabīb, Kitāb al-​Naqāʾiḍ, ed. Anthony A. Bevan, vol. 2 (Leiden: Brill, 1905–​12), 638–​48; al-​Ṭabarī, Tārīkh, 2:188–​93; Aḥmad ibn Muḥammad Ibn ʿAbd Rabbihi, al-​ʿIqd al-​farīd, vol. 5, ed. Ibrāhīm al-​Abyārī (Beirut: Dār al-​Kitāb al-​ʿArabī, n.d.), 246–​52; Abū al-​Faraj al-​Aṣbahānī, Kitāb al-​Aghānī, vol. 24,

36

36

Peter Webb

by the early seventh century CE, two names—​Iyās ibn Qabīṣa and Qays ibn Masʿūd—​ are memorialized as key frontier leaders, guarding the middle and lower reaches of the Euphrates, respectively. Nomads from several lineage groups of the Bakr ibn Wāʾil made incursions; the Sasanian shah was aggrieved, and summoned his frontier agents to action. Iyās ibn Qabīṣa mobilized Arabian groups of the Taghlib, Iyād, the Namir, the Quḍāʿa, the Ṭayyiʿ, and others; he was reinforced with Sasanian heavily armed cataphract cavalrymen under the Sasanian commanders Hāmarz and Hormuz Kharād,11 and the combined force marched against the Bakr. The sides met at Dhū Qār. The Bakr were offered the option of either submitting or retreating beyond the frontier, but a warrior leader named Ḥanẓala of the Bakr’s ʿIjl clan urged resistance, the peace terms were refused, and battle began. The fighting’s duration is unclear; it may have spanned two days, but whatever the length, the Bakr emerged victorious, killing Hāmarz and Hormuz Kharād and some of the Persian-allied Arabian frontier leaders too. The prose accounts elaborate upon the above framework with numerous details, yet with little concord and considerable contradiction. The precise trigger of the battle is a significant disagreement. The composite source of al-​Aṣbahānī (d. 356/​967) opens by narrating the battle firmly within the politics of frontier control: the Sasanian shah, Khosrow Parviz, executed the frontier guardian al-​Nuʿmān ibn al-​Mundhir, and into the ensuing power vacuum nomads of the Bakr began raiding. Khosrow Parviz charged Qays ibn Masʿūd to guarantee the border against the Bakr, but he was unable to control the most belligerent parties from the Shaybān and ʿIjl subgroups. The shah thus imprisoned Qays and commanded his other allies to fight the Arabian invaders.12 This account seems rather lucid, but the majority of sources narrate a second version which places more stress on a backstory that al-​Nuʿmān ibn al-​Mundhir, just prior to his execution, deposited weapons, herds, and perhaps his family in the safekeeping of Hāniʾ ibn Masʿūd (of Bakr lineage). Khosrow Parviz demanded Nuʿmān’s chattels, but Hāniʾ refused. Enraged, Khosrow decided to wipe out the Bakr in retribution.13 This second version renders the Persians more as aggressors, and moreover, it places Dhū Qār within ed. ʿAbd Allāh ʿAlī Muhanna and Samīr Jābir (Beirut: Dār al-​Kutub al-​ʿIlmiyya, 1992), 54–​70, Miskawayh, Tajārib, 1:149–​62; Abū Hilāl Al-​ʿAskarī, al-​Awāʾil, ed. Muḥammad al-​Miṣrī and Walīd al-​Qaṣṣāb, vol. 2 (Al-​Riyadh: Dār al-​ʿUlūm, 1981), 186–​90; Balʿamī, Tārīkhnāme, 2:812–​24; Ibn al-​Athīr, al-​Kāmil, 1:482–​90.

11 Hormuz Kharād’s name is recorded in the Persian Balʿamī’s Tārīkhnāme, 1:819; Arabic accounts render his name variously, most have “Khanābazīn.” 12 Abū al-​Faraj al-​Aṣbahānī, Kitāb al-​Aghānī, 24:55–​60.

13 Muḥammad Ibn Ḥabīb, Kitāb al-​Naqāʾiḍ, ed. Anthony A. Bevan, vol. 2 (Leiden, Brill, 1905–​12), 638–​39; Abū Hilāl al-​ʿAskarī, al-​Awāʾil, 2:186; al-​Andalusī, Nashwat, 1:285. The version of Ibn al-​Athīr is quite similar, but adds that a leader of the Taghlib, al-​Nuʿmān ibn Zurʿa was instrumental in urging Khosrow to attack the Bakr (al-​Kāmil, 1:488). Al-​Aṣbahānī also narrates this story, but precedes it with the narrative detailed above (al-​Aghānī, 24:60–​62); in essence, al-​Aṣbahānī narrates two separate triggers for the battle (this is not uncommon in composite Arabic literary texts).

37



Making War Ethnic

37

a more personalized revenge story surrounding the famous hero of Arabic pre-​Islamic Iraqi history, al-​Nuʿmān ibn al-​Mundhir, and the crux of this story turns upon vengeful Persian emotions and a stereotyped “Arab” trait of trustworthiness. Muslim literary accounts about pre-​Islamic Arabia are replete with examples of “Arab morality” which include the Arabs’ refusal to break promises—​refusal to hand over weapons deposited on trust is particularly paradigmatic of this trope.14 The differences between the first and second versions involve contrastive narrative techniques and historiographical perspectives. The first suggests that a border raid by groups of the Bakr happened to defeat a force sent to repulse them; the story typifies the dynamics of settled-​nomad relations, a story as old as the Euphrates frontier itself. It also fits Bosworth’s suggestion that Dhū Qār was a “skirmish,” not a momentous clash with wide historical ramifications.15 The second version, however, elevates Dhū Qār into a grander narrative of Arab–​Persian relations by linking the battle explicitly to important dramatis personae (al-​Nuʿmān ibn al-​Mundhir), and the invocation of stereotypes of Arab morality/​ fidelity fit the account ethically into an Arab story. According to the second version, Dhū Qār was no accident: it is glued to the history of the Arabs via intimate connection with al-​Nuʿmān, and by pitting Arab fidelity against the iniquitous Persian shah’s aggression, the narrative makes it clear where the heroism lies. The second version suggests narrative expansion that changed the signification of Dhū Qār. Muslim-​era storytellers converted memories of a raid by some subgroups of the Bakr into a key cog of pan-​Arab history, and the fingerprints of their narrative manipulation over a period of generations emerge from scrutiny of the details. For example, the sources exhibit significant dispute over names: Abū ʿUbayda is rather adamant that the border agent was not Qays ibn Masʿūd, but instead his grandson, Qays ibn Hāniʾ ibn Qays ibn Masʿūd;16 and al-​Nuʿmān’s chattels were either deposited with Hāniʾ ibn Qabīṣa ibn Masʿūd or Hāniʾ ibn Masʿūd ibn Hāniʾ.17 The names of the Bakr tribal leaders who invaded Iraq are also reported with different options.18 It is even reported that there were two battles of Dhū Qār; the first occurred when the Bakr were suffering from drought in the desert and invaded Iraq by necessity, defeating a Persian force thanks to the leadership of Ḥanẓala ibn Sayyār of the ʿIjl who staunchly defended his tent; whereas the famous Dhū Qār was a second battle in which a slightly differently named Ḥanẓala of the ʿIjl—​Ḥanẓala ibn Thaʿlaba ibn Sayyār—​was the hero who

14 For the archetype of this motif, see the story of al-​Samawʾal, Ibn Nubāta, Sarḥ, 102–103.

15 Clifford E. Bosworth, “Iran and the Arabs Before Islam,” in The Cambridge History of Iran Volume 3(1): The Seleucid, Parthian and Sasanian Periods, ed. Ehsan Yarshater (Cambridge: Cambridge University Press, 1983), 593–​612, at 608. 16 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:638.

17 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:639.

18 See al-​Aṣbahānī’s account of two different sets of raiding parties which do essentially the same thing (al-​Aghānī, 56–​58).

38

38

Peter Webb

staunchly defended his tent.19 The two-​Dhū-​Qār approach seems a rather blunt strategy of a later narrator to rationalize two versions of the one event by artificially separating them into separate battles. Other naming differences emerge in the narratives of the battle itself: for example, the first member of the Bakr to duel the Sasanian commander is named as either Yazīd ibn Ḥāritha, Burd ibn Ḥāritha, or Burayd ibn Ḥāritha—​each possible misreadings of the Arabic orthography of his name.20 All the above variations turn on such patently similar names that we can interpret that events of Dhū Qār were remote by the time Muslims began recording the story, and Muslim-​era historians faced difficulties ameliorating different strands of stories that had emerged in association with similar-​sounding names. Furthermore, most details of the battle come with variations, often contradictory. According to some, Qays ibn Masʿūd languished in prison during the battle, yet others have him fighting for the Sasanians, while others narrate his defection to the Bakr in the middle of the battle as a key turning point.21 Pluriform memories also shape the key motif explaining the Bakr’s stand. According to the story, Ḥanẓala ibn Thaʿlaba cut the ropes of the Bakr women’s palanquins from their camels, and so prevented the women from making an escape if the tide of battle turned against the Bakr: this was intended to motivate the men to fight to the grim end, but there are three distinct versions about the cutting of palanquin chords, and various sources narrate them all. Logically, Ḥanẓala could only have cut the chords once, but the memories about his act were pluriform, and our narrators amalgamated everything they had. While agreement across the extant accounts that palanquins were cut from the backs of camels indicates that the Dhū Qār was famed for this act of hazarding the Bakr’s women, the details as to precisely how and when it transpired had already received multiple treatments by the time the stories were collected. The organization of the Bakr army is also variably reported, and while the Sasanian force is more consistently described, there are disagreements: some say it included units of the Ṭayyiʾ, others not, and there are different identifications regarding the identity of the men who killed the Sasanian commanders, and disagreements over which Sasanian allies were killed as well. Deeper disagreements surround which of the Bakr’s subgroups were actually present at Dhū Qār: the Shaybān and the ʿIjl are unanimously mentioned, and while some narrators claim that no other subgroups participated, others include reference to warriors of the Qays and Yashkur subgroups. Another point of dispute concerns whether the famous warrior al-​Ḥawfazān participated in the battle or not: most accounts omit him, yet those which do include him ascribe him the pivotal 19 Abū ʿUbayd ʿAbd Allāh al-​Bakrī, Muʿjam Mā istaʿjam, ed. Muṣṭafá al-​Saqā, vol. 3 (Cairo: Lajnat al-​Taʾlīf wa-​l-​Tarjama wa-​l-​Nashr, 1947), 1042–​43.

20 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:643 has “Burayd” or “Yazīd”; al-​Aṣbahānī, al-​Aghānī, 24:69 has “Yazīd”; Miskawayh, Tajārib, 1:162 and Balʿamī, Tārīkhnāme, 2:821 have “Burd.”

21 For elaboration of the imprisonment narrative, see al-​Aṣbahānī, al-​Aghānī, 24:58–​59; for the defection narrative see Balʿamī, Tārīkhnāme, 2:822.

39



Making War Ethnic

39

role in the victory.22 Concerning the tipping point of the Bakr’s victory, some texts ascribe it to the flight of one of the Sasanian allies, in another, it was the charge of al-​Ḥawfazān after an inconclusive day of fighting, and in a third, it was a succession of assaults first by the ʿIjl and then by the Shaybān against the exhausted and parched Sasanians; however, other accounts place this joint charge first, and identify single combat duels against the Sasanian commanders as the decisive turning point.23 In sum, the Dhū Qār narrative possesses a recognizable set of characters and events shared between all sources, but it is impossible to overlook the differences in the details. The sources proffer far too many permutations to enable us to know what happened at Dhū Qār beyond the generality of a successful incursion by some groups from the Bakr against the Sasanian border which resulted in a clear (and perhaps unexpectedly clear-​ cut) victory arousing local excitement. Beyond this, an array of different things may or may not have happened. Overall, such variations indicate the battle’s importance as a site of memory in early Islam—​many different voices wanted to tell Dhū Qār stories—​ but the memory was highly plastic, and narrators employed a free hand to portray the battle in ways that suited their purposes.24 By the time the different versions were committed to writing in the third/​ninth century, Dhū Qār was too remote for historians to know how to differentiate the panoply of variation, and they created composite narratives according to their own styles. In terms of peoplehood, however, the prose accounts do all emphasize the Arabness and Persian-​ness of the opposing sides, but even this is deeply problematic on several counts.

Dhū Qār Narratives: The “Arabness Façade”

We noted that Muslim-​era historians open their Dhū Qār narratives with the remark that the battle was the “first time the Arabs equalled the Persians.” Ethnic difference is thus asserted upfront, and the narratives follow suit: the actions of the Bakr’s foe are homogenously written as being undertaken by “Persians” (ʿAjam or Furs): for example, “the Persians advanced,” “the Persians were defeated,” “they killed the Persians.”25 As a matter of syntax, therefore, the texts compel interpretation of Arabs on one side fighting Persians on the other. Likewise, the Sasanian army is described in signature “Persian” and clearly “non-​Arab” terms: for example, it has elephants, and its warriors 22 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:646–​48; Ibn ʿAbd Rabbihi, al-​ʿIqd, 5:248.

23 The three options are dispersed in the sources, see for the first, al-​Andalusī, Nashwat, 1:285, for the second, al-​Aṣbahānī, al-​Aghānī, 24:69, for the third, Ibn Ḥabīb, al-​Naqāʾiḍ, 2:644. See Miskawayh, Tajārib, 1:161 for the placement of the ʿIjl and Shaybān charges at the outset, with the single combat death of the Persian leader last. 24 Arabic literature often names the sources of historical anecdotes via detailing chains of authority of the stories’ transmission, but in the case of Dhū Qār, these are unfortunately lacking, and so we cannot identify each of the original sources nor ascribe them to particular tribal narrations.

25 All narratives adopt this style of homogenously referring to “the Persians”; the examples cited here are from the earliest extant narrative, Ibn Ḥabīb, al-​Naqāʾiḍ, 2:643–​44.

40

40

Peter Webb

are described as asāwira (the Sasanian Cataphracts/​knights).26 However, despite the textual ethnic homogenization, the more detailed descriptions of the composition of the Sasanian force reveal that it was, at most, only one-​third Sasanian soldiers under Persian-​ speaking commanders: the majority were Arabic-​ speaking frontier guards gathered from at least four tribal groups.27 The Arabian presence in the Sasanian army accords with the prevailing Sasanian policy of maintaining the frontier via agents, but that practice contradicts the expressly ethnic divide asserted by the Muslim-​era sources, and we can discern manifold narrative strategies employed to downplay the Sasanians’ Arabian supporters. For example, in narratives of the battle scenes, there is no mention of Arabian auxiliaries fighting alongside Sasanians; Balʿamī’s history is even explicit that all Arabs mobilized by the Sasanians defected during the battle, leaving the frontier agent, Hāniʾ ibn Qabīṣa, “alone.”28 Al-​Aṣbahānī’s account (which maintained the frontier-​incursion narrative as Dhū Qār’s trigger) is the only source to name Arabian tribal leaders killed fighting for the Sasanians;29 all other sources shift the focus squarely onto the deaths of Hāmarz and Hormuz Kharād, the Sasanian commanders. To further distance Arabness from the Sasanian ranks, some versions include a story that the Iyād tribe, which had been mobilized by the Sasanians, secretly informed the Bakr that they would desert once the battle began, thus assuring victory for their Arab brethren. That story, however, is not unanimously reported,30 and there is no poetry supporting the Iyād’s solidarity with the Bakr, which, given the salience of poetry in memorializing and communicating politicized messages, is suggestive that the side-​switching did not really occur. The Iyād’s defection is more likely a later device added to ethnically homogenize the belligerents, enabling narrators to eschew memorializing any killing between “Arabs,” and focus exclusively on Arabs killing Persians. In the same vein, some sources report that Qays ibn Masʿūd, one of the principal frontier agents of the Sasanians, also slipped away to join the ranks of Bakr (either before the fighting or between the battle’s first and second days).31 “Defection” deftly enabled historians to neutralize memories of Arabians allied with the Sasanians: memories such as Qays’ presence with the Sasanians were thereby not obliterated, but instead reoriented to show how “Arabs” innately wished to aid their “brethren” against the Persians. In the case of Qays, however, pre-​Islamic poetry (considered below) specifically chided him for not siding with the Bakr, and there are also reports that he led the centre of the Sasanian 26 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:640; Miskawayh, Tajārib, 1:160.

27 See, e.g., al-​Aṣbahānī, al-​Aghānī, 24:61–​62. 28 Baʿlamī, Tārīkhnāme, 2:823.

29 Al-​Aṣbahānī, al-​Aghānī, 24:70.

30 The Iyād’s flight is central in the narrative of Ibn Ḥabīb, al-​Naqāʾiḍ, 2:642, 644, al-​Andalusī, Nashwat, 1:285 and Ibn al-​Athīr, 1:489–​90; whereas neither the long account of al-​Aṣbahānī, al-​Aghānī, 24:67–​71 nor Ibn ʿAbd Rabbihi’s al-​ʿIqd, 5:248 report it. 31 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:640–​41; Balʿamī, Tārīkhnāme, 2:822.

41



Making War Ethnic

41

army,32 alongside contradicting reports that he was absent at the battle and died in Sasanian captivity.33 What Qays actually did is thus irretrievably lost, but the plethora of options reveals that the defection story was one of several retellings, and it assisted textual “ethnic cleansing” of Arabs from the Sasanian force. While Muslim-​era narrators contrived these stories to ensure that the narrative appeared as a binary Arab versus Persian contest, the original Battle of Dhū Qār, as has been argued by other scholars too, pitted Arabic-​speaking nomads on one side against a Sasanian frontier force which was itself largely composed of Arabic-​speaking patrols on the other.34 A deeper challenge to the putative Arabness of the Bakr forces is an abiding uncertainty over the nature of Arab identity itself at the dawn of Islam. Medieval-​era Muslim historiography does not question the Arabness of any pre-​Islamic Arabian group, and modern scholarship on Muslim historiography has been surprisingly lax in accepting those ethnic designations at face value. Despite the theoretical advances which have revealed the social constructedness of identities in manifold contexts across the globe and history, most studies until very recently did not seriously consider the constructedness of Arabness, nor did they use the available theory to investigate whether, or even how, Arab ethnic identity functioned among Arabian populations before Islam. Elsewhere, I examine the nature of Arab identity in pre-​Islamic Arabia and early Islam via applying anthropological theories of ethnogenesis to the evidence, and at present it seems quite clear that the circumstances necessary to create a pan-​Arabian sense of Arab community were absent at the time Dhū Qār was fought, and in support of, and in correspondence to the theory, there are almost no references to Arabness as an identity of “self” in pre-​Islamic Arabia.35 The form of Arabness which is familiar to us today was a creation of the Islamic period, emergent contemporaneous with the maturation of the source literature on Dhū Qār. The Muslim-​era texts’ adamant claims of the Arabness of the battle’s participants accordingly need testing via evidence more contemporary with the battle—​that is, the poetry of the combatants—​and it should not be assumed that “Arabness” was, or perhaps could have been, on the minds of warriors when they appraised their foe. The traditional presumption that pre-​Islamic Arabians were “Arabs” also overlooks a yet more significant issue concerning the identity of pre-​Islamic Arabians. Pre-​ Islamic Arabic-​language poetry did have a term which poets used to refer to their own community, but it was not ʿArab; instead, it was called Maʿadd.36 The contours of Maʿadd 32 Compare Balʿamī, Tārīkhnama, 1:822–​23 with Ibn ʿAbd Rabbihi, al-​ʿIqd, 5:248; Ibn al-​Athīr names Qays as amongst the Persian force, but says nothing of his involvement in the battle itself (al-​Kāmil, 1:489).

33 Al-​Aṣbahānī, al-​Aghānī, 24:75.

34 Fred Donner, “The Bakr b. Wā’il Tribes and Politics in Northeastern Arabia on the Eve of Islam,” Studia Islamica 51 (1980): 5–​38; Landau-​Tasseron, “Ḏū Qār.” 35 Webb, Imagining the Arabs.

36 Webb, Imagining the Arabs, 70–​77.

42

42

Peter Webb

as an Arabian social group at the dawn of Islam are detailed at length elsewhere,37 but a brief summary is in order since Maʿaddite identity is essential to make better sense of the Dhū Qār poetry. Central Arabian nomads, including the Bakr, regularly summoned the name “Maʿadd” to refer to the largest imagined collective group, and as an express form of kinship among themselves. Importantly, the communal boundaries of Maʿadd excluded the main Arabic-​speaking, semi-​nomadic groups who served as imperial frontier guards. The Syrian Ghassān in the service of Byzantium, the Iraqi Lakhmids and Ṭayyiʾ allied to the Sasanians, and the southern Arabian Kinda allied to Ḥimyar were all deemed genealogically non-​Maʿaddite. From the evidence of community expressed in pre-​Islamic poetry, the Bakr would have been part of the Maʿaddite people, whereas their foes at Dhū Qār were a mix of Persians, non-​Maʿaddite frontier guards, and other nomadic groups who would have been Maʿaddite too, but not close kin with the Bakr.38 By the time Arabic literature began to be recorded in the second/​eighth century, however, Maʿadd was no longer a functional label for a collective identity: it was being supplanted by “Arab,”39 and hence the ethnicity of Dhū Qār’s combatants could be retrospectively reworked and thoroughly Arabized. But quite how ethnicity functioned on the field at Dhū Qār itself is complex, and analysis in the following sections of Arabic poetry about the battle from pre-​Islamic times to the Abbasid era will facilitate suggestions. One final background consideration on identity concerns the Bakr itself. Heretofore, we have labelled the Sasanians’ opponents as “Bakr,” a tribal name connoting a northeast Arabian lineage group comprising several subgroups, notably the Shaybān, the Qays, the Dhuhl, the Taym Allāh, the ʿIjl, the Ḥanīfa, and the Yashkur. The subgroups were spread from the Gulf littoral to the fringes of the Syrian Desert: it is a vast region, and it bears questioning whether all the technical subgroups of the Bakr actually considered themselves kin and/​or were capable of acting in concert. Fred Donner rejects the Bakr’s putative pre-​Islamic unity, arguing that the above-​named clans were only collected under one “Bakr” tribal umbrella in the Umayyad era; thus Donner considers it is anachronistic to speak of “Bakr” as being present at Dhū Qār, and he would conceptualize the Sasanians’ foe as an ad hoc coalition of the Shaybān, the ʿIjl, and perhaps other elements.40 There is merit in Donner’s argument: scholarship has too readily accepted the corporate unity of Arabian tribal groups, and in practice, subgroups likely felt autonomous enough to act according to their own interests before those of the macro-​tribe.41 In support of 37 Peter Webb, “Ethnicity, Power and Umayyad Society: The Rise and Fall of the People of Maʿadd,” in The Umayyad World, ed. Andrew Marsham (London: Routledge, 2020), 65–​102.

38 A prime example of this is the Taghlib who were Maʿaddite and technically related to the Bakr, though they were inimical, and a story of an ancient pre-​Islamic conflict, the Basūs War, was memorialised as the explanation for the conflict between them notwithstanding their putative genealogical relation. For the Taghlib’s aligments before Islam, see Lecker, “Taghlib b. Wāʾil,” EI2. 39 Webb, “Ethnicity, Power and Umayyad Society,” 80–​87.

40 Donner, “The Bakr b. Wā’il Tribes,” 28–​36.

41 See Brian Ulrich, Arabs in the Early Islamic Empire (Edinburgh: Edinburgh University Press, 2019), for a theoretical discussion of tribes and tribal unity in Arabia at the dawn of Islam.

43



Making War Ethnic

43

Wā’il Bakr ‘Alī

Badan

Yashkur

Sa‘b •

‘Ukāba

Lujaym Hanīfa •

‘Ijl

Tha‘laba

Shaybān

Dhuhl

Qays

Figure 3.1 Outline genealogy of the Bakr ibn Wāʾil. Image by Peter Webb.

this view, it is noteworthy that al-​Ṣanāʾiʿ, one of the five fabled squadrons of Lakhmid cavalry (katāʾib), were (according to Muslim sources) recruited from Bakr lineages,42 and in the likelihood that they participated at Dhū Qār with the Sasanians, it follows that members of the same macro-​lineage group fought for both sides. While Donner’s argument could be extended to deny that the combatants at Dhū Qār were cognisant of sharing one “Bakr” tribal identity, pre-​Islamic poets do proffer some evidence that the name “Bakr” connoted a group identity among subgroups. “Bakr” therefore did mean something before Islam, but the extent to which Bakr as a tribal identity functioned remains an open question.43 At the present state of research, it seems that technical membership to “Bakr” via lineage was not of itself a decisive element of identity, nor did the Bakr qua tribe constitute a cohesive corporate body capable of effective collective action; however, members of several separate lineage groups did recognize some form of linkage via mutual claims of belonging to “Bakr.”

Dhū Qār in Pre-​Islamic Poetry: The Arabian Voices

To affirm Dhū Qār’s lofty status in Arabian battle history, Muslim-​era prose sources often cite Abū ʿUbayda (d. ca. 210/​825), the early Abbasid-​era collector of Arabian history and poetry, who counted Dhū Qār as one of the three greatest battles fought by 42 Muḥammad ibn Yazīd al-​Mubarrad, al-​Kāmil, ed. M. Aḥmad al-​Dālī, vol. 2 (Beirut: Muʾassasat al-​Risāla, 2008), 606. 43 For further discussion on the Bakr, see Webb, “Bakr b. Wāʾil”, EI3.

44

44

Peter Webb

the pre-​Islamic Arabs.44 The opinion, however, does not tally with critical study of pre-​ Islamic Arabic poetry. According to my survey of poets contemporary with the battle, Dhū Qār is sparsely mentioned: famous poets who lived at a remove from the Iraqi frontier, such as Labīd, ʿAmr ibn Maʿdī Karib, al-​Huṭayʾa, and al-​Shammākh al-​Dhubyānī, are silent on the battle, and even poets directly connected with politics and the rise of Islam such as al-​Ḥassān ibn Thābit and Kaʿb ibn Zuhayr make no mention of it either. These silences are noteworthy: if Dhū Qār really was as significant on a pan-​Arab, pan-​ Arabian scale as subsequently claimed, we might expect Arabian poets to summon the glory of victory or at least allude to the successful expedition into Iraq. The fact that they do not suggests that inner Arabian poets did not attach significance to Dhū Qār, and, perhaps, at a remove of considerable distance from Iraq, some may have been unaware of the battle’s occurrence.45 Another avenue to gauge the importance of an Arabian battle is to consider whether its victors were praised by itinerant court poets. Prominent leaders on Arabia’s frontiers attracted poets who sought patronage by composing praise poems memorializing their patrons’ battles. But again, in the case of Dhū Qār, there is scant evidence: my searches found neither praise of the men named as the heroes in the prose accounts (e.g., Hāniʾ ibn Qabīṣa or Ḥanẓala ibn Thaʿlaba), nor praise of the Bakr generally.46 In its immediate aftermath, therefore, Dhū Qār did not resonate very widely, nor did the victory elevate the Bakr to dominate the frontier and attract praise poets to their assemblies. When searching for pre-​Islamic poetry that does mention Dhū Qār, we find references restricted to a few poems by poets from the Bakr’s subgroups. Al-​Aʿshā Maymūn ibn Qays, a celebrated pre-​Islamic poet of the Qays branch of the Bakr, composed three poems mentioning the battle,47 and a clutch of verses by poets of the ʿIjl and Taym Allāh 44 Maʿmar ibn al-​Muthannā Abū ʿUbayda, Kitāb al-​Dībāj, eds. ʿAbd Allāh ibn Sulaymān al-​Jarbūʾ and ʿAbd al-​Raḥmān ibn Sulaymān al-​ʿUthaymīn (Cairo: al-​Khānjī, 1991), 78–​79.

45 Al-​Aṣbahānī reports a three-​line poem by an unnamed poet of the Rabīʿa mentioning Dhū Qār (al-​Aghānī, 24:74), but the poem lacks reference to the Bakr or the Persians, and instead ties Dhū Qār whimsically with the memory of al-​Nuʿmān. It is impossible to tell when this poem was composed: al-​Aṣbahānī offers no indication, but given that the poet is unknown and given the proliferation of poetry about Dhū Qār in the Umayyad period (noted below), the anonymous lines may have emerged later. Ibn ʿAbd Rabbihi, al-​ʿIqd, 5:251–​52 narrates a poem “sent by Laqīṭ al-​Iyādī to the Shaybān at the Battle of Dhū Qār,” but the ascription is mistaken: Laqīṭ is famous in Arabic literature for warning his own people about a Sasanian attack, yet this occurred some hundred years before Dhū Qār; Ibn ʿAbd Rabbihi confused these facts. 46 There is a reference to Wāʾil’s victory at the Battle of al-​Ḥinw in a pre-​Islamic poem ascribed to Shuraḥbīl ibn al-​Ḥārith: Abū Tammām, al-​Waḥshiyyāt, ed. ʿAbd al-​ʿAzīz al-​Maymanī al-​Rājkūtī (Cairo: al-​Maʿārif, 1987), 134; according to Abū ʿUbayda, al-​Ḥinw was one of the alternative names for the Battle of Dhū Qār (Ibn Ḥabīb, al-​Naqāʾiḍ, 2:638), but Shuraḥbīl was a much more ancient figure: he is associated with battles about a century before Islam, hence his “al-​Ḥinw” must intend a different battle.

47 Maymūn ibn Qays al-​Aʿshā, Dīwān, ed. M. Muḥammad Ḥusayn (Beirut: Dār al-​Nahḍa al-​ʿArabiyya, 1974), 233–​35, 277–​83, 309–​11. There is a fourth poem (al-​Aʿshā, Dīwān, 358–​61) which contains boasts about Dhū Qār, but Abū ʿUbayda considered this a false ascription, and he associates

45



Making War Ethnic

45

(a.k.a. Taym al-​Lāt) articulate similar boasts. Thus, the victors of Dhū Qār themselves considered the battle worthy of self-​praise, but the battle’s memory did not immediately escalate into an event cited by poets outside the Bakr. It thus appears that Dhū Qār was meaningful for the Bakr’s own story, but that is as far as its fame spread within one generation. Our hunch is supported from closer examination of the poetry and the ways it articulates senses of peoplehood. Al-​Aʿshā’s Dhū Qār poetry praises both his own subgroup, the Qays,48 and, principally, the Shaybān in typical pre-​Islamic warrior poetry style, whereby specific details are lacking in favour of generic tropes of vaunting bravery: Helms glistening, the host emerged from the valley Raising their banners aloft, They charged, we charged; Death was there Calamity swirled in full view.49

From the perspective of identity and peoplehood, al-​Aʿshā once refers to “a furious wave of Wāʾil” to describe the Bakr’s warriors:50 Wāʾil connotes the Bakr’s putative ancestor, and thus al-​Aʿshā identifies the combatants via macro-​tribal identity. “Clan Bakr” and “Bakr” are also specified by name in two Dhū Qār poems narrated by al-​Aṣbahānī, and if these are authentic voices from pre-​Islam, they further affirm that the combatants could be known by a “Bakr” identity. But al-​Aṣbahānī’s two poems are problematic: they are intimately tied with his prose narrative, and such poems bear higher risk of later fabrication to assist the narrative manipulation by subsequent storytellers.51 It is difficult to ascertain, but what is clear from survey of the poetry is that the term “Bakr” these lines with a different poet of the Shaybān; we consider this poem presently. A fifth poem (al-​Aʿshā, Dīwān, 349–​53) is of doubtful authenticity. It contains unusual vocabulary only common in Yemen [for example, the poem refers to the Sasanian commander as qayl (lines 12, 19]), and other sources actually ascribe the poem to Yemenis, particularly Sayf ibn Dhī Yazan, a Yemeni leader who opposed the Sasanian conquest of Yemen (Ibn Hishām, al-​Sīra al-​nabawiyya, ed. Muṣṭafā al-​Saqqā, Ibrāhīm al-​Abyārī, and ʿAbd al-​Shāfī Shalabī, vol. 1 (Beirut: Dār al-​Maʿrifa, n.d.), 88). Effectively, only one early narrator, Abū ʿAmr Isḥāq al-​Shaybānī (d. 206/​821), ascribed the poem to al-​Aʿshā (see al-​Aʿshā, Dīwān, 348), but Abū ʿAmr’s Shaybanid kinship is noteworthy here! 48 The Qays do not feature expressly in prose accounts of Dhū Qār, but one of al-​Aʿshā’s verses from this poem was adduced to prove the Qays were present (Ibn Ḥabīb, al-​Naqāʾiḍ, 2:644). The evidence is not very strong, however: the first-​person plural verb reported in al-​Naqāʾiḍ as proof that al-​Aʿshā refers to his people as “us” at the battle is recorded as a third-​person verb in al-​Aʿshā’s poetry collection where it reads “them”, i.e., the Shaybān. 49 Al-​Aʿshā, Dīwān, 309 (translation by Peter Webb, as are all poems translated herein).

50 Al-​Aʿshā, Dīwān, 283.

51 Al-​Aṣbahānī, al-​Aghānī, 24:63, 70. Al-​Aṣbahānī narrates that a third poem mentions the Bakr by name (al-​Aghānī, 24:64), but he expressly doubts its connection to Dhū Qār as it is ascribed to a poet whom al-​Aṣbahānī considers had died some time before the battle was fought. Al-​Aṣbahānī operated under the impression that the battle occurred in 624 CE, but most modern historians date it closer to 610, and hence the poem may actually be authentic, and constitute a real reference to “Bakr” as the means to identify members of the large tribe, but it is not certain.

46

46

Peter Webb

was not the most common form by which pre-​Islamic poets referred to the identity of the combatants: the majority of poems name individual subgroups. We have noted that al-​Aʿshā particularly singles out the Shaybān, and this subgroup is one of the two dominant names associated with the battle,52 the other is the ʿIjl. The preference for the clan names is suggestive of a more restrictive sense of identity whereby poets at the battle saw themselves representing their immediate kin, not necessarily the Bakr as a whole. For example, al-​Aghlab of the ʿIjl only mentions his own subgroup in his Dhū Qār poems: On the Day of Hormuz Kharād, they knew: They knew when the tribes advanced, They knew when scabbards flew and swords drawn, They knew when we met: we are the ʿIjl!53

We accordingly encounter two levels of identity at play: (i) in limited cases, unity of a broad group under the macro-​tribal label of Bakr, and (ii) in the main, the combatants are identified by narrower lineage bonds of Bakr subgroups. Another feature of the pre-​Islamic Dhū Qār poetry is the quantity of poems composed by poets from subgroups of the Bakr in praise of warriors from different Bakr subgroups. For example, al-​Aʿshā, from the Qays, devotes the majority of his Dhū Qār poetry to the Shaybān: May my camel and I be ransom for Dhuhl ibn Shaybān (Though we be meagre!) on the day of battle. At al-​Ḥinw, Ḥinw Qurāqir,54 they crashed blows Down upon Hāmarz’s ranks until the rout. Blessed are the eyes of those who saw this band, As they beat down foes thrusting from the plain With gleaming white helmets under high flags.55

Poets from the Bakr’s Taym Allāh praise the ʿIjl, such as Ḥuraym ibn al-​Ḥārith of the Taym Allāh:56 52 See Ibn Ḥabīb, al-​Naqāʾiḍ, 2:643.

53 Al-​ʿAskarī, al-​Awāʾil, 2:189.

54 These are apparently names for Dhū Qār. Ibn Ḥabīb, Naqāʾiḍ, 2:638 lists eight different names by which the battle was known. 55 Al-​Aʿshā, Dīwān, 309.

56 The authenticity of Ḥuraym’s poem is not straightforward. An almost identical poem is cited in Arabic literature in a totally different Hijazi political context ascribed to ʿAbd Allāh ibn al-​Ziʿbarā of the Quraysh: ʿAbd Allāḥ Ibn al-​Ziʿbarā, Dīwān, ed. Yahyā Wāhib al-​Jabūrī (Beirut: Muʾassasat al-​Risāla, 1981), 50. The similarity between the two poems has been noted, with an opinion that both may be authentic, on the basis that the line constituted a praise trope circulating amongst Arabian poets at the dawn of Islam: Johanna M. Coster, The Good, the Bad, and the Ugly: Allegiance and Authority in the Poetical Discourse of Muhammad’s Lifetime (PhD diss., University of Groningen, 2018), 92, 186–​87. Alternatively, the correspondences could have been added during the Muslim-​ era process of collection, especially since the extent of crossover between Ḥuraym’s poem and

47



Making War Ethnic

47

The Lujaym are a folk of power and wealth,57 A clan of means, their ancient merits unassailable, They protected our women at Dhū Qār, As lords defend pregnant pure-​bred camels. Always answering the cry ‘Advance!’ Can any but the Lujaym’s best prevent ignominy?58

In terms of peoplehood, the corpus of intra-​Bakr poetry suggests that members across subgroups of the Bakr recognized one shared kinship, and that members of a subgroup which did not fight at Dhū Qār felt that the victory of their combating kinsmen from the ʿIjl and Shaybān entitled all subgroups to share the honour, piggy-​backing off the combatants’ glory, so to speak. This suggests a degree of emotive significance attached to the idea of “Bakr” as a large-​scale identity, but equally, the pre-​Islamic poetry also reveals pushback from the ʿIjl and Shaybān against those non-​combatant groups claiming a share in the glory. For example, Abū Kalba of the Taym Allāh expressly lauds the ʿIjl and Shaybān, but also mentions the name “Lahāzim” as part of the victorious force—​the Lahāzim were an alliance that included the Taym Allāh alongside the ʿIjl and Shaybān. These verses apparently piqued the ire of al-​Aʿshā, who composed a counter: Someone inform Abū Kalba of the Taym (By God they’re an iniquitous group): The Shaybān protected you from the fight, While you were like a dog in a cave, barking.59

Herein is a noteworthy nebulosity of Bakr identity: on the one hand, it had a recognized set of subclans who knew of themselves as being related to each other as kin-​Bakr, and some capitalized on this network by attempting to absorb credit from Dhū Qār for themselves by praising other clans, whereas the clans which did participate in the fighting seem less willing to share the glory with non-​combatants, even if those were technically kin-​Bakr. The tension tallies with anthropological observations of contemporary tribes, whereby shared lineage operates primarily as a potential force: it enables groups to act in concert on an ad hoc basis when necessary, but in no way does it guarantee cohesion or necessitate that one subgroup share equally with others. In the same vein, Ibn ʿAbd Rabbihi reports that the ʿIjl and the Shaybān contended that only their members fought at Dhū Qār, and they rejected claims of other subclans of the Bakr.60 This squabbling occurred in the Abbasid period, and over the effluxion of Ibn al-​Ziʿbarā’s is inconsistently recorded, perhaps suggestive that the additions were added anachronistically to Ibn al-​Ziʿbarā’s poem. The question remains open, but at least Ḥuraym’s verses bear no obvious signs of anachronism: they reflect the tenor of the more securely datable pre-​ Islamic poetry about Dhū Qār. 57 “Lujaym” here refers to the ʿIjl; their genealogy was ʿIjl ibn Lujaym; see Figure 3.1. 58 Al-​Aṣbahānī, al-​Aghānī, 24:72.

59 Al-​Aṣbahānī, al-​Aghānī, 24:73. Ibn Ḥabīb, al-​Naqāʾiḍ, 2:645 narrates another version of this exchange, with different poems. 60 Ibn ʿAbd Rabbihi, al-​ʿIqd, 5:248.

48

48

Peter Webb

some 200 years since the battle, various poems were circulating (considered below) that expanded the range of subgroups alleged to have fought at Dhū Qār. The ʿIjl and Shaybān’s “monopoly” was accordingly rejected by the later scholars, but if we consider the pre-​Islamic poems alone, the names most commonly referenced are indeed just ʿIjl and Shaybān.61 Moreover, the ʿIjl and Shaybān are also the chief protagonists in the prose accounts, and hence it seems most likely that they were in fact the only clans which actually fought at Dhū Qār. Perhaps this also explains why al-​Aʿshā refers to the large-​scale tribal identity of “Wāʾil” in his poetry: his immediate kinsmen of the Qays were not represented at Dhū Qār in sufficient numbers to legitimately claim honour for himself, hence he enthusiastically lauded the Shaybān both explicitly by name and implicitly along with the other Bakr kin-​groups in a targeted attempt to curry favour and build bridges of real alliance from the theoretical bonds of kinship. The poets of the ʿIjl, however, had no need to appeal to the macro-​tribal Bakr to affirm their glory, since their own warriors had physically fought at the battle. Poetry’s testimony thus reveals that the combatants at Dhū Qār conceptualized themselves as members of their particular clans above all else, but they were aware of a select body of other groups with whom they perceived kinship under an umbrella identity of Bakr. Belonging to Bakr opened a network of interrelations, though this did not translate into one cohesive corporate identity in practice. Dhū Qār’s initial memorialization was as a triumph of the Shaybān and ʿIjl. In keeping with poetry’s expression of identity at clan levels of lineage, none of the pre-​Islamic Dhū Qār poetry refers to an “Arab” identity. In contrast to the later Muslim-​ era prose claims of the battle’s Arab–​Persian ethnic binary, the pre-​Islamic poets never invoke that dichotomy. The absence of any express large-​scale identities in the pre-​ Islamic poetry is a crucial observation: it underlines that those who fought at Dhū Qār did not consider themselves representatives of the whole macro-​tribe of Bakr, let alone any form of peoplehood greater than their own immediate clan. Whether or not they ever recognized themselves as “Arabs” is a question extending beyond this chapter, but the poetry is patently clear that the combatants left no express indication of Arabness when memorializing their victory. Given that many on the Sasanian side were Arabians too, the absence of perception of an ethnic divide between the opposing forces is not surprising, and must stand as a key corrective to the later ethnic interpretation of Abbasid-​era historiographers. As noted above, central Arabian poets at the time of Dhū Qār did widely express shared belonging to “Maʿadd” in terms indicative that Maʿadd was the “people” with whom they identified as a super-​tribal identity.62 In the case of Dhū Qār poetry, however, even Maʿaddite identity is muted: none of our poems refer to the combatants as representatives of Maʿadd, no poet claims victory in the name of the Maʿadd, and 61 They are the only groups repeatedly mentioned in the battle poetry recorded in Ibn Ḥabīb, al-​Naqāʾiḍ, 2:641–​43; see also the notes above. 62 For details of Maʿaddite peoplehood, see Webb, “Ethnicity, Power and Umayyad Society,” 66–​71.

49



Making War Ethnic

49

the name is in fact absent in all poetry barring one line by al-​Aʿshā, which is itself of problematic ascription. It reads: If all of Maʿadd had mustered with us at Dhū Qār, There would have been glory for them.63

The verse intends that the honour won by the Shaybān at Dhū Qār was so great that even had all of Maʿadd fought at the battle, there would be enough glory to go round. This is of course hypothetical: only some clans of the Bakr were actually present, and hence the rhetorical purpose is to emphasize the prodigious amount of excess glory won by the Shaybān. The sentiment expressly renders Maʿadd as an audience: the poet invites them to behold his own clan’s glory. While Maʿadd is invited to marvel at the victory, the actual combatants were representing themselves, and they do not intend to share the merit: their victory was for themselves alone. The reference to Maʿadd does provide an indication of the group which the poet considers to comprise the largest possible collective. Rhetorically, the poet seeks to articulate that the honour of Dhū Qār was great enough to be shared among “everyone,” and the fact that he summoned the name “Maʿadd” as the byword for “everyone” reveals the status of Maʿadd as the super-​tribal community in the eyes of the Arabian poet. The verse’s ascription to al-​Aʿshā, however, is not certain: it occurs within a group of lines about Dhū Qār at the end of a long poem on a different topic,64 and the modern editor of al-​Aʿshā’s poetry considers that the Dhū Qār section was originally composed by an Umayyad-​era poet of the Shaybān, ʿAbd Allāh al-​Nābigha al-​Shaybānī, and only subsequently appended to al-​Aʿshā’s ode.65 Evaluating authenticity becomes difficult since the fourth/​tenth century al-​Aṣbahānī does ascribe these verses to al-​Aʿshā,66 and so we are left with an open question. For our purposes of examining peoplehood, however, the verse attests that the largest social group mentioned in pre-​Islamic Dhū Qār poetry is possibly Maʿadd, and if that line is anachronistic, then every reference to “self” in authentic pre-​Islamic poetry revolves around the narrow terms of immediate kin groups. The operative identity of the Arabian combatants thus clearly skews toward clan, not macro-​tribe (Bakr), let alone ethnos (Maʿaddite or Arab).

Pre-​Islamic Poetry: The “Enemy” Identity

Given that recognition of ethnic identity takes shape via awareness of an “other” opposed to “self,” the investigation into the identity of the pre-​Islamic combatants at Dhū Qār need also consider how they depicted their foe, and herein the poetry proffers intriguing descriptions of the Sasanian force. Overall, there are few specifics: the poets 63 Al-​Aʿshā, Dīwān, 361.

64 Al-​Aʿshā, Dīwān, 361.

65 See al-​Aʿshā, Dīwān, 358. Such reworkings, and the adding and subtracting of lines, are quite common phenomena in early Arabic poetry. 66 Al-​Aṣbahānī, al-​Aghānī, 24:75.

50

50

Peter Webb

do not name the Arabian tribes whom they faced, and their handling of Persian-​ness is both vague and tantalizing. The pre-​ Islamic poems do not hearken Persian-​ ness in a materially “ethnic” sense, that is, actively “othered” to the Arabian warriors. There is neither intimation that the Sasanians lost because of ethnic inferiority, nor that Dhū Qār represented a culmination of protracted conflict between two different peoples, or a first Arabian victory over Persians. Such macro-​historical vision was beyond the horizons of the pre-​Islamic poets contemporary with the battle, and, pertinently, no poems represent the Sasanians as culturally different from the Bakr in appearance, habits, weaponry, tactics, or otherwise. The Bakr poets describe the battle in terms redolent with any inter-​Arabian conflict, without evident “othering” of their foe. This is perhaps a function of (i) the large Arabian contingent allied to the Sasanians, and (ii) the fact that the poets of the Bakr subgroups do not express their identity in “Arab” or “Maʿaddite” terms. Arabness versus non-​Arabness was not their concern: they were thinking primarily about clan, and took no discernible opportunity to remark upon the “ethnic” separateness of their foe. The ways Bakr poets name the battle do indicate that they were fighting the Sasanians, however. The poets only infrequently summon the name “Dhū Qār” itself: they instead allude to it via terms such as Yawm Kisrā—​the “Day of Khosrow,”67 or (once) Yawm Khanābazīn—​the “Day of Hormuz Kharād.”68 “Day” is the common Arabian byword for “battle,” and so the poets are clear that they were engaged in a fight against the Sasanians, referenced either by the title of the shah or the name of his commanders. Similarly, Khosrow is the byword for the foe in two poems: first, by Abū Kalba of the Taym Allāh: The horsemen of the ʿIjl disdained To leave the field to Khosrow.69

The other, by al-​Aʿshā relates:

Who will inform Khosrow when my Dismaying messages come in: ‘I say we will not surrender our boys As hostages to corrupt as he has done before.’70

Note here that the Sasanian monarch is a distant figure: neither poet intends that the Sasanian Empire was threatened by the battle, that Khosrow was the Bakr’s intended target, and neither refer to traits of Persian-​ness—​Khosrow is simply the nomenclature for the foe. Similarly, the Sasanian commander Hāmarz is named in a pair of al-​Aʿshā’s praise poems of the Shaybān: 67 Al-​Aṣbahanī, al-​Aghānī, 24:74.

68 Al-​ʿAskarī, al-​Awāʾil, 2:189. For explanation of the name, see above, note 11. 69 Al-​Aṣbahānī, al-​Aghānī, 24:73.

70 Al-​Aʿshā, Dīwān, 279.

51



Making War Ethnic

51

They were sufficient, when Hāmarz charged, His flag fluttering above him, a diving eagle.71

Again, the poetry engages the names of Sasanian leaders without depicting them as different from the warriors of the Bakr. The knowledge of the names of the Sasanian generals, coupled with the otherwise abiding lack of engagement with Persian-​ness, entails that the poets articulate Sasanian politics, not Persian ethnicity. The poets knew the Sasanian Empire as the land of its shah and his commanders, but since their actual foe were a mixture of Sasanian and Arabian troops, the opportunity to perceive Dhū Qār as ethnic did not arise, and instead they chose to memorialize the conflict as one against the power of the Sasanian regime as a political entity. The hypothesis finds further support in that terms expressly connoting “Persian” as people are strikingly absent in the pre-​Islamic Dhū Qār poems. Arabic primarily refers to Iranians as either Furs (Persian) or ʿAjam/​Aʿājim (non-​Arabic speakers); the first term is absent, and a single reference to the second appears in the problematic poem noted above by al-​Aʿshā, which is said to have been composed in the Umayyad era. It reads: A mighty army of the vainglorious king Of the Aʿājim with pearls in their ears When they drew their hands to their bows We clasped our swords and heads flew.72

The view of the editor of al-​Aʿshā’s poetry that deems the lines anachronistic seems right: the tropes invoked—​reference to Khosrow’s villainy and physical descriptions of Persians’ axiomatically “non-​Arab” garb—​become established in Umayyad-​era poems considered below, whereas such elements are lacking across all other pre-​Islamic poems. The style in which al-​Aʿshā derides Qays ibn Masʿūd for siding with Khosrow is also noteworthy. We noted that Qays was a Sasanian border agent, and his status at the battle is unknown thanks to contradictory prose accounts; as for al-​Aʿshā, he castigates Qays: If you had been satisfied with Shaybān, You would have spacious tents, a thronging tribe, and massed cavalry, … But you foolishly left them, though you were their leader. I hope I hear no more from you!73

These lines from a well-​established poem suggest that Qays did fight for the Sasanians, and for our investigation, it is noteworthy that al-​Aʿshā neither depicts him as a “traitor” who crossed ethnic boundaries, nor as a traitor at all; rather, Qays emerges as a fool for not trusting the might of Shaybān’s warriors. And so the survey of poetry contemporary with Dhū Qār ends upon its most consistent theme: the Arabian 71 Al-​Aʿshā, Dīwān, 309. The second instance is noted above, note 54.

72 Al-​Aʿshā, Dīwān, 361.

73 Al-​Aʿshā, Dīwān, 233–​34.

52

52

Peter Webb

combatants at the battle default to identifying themselves via clan allegiances. Even the macro-​tribe of Bakr and the people of Maʿadd are not the primary labels which they used, and the putative difference between Arabian and Persian is not at all a salient feature. The magnification of Dhū Qār’s significance on an ethnic level would take time and a seismic geopolitical change that shook the region over the following generations.

Dhū Qār in Early Islamic and Umayyad Poetry

Within two decades of Dhū Qār, Arabian armies again crossed the Euphrates, and this time, within the framework of Muslim conquest, they came to stay. A series of battles in Iraq, culminating at al-​Qādisiyya in ca. 15/​636 or 16/​637 and Nihavānd in 21/​642, drove the Sasanians over the Zagros, opened Iraq for the conquerors’ settlement, and signalled the end of the Sasanian Empire. As noted at the outset of this chapter, medieval Muslim historiographers interpreted Dhū Qār as the portentous foreshadowing for the Muslim conquerors’ success, but from the evidence of the sizable quantity of conquest-​ era poetry, it becomes difficult to maintain that the conquerors themselves actually had Dhū Qār on their minds or were cognisant of consummating the Bakr’s earlier victory. According to my searches of al-​Qādisiyya poetry, none draw connection with Dhū Qār: the oeuvre of the celebrated warrior poet ʿAmr ibn Maʿdī Karib, for example, contains two poems vaunting his prowess at al-​Qādisiyya, and refers to the ethnonym ʿajam to express the Persian-​ness of his foe, yet the poet makes no allusion to Dhū Qār.74 The absence of Dhū Qār in the conquest poetry tallies with our observations of Dhū Qār’s status in pre-​Islam. We saw that pre-​Islamic poets did not memorialize the battle as an Arab victory over Persians, and at the dawn of the conquests, Dhū Qār simply did not yet exist in memory as an exemplar of Arabian frontier victory. Moreover, Dhū Qār was memorialized among the subclans of the Bakr, and during the Muslim conquest of Iraq, a sizable contingent of Bakr forces fought with the Sasanians.75 Accordingly, the identity of the Muslim conquerors was not in fact aligned with the identity of the victors of Dhū Qār: the conquerors were not from the Bakr, and therefore they had no actual connection to or claim over Dhū Qār’s memory. Though a grand sweep of Arabian-​Persian history sees superficial similarities between the battles of Dhū Qār and al-​Qādisiyya, the crucial consideration of peoplehood differentiates them. The battles involved different communities: Dhū Qār engaged clans of the Bakr, whereas al-​Qādisiyya was won by Muslims from an array of different lineage groups. While the Bakr and the Muslims seem sufficiently homogenous as all “Arabians” from an outsider’s perspective, they possessed different traditions and senses of community, and the absence of reference to Dhū Qār in poetry composed contemporaneously with al-​Qādisiyya illustrates the

74 ʿAmr ibn Maʿdī Karib, Shiʿr ʿAmr ibn Maʿdī Karib al-​Zubaydī, ed. Muṭāʿ al-​Ṭarābīshī (Damascus: Majmaʿ al-​Lugha al-​ʿArabiyya, 1985), 114–​15, 172–​74. 75 Donner, “The Bakr b. Wā’il Tribes,” 28–​30.

53



Making War Ethnic

53

separateness of communal memories, and thereby the differing identities of Arabian groups in nascent Islam. Dhū Qār appears to have remained proprietary to the Bakr in the early Umayyad era too. The battle is not frequently mentioned in poetry, and is only found in poems ascribed to poets from the Bakr subgroups. Akin to the pre-​Islamic context, the early Umayyad-​ era Bakr subgroups made competing claims about participation at Dhū Qār, epitomized in the example of a poet of the Yashkur, Suwayd ibn Abī Kāhil, who, around the time of ʿĀmir ibn Masʿūd al-​Jumaḥī’s governorship in Kufa (ca. 60–​65/​680–​684), lampooned the Shaybān, claiming, among other boasts, that a warrior of the Yashkur killed Hāmarz, the Sasanian commander at Dhū Qār.76 As noted above, the identity of Hāmarz’s killer was unevenly reported in the sources, and the only evidence that he was felled by a member of the Yashkur is this poem. The Yashkur are otherwise essentially absent in the other battle narratives, and so this seems yet another attempt of a Bakr subgroup to muscle into memories of Dhū Qār. The continued dispute over the battle’s memory demonstrates that Dhū Qār remained a memorable part of the Bakr’s story in early Islam, though the details were sufficiently open-ended to permit creative reinterpretation. By the late first/​seventh century, however, and contemporaneous with the rise of the Umayyads’ Marwanid dynasty, poetry does begin to reveal a novel proliferation of Dhū Qār’s memory with new significations. For example, al-​ʿUdayl ibn Farkh, a Marwanid-​era poet of the ʿIjl, boasts: When the people recount all the battle days, None I hear are more glorious than Dhū Qār.77

In contrast to the pre-​Islamic poets of the Bakr, al-​ʿUdayl explicitly compares Dhū Qār to other Arabian battles. ʿAdīl’s choice to compare Dhū Qār with the victories of others is indicative of a novel communal context whereby groups marshalled their past victories in efforts to jockey for status. The Marwanid-​era Bakr are no longer alone and Dhū Qār is no longer an intra-​Bakr matter: the battle became a means to boast against non-​Bakr groups.78 Dhū Qār thereby becomes part of a larger pool of collective memory, and this process finds further expression in a line composed by a Marwaind-​era poet of non-​Bakr lineage, al-​Marrār ibn Saʿīd of the Faqʿas, who cites Dhū Qār metaphorically, comparing his pangs of lovesickness to the travails of fighting at Dhū Qār.79 No previous poet cited Dhū Qār apolitically or proverbially as a generic “great struggle”; the fact that the battle can stand as a decontextualized metaphor indicates a newfound currency of its memory among non-​ Bakr groups. 76 See al-​Aṣbahānī, al-​Aghānī, 14:117–​20. 77 Al-​ʿAskarī, al-​Awāʾil, 2:190.

78 The importance of the Marwanid period for early consolidation of different Arabian identities into a more cohesive sense of Arabness is discussed in Webb, Imagining the Arabs, 126–​56.

79 Abū al-​ʿAbbās Aḥmad Thaʿlab, Majālis Thaʿlab, ed. ʿAbd al-​Salām Muḥammad Hārūn, vol. 1 (Cairo: al-​Maʿārif, 2016), 208.

54

54

Peter Webb

Reference to Dhū Qār by poets from non-​Bakr lineages becomes a salient feature in the Marwanid period. One of the era’s greatest poets, al-​Farazdaq of the Dārim (d. ca. 110/​728) expressly invokes memory of Dhū Qār to praise the Shaybān, and al-​Qaṭāmī of the Taghlib (d. ca. 101/​719) likewise expressly names Dhū Qār when extolling the Bakr as a macro-​lineage.80 The pair of poems is doubly noteworthy for our purposes. First, both poets use the exact same half-​line to open their praise verses: They [the Shaybān] dismounted at Dhū Qār and fought steadfastly …

Both poets continue the line with their own compositions, but we witness here the emergence of a poetic stock phrase to signal praise of the Shaybān, revealing that Marwanid-​era poets were sharing material about Dhū Qār, entailing that the battle was becoming established in cultural repertoire. Second, al-​Qaṭāmī’s clan, the Taghlib, had been enemies of the Bakr at the time of Dhū Qār and fought with the Sasanians, yet al-​Qaṭāmī makes no reference to his own people’s historic opposition to the Bakr: his lines only mention the “battalions of Khosrow” (katāʾib Kisrā) as the Bakr’s enemy. In the same poem, al-​Qaṭāmī cites other past events that evidence brotherly relations between the Taghlib and the Bakr, and thus, at a remove of only seventy years from the battle, the memory of enmity between Taghlib and Bakr was being overlooked, and emphasis instead placed on unity between the former foes, abetted by memorializing common antipathy toward the Sasanians. Further seminal changes in Dhū Qār’s status in memory and community appear from a poetic duel recorded between the two great poet-​rivals, Jarīr (d. 111/​729) and al-​Akhṭal (d. ca. 92/​710), wherein al-​Akhṭal chides Jarīr’s kin: Did you assist Maʿadd on the ferocious day, Like we supported Maʿadd at Dhū Qār?81

Al-​Akhṭal elevates Dhū Qār to the status of Maʿadd’s signature collective victory. Akin to Shakespeare’s transformation of Henry V’s St Crispin’s Day escapade into a retrospective national triumph which any able-​bodied Englishman should wish to have attended, al-​Akhṭal’s version of the battle presents it as waged by Maʿadd, in order to deride Jarīr’s Ma’addite tribe for not participating. It is an obvious departure from the earlier poetry which clearly restricts Dhū Qār to subgroups of the Bakr, and moreover, al-​Akhṭal was from the Taghlib—​they had fought against the Bakr at Dhū Qār, but al-​Akhṭal reverses this since the Marwanid-​era Taghlib and the Bakr were both part of Maʿadd and their interests were unified under the Caliphate; hence Dhū Qār was reimagined as a victory of Maʿadd. Jarīr’s response to al-​Akhṭal engages an even more extraordinary twist by inserting his own tribe, the Tamīm, into the memory of the battle: 80 Al-​Aṣbahānī, al-​Aghānī, 21:385; Al-​Qaṭāmī, Dīwān, ed. Ibrāhīm al-​Sāmarrāʾī and Aḥmad Maṭlūb (Beirut: Dār al-​Thaqāfa, 1960), 125–​26. See also ʿAli ibn al-​Faraj al-​Baṣrī, al-​Ḥamāsa al-​Baṣriyya, ed. ʿĀdil Sulaymān Jamāl, vol. 1 (Cairo: al-​Khānjī, 1999), 75. 81 Al-​Akhṭal, Dīwān, 421.

55



Making War Ethnic

55

I am a Muḍarī at root. You cannot hope to vie with me and my prestige! We sent the horsemen to battle at Dhū Bahdā and Dhū Najab And we stood out on the morn of Dhū Qār.82

Jarīr’s claim puzzled the later commentator Abū Tammām (d. 231/​845)83 who wondered how a Tamīmī tribesman could assert connection to the battle. Abū Tammām reasoned that there must have been a separate battle of the same name pitting the Tamīm against the Bakr,84 but Abū ʿUbayda (d. ca. 210/​825) attempted a bolder justification, citing a narrative that some Tamīm tribesmen were captured by the Shaybān before Dhū Qār, and, on the eve of battle, they offered to fight in return for their freedom, and, according to Tamīmī partisans like Jarīr, they acquitted themselves manfully.85 This anecdote harmonizes with the other poetry of the Marwanid era evidencing a novel allure of Dhū Qār alongside wholesale reworking of its memory that raised the battle’s reputation to a pan-​Maʿaddite achievement into which different lineage groups sought to insert themselves in any way possible, even by claiming that they arrived at the battle as prisoners. Extending the St. Crispin’s Day analogy, Jarīr’s poem is reminiscent of Pistol’s duplicitous intention to rebrand ignoble wounds as scars from Agincourt! Al-​Akhṭal’s poetry also contains embellishment intoning that Dhū Qār achieved total victory over the Sasanians: The squadrons of Khosrow approached enraged, But we annihilated them, destroying all tyrants!86

The Khosrow reference reflects the language of pre-​Islamic poetry in which the shah’s name was a byword for the Sasanian army, but al-​Akhṭal’s tone is more vehement. His reference to the Sasanians’ annihilation could be standard poetic bombast, but the engagement of stronger emotions and stress on Khosrow’s tyranny are themes which pre-​Islamic poets did not elaborate; with al-​Akhṭal enmity against the Sasanians is elevated, and the fact that the non-​Bakr poet al-​Akhṭal counts Dhū Qār as his people’s primary victory underlines that the battle’s memory was shifting toward pan-​Maʿaddite glory against a now more uniformly projected Persian-​Sasanian foe. Explaining the shift has a straightforward logic. Marwanid-​era poets lived at two generations’ remove from the Muslim conquest of the Sasanian Empire, and poets such as al-​Farazdaq, Jarīr, and al-​Akhṭal were employed by the descendants of the conquerors. As the Marwanids looked back into the past, they possessed sufficient 82 Abū Tammām (attrib.), Naqāʾiḍ Jarīr wa-​l-​Akhṭal, ed. Anṭūn Ṣāliḥānī (Beirut: al-​Maṭbaʿ al-​Kāthulīkiyya li-​l-​Ābāʾ al-​Yasūʿiyyīn, 1922), 143. 83 Though ascribed to Abū Tammām, the book in which this opinion is contained is problematic, and may instead have been written by his contemporary al-​Aṣmaʿī. 84 Abū Tammām (attrib.), al-​Naqāʾiḍ, 143–​44. 85 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:646.

86 Abū Mālik Ghiyāth ibn Gawth al-​Taghlibī al-​Akhṭal, Shiʿr al-​Akhṭal (Riwāyat al-​Sukkarī ʿan Ibn Ḥabīb), ed. Fakhr al-​Dīn Qabāwa (Damascus: Dar al-​Fikr, 1996), 421.

56

56

Peter Webb

hindsight to alight upon Dhū Qār as marking the “beginning of the end” of Sasanian power, and hence Marwanid-​era poets and patrons were in a position to elevate the battle’s significance beyond what its actual combatants could have imagined. In terms of peoplehood, it is noteworthy that al-​Akhṭal and Jarīr invoke Maʿadd without mention of “Arab” to describe the victors at Dhū Qār. This supports the hypothesis that Maʿaddite identity only gradually evolved into “Arab” in early Islam: communal boundaries had changed by the Marwanid era, such that glories associated with individual tribes in pre-​ Islam were now counted as the property of a much wider super-​tribal identity, but the primary sense of Marwanid-​era peoplehood was Maʿaddite, not Arab.87 Other Marwanid-​era poems exhibit similar shifts, emphasizing Dhū Qār as a grand Persian defeat without assertion of the Arabness of its victors. Ibn Ḥabīb records two poems attributed to minor Muslim-​era poets from the ʿIjl that refer to Khosrow as jabbār (the despot), reflecting influence of the Qur’anic portrayal of Moses’s Pharaoh to whom Khosrow was linked in Muslim literature.88 The poets also paint the victory as a crushing blow to Persian imperial might: We took their booty, our cavalry was grim, On the day we stripped all Khosrow’s knights (iswār) of their armour.89

Pre-​Islamic poets neither referenced the evil of Khosrow, nor the whole might of Sasanian Iran, and they did not detail the Sasanian army in terms that “othered” it from the Bakr; the Marwanid-​era poetry’s iswār, asāwira cataphracts hearkens a term which Muslim-​era literature summoned as a byword for Sasanian nobility. The ascent of express Persian-​ness and reference to crushing defeat of the Sasanian Empire situate Dhū Qār within a sense of conquest history, and the poetry’s vocabulary embedding the stereotypical topos of vainglorious Persian shah versus plucky Arabian warrior (Maʿaddite, not yet Arab) helps propel the foreshadowing of conquest in the memorialization of the pre-​Islamic battle. A final consideration relevant to late Marwanid-​era poetry arises in a poem by Abū al-​Najm of the ʿIjl (d. ca. 130/​748). Abū al-​Najm enumerates his clan’s glory, and, in keeping with the venerable tradition of the ʿIjl dating back to pre-​Islam, he accords Dhū Qār pride of place, with these verses: On the morn of battle, when ranks are drawn We are leading from the back, and defending the back. On the day of Dhū Qār, we outshone all the Arabs.90

His summoning of the ethnonym “Arab” to boast of the ʿIjl’s performance appears to be the earliest verse in which Dhū Qār is explicitly associated with Arabness. Drawing

87 For more details on the transition of Maʿadd to ʿArab as an ethnonym and term of self-​identity, see Webb, “Ethnicity, Power and Umayyad Society,” 74–​84. 88 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:646.

89 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:646.

90 Abū al-​Najm al-​ʿIjlī, Dīwān, ed. Muḥammad Adīb ʿAbd al-​Wāḥid Jumrān (Damascus: Majmaʿ al-​Lugha al-​ʿArabiyya, 2006), 87.

57



Making War Ethnic

57

conclusions is difficult, however, since the entire poem runs to twenty-​two lines, and it is only reported in a fragmentary fashion in premodern Arabic literature. No source contains the whole poem, and the poem’s modern editor consolidated its verses from a wide array of sources. While many of the poem’s lines were recorded in numerous sources, abetting confidence of their authenticity, the line mentioning Dhū Qār and “Arabs” is solely narrated in the late source, al-​Ḥamāsa al-​Shajariyya, complied by Hibat Allāh ibn ʿAlī ibn al-​Shajarī (d. 542/​1148).91 As the line is a unicum from a sixth/​twelfth-​ century text, whereas much of the rest of the poem is well and repeatedly represented in earlier literature, there is a strong possibility that the particular line was not originally composed by Abū al-​Najm, and its reference to “Arabs” is anachronistic, and not the earliest poem to memorialize Dhū Qār as an Arab battle; especially since all other Marwanid-​era poets associate Dhū Qār with Maʿadd. However, the end of the Marwanid-​ era was a period of shifting ethnonyms in Arabic poetry, and the first mentions of “Arabs” in poetry to intend a people do legitimately date from the early second/​eighth century too, when “Arab” was also beginning to supplant “Maʿadd” as the term for the Caliphate’s elite community. It would not have been unprecedented, therefore, for Abū al-​Najm to identify the people at Dhū Qār as “Arabs,” but even if the line is genuine, he would nonetheless be the only extant Umayyad-​era poet to do so: the effective and express association of Dhū Qār with Arab identity was achieved under the Abbasids.

Dhū Qār in the Abbasid Period

The first explicit Arabness association with Dhū Qār can be dated to the late second/​ eighth century. The compiler of Arabian history, Abū ʿUbayda, is identified as the narrator of a hadith reporting the Prophet’s words that Dhū Qār “is the first battle in which the Arabs have become the Persians’ equal.”92 The hadith lacks isnād (chain of authorities), and so it is almost certainly an Abbasid-​era invention, but it spread during the third/​ninth century, acquiring both isnād and narrative additions along the way. Ibn Saʿd (d. 230/​845) reports Muhammad saying: “On this day the Arabs diminished Persian kingship,”93 and Khalīfa ibn Khayyāṭ (d. ca. 240/​853–​854) further elaborates Muhammad’s words: “Dhū Qār is the first battle in which the Arabs became the Persians’ equal; they were granted victory through me.”94 This Prophet-​assisted representation of Dhū Qār was included in the hadith collectors Ibn Ḥanbal’s Faḍāʾil al-​Ṣaḥāba95 and al-​Bukhārī’s al-​Tārīkh al-​kabīr, though neither collector included the hadith in their 91 Hibat Allāh ibn al-​Shajarī, al-​Ḥamāsa al-​Shajariyya, ed. ʿAbd al-​Muʿīn al-​Malūḥī and Asmāʾ al-​Ḥimṣī, vol. 1 (Damascus: Wizārat al-​Thaqāfa, 1970), 147. 92 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:640.

93 Ibn Saʿd, Muḥammad, al-​Ṭabaqāt al-​kubrā, ed. Muḥammad ʿAbd al-​Qādir ʿAṭā, vol. 7 (Beirut: Dār al-​Kutub al-​ʿIlmiyya, 1997), 54. 94 Ibn Khayyāṭ, Tārīkh Khalīfa ibn Khayyāṭ, ed. Suhayl Zakkār (Beirut: Dār al-​Fikr, 1993), 43.

95 Aḥmad ibn Ḥanbal, Faḍāʾil al-​Ṣaḥāba, ed. Waṣī Allāh ibn Muḥammad ʿAbbās, vol. 2 (Mecca: Markaz al-​Baḥth al-​Islāmī, 1983), 1045–​46.

58

58

Peter Webb

Musnad and Ṣaḥīḥ (collections which were compiled according to stricter standards of hadith authenticity).96 Despite the dubious authenticity, subsequent historians would repeat the hadith, giving it a canonical footprint in the battle’s memorialization.97 The explicit Arabization of Dhū Qār’s memory was thus expressed in tandem with prophetic history, and by the fourth/​tenth century, this dual underpinning became more salient, for instance in a new and colourful anecdote (without isnād) which al-​Aṣbahānī narrates: The battle was made manifest before Muhammad’s eyes while he was in Medina, and he raised his hands and prayed for victory for the Shaybān (or the Rabīʿa). He continued making the prayer until he was shown the Persians’ [furs] defeat.98

Al-​Aṣbahānī also narrates a new account for the battle’s beginnings, claiming it was sparked by the Arabs’ collective “anger” (ghaḍab) at Khosrow’s murder of the Arab frontier guard al-​Nuʿmān ibn al-​Mundhir.99 The anachronisms are patent: no pre-​Islamic poetry mentions that the Bakr’s warriors were seeking revenge for al-​Nuʿmān, and most of the Sasanian force were Arabians in any event; al-​Aṣbahānī’s rationalization of Dhū Qār as a manifestation of Arab anger could only have emerged once memories of the battle had been wholly Arabized, and once ethnic lines of division had been back-​ projected into the pre-​Islamic past. To further enhance the Arabization of Dhū Qār, Abū ʿUbayda and later narrators insert into the Dhū Qār narrative a poem attributed to the otherwise unknown pre-​ Islamic poet Bukayr al-​Aṣamm, which includes the verse:100 They attacked the Banū Aḥrār101 on that day With sword thrusts to their heads; Three hundred Arabs against a squadron Two thousand-​strong: Persians (aʿājim) from Banū Faddām.102 Ibn Qays found a battle The fame of which was heard from Iraq to Syria.103

96 Muḥammad al-​Bukhārī, al-​Tārīkh al-​kabīr, vol. 2 (Hyderabad: Dāʾirat al-​Maʿārif al-​ʿUthmāniyya, 1941–​64), 63. Pointedly, he also reports the hadith without Muhammad’s promise of future victory (8:313).

97 al-​Yaʿqūbī, al-​Tārīkh, 1:215, 225; al-​Ṭabarī, Tārīkh, 2:193 (in a second version of the narrative copied from Abū ʿUbayda, al-​Ṭabarī relates the hadith without Prophetic promise of victory, 2:207); see also al-​Masʿūdī, ʿAlī ibn al-​Ḥusayn, Murūj al-​dhahab wa maʿādin al-​jawhar, ed. Charles Pellat (Beirut: al-​Jāmiʿa al-​Lubnāniyya, 1966–​79), § 648; Miskawayh, Tajārib, 1:160; al-​Andalusī, Nashwa, 1:286; Ibn al-​Athīr, al-​Kāmil, 1:482.

98 Al-​Aṣbahānī, al-​Aghānī, 24:76.

99 Al-​Aṣbahānī, al-​Aghānī, 2:120.

100 I found no mention of Bukayr al-​Aṣamm in the major poetry anthologies nor in the biographical dictionaries of poets. Reference to him in al-​Aghānī is restricted to poems about Dhū Qār.

101 The “free born,” a sobriquet of the Persians and a reference to their stereotyped nobility.

102 Faddām allegedly refers to the veils (singular fidām) which Zoroastrian wine-​servers would wear when pouring wine, and it became a (rare) sobriquet for “Persian.” 103 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:645; see also al-​Ṭabarī, Tārīkh, 2:211; al-​Aṣbahānī, al-​Aghānī, 24:73.

59



Making War Ethnic

59

While it is impossible to prove the verse’s Abbasid-​era fabrication, there are several red flags. We have seen that “Arab” was not the ethnonym associated with the battle in earlier poetry, and certainly not in the pre-​Islamic poems about Dhū Qār. Bukayr is also an entirely unknown figure, and an easy target for false ascriptions of poetry. Moreover, on a lexical level, the references to the Islamic-​era nomenclature Banū Aḥrār/​Faddām for Persians and to Iraq and Syria (a specifically Islamic-​era division of space) resonate with Muslim-​era tropes,104 suggesting a Muslim-​era date for the poem’s creation. Accordingly, the poem’s reference to “Arab,” in distinction to the narrow tribal poetry of pre-​Islam and the repeated Maʿaddite references of the Marwanid-​era poets, seems yet another indication of its Abbasid-​era invention to facilitate the Arabization of Dhū Qār’s memory. Dhū Qār’s association with Muhammad was further entrenched via a new date postulated for the battle by the fourth/​tenth century al-​Aṣbahānī. Like all Abbasid-​era narrators, al-​Aṣbahānī is explicit in placing the battle within a narrative of Arab versus Persian, but he also dates Dhū Qār as contemporary with Muhammad’s victory over the pagan Meccans at Badr in 2/​624, elevating Dhū Qār to a pendant piece of Islam’s most famous victory.105 The story is almost certainly anachronistic—​historians before al-​Aṣbahānī were less specific and less symbolic as to the battle’s chronology: Abū ʿUbayda dates it loosely to the period of Muhammad’s prophecy (with no mention of Badr or Muhammad’s hijra in 1/​622), neither al-​Yaʿqūbī (d. ca. 284–​292/​897–​905) nor al-​Ṭabarī (d. 310/​922) date the battle, but al-​Ṭabarī precedes the battle narrative with a telling section detailing signs of the Arabs’ impending destruction of the Persian Empire. Al-​Ṭabarī’s placement of Dhū Qār as a significant way-​marker in world history was replicated: both Miskawayh (d. 421/​1030) and Ibn al-​Athīr (d. 630/​1233) situate the battle within a broader narrative of Persian doom. Al-​Masʿūdī (d. 346/​956), a historian contemporary with al-​Aṣbahānī, likewise connects Dhū Qār to symbolic dates of the Prophet’s career: either forty years after Muhammad’s birth, shortly after the hijra, or four months after Badr.106 In sum, the fourth/​tenth century marked the period when Dhū Qār’s association with both prophethood and conquest had been comprehensively articulated. Dhū Qār was also memorialized in the Abbasid era in entirely non-​historical, non-​ political contexts, which augment our considerations of the battle and peoplehood. The libertine poet Abū Nuwās (d. 195/​814) twice names Dhū Qār, first in irreverent fashion: What’s better than having camped at Dhū Qār? Camping at a tavern in the Anbār!107

104 For the signature Muslim-​era emphasis on spatial narratives dividing along the Euphrates between al-​Shām and al-​ʿIrāq, see Peter Webb, “Pre-​Islamic al-​Shām in Classical Arabic Literature: Spatial Narratives and History-​Telling,” Studia Islamica 110 (2015): 135–​64, at 158–​60. 105 Al-​Aṣbahānī dates Dhū Qār “a few months” after Badr (al-​Aghānī, 24:72).

106 Ibn Ḥabīb, al-​Naqāʾiḍ, 2:640; al-​Yaʿqūbī, Tārīkh, 1:215, 225; al-​Ṭabarī, Tārīkh, 2:188–​93; al-​Masʿūdī, Murūj, § 648. 107 Abū Nuwās, al-​Ḥasan ibn Hāniʾ, Dīwān, ed. Ewald Wagner, vol. 3 (Beirut: Orient-​Institut Beirut, 2012), 168.

60

60

Peter Webb

The sentiment reflects the anti-​heroic discourse typical of Abū Nuwās, who so frequently played the role of ritual clown and composed subversive poetry inverting heroic ideals and boasting of the glory of his bouts of drinking over the prototypical warrior poets’ boasts of bouts of fighting.108 The lines follow this trope: Abū Nuwās mocks poets who had boasted about Dhū Qār by countering that drinking at a tavern in al-​Anbār (the region near Dhū Qār) would have been a worthier pursuit. Implicitly, however, we can tell that Dhū Qār must have enjoyed the status as one of the great battles in Abū Nuwās’s cultural context, such that he would choose it as the specific vehicle for his satire. Abū Nuwās’s second Dhū Qār reference is equally subversive: The wine remembers Ṣiffīn, the Battle of the Camel And Dhū Qār too, with Hormuz the hero.109

He intends his wine’s exquisite age: that is, it was already in the cask during the battles of the Camel and Ṣiffīn (36–​37/​657), even since Dhū Qār. Ṣiffin and the Camel were two of the most seminal battles of early Muslim history, and thus their citation to describe the antiquity of wine is highly irreverent parody, and again, the fact that Abū Nuwās chose to include Dhū Qār in the list indicates that the battle had attained the loftiest levels in the community’s memory of its military past. Dhū Qār’s transition into an object of satire by the late second/​eighth century is, ironically, perhaps the clearest indication of the point when the battle had at last matured in collective memory as a pan-​communal symbol of “great battle.” Abū Nuwās was contemporary with Abū ʿUbayda’s narration and the first appearance of the prophetically endorsed Dhū Qār hadith, and hence we can propose that the integration of Dhū Qār within a cohesive sense of Arab communal history was established some 175 years after the original battle.

Dhū Qār and Persian-​ness

In closing, we visit the other side of Dhū Qār’s ethnic binary: the Persians. Heretofore we have relied on Arabic testimony, but Abū Nuwās’s poems connect us with counternarratives, particularly via his mention of Hormuz as the “hero” of Dhū Qār. Abū Nuwās presumably meant Hormuz Kharād, one of the Sasanian leaders killed at the battle, and while this seems a curious choice, it does squarely harmonize with Abū Nuwās’s penchant for subversion. Counter to the energies expended by early Abbasid-​ era Iraqi historians to emphasize Dhū Qār’s significance in both the Arab and Muslim story that established the battle as a cornerstone in the origin mythology of the ethnic and confessional identity of the Abbasid elite, Abū Nuwās again acts the ritual clown and aligns himself and his wine with memory of the Persians. The effort accentuates his 108 Andras Hamori, On the Art of Medieval Arabic Literature (Princeton: Princeton University Press, 1974), 44–​90. 109 Abū Nuwās, Dīwān, 5:528.

61



Making War Ethnic

61

oft-​repeated favouring of libertine life in taverns to Arabic poetry’s stereotyped ways of the Arabian warrior hero. With a wink between the lines, Abū Nuwās’s reference to the Persian commander at Dhū Qār as his “hero” betrays this rejection of Arab heroism, and thus again indicates, in reverse, what must have been the maturation of a “canonical” approach to conceptualizing Dhū Qār. The satire could not have worked unless a pro-​ Arab narrative had already been established, and his verse accordingly marks a cogent terminus ante quem for the widespread recognition of Dhū Qār as an “Arab victory.” Abū Nuwās’s professed support of the Persians also uncovers a role of Persian-​ness in resistance against the norms of early Abbasid-​era Iraq, thereby affirming that the status quo had been predominantly structured around Arabness and its triumphant opposition of Persians epitomized in the Dhū Qār story. Abū Nuwās’s satirical nostalgic reference to the bygone Persian heroes at Dhū Qār is a witty means to chide elite Arabness, but it remained a minority voice among all other poems and historical narratives of the Abbasid era which established Dhū Qār at the “beginning of the end” of the Sasanians, and at the beginning of triumphant Arab and Muslim hegemony. Against this backdrop, the next Persian-​partisan voice emerges in the late fourth/​tenth century with the Eastern Iranian historian Muḥammad Balʿamī (d. before 366/​977). Balʿamī composed a New Persian translation of al-​Ṭabarī’s Arabic Tārīkh al-​rusul wa-​l-​mulūk, the universal history from creation to the early fourth/​tenth century, which, as noted above, situated Dhū Qār within narratives of the Sasanian Empire’s collapse. Balʿamī’s choice to translate al-​Ṭabarī into Persian is an early sign of the rise of new Persian identity within the Muslim world, and his approach to al-​Ṭabarī’s Dhū Qār narrative offers unique insight into how Persians confronted the centuries of Arabic memorialization of the battle. Balʿamī does not follow al-​Ṭabarī verbatim: he adds two highly pertinent passages to explain the Persian defeat. First, Balʿamī explains that Khosrow selected the commander Hāmarz to lead the attack against the Bakr commander Hāniʾ ibn Qabīṣa on the basis of divination. According to Balʿamī, the divination of the name Hāmarz equalled “rise up” in Middle Persian (Pārsī), whereas Hāniʾ equalled “sit down”; hence, by appointing Hāmarz, Khosrow felt sure of victory.110 Such forms of divination were common in the courts of rulers, both pre-​Islamic and Muslim, but the insinuation in Balʿamī’s narrative is that human divination was not capable of understanding what would transpire at Dhū Qār: a higher power was involved. Balʿamī is explicit about this higher power in his narrative of what transpired in the Bakr’s camp between the first and second days of the battle. He relates that the Bakr leaders, Hāniʾ ibn Qabīaṣa and Ḥanẓala ibn Thaʿlaba, received news of Muhammad’s victory at Badr, and while neither Hāniʾ nor Ḥanẓala were yet Muslim, they were made aware that anyone who enters battle declaring Muhammad’s name would be assured victory. In the morn, they instructed the Bakr to proclaim Muhammad’s name as 110 Balʿamī, Tārīkhnāme, 2:822.

62

62

Peter Webb

they charged the Persians, and this carried the day.111 We can discount the account’s historicity, but the effect of the story firmly places Dhū Qār within Arab salvation history, and Balʿamī’s succeeding pages continue the theme: they report Gabriel informing Muhammad about the victory, and Muhammad’s dispatch of a letter to Khorsow urging him to embrace Islam.112 Intriguingly, no Arabic narrative, either before or after Balʿamī, is so explicit that the Bakr’s warriors themselves knew about Muhammad and/​or expressly invoked the new faith at Dhū Qār. Perhaps this blatant Islamisation was a step too far and too unbelievable for Arabic historiography, but Balʿamī is unabashed in ascribing the victory to the Arabs’ pseudo-​embrace of Islam. Why Balʿamī chose this unique narrative is perhaps best explained from considerations of peoplehood. For a Persian speaker who evidently prided his Persian identity (given his decision to write history in Persian for Persian-​speaking courtly patrons), Dhū Qār is technically an uncomfortable memory: it saw nomadic Arabians defeat the Sasanian Empire, and, as had been established by Balʿamī’s day, it was the turning point after which Persian power was utterly eclipsed. The whole enterprise is unavoidably embarrassing if one was a Persian partisan; however, Balʿamī was also Muslim, and his narrative offers a reconciliation of two forms of identity. By projecting Dhū Qār as an essentially Muslim victory, Balʿamī can overlook the ethnic issues involved in Sasanians being defeated by Arabs—​his version makes it clear that the Persians were not really fighting Arabs, they were instead on the wrong side of God’s will. God had chosen the Arabs for victory under the name of Muhammad, and by invoking Muhammad’s name, the Bakr appealed to divine power before which even the most august earthly power of the Sasanians was impotent. Balʿamī thus saves face. He presents Dhū Qār as a battle between Arab and Persian, but he intones that the victor was God, and hence there was no ethnic winner and loser, but rather the religious force carried the day, the same religious force which the conquerors at al-​Qādisiyya and Nihavānd would use again to defeat the Sasanians for the final time. Balʿamī thus reveals Persian partisanship: he evidently was uncomfortable enough with the established narrative of the Arabs’ martial superiority such that he added an overt Islamisation of the victory. He thereby enables Dhū Qār to sit more comfortably within the collective memory of Muslim Persians of his day: they need feel no shame from memory of the defeat of their “ancestors” since Muslim Persians, via their embracing of Islam, had entered on the divine side too, and are thus assured to avoid the fate of their ethnic, but non-​Muslim, forebears. Moreover, the lessons of Balʿamī’s Dhū Qār narrative imply that once anyone (including Persians) embraces Islam, they join the right side of faith, and, at such a point—​perhaps—​the historic ethnic superiority of Persian-​ness can come into play anew.

111 Balʿamī, Tārīkhnāme, 2:823.

112 Balʿamī, Tārīkhnāme, 2:824–​25.

63



Making War Ethnic

63

Conclusion The stages of Dhū Qār’s memorialization reveal a 300-​year process whereby dim and mutable reflections of a pre-​Islamic battle were gradually reshaped to convert a frontier clash waged by a limited group of subgroups of the Bakr against a composite force of Sasanian border guards into new guises that changed in step with the evolving organization of elite community in the Middle East. First, in the Umayyad period, the Caliphate’s military elite imagined themselves as the people of Maʿadd, and they harnessed memory of Dhū Qār, reshaping it into a collective achievement, amenable to their efforts to unify the Caliphate’s elite. Hence they could reorient memory of the battle from the narrow confines of the glory of a single tribe, or even just particular branches of that tribe, into a glory of Maʿadd, and as a harbinger of the Sasanian collapse from which the Marwanid-​era elite derived their worldly authority. Then, as the Marwanid dynasty gave way to the Abbasids, senses of peoplehood were changing, as Maʿaddites and others were being reorganized into a new form of Arab community, underwritten by the now well-​articulated doctrines of Islam. In turn, Dhū Qār was re-​memorialized as a victory for the Arab people with explicit connections to the Prophet Muhammad and Islam’s rise. As groups who memorialized Dhū Qār and made its memory their own became more powerful and better articulated the trappings of ethnic identity, so the macro-​historical significance of the battle increased. The story of Dhū Qār’s historiography thus takes us to the heart of early Muslim-​era myth-​making which reconfigured memories of the past not just to explain the rise of Islam, but also to create an antiquity for Arab identity formed by consolidating tribal histories into a composite ethnic story of all Arabs. And in sum, Dhū Qār is a prototypical site of memory: its permanence in history has little relation to what actually transpired, but instead derives from the staying power of the people who later laid claim to it.

64

65

Chapter 4

CAPTIVE IDENTITIES: INSCRIBING ARMENIANNESS FROM – SEBEOS TO AYRIVANEC‘I Sergio La Porta In ca. 600

CE the Sasanian Shah Ḵosrow II (r. 590–​628) appointed the Armenian nobleman Smbat Bagratuni marzpan (marzbān) or governor of the province of Vrkan (Gorgān/​Hyrcania) to the east of the Caspian Sea as a reward for his successful military suppression of revolts in the empire.1 Upon his arrival in Vrkan, Smbat came across a colony of captives who had been deported from their native lands at some point in the past to the edge of the desert that stretched from Turkistan to Dehestān.2 The colony consisted of peoples captured from Armenia, the Byzantine Empire, and Syria, and included Christians as well as “infidels,” possibly adherents of some form of Zoroastrianism or northern Mesopotamian Semitic cult. The episode is first recorded in Armenian historiography in the seventh-​ century History of Armenia attributed to Sebēos, who focuses on the captives from Armenia.3 He recounts that the captives had lost their language, literacy, and religious hierarchy, but he does not explain what identified them as “Armenian,” given that these people had lost the ability to speak, read, and write Armenian, and did not have a priest. Nonetheless, they were subsequently “confirmed in the faith,” taught to speak and become literate in Armenian, and had one of their elders, a certain Abēl, appointed as priest for their community. The story was subsequently repeated in historiographical works up through the thirteenth century.4 The passage is particularly suggestive because the notion of “lost Armenians” opened a space for the narrators who retold it to inscribe their own ideology of what constituted “Armenianness” and who had the authority and 1 On the Bagratuni family, see Cyril Toumanoff, Studies in Christian Caucasian History (Washington, DC: Georgetown University Press, 1963), 201–​3; Christian Settipani, “The Seventh-​Century Bagratids between Armenian and Byzantium,” Travaux et mémoires 17 (2013): 559–​78. 2 Dehestān sits on the edge of the Kara Kum desert, see Clifford E. Bosworth, “Dehestān.” Encyclopaedia Iranica Online: www.iranic​aonl​ine.org/​artic​les/​dehes​tan.

3 Sebēos, Patmut‘iwn Sebēosi, ed. Gêvorg V. Abgaryan (Yerevan: Academy of Sciences, 1979), 97; Sebēos, The Armenian History Αttributed to Sebeos, trans. with commentary Robert W. Thomson and James Howard-​Johnston with Tim Greenwood, 2 pts. (Liverpool: Liverpool University Press, 1999), 1:44. 4 Mik‘aēl Č‘amč‘ean (d. 1823) also included this story in his monumental Patmut‘iwn Hayoc‘, 3 vols. (Venice: Petros Vałvazeanc‘, 1784–​86), 2:300; Gérard Garitte, La Narratio de rebus Armeniae, CSCO 132, subs. 4. (Louvain: Durbecq, 1952. Reprint, Peeters, 2003); cf. note 68 below.

66

66 Sergio La Porta legitimacy to confer that identity. In other words, the story persisted as a lieu de mémoire,5 a literary site whose conscious remembering provided an opportunity for each author to stake claims to “Armenianness.” This process of remembering did not comprise the mere repetition of information, but entailed strategies of unremembering, addition, and alteration. Beginning with a close reading of this passage in Sebēos, this chapter interrogates from a mnemohistorical perspective how later narrators modulated the story.6 It argues that changes to the text reflect the positionality of each of the authors and foreground multiplicities of ideologies of Armenian identity. In articulating these arguments, it further seeks to complicate chronological models of identity formation or processes of ethnicization and destabilize static conceptualizations of “Armenianness.”7 This chapter thus does not aim to identify the components of Armenian identity or a moment of reification of an Armenian identity; rather, it poses “Armenianness” as a historically contingent construct that required—​and was subject to—​continual adaptation, rearticulation, and reinforcement. The readings of the “lost Armenians” presented underscore that ideologies of identity were fluid, contested, and dependent upon the positionality and rhetorical aims of the authors.

Smbat’s Lost Armenians The History attributed to Sebēos addresses the 180-​year period between the Armenian rebellion of the 480s that resulted in the treaty of Nuarsak and the first fitna of the 5 The concept of a lieu de mémoire, or memory site, is indebted to the work of Pierre Nora, see “Between Memory and History: Les Lieux de Mémoire,” trans. Marc Roudebush, Representations 26 (1989): 7–​24; cf. Sarah Bowen Savant, “Shaping Memory of the Conquests: The Case of Tustar,” in Violence in Islamic Thought from the Qur’ān to the Mongols, ed. Robert Gleave and István Kristó-​ Nagy (Edinburgh: Edinburgh University Press, 2015), 70–​89, and Antoine Borrut, “The Future of the Past: Historical Writing in Early Islamic Syria and Umayyad Memory,” in Power, Patronage, and Memory in Early Islam: Perspectives on Umayyad Elites, ed. Alain George and Andrew Marsham (New York: Oxford University Press, 2018), 275–​300, at 281–​83. 6 On the aims and methods of mnemohistory, see, in particular, Jan Assmann, Moses the Egyptian: The Memory of Egypt in Western Monotheism (Cambridge, MA: Harvard University Press, 1998), esp. 8–​16.

7 For example, in her study on the period of Armenian history between 600 and 750, Garsoïan found that era to constitute “the crucible in which the true constants of Armenian history appeared, or were consolidated, in the context of a period of statelessness, one in which the national identity achieved its true synthesis,” Nina Garsoïan, Interregnum: Introduction to a Study on the Formation of Armenian Identity (ca. 600–​750), CSCO 640, subs. 127 (Leuven: Peeters, 2012), xi. See also Gregory Areshian, “Historical Dynamics of the Endogenous Armenian, i.e., Hayots, Identity: Some General Obervations,” in Reflections of Armenian Ιdentity in Ηistory and Ηistoriography, ed. Hourai Berberian and Touraj Daryaee (Irvine: UCI Jordan Center for Persian Studies, 2018), 15–​35, and the studies listed on p. 22, note 29 for other approaches to Armenian identity formation.

67



Captive Identities

67

Caliphate that ended in 661.8 The author’s perspective is therefore that of someone living in the second half of the seventh century after the collapse of Sasanian power. A large portion of the text is devoted to the Byzantine–​Sasanian conflicts of the sixth and early seventh centuries and to the reign of the Sasanian Shāh Ḵosrow II (r. 590–​628) in particular. The episode of the “lost Armenians” is included among the material that Sebēos devotes to the career of Smbat Bagratuni, whom Ḵosrow II appointed the marzpan (governor) of Vrkan in ca. 600 as a reward for his military success against various rebellious forces.9 Smbat was the head of the Bagratuni noble family who had held the hereditary office of royal coronant within the Arsacid Kingdom of Armenia, which had come to an end in 428. According to Sebēos, Smbat was so respected and honoured by the Sasanian monarch that the latter bestowed upon him the title of Ḵosrow Šnum, that is, “the joy of Ḵosrow.”10 Sebēos records that Smbat first served the Byzantine Empire in the Balkans, but was later arrested on suspicion of leading an Armenian defection from imperial service.11 Thrown into the arena by the Emperor Maurice as a punishment, Smbat managed to defeat the wild animals sent to devour him and won the support of the crowd. Through the intercession of the Empress, Smbat was granted mercy. Nevertheless, he was shortly thereafter exiled to Africa. Sebēos does not enlighten us as to how Smbat managed to extricate himself from Africa and come to play a role in Sasanian affairs as he omits any information about him until his appointment as marzpan of Vrkan. Smbat subsequently embarked on a series of military exploits, such as defeating the king of the Hepthalites (kušank‘) in single combat, and pious acts, including rebuilding the Cathedral of St. Grigor at Duin and convening a council to elect a new catholicos for Armenia, and amassed a heap of royal honours and gifts. As Thomson and Howard-​Johnston point out, Sebēos likely drew upon a laudatory gesta of Smbat for his information about this charismatic figure that contained the story about the 8 The narrative very speedily covers the century from the 480s to 572 in the second chapter of the work [=​chap. 8]; the original conclusion of the text was in 655, but four later-​appended notices bring the account to 661, see Sebēos, Armenian History, 2:281–​87.

9 On Smbat’s career, see Nina Garsoïan, “Smbat Bagratuni,” Encyclopaedia Iranica Online. www. iranic​aonl​ine.org/​artic​les/​smbat-​bagrat​uni; Garsoïan, “Frontier—​Frontiers? Transcaucasia and Eastern Anatolia in the Pre-​Islamic Period,” in La Persia e Bisanzio: Atti dei convegni Lincei (Rome: Academia nazionale dei Lincei, 2004), 327–​52; Parvaneh Pourshariati, Decline and Fall of the Sasanian Empire: The Sasanian-​Parthian Confederacy and the Arab Conquest of Iran (London: I.B. Taurus, 2008), 136–​40; Scott McDonough, “The ‘Warrior of the Lords’: Smbat Bagratuni at the Center and Periphery of Late Sasanian Iran,” Iranian Studies 49 (2016): 233–​45; and literature cited there. These studies concentrate on Smbat’s military career within the Sasanian Empire and do not delve into the question of the captives. 10 In the text of Sebēos, Smbat is referred to as Ḵosrow Šum. Scholars have restored the second word to Šnum based on the Middle Persian word for “joy.”

11 This was likely in 589, Sebēos, Armenian History, 2:179; Theophylact Simocatta, History, trans. Michael Whitby and Mary Whitby (New York: Claredon Press, 1986), 3.8.4–​8, also recounts Smbat’s exploits in the Byzantine Empire.

68

68 Sergio La Porta “lost Armenians.”12 As the original gesta has not survived, it is impossible to determine how much material Sebēos may have changed, but it is clear that the story of the “lost Armenians” was considered significant enough for the memory of Smbat to be inscribed by both the composer/​transmitter of the gesta and Sebēos. According to Sebēos: there were in that realm [i.e., Vrkan] the peoples captured (azgn gerealk‘)13 from Armenian land (i hayastan erkrē) and put on the edge of the great desert which extends from T‘urk‘astan and Delhastan.14 They had forgotten their own language, and were deficient in literacy (dprut‘iwn), and lacked the order of priesthood.15 There was also there a group of Kodrik‘16 who had been taken captive with our own men; and furthermore not a few from the Greek empire and from the region of Syria. The community of Kodrik‘ were infidels. But over the Christians there shone a great light. They were confirmed in the faith and learned to write and speak their language. A certain presbyter among them who was named Abēl was appointed to the priestly rank in that land.17

Sebēos’s story links forced geographic displacement, deprivation of sociopolitical institutions, and admixture with others to loss of the significant cultural markers of language, literacy, and specific religious identity; but the laconic passage raises several questions. As mentioned, Sebēos unfortunately does not reveal how the community was recognized as being Armenian if they could not communicate in Armenian and had no institutionalized hierarchy. He does note that a bright light shone above the Christians, 12 Sebēos, Armenian History, 1:lxvii–​lxviii; on the epic nature of Smbat’s gesta, see also Tim Greenwood, “Sasanian Εchoes and Αpocalyptic Εxpectations: A Re-​evaluation of the Armenian History Attributed to Sebeos,” Le Muséon 115 (2002): 323–​97, at 347–​53.

13 The phrasing is unclear as azg, “people,” here is in the singular, but “captured,” gerealk‘, is plural. The verb of the sentence, “there were,” kayin, is also plural, so I have taken it to refer to the different peoples that may have been captured, but it is possible that Sebēos intended the many individuals that formed the “people.” Thomson translates it as a “captured community,” noting that it takes the use of the singular to refer the group’s being of “one ethnic origin,” Sebēos, Armenian History, 1:44n273. Later authors attempt to resolve this discrepancy in different ways. 14 That Delhastan is to be understood as Dehestān; see the literature cited in Sebēos, Armenian History, 1:44n274.

15 The grammar of this sentence is ambiguous, and the textual variants do not really help. According to Abgaryan (Sebēos, Patmut‘iwn, 259n279), all mss. read “they forgot” moṙac‘an, although he chose to emend the text to a participle, moṙac‘eal. The subsequent verbs are also plural participles, “were deficient in,” nuazealk‘, and “lacked,” pakasealk‘. Neither “literacy” nor “order of priesthood” bears the marker of the accusative, while “language” does. Thomson translates dprut‘iwn nuazealk‘ as they “lost writing,” which is the intent of the phrase given that they needed to be taught to write, but later authors also found this troubling and sought to resolve it. The word rendered above as “literacy” (dprut‘iwn) intimates the letters, the act of writing, and literature. The word dpir means “scribe.”

16 Who exactly the Kodrik‘ are is unclear. In the notes to his translation, Thomson (Sebēos, Armenian History, 1:44n275) queries whether this could refer to “Kotri in the south-​east of Iran … or the Kordrik‘ on the border with Asorestan?” The latter seems more likely. 17 In this and the following passages, my translation differs from that of the English translations listed, as I have modified them to harmonize the correspondences and follow the Armenian more closely.

69



Captive Identities

69

but this would have presumably included Greek and Syriac Christians, and not just the Armenians. Further complicating their identity is that the text does not refer to them as Armenians, but as a captured people(s) from the land of Armenia, who may have self-​identified in a variety of ways. More frustratingly, he does not tell us what language they spoke to each other or to the governor or his men—​was it a form of Middle Persian, Aramaic, Greek? Sebēos does not inform us or does not know how long this “Armenian” community had been there. It is tempting to think that they were imagined to have been the descendants of the Armenian nobles who were deported to the region after the suppression of the Armenian revolt of 451 by Yazdegird II—​thereby evoking the famous battle of Awarayr—​but there is nothing that allows us to definitively do so.18 Although the overall plan of “re-​Armenizing” these people is clearly meant to be attributed to Smbat as it is included among his deeds, Sebēos does not explicitly tells us this nor who taught them Armenian and “confirmed them in the faith,” nor what he means by the latter phrase, though presumably he intended baptism. Interestingly, Sebēos never brings up the possibility that these “Armenian captives” would be or should be returned “home.” Instead, the captives were “recaptured” or colonized through instruction in Armenian Christianity, language, and literature, and the consecration of a priest. Although geographical displacement was predicated as the primary cause of the loss of cultural identity, it does not seem to have been considered an obstacle to the (re)assimilation of the captives within an Armenian cultural sphere. Similarly, the passage does not limit those people who were “Armenized” to the captives from Armenia, although it does imply that the process was confined to the Christian population. We may speculate that various individuals or groups may have thought it advantageous to participate in a program initiated by the new marzpan of the area. The episode of the “lost Armenians” was clearly intended to magnify the stature of Smbat Bagratuni. Indeed, the finding of lost compatriots may even have been a literary trope of sorts. Nabel has shed light on an analogous passage in Philostratus’s Life of Apollonius of Tyana (third century CE),19 in which Apollonius (first century CE) comes across a community of Greek exiles on the far side of the Euphrates in Mesopotamia who had been deported by Darius I in 490 BCE.20 According to the text, Apollonius successfully intervenes with the Parthian king on behalf of this Greek community in order to improve their security from attacks by indigenous tribes. As Nabel observes, regardless of the historicity of the actual deportation and survival of this community, 18 Łazar P‘arpec‘i, The History of Łazar P‘arpec‘i, trans. Robert W. Thomson (Atlanta: Scholars, 1991), 132–​33. Alison Vacca, Non-​Muslim Provinces Under Early Islam: Islamic Rule and Iranian Legitimacy in Armenia and Caucasian Albania (Cambridge: Cambridge University Press, 2017), 155, similarly observes the ambiguous definition and origin of this community. 19 Jake Nabel, “Fish out of Water: Greek Deportees, Persian Empires, and the Classical Mediterranean,” paper presented at Ancient Persia and the West (Los Angeles: Getty Research Institute/​UCLA Symposium, April 25, 2018).

20 Philostratus, Apollonius of Tyana, Volume I: Life of Apollonius of Tyana, Books 1–​4, ed. and trans. Chris P. Jones, Loeb Classical Library 16 (Cambridge, MA: Harvard University Press, 2005), 1:23.

70

70 Sergio La Porta Philostratus’s presentation of the story foregrounds the heroic status of his protagonist. Even so, significant differences between the two stories should be recognized. Nabel justly argues that the ideology reflected in Philostratus underscores the differences between the Iranian world and the Greek. The Greek exiles are described as “fish out of water,” completely at odds with their “new” surroundings. By contrast, neither Smbat nor the “lost Armenians” are out of place in the Sasanian Empire. Smbat does not intervene with the monarchy on the community’s behalf; he is the imperial representative in that area. Moreover, Sebēos’s discourse does not posit alterity as dangerous to the captives; their near complete loss of identity does not derive from external pressure or persecution from the empire nor from their co-​exiles with whom they apparently live in peace. Rather, it was precisely the lack of defined boundaries that allowed for the cultural slippage they experienced. It is Smbat who erects those boundaries and creates an extraterritorial ethnoreligious community by “restoring” their Armenian identity. The gesta of Smbat Bagratuni was composed between the protagonist’s death in 617 and the middle of the seventh century when it was available to the author of the History of Armenia. Like its original form, whether it was articulated immediately after his death or two decades later cannot be determined; but how Smbat’s life and his actions as marzpan were received by audiences must have changed as Sasanian power waned and disappeared, to be replaced by caliphal forces. Writing at the time of the first fitna, Sebēos’s renarration of the life of Smbat Bagratuni and of his finding of “lost Armenians” in particular, holds forth a model of Armenian identity and power that is not reliant upon Byzantine or Sasanian legitimacy. Although an imperial appointee, Sebēos’s account suggests that Smbat himself without Sasanian approval created an Armenian community outside of any territorial borders associated with Armenia. By way of comparison, we may introduce the figure of the Albanian hero, J̌uanšir, as represented in the tenth-​century History of Caucasian Albania by Movsēs Dasxuranc‘i.21 As Vacca has elucidated, through a selective process of historical rewriting, J̌uanšir serves as a symbol of stability who successfully transitions from Sasanian to caliphal service and maintains Albanian political relevancy.22 Chronological differences between Smbat and J̌uanšir naturally resulted in different narrative possibilities for the composers of the gestae and the later retellings, but their composers employed these heroic actors as a means of mediating the political transformations of the seventh century in the region. The significance of each story for the creation of Armenian and Albanian political identities 21 A comparison between the epic figures of Smbat and J̌uanšir is also made in Sebēos, History, lxviii; and Greenwood, “Sasanian Echoes,” 348, 352–​53.

22 Alison Vacca, “J̌uanšir and the End of Sasanian Rule in the Caucasus,” paper presented at Ērānshahr in the Age of Transition: West and Central Asia Between Sasanians and Islam (Princeton: Princeton University, May 10, 2019). For another example of the heroic transformation of a transitional figure, Vard Ṙštuni, see Zara Pogossian, “Relics, Rulers, Patronage: The True Cross of Varag and the Church of the Holy Cross on Ałt‘amar,” in The Church of the Holy Cross of Ałt‘amar: Politics, Art, Spirituality in the Kingdom of Vaspurakan, ed. Zara Pogossian and Edda Vardanyan (Leiden: Brill, 2019), 126–​206, at 143–​50.

71



Captive Identities

71

and memory may be exposed in the omission of any mention of J̌uanšir in Sebēos, and of Smbat in Dasxuranc‘i. Later Armenian sources, however, clearly felt an uneasiness with the abovementioned ambiguity in Sebēos’s account of the “lost Armenians,” for when remembering this episode they attempt to elucidate the meaning of the passage as well as employ it for their own purposes.

Remembering Lost Armenians in the Armenian Kingdoms

In the centuries following the inclusion of Armenia within the Caliphate, the fortunes of the Bagratuni continued to rise and they emerged as the pre-​eminent noble family in Armenia. As ‘Abbsāsid power waned at the end of the ninth century, Prince Ašot Bagratuni capitalized upon the situation and in 884 was recognized as king by the Caliphate and, later, by the Byzantine Empire.23 In 908, the Arcruni princes of Vaspurakan similarly established a separate kingdom to the south centred around Lake Van. The next two iterations of Smbat’s finding of lost Armenians occur in works composed by authors living in this time of the Armenian monarchies: the tenth-​century History of catholicos Yovhannēs V Drasxanakertc‘i (sed. 897–​925/​29), and the Universal History of Step‘anos Tarōnec‘i (ca. 1004). During Yovhannēs’s tenure as catholicos, the young Bagratuni kingdom came under pressure.24 A rivalry between Ašot II Erkat‘ (Iron) and his cousin Ašot, the son of Šapuh, threatened the cohesion of the kingdom for many years, while the emergence of a separate Arcruni kingdom further created a rival political power in neighbouring territory. In addition, Bagratuni lands were under assault by the Sājid emirs of Azerbaijan, who for much of this period were also in control of the city of Duin, where the seat of the catholicos was located. Yovhannēs Drasxanakertc‘i actively laboured to solidify the Bagratuni kingdom, trying to reconcile the cousins, participating in diplomatic negotiations, and creating a royal ideology to legitimize Bagratuni rule. It is not surprising, therefore, that he includes the exploits of Smbat Bagratuni. His memory of Smbat, however, begins with an act of unremembering. Unlike Sebēos, he does not include Smbat’s service to the Byzantine Empire or his dramatic appearance in the arena. It is possible that Yovhannēs did not identify the two mentions of Smbat, but he may have intentionally omitted Smbat’s earlier career so as to not have him serve a Chalcedonian emperor. On the other hand, he does recall Smbat’s visit to Armenia from Vrkan and rebuilding of the cathedral at Duin, an event of understandable interest for a catholicos. Yovhannēs starts his account of Smbat 23 Although the Caliphate sent Ašot a crown, Arabic and Greek sources do not title him king, but “Prince of princes”; see Krikor Yužbašyan, “ ‘Bagratuneac‘ šrǰani’ Hayastaně` miǰazgayin iravunk‘i tesankyunic‘,” Patma-​banasirakan Handēs 1 (1975): 33–​53 for a discussion.

24 See Maksoudian’s biographical introduction to his translation, Yovhannēs Drasxanakertc‘i, History of Armenia, trans. Krikor Maksoudian (Atlanta: Scholars, 1987), 7–​23.

72

72 Sergio La Porta with the Bagratuni’s appointment as marzpan of Vrkan and his discovery of the “lost Armenians.” According to him, Smbat: found peoples25 captured (azgs gereals) from Armenians (i hayastaneayc‘) and settled in the region of the great desert which borders on T‘urk‘astan and is called Sagastan. They had forgotten their language and their literacy (dprut‘iwn) had greatly decreased.26 When they saw Smbat, they were greatly overjoyed, and at his order they received instruction anew in the pronunciation of Armenian syllables and renewed the language. And having learned Armenian writing (dprut‘iwn), they were confirmed in the faith. Then Smbat ordered our great patriarch Movsēs to set a certain presbyter by the name of Abēl, who was one of them, as their bishop. And thus he organized those who were in a distant land into a prelacy of the great see of St. Grigor [that has lasted] until the present time.27

As is evident, the general tenor of the story is the same, but Drasxanakertc‘i has altered its specifics. We may first observe that he deemphasizes a geographically defined notion of “Armenian.” Sebēos uses the phrase “from Armenian land” (i hayastan erkrē), an expression that explicitly invokes the concept of geography and does not necessarily refer to a particular people.28 Yovhannēs, however, accentuates the ethnic identity of the captives. He retains and grammatically clarifies the plural “peoples,” but claims that they were captured from “Armenians,” using the collective designation hayastaneayk‘, rather than the expression “from Armenian land.” At the same time, he applies the concept of land to the Armenians’ new settlement “in a distant land” (i heṙastan erkri). Second, he changes the geographic description of the captives’ location from the edge of the desert between T‘urk‘astan and Delhastan to the desert that borders T‘urk‘astan and is called Sagastan, suggesting that Delhastan was not a commonly recognized toponym at this time.29 Moreover, Yovhannēs seems careful to diminish the amount of linguistic and cultural loss that had occurred. He describes their ability to read and write Armenian as greatly reduced rather than as completely lacking. Similarly, he remarks that these captives only needed to be reinstructed in pronouncing Armenian, not that they had no linguistic memory of the language altogether. By attributing some linguistic and literary ability to these captives, Yovhannēs endows the community with linguistic markers 25 “peoples,” azgs: Maksoudian translates as “families.”

26 “greatly decreased,” nuazeal yoyž. Drasxanakertc‘i uses the same verb, nuazim, as Sebēos to indicate the captives’ literacy. However, Yovhannēs has understood it as modifying “literacy” and added the adverb, suggesting that he did not think it meant that their literacy was completely gone.

27 Yovhannēs Drasxanakertc‘i, Hayoc‘ Patmut‘iwn, ed. Georg Tēr Vardanean, in Matenagirk‘ Hayoc‘ vol. 11 (Antelias: Armenian Catholicosate of Cilicia—​Matenadaran Mashtots Institute of Ancient Manuscripts, 2011), 399; Yovhannēs Drasxanakertc‘i, History, 95.

28 The word hayastan is composed of the Armenian designation for themselves, hay, plus the ending –​astan (Persian -​stān), indicating a place and thus means the place of the hay. By the fifth century, however, the term seems to have served mainly as an adjective meaning “Armenian” and to have lost its sense as a toponym. Hence Sebēos employs it to describe the land from which the captives were taken. 29 The seventh-​century Ašxarhac‘oyc‘ (Geography) of Anania Širakac‘i places Turkistan in Sagastan, The Geography of Ananias of Širak (Ašxarhac‘oyc‘): The Long and Short Recensions, trans. Robert H. Hewsen (Wiesbaden: Dr. Ludwig Reichert, 1992), 74–​75.

73



Captive Identities

73

of identity that provide a means of mutual recognition on the part of the captives and Smbat. Drasxanakertc‘i further privileges linguistic markers of identity when he relates that they were first trained in speaking and writing Armenian and only afterwards instructed in the faith, thus reversing the order of events as recorded in Sebēos. Yovhannēs also attributes a priest to the community, minimizing the degree to which they had been separated from the Armenian Church. Moreover, he completely omits any reference to the other communities mentioned by Sebēos that had also been deported by the Persians, thereby simplifying the sociolinguistic features of the community and eliminating the possibility of ethnic or cultural mixing. Through these selective moves of unremembering, Drasxanakertc‘i significantly de-​Others the captive community and presents a very different memory of them from that portrayed in Sebēos. In addition, Yovhannēs foregrounds Smbat’s agency in the narrative, making explicit the implicit connections in Sebēos. Thus, Smbat “finds” the captives; the captives are overjoyed at seeing Smbat; he orders them to be instructed; and he commands the catholicos to appoint Abēl a bishop. Finally, in contrast to Sebēos, Yovhannēs introduces the role of the Armenian catholicos Movsēs in assimilating this community within the hierarchy of the Armenian Church by personally appointing Abēl its bishop, not just a priest, and establishing this extraterritorial diocese. Likewise, in contrast to Sebēos who only generally remarks on the community’s inclusion within the Armenian Apostolic Church through the ordination of Abēl as a priest, Drasxanakertc‘i far more explicitly delineates that position by designating the community a prelacy within “the great see of St. Grigor,” thereby foregrounding the role of Grigor Lusaworič‘ (the Illuminator) as the patron saint of Armenian Christianity. Drasxanakertc‘i’s insistence on the role of the Catholicosate working in tandem with political authority speaks to the position of the Catholicosate both in the late sixth to early seventh centuries and in his own time, the tenth. As he was well aware, the jurisdiction of the Armenian Church had been severely curtailed by the partition of Armenia in 591 after the Emperor Maurice had helped Ḵosrow II secure the throne. Discussions between the catholicos Movsēs II and the Imperial Church did not go well and Maurice established an anti-​Catholicosate on Byzantine territory. The Armenian Church was able to slowly regain its “lost bishoprics” after the death of Maurice in 602 with the election of Abraham I at the council of Duin in 607 convened by Smbat Bagratuni, and the capture and exile of the anti-​catholicos Yovhan by the Persians in 610.30 Drasxanakertc‘i’s retelling of the episode of the “lost Armenians,” however, diverts attention away from the Catholicosate’s temporary loss of these dioceses and regional prestige to its ability to augment its authority and stature by extending its influence extraterritorially—​“in a distant land”—​through the establishment of a diocese in Vrkan with Smbat Bagratuni’s 30 On the ecclesiastical conflict between Maurice and Movsēs II and its repercussions, see Nina Garsoïan, L’Église arménienne et le Grand Schisme d’Orient, CSCO 525, subs. 100 (Louvain: Peeters, 1999), 267–​82; Jean-​Pierre Mahé, “L’Église arménienne de 611 à 1066,” in Histoire du christianisme des origines à nos jours. IV: Évêques, moines, et empereurs (610–​1054), ed. Gilbert Dagron, Pierre Riché, and André Vauchez (Paris: Desclée, 1993), 457–​547, at 462–​65.

74

74 Sergio La Porta assistance. This picture of political cooperation in support of ecclesiastical expansion and authority over Armenians regardless of their place of inhabitation invokes the ideal relationship of King Trdat and St. Grigor in the conversion of Armenia to Christianity, as articulated in the History attributed to Agat‘angełos,31 while simultaneously supporting Yovhannēs’s own vision of the relationship between the two spheres of power.32 Similarly, Yovhannēs’s explicit evocation of St. Grigor manifests his general interest in and patronization of the cult of St. Grigor. The figure of St. Grigor was a contested symbol who was neither unambiguous nor the sole property of adherents of anti-​ Chalcedonianism in Armenia. For example, the Narratio de rebus Armeniae, composed in Armenian ca. 700 and subsequently translated into Greek, claims that the Armenians who rejected Chalcedon discarded the tradition of St. Grigor,33 as does the independent List of Catholicoi of Armenia preserved in Greek and Georgian, but also translated from Armenian.34 Arsen Sapareli, the tenth-​century composer of a treaty on the Armeno-​ Georgian schism likewise remarks that the Armenians separated themselves from “the faith of St. Grigor.”35 While Drasxanakertc‘i’s own theological position regarding Chalcedonism is debated—​he seems to have been more of an ecumenicist than an ardent anti-​Chalcedonian36—​he was deeply invested in the saint and the propagation of his cult.37 His association of Armenianness with St. Grigor further promoted the saint’s valence as a marker of Armenian Apostolic confessional identity. 31 See Thomson’s remarks in Agat‘angełos [Agathangelos], History of the Armenians, trans. Robert W. Thomson (Albany: SUNY, 1976), xci–​xciii; cf. Sergio La Porta, “The Vision of St. Grigor Lusaworič‘ and the Role of Apocalyptic in the Conversion of Armenia,” in The Armenian Apocalyptic Tradition: A Comparative Perspective, ed. Kevork Bardakjian and Sergio La Porta, SVTP 25 (Leiden: Brill, 2014), 296–​312, at 301–​5. 32 See Maksoudian’s comments in Yovhannēs Drasxanakertc‘i, History, 14–​18; Mahé, “L’Église arménienne,” 497–​505.

33 ἀποβαλόντας τὴν παράδοσιν τοῦ ἁγίου Γρηγορίου, Gérald Garitte, La Narratio de rebus Armeniae, CSCO 132, subs. 4 (Louvain: Durbecq, 1952. Reprint, Peeters, 2003), 35 (65.158–​59). 34 Garitte, Narratio, 404 (36); 409 (36).

35 Zaza Aleksidzé and Jean-​Pierre Mahé, “Arsen Sap‘areli: Sur la séparation des Géorgiens et des Arméniens,” Revue des études arméniennes 32 (2010): 59–​132, at 88 (5:2.26–​7), 92 (9:3.18), 106–​ 7, 112. 36 See Maksoudian’s remarks, Yovhannēs Drasxanakertc‘i, History, 18–​20.

37 The relics of St. Grigor have a varied history, but their cult was bolstered in the late ninth century when they were “rediscovered” during the reign of Basil I and King Ašot I of Armenia instituted a feast for their invention, Kirakos Ganjakec‘i, Patmut‘yun Hayoc‘, ed. Karapet Melik‘-​ Ōhanǰanyan (Yerevan: Academy of Sciences, 1961), 14; cf. Vardan Arewelc‘i, Chronicle: Vardan Arewelc‘i. A Facsimile Reproduction with an Introduction by R. W. Thomson (Delmar, 1991), 85. Yovhannēs himself went on a pilgrimage to the burial site of Grigor on Mt. Sepuh. Despite being the locus of the death of Grigor, Mt. Sepuh does not seem to have served as a major centre of pilgrimage prior to the tenth century. Drasxanakertc‘i spent ten months there, and in his History he provides a vivid picture of ascetic life on the mountain, Yovhannēs Drasxanakertc‘i, Patmut‘iwn, 534–​35; Yovhannēs Drasxanakertc‘i, History, 198–​99. In addition to personally engaging with so much of

75



Captive Identities

75

At the beginning of the turn of the millennium, another historian, Step‘anos Tarōnec‘i, once again included the discovery of this Armenian community in his Universal History. Even before examining how Step‘anos remembers this story, two striking differences between his and the earlier accounts of Smbat’s encounter are apparent. First, Step‘anos refers to Smbat as the marzpan of Armenia, not of Vrkan. According to Tarōnec‘i, Smbat discovered the “lost community” while fighting in Vrkan, not when he was assigned to be its governor. Second, he has unremembered every other aspect of Smbat’s career except for this. Thus, he does not mention any of Smbat’s other military or pious activities. In Step‘anos’s retelling: This Smbat went to the realm of Vrkan in battle and found there peoples captured from Armenia (azgs gereals i hayoc‘) and dwelling38 in the desert of T‘urk‘astan, which is called Sagastan; they had forgotten their Armenian language and literacy (dprut‘iwn). When Smbat saw this, he rejoiced greatly and appointed (kargē) a presbyter named Habēl as their vardapet. They learnt Armenian literature (hayerēn dprut‘iwn) from him and continue to be a diocese of St. Grigor to today.39

Although Tarōnec‘i’s account is based on that of Yovhannēs Drasxanakertc‘i, there are once again subtle but indicative differences. In this passage, hayastaneayk‘, “Armenians,” has been replaced by the use of the simpler and more common term, hayk‘, “Armenia(ns),” a pluralis tantum in the ablative case. The ethnonym was understood to refer to the common ancestor of the Armenians, the hero Hayk, although linguistically the two words are not related.40 As in Drasxanakertc‘i, it is difficult to detect complete consistency in Step‘anos’s use of terms referring to Armenia and Armenians. He uses the term hayk‘ both as an ethnonym and a toponym;41 but he also sometimes uses the term hayastaneayk‘ to refer to Armenians42 and the phrase “realm of Armenia” (ašxarh Grigor’s life and relics, Drasxanakertc‘i provided for the first building program on the mountain by commissioning the construction of a church at the site of Grigor’s burial. 38 “dwelling,” bnakeal, against Drasxanakertc‘i’s use of the factative bnakec‘uc‘eal, “settled.”

39 Step‘anos Tarōnec‘i, Tiezerakan Patmut‘iwn, ed. Gor Manukean, in Matenagirk‘ Hayoc‘ vol. 15 (Yerevan: Armenian Catholicosate of Cilicia—​Matenadaran Mashtots Institute of Ancient Manuscripts, 2011), 619–​829, at 692; Step‘anos Tarōnec‘i, The Universal History of Step‘anos Tarōnec‘i, trans. Tim Greenwood (Oxford: Oxford University Press, 2017), 154.

40 The legend of Hayk is of ancient origin, but he is first attested as the eponymous ancestor of the Armenians in the seventh-​century Ašxarhac‘oyc‘ (Geography) of Anania of Širak and the anonymous Primary History, possibly from the same century. Hayk’s story was further developed in Movsēs Xorenac‘i’s History of the Armenians, trans. Robert W. Thomson (Cambridge, MA: Harvard University Press, 1978), 1.10–​12. The dating of this text remains controversial, but was likely composed in the form we now have it in the eighth–​ninth centuries. See Movsēs Xorenac‘i, History, esp. 88n6. 41 For example, he refers to “Armenian forces” (zōrk‘ hayoc‘) as well as someone being sent “to Armenia” (i hays), Tiezerakan Patmut‘iwn, 679. These examples are not intended to be exhaustive by any means. 42 Cf., e.g., Step‘anos Tarōnec‘i, Tiezerakan Patmut‘iwn, 639, 675, 676.

76

76 Sergio La Porta hayoc‘) when he wants to emphasize territoriality or the political construct of Armenia.43 The replacement, then, of i hayastaneayc‘ by i hayoc‘ arguably does not indicate an intensification of ethnic identity, but the employment of a more conventional term that also signified both place and peoplehood. In contrast with Drasxankertc‘i, this Armenian community had completely forgotten their language and writing, which Tarōnec‘i explicitly designates as Armenian. Step‘anos makes no mention of their religious identity and their Armenian Christian faith appears assumed. In his version, Smbat rejoices at having found the captives and not the community at meeting Smbat. Habēl is not ordained a priest or bishop by the catholicos, but is appointed by Smbat as their vardapet, or monastic teacher, who is to instruct them in acquiring literacy in Armenian. Step‘anos makes no mention of whether they also learned how to speak the language. Similar to Drasxanakertc‘i, however, Tarōnec‘i foregrounds St. Grigor as the defining symbol of Armenian ecclesiastical identity. The inflections in Step‘anos’s narrative primarily reproduce his own position as a scholar in the Armenian Church.44 In addition, undertones of the account of the life of the inventor of the Armenian alphabet, Maštoc‘, rest behind his description and in Step‘anos’s changing Abēl’s name to Habēl. According to Koriwn, a student of Maštoc‘ who wrote his vita, a certain priest Habēl introduced the Danielic letters to the Armenian court.45 Although these letters were rejected, the connection between the name Habēl and the process of the creation of the Armenian alphabet was a natural one for Step‘anos to make. Tarōnec‘i redirects the “joy” Drasxanakertc‘i attributes to the community’s reception of Smbat to Smbat’s anticipation of educating the captives. Again, Koriwn’s work may be at play here as Smbat’s rejoicing (uraxac‘eal) echoes more Koriwn’s description of the joy (urax) King Vṙamšapuh, Maštoc‘, and the Patriarch Sahak experienced at receiving the letters from Habēl than Drasxanakertc‘i’s description of the community as overjoyed (xrax linēin). Furthermore, in assigning the joy to Smbat rather than to the community, Tarōnec‘i creates a parallelism between the marzpan and the Armenian king. Finally, in comparison with Drasxanakertc‘i, who is more concerned with a jurisdictional definition of Armenian identity within the Armenian Church, Step‘anos envisions an intellectual community whose members share a literary heritage and are bound together through teacher–​student relationships. Drasxanakertc‘i and Tarōnec‘i show that, already in its earliest attested retellings, Sebēos’s story was strategically remembered as well as unremembered and Armenianness circumscribed in disparate ways contingent upon the personal and historical context of the authors. As the sociopolitical environment of the region, and of the Bagratuni kingdom in particular, shifted over the course of the eleventh and twelfth 43 Cf., e.g., Step‘anos Tarōnec‘i, Tiezerakan Patmut‘iwn, 669, 672, 674, 681.

44 Greenwood emphasizes well Tarōnec‘i’s scholarly perspective in the introduction to his translation.

45 Koriwn, Vark‘ Maštoc‘i, ed. Manowk Abełyan, rev. Karên Iwzbašean and Paruyr Muradyan, in Matenagirk‘ Hayoc‘, vol. 1 (Antelias: Armenian Catholicosate of Cilicia, 1930), 227–​72, at 236–​37.

77



Captive Identities

77

centuries, later writers again reshaped the narrative to suit their own rhetorical purposes and idealization of Armenianness.

Remembering Lost Armenians in Ani

During the eleventh century, the Macedonian dynasty of the Byzantine Empire successfully enacted a policy of annexing the Bagratuni and Arcruni kingdoms.46 The last Arcruni king, Yovhannēs-​Senek‘erim, exchanged his territory for Byzantine lands in 1021. The Bagratuni kingdoms centred at Ani and Kars were captured by Byzantium in 1045 and 1064, respectively, and subsequently taken by the Seljuks in 1064 and 1065. The Seljuk sultan Alp Arslan then sold the former capital city of Ani and its environs to the Kurdish Shaddādids, who had been displaced from Ganja. The Shaddādid emirs retained control of the city for most of the twelfth century, but the rising Georgian monarchy fought for its possession on several occasions.47 Two authors from Ani remember the story of the lost Armenians at the end of the twelfth century during a fraught moment for the urban centre. The first is a chronicle from Adam until 1173 composed by Samuēl, priest of the Cathedral of Ani, between 1173 and 1176 at the request of catholicos Grigor IV Pahlawuni (sed. 1173–​1193). Smbat appears under two notices. First, under 585, he relates Smbat’s having been maligned and sent to fight wild animals. However, in marked contrast to Sebēos, Samuēl does not attribute this to the Byzantine emperor, but to a Sasanian, Šapuh-​Ḵosrow. According to Samuēl, Smbat was once again elevated by the Persian king after defeating the animals. As Samuēl had just noted that Hormizd IV had come to the throne in the previous entry, Hormizd must be the king to whom Samuēl refers. Who Šapuh-​Ḵosrow might be is unclear, but Samuēl may mean the future king Ḵosrow II. The second mention of Smbat occurs under the year 596, when Samuēl notes the accession of Ḵosrow II to the throne. He then continues: Here, Smbat Bagratuni waged many battles with superior bravery against the enemies of Ḵosrow, on account of which he honored Smbat and made him marzpan of Vrkan. And he went and found peoples captured (azgs gereals) from Armenia (i hayoc‘) there, which is Sagastan, who had forgotten [their] language and literacy (dprut‘iwn). Smbat renewed them, and had the catholicos of Armenia ordain a certain Habēl bishop, and established it as a diocese of St. Grigor.48

Samuēl seems to have based his account on that of Yovhannēs Drasxanakertc‘i. Like Drasxanakertc‘i, Samuēl claims that Smbat “renewed” (norogeal) their language, had Habēl ordained a bishop and established the community as a diocese of St. Grigor. Nonetheless, there are divergences where Samuēl apparently preferred the story 46 Hrač Bartikian, “La conquête de l’Arménie par l’Empire byzantine,” Revue des études arméniennes 8 (1971): 81–​92. 47 Vladimir Minorsky, Studies in Caucasian History (London: Taylor’s, 1953), 79–​106.

48 Samuēl Anec‘i, Hawak‘umn i groc‘ patmagrac‘, ed. A. Tēr-​Mik‘elēan (Vałaršapat: Eǰmiacin, 1893), 77.

78

78 Sergio La Porta as related in Step‘anos. For example, Samuēl treats Smbat’s military successes only briefly, and makes no mention of his rebuilding of the cathedral at Duin, highlighting his work with the captives instead. Moreover, he describes this people as having been “from Armenia” (i hayoc‘), like Step‘anos Tarōnec‘i, rather than “from Armenians” (i hayastaneayc‘), and notes that the captives had forgotten both their language and writing. Contrary to both Drasxanakertc‘i and Tarōnec‘i, Samuēl neglects the reference to T‘urk‘astan, leaving Vrkan to be identified with Sagastan, and does not attribute any prior status to Habēl. Moreover, while Movsēs still ordains Habēl a bishop, Samuēl attributes the creation of the diocese to Smbat, blurring the lines of political and religious authority. Finally, he does not mention that the diocese created in Sagastan continued to exist in his day. If Drasxanakertc‘i’s and Tarōnec‘i’s information is accurate, the diocese apparently disappeared at some point between 1000 and 1170. The second twelfth-​century appearance of the story is in the historical work of Mxit‘ar Anec‘i, who was also a priest associated with the Cathedral of Ani. This work was finished in 1193 or shortly thereafter, but the surviving text is incomplete.49 Mxit‘ar refers to it as a “book of epic world events” (matean ašxarhavēp handisaranac‘). It seems to have been written for a certain father Grigor, who may have been the abbot of the monastery of Haṙič, if the later subtitle in the manuscript is correct, but he is not designated such by Mxit‘ar himself. Like Samuēl, Mxit‘ar was aware of the historiographical tradition that preceded him, including Samuēl’s chronicle. In keeping with the “epic” theme of his work, Mxit‘ar devotes a lot more attention to Smbat’s career than Samuēl or Step‘anos, and draws on both Sebēos and Drasxanakertc‘i. Thus, he discusses Smbat’s rebuilding of the cathedral at Duin and his fighting in the arena and being sent to Africa, although he places these latter two at the end of Smbat’s career, rather than at the beginning like Sebēos and Samuēl. Possibly due to the contradiction between Sebēos’s and Samuēl’s account over who condemned Smbat, Mxit‘ar does not indicate which king in fact relegated him to the arena. Mxit‘ar starts his story of Smbat with Ḵosrow II’s rewarding him with the appointment as marzpan of Vrkan for his victories against the king’s enemies: When he [Smbat] went there he found one people captured from Armenia (azg mi gereal i hayoc‘) and settled in T‘urk‘astan, which is called Sagastan, who forgot the Armenian language and writing (gir). When they saw Smbat, they were happy. And he renewed the Armenian language and script (gir) through the presbyter Abēl and made that same one their prelate. And afterwards Movsēs ordained Abēl a bishop.50

It is important to observe that Mxit‘ar has not simply harmonized the different versions of the story known to him, but has made editorial adjustments to the narrative. For example, he has altered the “peoples captured” (azgs gereals) in the plural to “one 49 The work is preserved in a single manuscript, cod. 2678, held at the Maštoc‘ Institute of Ancient Manuscripts (Matenadaran) in Yerevan, Armenia, copied between 1426 and 1476.

50 Mxit‘ar Anec‘i, Matean Ašxarhavēp Handisaranac‘, ed. Hasmik Margaryan (Yerevan: Armenian Academy of Sciences, 1983), 91.

79



Captive Identities

79

people captured” (azg mi gereal) from Armenia, thereby eliminating the possibility of multiple peoples in Armenia and reinforcing Drasxanakertc‘i’s move of unremembering the others who were captive with them. He also replaces the abstract dprut‘iwn, which I have rendered as “literacy,” with the more concrete gir, “letter” or “writing,” conjuring the physical image of the Armenian alphabet. He emphatically repeats that the language and writing lost and regained was Armenian (hayerēn) so that the ambiguity of Sebēos’s text is removed. He restores the joy of the captives at seeing Smbat, as in Drasxanakertc‘i, but reversed in Tarōnec‘i and ignored by Samuēl. On the other hand, he has Smbat appoint Abēl, a prelate to whom he ascribes the instruction of the captives in speaking and writing Armenian, thus echoing Tarōnec‘i who calls him a vardapet. According to Mxit‘ar, Movsēs later ordains Abēl a bishop, as in Drasxanakertc‘i and Samuēl, but without the intervention from Smbat recounted in the other histories. Mxit‘ar thus discernibly reduces Smbat’s prominence in the narrative, and in contrast to Samuēl clearly inscribes a boundary between secular and religious authority. He attributes the instruction of the people to the prelate Abēl and reserves the right to ordain him a bishop to the catholicos alone. Finally, like Samuēl, Mxit‘ar also does not indicate that the diocese existed in his lifetime. Indeed, he neglects to mention even the creation of the diocese, lending further support to it having disappeared by the twelfth century. It is not surprising that two authors residing in the former Bagratuni capital would summon the past glories of the Bagratuni in the later twelfth century. At the time, the city was contested between the Kurdish Shaddādid emirs with Eldigüzid support and the Georgian Bagrationi monarchy. The Georgians captured the city in 1174, but it was apparently lost once again to the Shaddādids until 1198/​1199 when it was captured by the amirspasalar (commander-​in-​chief) of the Georgian army, Zak‘arē.51 Samuēl possibly composed his text at the moment of the restoration of Georgian power over Ani, but this was also a tense period. According to Vardan Arewelc‘i, the Georgian king Giorgi III entrusted Iwanē Ōrbēlean with rule of the city. However, the bishop of Ani, Barseł, persuaded Giorgi to remove Iwanē and install his own brother, Apirat, instead. His removal eventually led Iwanē to launch a rebellion against Giorgi in 1176, which the monarch successfully put down.52 Bagratuni genealogies and symbolism were employed by many of the actors as a means to legitimize their control of Ani and Armenia. The Georgian royal Bagrationi family was descended from a junior branch of the Bagratuni family, who had moved into Iberia in the late eighth century to become kings of Iberia in 888 and of the unified Georgian kingdom beginning in 1008.53 Iwanē Ōrbēlean’s spouse, Ṙusudan, was a 51 Sergio La Porta, “ ‘The Kingdom and the Sultanate Were Conjoined’: Legitimizing Land and Power in Armenia During the 12th and Early 13th Centuries,” Revue des études arméniennes 34 (2012): 73–​118, at 77–​86.

52 Sergio La Porta, “Lineage, Legitimacy, and Loyalty in Post-​Seljuk Armenia: A Reassessment of the Sources of the Failed Ōrbēlean Revolt against King Giorgi III of Georgia,” Revue des études arméniennes 31 (2008–​9): 127–​65. 53 Toumanoff, Studies, 203.

80

80 Sergio La Porta member of the Bagratuni royal family and Iwanē staked claims of hereditary rights to Bagratuni property. Similarly, Zak‘arē and his family gradually advanced a Bagratuni lineage for themselves, including the adoption of Bagratuni titles and artistic models, and restoring Bagratuni churches and monasteries, among them that of Haṙič.54 The Ṙupenian princes, later kings, in Cilicia, too, asserted a Bagratuni heritage, as recorded in Samuēl Anec‘i.55 In this climate of Bagratuni legitimization, the genealogy and acts of the Bagratuni engendered great interest. Indeed, Mxit‘ar spends much of his work precisely on that topic, mining his sources for information on the family, and Samuēl’s Chronicle, too, points out significant moments in Bagratuni history. While a general contemporary fascination with the Bagratuni may account for the presence of this story in both accounts, it does not explain why these two authors handle it differently. One immediately notices that Samuēl de-​emphasizes the “Armenianness” of the cultural markers of language and literature present in Drasxanakertc‘i and Tarōnec‘i—​he omits any mention that the language and literacy that was lost and regained was Armenian—​whereas Mxit‘ar reinforces it. While it is possible that the “Armenianness” of both the people captured and their language was assumed by Samuēl, that certainly was not the case with Mxit‘ar. As noted, he reduces the “peoples” to a single “people,” repeats the attribution “Armenian,” and accentuates the concrete sign of the letter. Are we to understand Mxit‘ar’s foregrounding of “Armenianness” to signal anxiety on the author’s part, stemming from the ambiguity latent in the text? The diverse and competing ethnoreligious communities living in Ani and Armenia generally during his life and the lack of a hegemonic regional power at that moment may have provoked Mxit‘ar to define more specifically the cultural identity of the lost Armenians. The absence of a dominant Armenian secular authority may similarly have led Mxit‘ar to downgrade Smbat’s agency and affirm the position of the institutional hierarchy in instruction and ordination. With Samuēl and Mxit‘ar we have two people—​both attached to the same cathedral, with access to the same literary heritage, writing approximately twenty years apart in the same place, for similar if not identical patrons—​who have made very different choices in how they recall this episode. Our knowledge of the authors, the state of their texts, and their patrons are too minimal at present to determine completely what motivated their choices, but their divergence underscores how personal ideologies of identity were and should cause us to pause before projecting generalizations about identity onto premodern societies. 54 La Porta, “Lineage, Legitimacy, and Loyalty,” 127–​65; La Porta “ ‘The Kingdom and the Sultanate,’ ” 89–​102.

55 Samuēl Anec‘i, Hawak‘umn, 116; Nicholas Adontz, “Notes Arméno-​Byzantins. VI. L’aïeul des Roubéniens,” Byzantion 10 (1935): 185–​203, at 195–​97; Zara Pogossian, “The Last Roman Emperor or the Last Armenian King? Some Considerations on Armenian Apocalyptic Literature from the Cilician Period,” in The Armenian Apocalyptic Tradition: A Comparative Perspective, ed. Kevork Bardakjian and Sergio La Porta, SVTP 25 (Leiden: Brill, 2014), 457–​503, at 470–​71.

81



Captive Identities

81

Lost Armenians beyond the Bagratuni Two witnesses to the story of the lost Armenians that date from the late thirteenth century show the currency the episode retained beyond the immediate confines of a Bagratuni orbit. They also provide us with another opportunity to compare how near contemporaries could receive and remember the episode. Vardan Arewelc‘i (ca. 1200–​1271) repeats this incident in his Historical Compilation, completed sometime soon after 1267 at the monastery of Xor Virap. Vardan was a vardapet (monastic scholar and teacher), a biblical exegete, and very involved in contemporary theological issues.56 He travelled widely, working in Cilicia as well as in various monasteries in Armenia, and even journeyed to the court of the Il-​khān Hülegü at the latter’s request.57 Basing himself primarily upon Yovhannēs Drasxanakertc‘i’s account of Smbat, Vardan records the latter’s subduing of Ḵosrow II’s enemies and his reward of the marzpanate of Vrkan, although he makes no mention of Smbat’s earlier contest against the wild animals. According to Vardan, when Smbat arrived in Vrkan, he: found peoples of Armenia captured (azgs hayoc‘ gereal), and dwelling in the great desert of T‘urk‘astan called Sagastan, who had forgotten their language and literacy (dprut‘iwn). When they saw Smbat they rejoiced and they restored their language and writing (gir). He had a presbyter of theirs, Habēl by name, ordained bishop by the great patriarch Movsēs; he confirmed them as a diocese of St. Grigor.58

Vardan also repeats the information that subsequently Smbat was permitted to visit “his own land,” where he rebuilt the cathedral of Duin. Although it is evident that Vardan’s rendering of the story is based on Drasxnakertc‘i’s and Samuēl Anec‘i’s accounts,59 he was also aware of Mxit‘ar’s reading, as intimated from his substitution of gir, “writing,” for dprut‘iwn, “literacy”; as well as from his saying that they had forgotten their literacy, not that it had been greatly reduced.60 Nonetheless, he chose not to follow Mxit‘ar’s clear separation of secular and religious authority, adhering instead to Drasxanakertc‘i’s and Samuēl’s articulation of that dynamic. We also notice that Vardan chose not to designate the captives with the collective noun hayastaneayk‘, but instead followed Step‘anos, Samuēl, and Mxit‘ar in using the ethno-​toponym hayk‘. In contrast to these earlier authors, however, Vardan employs the noun in the genitive rather than in the ablative case. Thus, the captives are not “from Armenia,” but “of Armenia,” or “Armenian,” which emphasizes the ethnic dimension over the geographic, and brings his reading closer to that of Drasxanakertc‘i in meaning, if not lexically. The episode may have also borne 56 Vardan Arewelc‘i, “The Historical Compilation of Vardan Arewelc‘i,” trans. Robert W. Thomson, Dumbarton Oaks Papers 43 (1989): 125–​226, at 126–​28. 57 Vardan Arewelc‘i, Chronicle, 155–​56; Vardan Arewelc‘i, “Historical Compilation,” 220. 58 Vardan Arewelc‘i, Chronicle, 60; Vardan Arewelc‘i, “Historical Compilation,” 173.

59 Vardan’s phrasing more closely echoes that of Samuēl Anec‘i in the final sentence of the text, including naming the presbyter Habēl rather than Abēl.

60 Vardan refers to Mxit‘ar in his work, Vardan Arewelc‘i, Chronicle, 92, 94, 137; Vardan Arewelc‘i, “Historical Compilation,” 190, 191, 211.

82

82 Sergio La Porta other resonances for Vardan. Later in his history, he reports that a Scythian (i.e., Türkic) tribe led by a certain Il-​Ghāzī, possibly to be identified with the Artuqid Il-​Ghāzī, who “in earlier times” had been hiding among the Vrkanians, attacked Ani between 1092 and 1095. Vardan may have imagined that the Armenians settled in Vrkan found themselves in a foreign land among such hostile “Scythians.”61 Finally, Mxit‘ar Ayrivanec‘i (1222–​ca. 1300) records the episode in his Chronicle, which ends in 1289. Ayrivanec‘i was, like Vardan, a vardapet (monastic scholar and priest), but also an archbishop and the abbot of the monastery of Ayrivank‘ (“monastery of the caves”). In addition to his Chronicle, he composed hymns and vitae for the Church, compiled a giant miscellany of biblical books and educational texts, and was particularly interested in calendrical systems. Ayrivanec‘i presents us with the briefest account of the lost Armenians. Ayrivanec‘i does not mention that Smbat served Ḵosrow II nor that he was appointed marzpan. Instead, he whittles the story down—​almost as if posting a tweet62—​to only Smbat’s finding of the Armenians, which he places between events dated to 571 and 601: Smbat Bagratuni goes to Vrkank‘ (sic), which is Sagastan, and finds Armenians (hayk‘) [who] forgot [their] language (zlezu), and he appoints Habēl their bishop.63

In addition to omitting the context of Smbat’s appointment, Ayrivanec‘i did not think it significant to remember that the Armenians in Vrkan had been captured and settled there; that not just their language but their literacy was also lost; that they relearned their language and writing; or that the catholicos may have been involved. The toponym Vrkan seems to have been unfamiliar to him as he adds the marker of the plural to it,64 though Sagastan apparently retained its comprehensibility. Furthermore, he refers to the “people from/​of Armenia” simply as “Armenians.” In sum, Ayrivanec‘i has completely simplified and decontextualized the story as well as discarded those troublesome words that were open to multiple interpretations and continue to cause readers such problems, such as azg (nation, people), gereal (captive), dprut‘iwn (writing, literature, literacy), and even norogel (renew).65 61 This identification would have been reinforced if Vardan understood “Sagastan” to mean “land of the Saka” or “Scythians,” but he nowhere gives an indication that he did read it in such a manner. As noted, Anania Širakac‘i, Geography, 74, includes Sagastan in Scythia, so Vardan may have made the association nonetheless.

62 In the printed edition, which resolves any abbreviations that likely appeared in the manuscript, this notice expends a total of 84 characters. It may be too far to speculate that Ayrivanec‘i’s obsession with calendars and time may have driven him to arrive at an economic efficiency of words at the cost of aesthetic considerations.

63 Mxit‘ar Ayrivanec‘i, Patmut‘iwn žamanakagrakan, ed. K̕erovbê Patkanean (St. Petersburg: Imperial Academy of Sciences, 1867), 64; Mxit‘ar Ayrivanec‘i, Histoire chronologique, trans. Marie-​ Félicité Brosset (St. Petersburg: Imperial Academy of Sciences, 1869), 74. 64 It is possible that this should be attributed to later scribal error.

65 The interpretation and translation of gereal (captive) and norogeal (renew) is not problematic lexically, but semantically. The concept of captivity raises the questions noted above of who captured

83



Captive Identities

83

Vardan’s and, to a greater degree, Ayrivanec‘i’s foregrounding of ethnic identity over other forms may speak to the dispersed nature of Armenian communities in the thirteenth century that prompted Armenians to conceptualize an Armenian identity beyond geographic boundaries. The shifting of the centre of Armenian political life from the Armenian plateau to Cilicia may have further contributed to a de-​territorialized sense of identity. Similarly, the establishment of the monarchy by Lewon I Ṙupenian in 1198 may account for Vardan’s and Ayrivanec‘i’s comfort with Smbat’s ordering the catholicos to ordain Abēl a bishop. Presenting Smbat as the primary actor in this endeavour echoes Samuēl’s expression of secular prerogatives and signals a rejection of Mxit‘ar Anec‘i’s desire to diminish Smbat’s authority and to separate religious and secular spheres of activity.

Conclusion

I have argued that as a lieu de mémoire the story of the lost Armenians provided a place for its narrators to inscribe ideologies of identity and authority. The variations between the rememberings of the story arose from deliberate choices made by each of the authors and did not result from simple textual slippage or scribal error. Whether it was through the act of choosing a particular source to foreground, of remembering or unremembering certain parts of the story, of creating “mnemonic additions,”66 or of changing the terminology, these authors contoured the story in their retelling that responded to their own position or rhetorical purposes. It is useful here to review what has been lost, added, or altered in the course of the story’s rewriting. A good deal of the story was successfully unremembered already by Yovhannēs Drasxanakertc‘i; successful, in that other authors who seem to have had access to Sebēos did not choose to remember these lost memories later. By far the biggest loss forged by Drasxanakertc‘i was the erasure of the other people—​the Kodrik‘, the Greeks, and the Syrians—​who had been settled in the desert of Vrkan. Delhastan, too, was wiped off the map. With the loss of the “others,” the great light of divine indication that shone over the Christians proved to be unnecessary. Drasxanakertc‘i also obscures that the captives did not have a priest and were not part of the Armenian Apostolic Church. Other acts of unremembering happened later, such as the captives being confirmed in their faith, which Step‘anos Tarōnec‘i first excluded and no subsequent retelling chose whom exactly, when, and why. Likewise, “renew” implies that that which has been lost was once there and raises questions of whether the loss was complete or not, of who did the renewing, and how was the renewal accomplished. By omitting these terms, Ayrivanec‘i closes off those avenues of inquiry. 66 On “mnemotic additions” in Arabic geographical literature see Vacca, Non-​Muslim Provinces, 50–​52, developed from the work of Sarah Savant, “Forgetting Ctesiphon: Iran’s Pre-​Islamic Past, c. 800–​1100,” in History and Identity in the Late Antique Near East, ed. Philip Wood (Oxford: Oxford University Press, 2013), 169–​86, and Savant, The New Muslims of Post-​Conquest Iran: Tradition, Memory, and Conversion (Cambridge: Cambridge University Press, 2013).

84

84 Sergio La Porta to reincorporate. Samuēl and Mxit‘ar Anec‘i forgot that the settlement was on the edge of a desert; while, as already noted, Mxit‘ar Ayrivanec‘i erased nearly the entire context and plot of the story.67 While unremembering shaped a good deal of the remembrance of this narrative, mnemonic additions also figure significantly in the repositioning of this story. Yovhannēs Drasxanakertc‘i created new memories by adding an explicitly prominent role to Smbat as the motivator for the restoration of the captives, defining the desert where the captives were found as Sagastan, attributing joy to the captives at their seeing Smbat, having the catholicos assume the responsibility of appointing Abēl a bishop, and remarking that this action created a new diocese of the see of St. Grigor. These additions did not remain completely stable, however, as they themselves were subject to their own erasure and inflection. For example, Tarōnec‘i modifies the joy of the captives at seeing Smbat to the joy of Smbat at seeing the captives; Samuēl and Vardan think of the diocese only as having existed in the past, while Mxit‘ar Anec‘i and Ayrivanec‘i do not remember its creation at all. Subtler shifts of language and meaning further adjusted the direction of the story. So, “from Armenian land” (Sebēos) became “from Armenians” (Drasxanakertc‘i), then “from Armenia” (Tarōnec‘i, Mxit‘ar Anec‘i, Samuēl), then “of Armenia” (Vardan), and finally, “Armenians” (Ayrivanec‘i). Literacy went from being deficient (Sebēos) to very reduced (Drasxanakertc‘i) to completely lost; “literacy” itself became “writing,” specifically in Mxit‘ar Anec‘i, a move echoed in Vardan. Yovhannēs Drasxanakertc‘i reversed the process of restoring the captives from Sebēos, who had them first be confirmed in the faith, then become literate, and finally able to speak. Abēl went from becoming a priest in Sebēos to becoming a bishop in Drasxanakertc‘i and all the others, except for Tarōnec‘i who makes him a vardapet and Mxit‘ar Anec‘i who has him first appointed a prelate and then a bishop. To a large extent, the question of Armenian identity, of “Armenianness,” lurks behind these narrative decisions of erasure, addition, or alteration. Of the four variables of identity adduced in the text—​place, ethnicity, language, and confession—​that of confession remained the most stable following Drasxanakertc‘i. Confession permitted the basic identification of the captives as Armenian and the episode culminates in a jurisdictional expansion of the Armenian Church. This is perhaps not surprising given that all of the authors were members of the ecclesiastical hierarchy. By contrast, language and literacy do not figure prominently as minimal indicators of identity as they could be partially or completely lost without impacting the identification of the captives as Armenian, although they remained significant for full inclusion within the group. 67 In addition to these acts of erasure within this narrative unit, we may note the unremembering of other units related to Smbat that surround and contextualize this one. For example, Drasxanakertc‘i, Tarōnec‘i, Arewelc‘i, and Ayrivanec‘i omit the fight in the arena, while Samuēl misattributes the condemnation to a Persian, and Mxit‘ar Anec‘i leaves it unclear as to who condemned Smbat. Tarōnec‘i, Samuēl, and Ayrivanec‘i make no mention of Smbat’s building program for the cathedral at Duin. Finally, Tarōnec‘i makes Smbat the marzpan of Armenia rather than of Vrkan.

85



Captive Identities

85

It is tempting to see in the movement from “from Armenian land” to “Armenians” a chronological process of ethnicization, whereby identity based on a common descent from a mythic ancestor became privileged over that of a common place of origin.68 One could cite in support of this development Movsēs Dasxuranc‘i’s composition of the History of the Ałuank‘.69 Although written in Armenian and based on earlier sources, it is the only surviving native history of that people. The problem of Ałuanian identity is complex, but the desire to compile such a work at the end of the tenth century likewise suggests a similar process of ethnic idealization. Further validation may be found in ter Haar Romeney’s observation that the clerical elite of the Syriac Orthodox community gradually foregrounded a sense of ethnic identity in the period between the tenth and thirteenth centuries.70 Nonetheless, three factors destabilize any proposed fixed notion of “Armenianness.” First, the terms employed by the authors cited above are not exclusive; that is, they all use each of them at different points in their works. It is generally not possible to discern any stark difference in intention between hayastaneayk‘ and hayk‘; or between the uses of the oblique cases of hayoc‘ meaning “Armenian” in a geographic as opposed to an ethnic sense. Such nuances are only available by examining the terms in context. Second, our evidence derives from a small social and intellectual elite, who was literate and very often associated with the Church. Finally, and most significantly, it is critical to emphasize that the above readings reveal that at no point did a single ideology of Armenian identity exist. When we contextualize “Armenianness” within the retellings of this story, its definition emerges to be situational and localized to the point of the personal. “Armenianness” contained multiple possible articulations into which these narrators invested their unconscious biases and anxieties as well as onto which they projected ideologies of identity and authority that supported their rhetorical aims. 68 Cf. Areshian, “Historical Dynamics,” 34. As noted above, at the end of the eighteenth century, the Mekhitarist historian Mik‘ayēl Č‘amč‘ean chose to recall the story of the lost Armenians in his monumental History of the Armenians. His remembering of the episode was based on the accounts of Drasxanakertc‘i, Tarōnec‘i, and Samuēl Anec‘i, but he accentuates the ethnic distinction between Armenians and Persians. Thus, according to him, “many Armenians” (bazum hays) had been previously captured by the Persians (i parsic‘) and settled in Sagastan. Having forgotten “the language and writing of the Armenians” (zlezu ew zgirs hayoc‘), Č‘amč‘ean adds they conducted their affairs in Persian (varēin parskakanōk‘), Č‘amč‘ean, Hayoc‘ Patmut‘iwn, 2:300. How this episode featured in Č‘amč‘ean’s construction of Armenian identity remains outside the limits of this study, but his inclusion of it underscores the resilience and utility of the episode for Armenian historiographers. 69 Movsēs Dasxuanc‘i, Patmut‘iwn Ałuanic‘ Ašxarhi, ed. V. Aṙak‘elyan (Yerevan: Academy of Sciences, 1983); Movsēs Dasxuanc‘i, The History of the Caucasian Albanians by Movsēs Dasxuranc‘i, trans. Charles Dowsett (London: Oxford University Press, 1961).

70 Bar ter Haar Romeny,“Ethnicity, Ethnogenesis and the Identity of Syriac Orthodox Christians,” in Vision of Community in the Post-​Roman World: The West, Byzantium and the Islamic World, 300–​1100, ed. Walter Pohl, Clemens Gantner, and Richard Payne (Farnham: Routledge, 2012), 183–​204, at 201–​2.

86

87

Chapter 5

WAR AND IDENTITY IN EARLY MEDIEVAL BULGARIA Panos Sophoulis Perhaps more than anywhere else in early medieval Europe, Bulgaria was a society organized for war. For approximately the first two hundred years of its existence, the Bulgar polity was forced onto the defensive in a desperate battle for survival. With far fewer resources than its aggressive Byzantine and Khazar neighbours, it was liable to political collapse in the face of inexorable pressure. Its rulers had therefore little choice but to keep their subjects on a permanent war footing and divert all of the state’s resources for military purposes. In the discussion that follows it will be argued that warfare, more than any other factor, had a decisive impact on the political and social development of Bulgaria in the Early Middle Ages. Specifically, the chapter will explore how: i) the high level of militarization resulted in sweeping changes that occurred at almost every social level, but mainly affecting the relationship between the various tribes and clans that constituted the political elite of the Bulgar state; ii) the intense military pressure exerted by its powerful sedentary neighbour, Byzantium, coupled with the growing influence of Christianity in the realm, strengthened the Bulgars’ connection with the steppe imperium and facilitated the development of an aggressive Bulgar “nationalism,” clearly traceable from the middle of the eighth century onwards; iii) the continual contact—​military, economic, and cultural—​which populations in Bulgaria had with the Byzantine world appear to have created both a high level of “Byzantine awareness” as well as a perception of the empire as “the other.” Such an awareness, among other things, strengthened political ties between the heterogeneous components of the state and reinforced the sense of collective Bulgar identity. The Emergence of the Bulgar State

The Bulgar state or khanate in the Balkans was established in the early 680s by a group of nomadic warriors of Türkic-​Oğuric origin, who had been present in the steppes north of the Black Sea at least since the fourth century.1 Around the first quarter of the seventh century, one of their leader, a certain Kubrat, is said to have thrown off Avar rule and to have created an independent polity in the Lower and Middle Dnieper region, “Old Great 1 For a discussion, see Peter B. Golden, An Introduction to the History of the Turkic Peoples: Ethnogenesis and State Formation in Medieval and Early Modern Eurasia and the Middle East (Wiesbaden: Harrassowitz, 1992), 97–​104; Veselin Beševliev, Die protobulgarischen Periode der bulgarischen Geschichte (Amsterdam: Hakkert, 1981), 314–​17.

88

88

Panos Sophoulis

Bulgaria,” which maintained close diplomatic relations with the Byzantine Empire.2 However, following Kubrat’s death, his confederation was pulled apart by the new dominant force in the region, the Khazars, and some of its elements were subsequently forced to migrate to the West. One of Kubrat’s sons, Asparuch, led his nomadic horde, comprising peoples with different anthropological and cultural characteristics as shown by grave finds, just north of the Danube Delta.3 From there the Bulgars began raids on the Balkan provinces of the Byzantine Empire, and following their victory over the army of Constantine IV in 680 they moved across the river, incorporating the Slav and Christian inhabitants of that region into a subject tribal union.4 Despite the fact that Byzantium was forced to recognize the new status quo by signing a treaty and pledging to pay an annual tribute to Asparuch, eighth, ninth, and tenth-​century emperors continued to regard the whole area settled by the Bulgars as imperial territory, and accordingly made repeated attempts to either drive the newcomers out of the Balkans or impose imperial suzerainty and a lasting peace—​something that was only achieved by Basil II at the beginning of the eleventh century.5 The members of the Bulgar ruling class, as already remarked, were culturally related to the peoples of the Eurasian steppes. Their religious beliefs and daily habits in their new homeland in the Lower Danube region were typical of those practised by other nomadic aristocracies.6 Likewise, their political, military, and social organization was 2 Theophanes, Chronographia 357.9–​11, ed. Carl de Boor (Leipzig: Teubner, 1883); Nikephoros Patriarch of Constantinople, Short History 22.1–​7, ed. and trans. Cyril Mango (Washington, DC: Corpus Fontium Historiae Byzantinae 13, 1990). For the burial assemblages in the Lower and Middle Dnieper region, which, among other things, included Byzantine golden coins and other valuable objects that served as gifts from Constantinople, see Joachim Werner, Der Grabfund von Malaia Pereščepina und Kuvrat, Kagan der Bulgaren (Munich: Verlag der Bayerischen Akademie der Wissenschaften, 1984), 31–​36, 35–​44, fig. 32, 1–​2; Péter Somogyi, “New Remarks on the Flow of Byzantine Coins in Avaria and Walachia during the Second Half of the Seventh Century,” in The Other Europe in the Middle Ages: Avars, Bulgars, Khazars and Cumans, ed. Florin Curta (Leiden: Brill, 2008), 83–​149, esp. 127–​28.

3 Theophanes, Chronographia 357.11–​358.3; Nikephoros, Short History 35.5–​25. For a summary of the archaeological evidence from Bulgar cemeteries that shows the culturally diverse composition of the migrating tribes, see Panos Sophoulis, Byzantium and Bulgaria, 775–​831 (Leiden: Brill, 2012), 66–​69. 4 Theophanes, Chronographia 358.11–​359.21; Nikephoros, Short History 36.1–​29.

5 Jonathan Shepard, “Slavs and Bulgars,” in The New Cambridge Medieval History, II: c. 700–​c. 900, ed. Rosamond McKitterick (Cambridge: Cambridge University Press, 1995), 228–​48; Florin Curta, Southeastern Europe in the Middle Ages, 500–​1250 (Cambridge: Cambridge University Press, 2006), 81–​90, 237–​47; Daniel Ziemann, Vom Wandervolk zur Großmacht: Die Entstehung Bulgariens im frühen Mittelalter (7. Bis 9. Jh.) (Cologne: Böhlau, 2007), 180–​84; Dennis P. Hupchick, The Bulgarian-​Byzantine Wars for Early Medieval Balkan Hegemony: Silver-​Lined Skulls and Blinded Armies (Cham: Palgrave Macmillan, 2017), 47–​122.

6 For more details, see Dimităr Ovčarov, Prabălgarskata Religija: Proizhod i Săshtnost (Sofia: Gutoranov i sin, 1997); Veselin Beševliev, Părvobălgarite. Bit i Kultura (Sofia: Nauka i Izkustvo, 1981; repr. Plovdiv: Bălgarsko Istoričesko Nasledstvo, 2008).

89



War and Identity in Early Medieval Bulgaria

89

structured along traditional nomadic lines that are well known to us from other “steppe empires.” To give just one example, in early medieval Bulgaria, as in the Türk and Avar qaghanates, military and political administration appear to have been inseparable, and senior office-​holders usually combined military and civilian duties.7 To be sure, warfare played a central role in nomad culture; in fact, it was an integral part of the nomadic way of life—​it was learned in childhood and was constantly practised—​and this certainly accounts for the creation of the topos of the steppe-​nomad as a “natural warrior” that is commonplace in Roman, Byzantine, and Chinese literary tradition.8 While warfare was so closely intertwined with the life of nomads, its intensification—​ that is, the increase in the level of militarization—​usually brought about significant changes at every level of society. In the case of Bulgaria, the growing military threat posed by Byzantium from the mid-​eighth century onwards seems to have facilitated the drive toward greater centralized political power. Exactly how this happened is not entirely clear. Nevertheless, the surviving sources point to the existence of intense political contention among rival clans, sparked by Constantine V’s repeated campaigns against the Bulgar state.9 As war took its toll, the Dulo, the leading “charismatic” clan, whose legitimacy rested on an alleged line of descent from the Western Türks, was unable to assert its authority over the nobility and was eventually toppled by a more powerful faction.10 Over the next few years, and as Constantine V’s offensive continued, the decision-​ making council or komventon in the Byzantine sources—​an institution that still yielded considerable power—​appointed and deposed several rulers who represented different clans.11 Then, toward the turn of the century, a new set of khans, who may well have marked the rise of a new ruling dynasty, were able to tighten their control over the military aristocracy, dramatically altering the relationship among the tribes and lineages that constituted the upper echelons of Bulgar society. Once again, although the exact sequence of events remains poorly understood, there is little doubt that these changes unfolded in response to significant military developments. Specifically, the period between the end of the eighth and the beginning of the ninth century witnessed, on the one hand, a new phase of active Byzantine involvement in the Balkans, which culminated in a series of campaigns of Emperor Nikephoros I (802–​811) against Bulgaria, and on the other, the disintegration of the Avar qaghanate, which opened up opportunities 7 Sophoulis, Byzantium and Bulgaria, 69–​71, 73–​76.

8 Denis Sinor, “The Inner Asian Warriors,” Journal of the American Oriental Society 101 (1981): 133–​44; Nicola Di Cosmo, “Introduction: Inner Asian Ways of Warfare in Historical Perspective,” in Warfare in Inner Asian History (500–​1800), ed. Nicola Di Cosmo (Leiden: Brill, 2002), 1–​29.

9 Theophanes, Chronographia 432.25–​27; Borislav Primov, “Bulgaria in the Eighth Century: A General Outline,” Byzantinobulgarica 5 (1978): 7–​41, esp. 24–​35; Ziemann, Vom Wandervolk zur Großmacht, 213–​34. 10 Hupchick, Bulgarian-​Byzantine Wars, 47–​62.

11 Theophanes, Chronographia 433.14–​22; Nikephoros, Short History 77.1–​11, 79.1–​12; Beševliev, Bit i Kultura, 297–​98.

90

90

Panos Sophoulis

for profitable expansion in that direction.12 There is good reason to believe that the early career of Krum, one of the most successful rulers in early Bulgar history, was closely bound up in his role as a leader in the offensive in the former terra Avarorum. His unprecedented successes in that region,13 but also in the southern frontier, in the wars against Byzantium, allowed him to secure a broad range of aristocratic support—​ arguably far broader than any of his predecessors. This support ultimately stemmed from his sheer military ability and personal charisma, but also from his own skilful manipulation of force within an elaborate tribal network. This, in turn, enabled him to monopolize revenues yielded by punitive raids, thereby becoming, in effect, the primary source of wealth—​and hence of political power—​in the Bulgar state. Indeed, a chronicle reporting on the Byzantine–​Bulgar wars of the early ninth century informs us that after capturing the Bulgar capital Pliska in 811, Nikephoros I discovered in Krum’s palace a large amount of treasure including stocks of metal, clothing, and other valuables.14 This strongly suggests that the khan had built up a reserve of wealth, on which a political mechanism of patronage and rewards would have been based. The evidence pointing to an increased centralization of authority in ninth-​century Bulgaria is abundant. Already during the reign of Krum the influence of the decision-​ making council had declined; it could no longer restrict the powers of the ruler, and played very little or no legislative role.15 No khan is known to have been killed or deposed after 815—​a far cry from the conditions present in the realm in the previous century. What is more, this period was marked by episodes of despotic behaviour. The construction of palaces and temples, the building of earthworks, the production of stone inscriptions which conveyed the official ideology of the state, as well as the orchestration of elaborate public rituals and ceremonies, are all interpreted as explicit statements of royal power.16 And thus, as Krum and his immediate successors were able to control the distribution of booty accumulated during military operations, the relationship between the ruler and the rest of the Bulgar aristocracy went from—​relatively—​egalitarian and horizontal to—​increasingly—​hierarchical and vertical. Access to the khan’s inner circle, which was inseparably linked to the holding of office in the civil and military administration, was no longer limited to members of the dynastic clan and the “traditional” nobility, 12 A detailed discussion in Sophoulis, Byzantium and Bulgaria, 180–​216 and Hupchick, Bulgarian-​ Byzantine Wars, 67–​122. For the collapse of the Avar qaghanate, see Walter Pohl, The Avars: A Steppe Empire in Central Europe, 567–​822 (Ithaca: Cornell University Press, 2018), 376–​89. 13 The tenth-​century compilation known as Suidas credits Krum with the destruction of the Avar state. Although this is inaccurate, the story clearly echoes the growing Bulgar involvement and success in that region. Suidae Lexicon 483–​84, vol. 1, ed. Ada Adler (Leipzi g: Teubner, 1928). 14 Ivan Dujčev, “La Chronique Byzantine de l’an 811,” Travaux et Mémoires 1 (1965 ): 212.19–​20.

15 Tsvetelin Stepanov, “The Bulgar Title KANASUBIGI: Reconstructing the Notions of Divine Kingship in Bulgaria, AD 822–​836,” Early Medieval Europe 10 (2001): 1–​19, at 3–​4. 16 Florin Curta, “Qagan, Khan or King? Power in Early Medieval Bulgaria (Seventh to Ninth Century),” Viator 37 (2006): 1–​31, at 23.

91



War and Identity in Early Medieval Bulgaria

91

but instead it was expanded to include several “outsiders” such as Slavs, Christians, and provincial elites.17 The creation of a special circle around the ruler is reflected in the appearance under Krum’s son Omurtag—​who, just like his father, spend most of his reign campaigning in the Byzantine Balkans, the Carpathian Basin, and the steppes north and northeast of the Danube Delta—​of a new institution, the so-​called “nurtured men” (threptoi anthropoi), known from a set of inscription carved by order of the khan to commemorate his fallen comrades.18 The union between the ruler and these nobles was forged through initiatory rituals involving food, drink, and gift exchange, which provided participants the opportunity to publicly communicate their status.19 Such interchanges are generally thought to have served as a means of establishing bonds of communal solidarity and accentuating the importance of aristocratic kinship networks.20 What is particularly interesting in Omurtag’s case is that the “nurtured men” mentioned in his inscriptions belong to different tribes or clans, perhaps a sign that this ruler tried to build some form of group identity that looked beyond the pre-​existing social boundaries. The successful prosecution of war and the subsequent territorial expansion of the Bulgar state in the first decades of the ninth century enabled its rulers to develop a more elaborate administrative hierarchy which further subordinated the nobility to the throne. A large number of titles—​much larger than those known up to the late 700s—​ is recorded during this period, a fact that by itself suggests a wholesale attempt to restructure state bureaucracy.21 Omurtag and his immediate successors created what could be described as a “khan-​centred” descending hierarchy aimed at controlling the powers of the nobility, especially in the rapidly expanding periphery of the realm. So far away from the capital, power could not be delegated unless it was institutionalized. Thanks to the increasing hierarchization of the state, Omurtag made his nobles directly answerable to the centre. In this context, literacy was employed as a tool by the ruler, in order to record and publicize the services extracted from the nobility. A number of inventory inscriptions discovered in or near the capital, each listing, under the title of an officer, the number of weapons that he was required to provide, often under penalty of death, clearly show how the khan’s orders came to be presented in formal and public terms, thereby underlining the legitimacy of his power.22 17 Sophoulis, Byzantium and Bulgaria, 289, 305–​6.

18 Veselin Beševliev, Părvobălgarski Nadpisi (Vtoro Predraboteno i Dopălneno Izdanie) (Sofia: Bălgarska Akademija na Naukite, 1992), nos. 59–​69. 19 Beševliev, Nadpisi, no. 58.

20 Aron J. Gurevich, Categories of Medieval Culture (London: Routledge & Kegan Paul, 1985), 222–​33, 244–​48; Florin Curta, “Iron and Potlatch: Early Medieval Hoards of Implements and Weapons in Eastern Europe,” Archivum Eurasiae Medii Aevi 10 (1998–​99): 15–​62, at 37–​38. 21 Beševliev, Nadpisi, nos. 50, 60, 64, 65, 69; Vasil Gjuzelev, Kavhanite i Ičirgu Boilite na Bǎlgarskoto Hanstvo-​tsarstvo (VI–​XI v.) (Plovdiv: Bălgarsko Istoričesko Nasledstvo, 2007), 43–​44. 22 For example, Beševliev, Nadpisi, no. 48: “bagatur bagain has altogether 53 coats of mail and 45 helmets.”

92

92

Panos Sophoulis

Protracted Conflict and Ethnocentrism Although Byzantine–​Bulgar relations were belligerent from the outset, it was only during the reign of the warlike emperors of the mid-​eighth and ninth centuries that the empire put up a concerted effort against its newly arrived neighbour. Characterized as it was by a small ruling elite, the Bulgar polity was almost bound to undergo stress once it faced the full military might of a superior power. The subsequent political crisis manifested itself in the form of bloody coups, shifting clan alliances, and the refusal of certain factions to accept the authority of others.23 The surviving sources that describe these events indicate that a particularly influential section of the Bulgar nobility regarded Byzantium as a hereditary foe and pushed for a more aggressive stance, as opposed to what seems to have been a less radical faction that saw Bulgaria’s future closely bound up with that of the empire.24 To be sure, at least two rulers seeking peace arrangements with Constantine V in the 760s were deposed or killed by members of the militant wing, who then went on to destroy a network of Byzantine agents and sympathizers operating at the time in Bulgaria.25 As we will see below, by the early ninth century, amid a fresh wave of Byzantine attacks, this hostile attitude toward the empire, which had earlier helped maintain cohesion among the warrior class, appears to have developed into an aggressive and violent anti-​Byzantine movement. However, simultaneously, the rulers of this period formulated anew their self-​awareness, consciously (re)asserting their identification with the Eurasian world. This comes as little surprise, for scholars have pointed out that when traditional Türkic institutions flourished among Eurasian nomads, they were subject to very strong cultural or political pressures.26 Indeed, finding themselves in an essentially non-​steppe environment, where they were vastly outnumbered by their agriculturalist Slav subjects, and being threatened by their Byzantine (and possibly Khazar) neighbours, the Bulgars’ connection with the steppe became far more pronounced. It was thus during the reign of Omurtag that the title kana sybigi—​ a qaghanal title emphasizing the idea of a sacral sanction of political rule—​becomes clearly visible. It is attested, usually together with its Greek equivalent ek theou archon (“ruler by God”), in building, memorial, and other stone inscriptions, which were meant to propagate the idea of Omurtag as a “divinely-​appointed monarch.”27 At the same time, the Bulgars actively promoted the cult of Tangra/​Tängri, the sky-​god and supreme divine being of many Central and Inner Asian nomads, which from the sixth century had become closely associated with the ruling clan of the Türk empire.28 Much like the qaghanal title, the cult of Tangra is first recorded in Bulgaria under Omurtag, and it is 23 Primov, “Bulgaria in the Eighth Century,” 24–​35; Hupchick, Bulgarian-​Byzantine Wars, 47–​62. 24 Hupchick, Bulgarian-​Byzantine Wars, 61.

25 Theophanes, Chronographia 433.14–​22, 448.4–​10.

26 Peter B. Golden, “Imperial Ideology and the Sources of Political Unity amongst the Pre-​Činggisid Nomads of Western Eurasia,” Archivum Eurasiae Medii Aevi 2 (1982): 37–​76, at 73. 27 Curta, “Qagan, Khan or King?” 28–​29.

28 Golden “Imperial Ideology,” 44–​45; Sophoulis, Byzantium and Bulgaria, 83–​86.

93



War and Identity in Early Medieval Bulgaria

93

hard to avoid the conclusion that its emergence enhanced royal power even further, for it endowed it with an aura of sanctity.29 It is clear, then, that the military pressure exerted on the Bulgar state in the early 800s—​ a pressure that nearly led to its complete collapse following the sack of the capital Pliska by the imperial forces in 811—​exacerbated the bitter anti-​Byzantine feeling among the elite, whose ancient culture became more self-​conscious and aggressive in the face of adversity. This resulted, among other things, in the adoption of a more militant stance toward Christianity in the Bulgar realm. Indeed, a significant number of Christians—​remnants of the late-​antique population north of the Haimos Mountains, as well as immigrants and refugees from Byzantium—​had been present there in the seventh and eighth centuries, but there is strong evidence that the Bulgars took a tolerant approach toward them.30 With their literary and organizational skills, these men were quickly incorporated into the central government and the army, and the Bulgar rulers tried to develop a shared sense of identity with them (see the section titled “The Byzantine Influence” below). During the wars of the early ninth century, the size of the Christian community grew, thanks to the arrival of thousands of prisoners from Byzantine Thrace and the environs of Constantinople. Their influence seems to have facilitated the spread of Christianity, particularly among the Slavs who appeared to be more receptive to the new faith than their Bulgar overlords.31 Still, it is highly unlikely that the severe persecution of Christians carried out early in 815 had anything to do with the progress of their religion in Bulgaria. Rather, it was a violent reaction to poor performance and, ultimately, failure on the battlefield. Indeed, just a few weeks prior to these events, the Bulgar army had experienced a crushing defeat at the hands of Emperor Leo V outside the city of Mesembria on the Black Sea coast. The outcome of the battle came as a major shock to the Bulgars, who since 811 had had the upper hand in the ongoing struggle against Byzantium.32 A number of Byzantine generals (possibly defectors—​we know at least four of them by name) had been assigned in important military commands in the territories annexed by the Bulgars in Thrace, and it is safe to infer that they had fought against the imperial forces at Mesembria.33 Omurtag, who had just come to power, is likely to have suspected 29 For the badly damaged inscription which commemorates a sacrifice made by Omurtag “to the god Tangra,” see Beševliev, Nadpisi, no. 6. The inscription was found at the rocky cliff of Madara, a site that is commonly associated with the cult of Tangra. See for example, Rašo Rašev, “Srednovekovnite postroiki severno ot skalnija relef v Madara,” Madara 3 (1992): 109–​26, at 117. 30 Shepard, “Slavs and Bulgars”, 232; Robert Browning, Byzantium and Bulgaria: A Comparative Study across the Early Medieval Frontier (London: Temple Smith, 1975), 281.

31 Robert Browning, “Byzantines in Bulgaria—​Late 8th–​Early 9th Centuries,” in Studia Slavico-​ Byzantina et Mediaevalia Europensia: In Memoriam I. Dujchev, ed. Petăr Dinekov et al., vol. 1 (Sofia: Tsentr dlja Slavjano-​Vizantijskih Ιssledovanij im. Ivana Duičeva, 1988), 32–​36; Rašο Rašev, “Vizantijtsite v Bǎlgarija do pokrǎstvaneto,” in Civitas Divino-​humana. V Čest na Profesor Georgi Bakalov, ed. Tsvetelin Stepanov and Veselina Vačkova (Sofia: Tangra, 2004), 151–​62. 32 Sophoulis, Byzantium and Bulgaria, 266–​70. 33 Beševliev, Nadpisi, no. 47.

94

94

Panos Sophoulis

these men of treason or may have simply felt it was no longer safe to have Byzantines in high positions on the border. Personal and political conflicts within the Bulgar elite undoubtedly contributed to the tenseness of the situation. At any rate, it is clear that the new ruler decided to associate himself with the interests of the militant warlike aristocracy, which would have welcomed any measures taken against the Christian population of Bulgaria. According to a number of hagiographical sources, hundreds were subsequently put to death, including two of the aforementioned generals, the archbishop of Adrianople, and several other prominent captives, who provided a handy scapegoat for the defeat.34 As was frequently the case with steppe-​nomad states, legitimization of political rule in Bulgaria rested largely on the vigorous and successful prosecution of warfare. In a bid to consolidate his hold on power early on in his reign, Omurtag decided to engage in a policy of intensifying confrontation with Byzantium, knowing that military action was bound to produce a higher degree of political cohesion and enhance the power of the ruler. He was thus able to secure the support of an influential section of the Bulgar elite which regarded Byzantium as a hereditary foe, but in keeping with this stance, his actions were overtly and at times violently anti-​Christian. Overall, this was a time when Bulgar culture emerged as more self-​conscious and aggressive, a time when the nobility’s connection with the world of the steppes and its traditions became stronger than ever before—​a phenomenon that, once again, should be interpreted as a response to the “outside” pressures generated by their powerful sedentary neighbour.

The Byzantine Influence It has already become clear that from the moment of their arrival in their new homeland in the Lower Danube region, if not earlier, the Bulgars maintained very close contact with the Byzantine world. The frequent and intensive warfare reported by our sources is only one side of the story. As a matter of fact, there is sufficient evidence pointing also to a growing economic and cultural interaction across the Byzantine–​Bulgar border. Thanks to the existence of a substantial Byzantine population in the northern Balkans, Christianity made slow inroads in Bulgaria where, with one notable exception discussed above, it appears to have been largely tolerated. In addition, traders from both countries engaged in a lively and wide-​ranging commercial activity that seems to have continued even during the most troubled times. Items of exchange included, on the one hand, foodstuffs, raw materials, and possibly slaves from the Black Sea steppes, and on 34 Synaxarium Ecclesiae Constantinopolitanea: Propylaeum ad Acta Sanctorum Novembris, ed. Hippolyte Delehaye (Brussels: Apud Socios Bollandianos, 1902), 415–​16; Menologion Graecorum Basilii Imp. 276D–​277A, in Patrologia Graeca 117, ed. Jacques P. Migne (Paris: Apud Garnier Fratres, 1864); Veselin Beševliev, “Eine nicht genügend anerkannte hagiographische Quelle,” in Polychronion: Festschrift Franz Dölger zum 75. Geburtstag, ed. Franz Dölger (Heidelberg: Carl Winter, 1966), 91–​104.

95



War and Identity in Early Medieval Bulgaria

95

the other, Byzantine-​manufactured goods.35 Finds of seals of Byzantine kommerkiarioi, the officers charged with the collection of the imperial sales tax, who were stationed in several cities in Thrace and the Black Sea coast (Mesembria, Adrianople, Didymoteichon, and Debeltos) until the mid-​eighth and again from the early ninth century onwards, but also of stray Byzantine cooper coinage (which was mainly used for the minor transactions of everyday life) in the territories controlled by the Bulgars, constitute significant evidence for the existence of direct contact between Byzantines and Bulgars, both on an official level and among the ordinary population.36 Unsurprisingly, the military, political, cultural, and economic contact which populations throughout the Bulgar lands had with the Empire created a high level of “Byzantine awareness,” as well as a powerful perception of Byzantium as “the other.” Such perceptions are most clearly manifested in a series of inscriptions (written in Greek) in which Bulgar rulers express both their profound distrust of the Byzantines and a distinct opposition between self and other. Thus, the inscription found in the Bassilica “B” of the city of Philippoi in northern Greece, khan Persian (836–​852) reminds onlookers that: when someone tells the truth, god sees. And when someone lies, god sees that too. The Bulgars did many favours to the Christians [i.e., the Byzantines], but the Christians forgot them. But god sees.37

In another inscription, carved by order of Krum on the right side of an ancient altar at Malamirovo in northern Thrace, Emperor Nikephoros I is accused of “setting off to war and burning our settlements, forgetting the oaths.”38 Elsewhere, Bulgar rulers employ diverse rhetorical devices to spar with the Byzantine emperors. An example of this can be observed most clearly in the inscription from Čatalar in northeastern Bulgaria, where 35 Theophanes, Chronographia 497.24–​26; Ljuben Berov, Ikonomičeskoto Razvititie na Bălgarija prez Vekovete (Sofia: Profizdat, 1974), 29–​30. The evidence suggesting that Bulgaria supplied the empire with slaves comes mainly from the distribution of Byzantine iron shackles that were probably used to restrain captives. Most specimens dated between the eighth and eleventh centuries came from the Bulgar lands and the Pontic steppes; Joachim Henning, “Gefangenenfesseln im slawischen Siedlungsraum und der europäische Sklavenhandel im 6. bis 12. Jh. Archäologisches zu Bedeutungswandel von ‘sklabos-​sakalibasclavus’,” Germania 70 (1992): 403–​26, at 416, fig. 8. Liljana Simenova, Pătuvane kăm Konstantinopol: Tărgovija i Komunikatsii v Sredizemnomorskija Svjat (Kraj na IX—​70-​te godini na XI v.) (Sofia: Paradigma, 2006), 137–​40.

36 For the seals of kommerkiarioi see Wolfram Brandes, Finanzverwaltung in Krisenzeiten: Untersuchungen zur byzantinischen Administration im 6.–​ 9. Jahrhundert (Frankfurt am Main: Löwenklau-​Gesellschaft, 2002), 393; Ivan Jordanov, “Vizantijski Komerkiarii za Bălgarija (681–​ 971),” in Dogovora, Hora, Sădbi, ed. Rumjana Mihneva, Stefka Petkova, and Vladimir Pavlov (Varna: VSU “Černorizets Hrabăr,” 2000), 17–​24. For an overviewft hee numismatic evidence, see Sophoulis, Byzantium and Bulgaria, 139–​41; Ernest Oberländer-​Târnoveanu, “La monnais dans l’espace rural byzantin des Balkans orientaux—​Un essai de synthèse en commencement du XXIe siècle,” Peuce, n.s., 1 (2003): 341–​412, at 344–​47. 37 Beševliev, Nadpisi, no. 14. 38 Beševliev, Nadpisi, no. 2.

96

96

Panos Sophoulis

Omurtag appeals to god to grant him the opportunity to perform the calcatio (the Roman ritual trampling of an enemy leader) on the emperor, thereby signifying to all onlookers the totality of his victory.39 Omurtag’s political message was not much different from that conveyed by Tervel in the fragmentary inscription accompanying the Madara relief: the khan mentions the military assistance consisting of 5,000 men that helped Justinian II regain the imperial throne in 705 (the latter had been deposed by the strategos Leontios in 695), but at the same time calls the emperor “slit-​nosed,” a derogatory nickname used primarily by Justinian II’s enemies.40 Some scholars have subsequently claimed that the Madara horseman itself appears not simply as a statement of Bulgar royal power, but as a politically motivated, almost sarcastic, reinterpretation of Byzantine art and administrative language.41 Recent research in the fields of history, sociology, and historical anthropology emphasize the importance of contact with other groups for the creation of ethnic identities and boundaries in the Middle Ages. To be sure, group definitions were relatively fluid and changed frequently in people’s minds. As the drawing of ethnic boundaries is subject to an ongoing process of communication and discourse, whereby the constructed communities were being defined and redefined in relation to varying contexts, some ethnic boundaries were maintained and reinforced, whereas others could have become less imperative, and even lose their significance. Overall, medieval societies defined themselves by describing what they were not, and they found their own definition by comparing themselves with “the others.” The defining of “the other” was therefore necessary, for it served as a catalyst to identify and characterize the self.42 In this light, there is little doubt that the Byzantine awareness—​especially pronounced among the majority of the Bulgar ruling stratum—​would have strengthened political ties between the heterogeneous (nomadic and sedentary) components of the realm and reinforced a sense of collective “Bulgar” identity. It is certainly no coincidence that from the reign of Omurtag onwards a significant amount of Christian symbolism becomes evident in Bulgaria. Court ceremonials and acclamations, insignia, titles, and other expressions of power that were clearly thought to safeguard and support the 39 “Let god give the ruler whom he has made to trample well with his feet the emperor for as long as Tucha flows and the sun [shines], to rule over the many Bulgarians and subjugate his enemies, and live in joy and delight one hundred years,” Beševliev, Nadpisi, no. 56. English translation of the text by Kiril Petkov, The Voices of Medieval Bulgaria, Seventh–​Fifteenth Century: The Records of a Bygone Culture (Leiden: Brill, 2008), 11. 40 Beševliev, Nadpisi, no. 1, Ic.

41 Curta, “Qagan, Khan or King?” 14; Tsvetelin Stepanov, The Bulgars and the Steppe Empire in the Early Middle Ages: The Problem of the Others (Leiden: Brill, 2010), 44–​45, 58–​59.

42 Albrecht Classen, “Introduction: The Self, the Other, and Everything in Between: Xenological Phenomenology of the Middle Ages,” in Meeting the Foreign in the Middle Ages, ed. Albrecht Classen (London: Routledge, 2002), xvi; Lars I. Hansen, “Interaction between Northern European Sub-​ Arctic Societies during the Middle Ages: Indigenous Peoples, Peasants and State Builders,” in Two Studies on the Middle Ages, ed. Magnus Rindal (Oslo: Research Council of Norway, 1996), 37; Richard Jenkins, Social Identity (London: Routledge, 1996), 65.

97



War and Identity in Early Medieval Bulgaria

97

ruler’s position of dominance over his non-​Bulgar subjects imitated Byzantine forms.43 The best example is offered by two golden medallions from Belogradets near Varna and Tsarevets Hill in Veliko Tărnovo, which combined steppe and Byzantine forms of imperial representation, and were probably struck for distribution among Omurtag’s supporters. Here the khan is depicted in the manner of a Christian ruler with a crown on his head, holding a ceremonial cross in his right hand and an akakia in his left hand. He is also wearing a chlamys, fastened with a brooch on his right shoulder. The medallions, which imitated golden coins struck for eighth and ninth-​century Byzantine emperors, bear the inscription “CAN-​ESY-​bHΓIωMORTAΓ” (“kana sybigi Omurtag”) in mixed Greek and Latin letters, pointing to a possible ideological interplay with the Frankish world.44 As noted already, the title kana sybigi is usually attested with its Greek equivalent ek theou archon (“ruler by god”), but it is certainly interesting that this god is never specified. By not mentioning his name, the Bulgars evidently tried to make that god recognizable to the different peoples of the realm. Thus, for the Greek-​speaking population living in Bulgaria the anonymous god invoked in the title of the ruler as the provider of his divine mandate is very likely to have been the Christian god. The imitatio imperii put into practice by Omurtag subsequently facilitated the incorporation of the Christian population into the state. By imitating the titles, ceremonials, and fashions of the Byzantine emperors, Omurtag sought to project his authority in a way that was recognizable to his Byzantine, Greek-​speaking subjects. To be sure, in the inscription found at Philippoi, khan Persian is designated “ruler by god of the many Bulgars,” an expression which is thought to suggest the existence of a new form of sacral kingship, suited to all Bulgar subjects regardless of their cultural or religious background.45 Although a shared sense of identity could hardly have derived from a concept of common descent and origin, it seems very likely that certain institutions, ritual practices, or customs forged a bond between the ruling stratum and at least some of the Christians living in Bulgaria. As in all medieval societies, in pre-​Christian Bulgaria identities were created or reinforced in relation to others. The fear and suspicion stirred up by the continual military, cultural, and economic interaction with the Byzantine Empire enabled the Bulgars to redefine themselves by shaping new forms of communal identity that looked beyond the pre-​existing social, religious, and ethnic boundaries. The imitation or adoption of elements from the imperial ideology, much like the “appropriation” of the Christian god in the proto-​Bulgarian inscriptions, not only made the khan’s position stronger vis-​à-​vis Constantinople, but also served as an instrument for strengthening 43 Beševliev, Die protobulgarischen Periode, 425–​29.

44 Beševliev, Nadpisi, nos. 86 a–​b; Stepanov, “The Bulgar Title,” 6–​7; Uwe Fiedler, “Bulgars in the Lower Danube Region: A Survey of the Archaeological Evidence and of the State of Current Research,” in The Other Europe in the Middle Ages: Avars, Bulgars, Khazars and Cumans, ed. Florin Curta (Leiden: Brill, 2008), 151–​236, at 192–​93. 45 Beševliev, Nadpisi, no. 14; Stefan Nikolov, “The Pagan Bulgars and Byzantine Christianity in the Eighth and Ninth Centuries,” Journal of Historical Sociology 13 (2000): 325–​64, at 334.

98

98

Panos Sophoulis

political ties and creating stronger bonds of allegiance with the Christian inhabitants of the Bulgar realm.

Conclusion

Warfare played a central role in the history of the early Bulgar state and had a decisive impact on its political and social development. As steppe-​nomads or semi-​nomads, the members of the aristocratic class were generally considered to be “naturally expert in war” and thus to have enjoyed an advantage over their sedentary enemies by virtue of their customs and way of life. In their new homeland in the Balkans their martial skills were constantly put to the test, as they found themselves surrounded by powerful and aggressive neighbours. The intense military pressure applied in particular by Byzantium from the mid-​eighth century brought about sweeping social and political changes; above all, it contributed to the increased centralization of political authority, as Bulgar rulers were able to monopolize revenues obtained through booty, radically transforming their relations with the nobility. At the same time, and largely in response to the intensification of war with Byzantium, the Bulgar elite appears to have strengthened its connection with the Eurasian steppes and its traditions, and to have subsequently adopted a more aggressive and, at times, violent stance toward Christianity in the Bulgar realm. Nevertheless, the treatment of Byzantine and Christians refugees, immigrants, and prisoners-​of-​war in Bulgaria was overall flexible and tolerant. Apart from that, it is clear that the growing fear of Byzantium’s cultural and military superiority, that is, the high level of “Byzantine awareness,” compelled the Bulgar elite to redefine itself by creating new forms of group identity that looked beyond the pre-​existing ethnic and religious boundaries. Accordingly, Bulgar rulers actively sought to project their authority in a way that would be recognizable to the Christian inhabitants of the state and thus develop stronger political ties with them.

99

Chapter 6

COLLECTIVE IDENTIFICATIONS IN BYZANTINE CIVIL WARS Yannis Stouraitis The relationship between warfare and identity formation in the medieval East Roman Empire, known under the terminus technicus Byzantine Empire, is a relatively understudied topic. In this chapter, I intend to explore collective identification as this emerged from and was conceptualized in Byzantine accounts of civil wars in the period from the seventh century onwards. This is the period when the imperial state of Constantinople was ultimately confined territorially to the eastern Mediterranean and its elite culture ceased to be divided between Latin and Greek as had been the case in the Roman Empire of late antiquity. After the seventh century, the merging of Hellenic cultural elements with the Roman political culture gave birth to a hybrid, non-​Latin Roman identity which has become an object of revisionist research in roughly the last decade.1 In my latest contributions to the topic, I have tried to show that the exploration of Byzantine Romanness needs to break free from a rigid methodological framework which, based on a cherry-​picking approach to the evidence of the sources, seeks to reveal or rather to reconstruct an East Roman ethnonational identity that was allegedly shared by the majority within that social order. Such approaches depart from a rather superficial constructivism which tends to reify ethnicity, arbitrarily turning the latter into the basis for political loyalty. Consequently, they tend to ignore or misrepresent three important issues: first, the boundaries of the East Roman political community were not imagined as delimited in cultural-​territorial terms. Second, ethnicity or indigeneity were not the principal criteria for full membership in the Roman political order. Third, the Greek-​ speaking Romans did not perceive themselves as a natural community of descent, the narratives about the community’s historical origins being malleable, something which is interrelated with the changing content and the situational nature of Roman ethno-​ cultural discourses over the centuries.2 1 The latest publications on the topic with all previous bibliography are: Yannis Stouraitis, “Reinventing Roman Ethnicity in Byzantium in the High and Late Middle Ages,” Medieval Worlds 5 (2017): 70–​94; Anthony Kaldellis, Romanland: Ethnicity and Empire in Byzantium (Cambridge: Cambridge University Press, 2019); Douglas Whalin, Roman Identity from the Arab Conquests to the Triumph of Orthodoxy (Cham: Palgrave Macmillan, 2020).

2 I have raised these issues in a number of recent papers: Stouraitis, “Reinventing Roman Ethnicity”; “The Lands of the Rhōmaíoi: Imagined Geographies in Byzantium before and after 1204,” in Imagined Geographies in the Mediterranean, Middle East and Beyond, ed. Dimitris Kastritsis, Anna Stavrakopoulou, and Angus Stewart (Cambridge, MA: Harvard University Press,

100

100

Yannis Stouraitis

In this light, collective attachments in the East Roman political community were determined by the idea of a hierarchical political order circumscribed by the fluctuating limits of enforceable authority of the Roman emperor of Constantinople. The notion of a divinely ordained and perpetual imperial rule seems to have been predominant among the various strata of imperial subjects whose ethnic/​cultural and even religious backgrounds varied throughout the centuries.3 Within this framework, there were of course various degrees of identification with and loyalty to centralized imperial rule. Roman law distinguished between first-​and second-​class Roman subjects, demonstrating that the main criterion for this distinction was not indigeneity or ethnic origin but religious doctrine.4 This pinpoints the divergence between Romanness as a discourse of membership in the political order of the imperial city-​state of Constantinople and Romanness as a discourse of elite culture or popular ethno-​cultural demarcation. The latter became more and more salient during the High Middle Ages as a discourse of distinction of the Rhōmaioi as a dominant ethnic category of indigenous Greek-​speaking Chalcedonians Christians from the other Roman subjects within the imperial order.5

Civil War and the East Roman Vision of Community

The exploration of notions of peoplehood in the East Roman Empire through the lens of warfare can help us further exemplify the situational character of various discourses of Romanness as well as the practices of identification related to those discourses. If we accept the sociological premise about the potential of warfare to have either a constructive or a destructive impact on the cohesion of various kinds of groups—​such as ethnic communities, kingdoms, empires, and nation-​states6—​the study of civil war may provide an insight into both developments. Internal armed conflicts aiming at the leadership of a community or pursuing ideological change within that community are typical examples of warfare which, despite dividing the group internally for a shorter or longer period of time, eventually lead to reasserting its cohesion and apparent homogeneity. On the other hand, civil wars aiming at separatism from the community in 2023), 45–​63; “What Did it Mean to Be ‘Roman’ in Byzantium?” in Byzantine Spheres: The Byzantine Commonwealth Re-​Evaluated, ed. Jonathan Shepard, Peter Frankopan, and Averil Cameron (Oxford: Oxford University Press, forthcoming); “Romanness and Chalcedonian Christianity in Byzantium,” in Ethnicity and Religion in the Middle Ages, ed. Walter Pohl, Ingrid Hartl, and Gerda Heydemann (Leiden: Brill, forthcoming). 3 John Haldon, The Empire That Would Not Die: The Paradox of Eastern Roman Survival, 640–​740 (Cambridge, MA: Harvard University Press, 2016), 72–​119; Stouraitis, “Romanness and Chalcedonian Christianity in Byzantium.” 4 Stouraitis, “Reinventing Roman Ethnicity,” 75–​76.

5 For a balanced approach to eastern Roman ethnicity, see Gill Page, Being Byzantine: Greek Identity before the Ottomans (Cambridge: Cambridge University Press), 72ff.

6 Sinisa Malešević, The Sociology of War and Violence (Cambridge: Cambridge University Press, 2010), 188–​89.

101



Collective Identifications

101

the name of ethnic difference or for the sake of control of resources are typical examples of the destructive impact that civil war can have on the unity of groups. In the case of the medieval East Roman world, civil war is a central aspect of its historical development, especially in the period between the seventh and fifteenth centuries.7 This period is divided in two major phases demarcated by the historical turning point of 1204. The sack of Constantinople by the Crusaders gave an end to a period of nearly nine centuries of a unitary political community in the east under the centralized authority of the emperor in Constantinople. As a result, the time from 1204 up to the final sack of Constantinople by the Ottomans in 1453 was marked by the East Roman world’s political disintegration and fragmentation, which endured even after the capture of Constantinople by the rulers of Nicaea, one of the Byzantine successor polities. After driving out the Latins in 1261, they created anew a Roman mini-​empire ruled from Constantinople, which proved rather weak and short-​lived. In light of this, the period between 600 and 1204 is particularly interesting because during that time the centralized political order of the Constantinopolitan emperors was marked by a large number of rebellions (around seventy), many of which escalated into a civil war.8 The majority of these rebellions aimed at the usurpation of the imperial throne, that is, they aimed at perpetuating a united and centralized imperial order through the change of ruler. A smaller number were civil wars of secession aiming or, at least, resulting in the permanent or contemporary separation of a part of the imperial state from centralized rule. Even more interesting in this respect is the fact that the vast majority of civil wars leading to secession occurred after the late eleventh century, with their number increasing considerably in the last quarter of the twelfth century. The aforementioned typological approach to Byzantine civil wars is based, of course, on the employment of the term as a modern category of analysis referring to armed conflicts that emerge within a political order, a state.9 However, the use of the term emphylios polemos (internal armed conflict) as a category of practice by Byzantine authors testifies to a differentiation of the Byzantine understanding from the modern analytical category. As I shall show below, this differentiation has clear implications for how Romanness as a discourse of political identification was perceived and functioned in an imperial order in which political loyalty was not bound to a shared ethnic identity. 7 On civil war in Byzantium, see Walter E. Kaegi, Byzantine Military Unrest 471–​843: An Interpretation (Amsterdam: Hackett, 1981); Jean-​Claude Cheynet, Pouvoir et contestation à Byzance (963–​1210) (Paris: Publications de la Sorbonne, 1990); Warren Treadgold, “Byzantium, the Reluctant Warrior,” in Noble Ideals and Bloody Realities: Warfare in the Middle Ages, ed. Niels Christie and Maya Yazigi (Leiden: Brill, 2006), 209–​33; Yannis Stouraitis, “Civil War in the Christian Empire,” in A Companion to the Byzantine Culture of War, ed. Yannis Stouraitis (Leiden: Brill, 2018), 99–​123.

8 On a definition of Byzantine civil war, see Treadgold, “Byzantium, the Reluctant Warrior,” 224; and the critical reassessment of the latter’s argument in Stouraitis, “Civil War in the Christian Empire,” 92–​94. 9 Here, I follow the typology of civil wars in Jan Angstrom, “Towards a Typology of Internal Armed Conflict: Synthesizing a Decade of Conceptual Turmoil,” Civil Wars 4 (2001): 93–​116, at 95–​96.

102

102

Yannis Stouraitis

Byzantine authors employed various terms to denote armed rebellions which could trigger a civil war, such as apostasia (apostasy/​defection), tyrannis (usurpation), neōterismos (change by force), epanastasis (rebellion), and stasis (insurrection). Most of those terms were equally applied to designate rebellions of usurpation of imperial rule as well as rebellions of secession from it. However, the Byzantine term for civil war, emphylios polemos (literally meaning war within a race or group), was employed to designate only armed conflicts within the empire caused by rebellions that aimed at the usurpation of centralized imperial rule, not at secession from it. The historiographer John Skylitzes provides a typical example of this when he reports on the contemporary rebellions of Samuel and Bardas Skleros at the beginning of the reign of Basil II (976–​1025).10 Both rebellions against imperial rule are designated with the same term, apostasia (apostasy/​defection), insofar as both magnates took up arms as Roman subjects from the viewpoint of the imperial power in Constantinople. In the course of his account, however, Skylitzes employed the term emphylios polemos to denote only the war between the emperor and the forces of the general Bardas Skleros, whose aim was to usurp the imperial throne. On the contrary, the war with Samuel, who had seceded from imperial rule in order to become tsar and resurrect an independent Bulgarian kingdom in the Balkans not long after its subjugation by the emperor John II Tzimskes in 972, was recounted in terms of interstate warfare. From a heuristic viewpoint, civil wars relating to ethnic separatism are considered as the ones that resemble interstate warfare the most and often take its form.11 In the case of Samuel, however, his different ethnic identity was not the primary cause that dictated the representation of his war with the emperor as an interstate conflict. This was rather owing to the fact that his apostasy aimed at breaking away from the empire through autonomous state building.12 That a rebels’ ethnic categorization played a marginal role in the conceptualization of a conflict with the imperial power as civil war becomes evident if we take a look at another rebellion within the empire that was instigated almost a century and a half earlier by an officer named Thomas against the then newly enthroned Emperor Michael II (820–​829). As has been recently argued, it is probable that in this case our sources merged two rebellions of two persons named Thomas into one. The Thomas who was able to fight a civil war for a whole three years probably fled from the empire at an early stage, went to the Abbasid Empire, and from there invaded again with the support of the Arabs to claim the Byzantine throne.13 The war between the emperor and the rebel is unanimously 10 John Skylitzes, Synopsis, ed. Johannes Thurn, Ioannis Scylitzae synopsis historiarum, CFHB 5 (Berlin: De Gruyter, 1973), 324, 317, 328–​29, 339; cf. John Zonaras, Epitome, ed. Ludwig August Dindorf, Ioannis Zonarae epitome historiarum, vol. 1 (Leipzig: Teubner, 1868), 548. 11 Angstrom, “Towards a Typology of Internal Armed Conflict,” 113.

12 That ethnic diversity should not be considered as the primary trigger and the cause of Samuel’s rebellion of secession, see Stouraitis, “Civil War in the Christian Empire,” 115–​16.

13 See Juan Signes Codoñer, The Emperor Theophilos and the East, 829–​842: Court and Frontier in Byzantium during the Last Phase of Iconoclasm (Farnham: Routledge, 2014), 40–​59, 183–​89.

103



Collective Identifications

103

designated a civil war (emphylios polemos) in Constantinopolitan historiography. If this comes as no surprise considering that the rebel aimed at usurping the imperial throne, what makes this case particularly interesting for our topic is the very fact that Constantinopolitan historiographers report that Thomas was a Roman subject of Slavic origin and of barbarian culture.14 Moreover, in the version of the story in which it is reported that Thomas invaded the empire in alliance with the Muslims and with a largely foreign army, the authors still used the term emphylios polemos to label the conflict. The historiographical discourse that highlighted Thomas’s ethnic origins and low culture as well as his alliance with the Muslims was, of course, intended to construct an image of him in retrospect as a person unworthy of the Roman throne. Still, the consistent conceptualization of the conflict as a civil war is indicative of the way political Romanness was conceptualized, pointing at the unimportance of a shared ethno-​cultural identity when it came to defining who rebelled as a Roman and who did not. What made the war between Thomas and Emperor Michael II an internal armed conflict was the fact that he had rebelled as a Roman subject loyal to the vision of a centralized empire, that is, with the aim not to break away from imperial rule but to reproduce the imperial system by usurping the Roman throne. The representation of Thomas’s identity in the sources as a Roman subject of Slavic origin did not differ from the representation of Samuel’s identity in his conflict with Emperor Basil II. The latter was also viewed as a Roman subject of Bulgar ethnicity at the time of his rebellion, as demonstrated by the fact that Byzantine historiography designated his uprising an apostasy from imperial rule. The main difference between the two apostates, which determined the image of their armed conflict with the imperial power as intrastate and interstate warfare, respectively, was the goal of their rebellion (usurpation vs. secession). These examples raise the issue of the meaning of the term emphylios in Byzantine discourse. The military treatise Taktika of Emperor Leo VI shows that the term was employed to refer to a political community of subjects, irrespective of their ethnic backgrounds. In the famous passage on just war in the text, the author remarked: We [i.e., the imperial power] always welcome peace both for our subjects (hypêkoous) as well as for the barbarians in the name of Christ the ultimate king and God. If the peoples (ethnê) want the same and stay within their own boundaries without committing any injustice, then you (general) should refrain from actions against them and should not befoul the land with the blood of our own people (emphyliois haimasi) or that of the barbarians.15

In this context, the notion of “our own people’s blood” is explicitly connected to the term hypêkooi, that is, to a vision of a community of Christian Roman subjects whose 14 Theophanes Continuatus (libri I–​IV), ed. Michael Featherstone and Juan Signes Codoñer, Chronographiae quae Theophanis Continuati nomine fertur Libri I-​IV, CFHB 53 (Berlin: De Gruyter, 2015), 50.

15 Leonis VI Tactica 2.30, ed. George T. Dennis, The Taktika of Leo VI. Text, Translation and Commentary, CFHB 49 (Washington, DC: Dumbarton Oaks, 2010), 34.

104

104

Yannis Stouraitis

ethnic backgrounds could vary, as the same text makes evident. The author of the Taktika reports, for instance, that the emperor’s father and predecessor had integrated large numbers of Christian Slavic populations in the Balkans into the Roman body politic some decades earlier, thus making them first-​class Roman subjects (hypêkooi) from the viewpoint of Roman law.16 It is this conceptual background that can provide us with an explanation for the absence of the term emphylios polemos from the discourse of Byzantine authors when they described conflicts that took place between the imperial power and secessionist groups of Roman subjects, whose leaders shared the dominant Roman ethno-​cultural identity of the imperial centre, that is, they were Greek-​speaking Chalcedonian Christians. A good case in point is the rebellion of Theodore Mangaphas, which led to the short-​ lived autonomy of the region of Philadelphia in Asia Minor during the reign of Isaak II Angelos (1185–​1195).17 The armed conflict between the emperor and the apostate is nowhere designated as an emphylios polemos by the historian Niketas Choniates, since the author confines the term in his history only to those rebellions that explicitly aimed at the imperial throne of Constantinople.18 The evidence from the period after 1261 provides further insights in this regard. As mentioned above, the disintegration of the so-​called Byzantine Empire after the sack of Constantinople during the Fourth Crusade resulted in the emergence of various polities under Byzantine magnates next to the Latin Empire of Constantinople and other Latin principalities. One of the main consequences of this development was the emancipation of the image of the Roman community from that of a centralized political order demarcated by the boundaries of imperial authority. In the period between 1204 and 1261, the East Romans were a group demarcated by abstract cultural boundaries and divided in various polities. Their shared ethno-​cultural markers evidently did not function as an ideological incentive for the elites of the Byzantine successor polities to pursue their political reunification and to fight side by side against the Latin foe.19 The capture of Constantinople by the ruler of the successor state of Nicaea, Michael Palaiologos, in 1261, gave birth to a discourse of restoration of the Roman imperial rule in its historical homeland, which was employed to retrospectively legitimize the Nicaean rulers as the sole legitimate Roman emperors in exile. Within this framework, Romanness as a political discourse of identification with the imperial order of Constantinople was fully restored and was used to exclude from the Roman political community all those Greek-​speaking Chalcedonian Christians who had not been subject and loyal to the 16 Leonis VI Tactica 18.95, ed. Dennis, 470–​71; cf. Stouraitis, “Romanness and Chalcedonian Christianity in Byzantium.” 17 On Mangaphas’s rebellion as a secessionist movement, see Cheynet, Pouvoir et contestations, 123, 134–​35, 454–​55.

18 Choniates, Historia, ed. Johannes van Dieten, Nicetae Choniatae historia, pars prior, CFHB 11.1 (Berlin: De Gruyter, 1975), 399–​400. 19 Stouraitis, “Reinventing Roman Ethnicity,” 82–​85.

105



Collective Identifications

105

rulers of Nicaea after 1204, and who remained not subject and disloyal to the rulers of Constantinople after 1261.20 In this context, Constantinopolitan authors used the term civil war (emphylios polemos) in this period to denote explicitly those armed conflicts that emerged within the boundaries of the state of the Constantinopolitan emperor.21 The most typical example that demonstrates the marginal role of ethnicity in configuring the image and boundaries of a Roman political community is provided by the historiographer George Pachymeres writing in Constantinople in the early fourteenth century. The author referred to an incident between the Alan mercenaries of Emperor Andronikos II Palaiologos (1282–​1328) and another part of the imperial forces in 1302 as follows: an altercation and an internecine battle (phyletikê machê) broke out, because even though both armies were of a different race (genos), they were under the same imperial authority, and that made it a civil war (emphylios polemos) as much as one may infer.22

The Alans were bearers of a different ethno-​cultural identity, but they were the emperor’s soldiers and therefore Roman subjects. That made it possible to conceptualize a conflict with them within the framework of the imperial state as a civil war. This case needs to be juxtaposed with the consistent representation of the armed conflicts between the imperial state of Constantinople (or previously that of Nicaea) and the other Byzantine polities as interstate warfare. Despite the common ethno-​cultural identity of the two sides, Byzantine historiographers never used the term civil war (emphylios polemos) to designate those conflicts. The reason was that political Romanness was first and foremost understood as a practice of identification and loyalty to the centralized imperial rule of the sole legitimate Roman emperor of Constantinople. All those that remained politically independent from the rulers of Nicaea after 1204 or from the rulers of Constantinople after 1261 were excluded from Romanness, understood as ideological attachment and loyalty to the political community under the centralized rule of the Roman emperor. Hence, they were deprived of the ethnonym Roman in the Constantinopolitan historiographical discourse. For George Akropolites, writing shortly after the establishment of Palaeologan rule in Constantinople in 1261, the ruler of Epirus, Michael II, was a rebel and a renegade.23

20 On historiographical discourse depriving the other post-​1204 Byzantine polities of their Roman identity, see Ruth Macrides, George Akropolites: The History. Introduction, Translation and Commentary (Oxford: Oxford University Press), 53, 94–​95; Page, Being Byzantine, 79–​85, 104–​6, 121, 135–​36, 157–​85. 21 Savvas Kyriakidis, “The Idea of Civil War in Thirteenth and Fourteenth-​Century Byzantium,” Zbornik Radova Vizantološkog Instituta 49 (2012): 243–​56, at 248–​54.

22 George Pachymeres, Historiae, ed. Albert Failler and Vitalien Laurent, Georges Pachymérès: Relations historiques, 2 vols., CFHB 24, 1–​2 (Paris: Belles Lettres, 1984), 321. 23 George Akropolites, Annales, ed. August Heisenberg, Georgii Acropolitae opera, vol. 1 (Leipzig: Teubner, 1903), § 70.

106

106

Yannis Stouraitis

The term rebel was also used to label the autonomous ruler of Rhodes, Leon Gabalas.24 This terminology was, of course, intended to delegitimize all those outside the current boundaries of the Roman imperial order as defectors, that is, as former subjects of the empire that broke away from centralized rule. However, it was not their Roman ethno-​ cultural identity that made them rebels but their disloyalty to centralized imperial rule, similarly to the previously presented case of the Bulgar rebel Samuel. The marginal role of shared ethnicity in shaping the boundaries of the Roman political community in the period after 1261 is further exemplified by the sole case in which a conflict between the Palaeologan imperial state of Constantinople and the ruler of Epirus was designated as an emphylios polemos. The statement comes from Patriarch Arsenios, who warned the emperor of Constantinople, Michael VIII Palaiologos (1259–​ 1282), to avoid provoking internal armed conflicts (emphylious polemous) with the ruler of the Epirotes because they were sharing the same faith with them as Christians.25 This discourse demonstrates that in the Patriarch’s view the cultural framework for an internal armed conflict (emphylios polemos) was not provided by the notion of a shared Roman ethnicity between the Constantinopolitans and the Epirots, but by a vision of Christian brotherhood. This discourse about an internal armed conflict among Christians was nothing new in Byzantine political thought. It was used already in the period before 1204 by the Byzantines to designate conflicts between the imperial state and Christian enemies, such as the Bulgars, thus applying the political notion of civil war to a supra-​ethnic vision of an imagined community of Christians.26 The evidence presented so far allows for some preliminary conclusions. The term emphylios polemos, despite its ethnic connotations in etymological terms (en =​ within, phylo =​ race/​group), was not employed by Byzantine authors to define an internal armed conflict within a community demarcated by boundaries of a shared ethno-​ cultural identity.27 The content of the term in this context was determined by the vision of an imperial-​political community, whose members’ principal identity trait was their subject status (underpinned by their Chalcedonian confessional identity) and loyalty to the emperor in Constantinople, that is, identification with centralized imperial rule. The ethno-​cultural backgrounds and identifications of the participants were irrelevant insofar as they remained faithful to the vision of a united imperial community. Rebellion, either as usurpation of or as defection/​secession from imperial rule, was equally regarded as an act of political apostasy, that is, disloyalty toward the emperor, irrespective of the rebel’s goals and ethnicity. Within this framework, the Byzantines distinguished the armed conflicts caused by rebellions as civil war or interstate warfare, 24 George Akropolites, Annales, § 27–​28. 25 Pachymeres, Historiae, 313–​15.

26 Yannis Stouraitis, “Byzantine War against Christians—​An emphylios polemos?” Byzantina Symmeikta 20 (2010), 85–​110, at 93–​101.

27 For the use of the term by Byzantine authors to describe internal armed conflicts in various kind of communities such as ethnic groups or communities of coreligionists, see Stouraitis, “Civil War in the Christian Empire,” 95.

107



Collective Identifications

107

respectively, according to the intention of the rebellious party to remain a part of the imperial order, that is to aim at the latter’s unity and perpetuation, or not.

Representations of Collective Identity in Byzantine Civil Wars

The interdependence between the Byzantine image of internal armed conflict and the perception of Romanness as a practice of political identification with the monocracy of the Christian Roman emperor of Constantinople sets the background against which historiographical representations of the collective identity of the participants in Byzantine civil wars should be examined. If Romanness as a dominant political discourse of loyalty to centralized imperial rule accommodated notions of ethno-​cultural diversity, rendering the ethnic categorizations of Roman subjects politically unimportant within the empire, it is still worth asking how the collective identity of those participating in a civil war was conceptualized. In particular, it is worth examining the discursive use of the language of metaphoric kinship by Byzantine authors and to which extent this reflected or contributed to the configuration of an image of ethnic Romanness underneath the predominant Roman imperial-​political discourse. It is well known that the conclusion of the process of the—​at least nominal—​ Christianization of the late Roman Empire’s populations by the late sixth century had provided Roman political discourse with a common cultural marker which transcended social strata, as well as the regional and ethnic subcultures within the empire.28 The impact of this development on the discourse of high-​level collective identification within the imperial community is made evident if one looks at the interchangeable employment of the ethnonym Rhōmaios (Roman) and the religious label Christianos (Christian) by Byzantine authors when they collectively categorized the imperial subjects.29 Of course, the label “Christian,” like the label “Roman,” often functioned as an umbrella term in Byzantine historiography, since it accommodated not only different ethnic identities but also both orthodox and heretic Christians, that is, first and second-​ class imperial subjects, overshadowing at the same time the existence of minorities of non-​Christian Roman subjects (for example, Jews). Nonetheless, it provided an ideal means for applying notions of metaphoric kinship to the Roman imperial community, insofar as early Christianity had fully adopted the discourse of ethnicity in the context of the representation of Christians as a new people that had succeeded Jews as God’s new chosen people.30 28 Hervé Inglebert, “Citoyennete romaine, romanitas et identites romaines sous l’Empire,” in Ideologies et valeurs civiques dans le monde romain. Hommage Claude Lepelley, ed. Hervé Inglebert (Paris: Picard, 2002), 241–​60.

29 Nikephoros Patriarch of Constantinople, Short History, ed. Cyril Mango, Nicephori archiepiscopi Constantinopolitani Breviarium historicum de rebus gestis post imperium Mauricii, CFHB 13 (Washington, DC: Dumbarton Oaks, 1990), 1. 30 Denise K. Buell, Why This New Race: Ethnic Reasoning in Early Christianity (New York: Columbia University Press, 2005); cf. the overview of all recent bibliography on the topic in Todd Berzon,

108

108

Yannis Stouraitis

The first civil war in the period examined here, in which Christian identity was highlighted as the main cultural maker that united the emperor’s subjects, is the conflict between Emperor Constantine V (r. 741–​775) and one of his generals, Artabasdos, which took place between 741 and 743. Our main source for this event, the chronicler Theophanes the Confessor writing in the early ninth century, reports that “two emperors were proclaimed and there ensued terrible battles and conflicts between the imperial subjects (hypêkoous).”31 In another part of his account the author refers to the Armenian ethnicity of Roman soldiers and officers that supported the rebel Artabasdos—​who probably had an Armenian ethnic background himself as well—​and adds that: the devil, the source of every evil, had roused in those days such fury and mutual slaughter among Christians that sons would murder their fathers without any mercy and brothers would murder their own brothers and pitilessly burn each other’s houses and homes.32

Both statements indicate that the collective identity of the participants in the civil war was determined by two main markers in the author’s view: first, their status as subjects of the empire and, second, the fact that they were all bearers of the same religious identity. The Christian label was not only employed to supplant the different ethno-​cultural backgrounds of Roman subjects, but also doctrinal diversity in this case, if we consider that Theophanes was keen to present the civil war as a conflict between an iconoclast—​that is, heretic—​emperor and an orthodox contender of the throne.33 Despite the fact that Constantine V and his supporters were labelled as heretics, that is, not as true (orthodox) Christians by the author, this seems not to have hindered him from categorizing them under the same religious label in his effort to highlight the image of the Romans as a community torn by internal conflict. In tenth-​century Byzantine historiography, the formation of an ethnoreligious background for Byzantine Romanness in civil wars becomes further evident. In his account of the civil war between the rebel Thomas and Emperor Michael II (820–​829), the mid-​tenth-​century author Joseph Genesios reports that “Thomas increased the numbers of his army in order to turn them against his own people (homophylous).”34 Homophylos means literally “of the same race or group,” but could also have the meaning “of the “Ethnicity and Early Christianity: New Approaches to Religious Kinship and Community,” Currents in Biblical Research 16 (2018): 191–​227. 31 Theophanes, Chronographia, ed. Carl de Boor, CSHB (Leipzig: Teubner, 1883), 414–​15; Engl. Trans. Cyril Mango and Roger Scott, The Chronicle of Theophanes Confessor: Byzantine and Near Eastern History, AD 284–​813 (Oxford: Oxford University Press, 1997).

32 Theophanes, Chronographia, 418.

33 On Theophanes’s biased representation of the civil war as an ideological conflict over the issue of the icons, see Leslie Brubaker and John F. Haldon, Byzantium in the Iconoclast Era c. 680–​850: A History (Cambridge: Cambridge University Press, 2011), 117–​35.

34 Genesios, Basileiai, ed. Anna Lesmüller-​Werner and Johannes Thurn, Iosephi Genesii regum libri quattuor, CFHB, 14 (Berlin: De Gruyter, 1978), 2.2.

109



Collective Identifications

109

same faith” in Byzantine usage.35 If the employment of the term by the author implies an intention to apply an image of common kinship to the East Roman community, we need to consider that Genesios, like his contemporary history of Theophanes Continuatus, highlighted in his text the Slavic (Scythian) ethnic origin of the rebel Thomas.36 From this we may conclude that what made Thomas an homophylos in the context of a vision of Roman peoplehood was not the notion of the Romans as a people of common ethnic origin. Instead, homophylos in this case seems to follow the non-​ethnic use of the term emphylios, as presented above, pointing to a shared Roman subject status and shared political identification between the two opposite parties, that of Thomas and that of the emperor. Moreover, the author’s discourse points to a religious-​cultural content of the term as well, since in the same context Genesios referred to Thomas’s countrymen as homopistous (literally meaning those of the same faith).37 Further, he reported that Emperor Michael received an offer from the Bulgar ruler for aid in the civil war “but (the emperor) declined it because he did not wish the war against the rebels to provide occasion for the Bulgars to shed Christian blood (christianōn haimata).”38 This report is incorrect, since the emperor did seek and receive the aid of the Bulgars, as testified by other authors.39 Genesios contradicts himself, in fact, when in the course of his account he presents the Bulgars as allies of the emperor. He reports that, after the rebel’s army had been defeated and had withdrawn from the walls of Constantinople, they were besieged by the emperor at Arkadioupolis, where they had sought refuge. The emperor avoided to employ siege-​engines against the city, however, because “he feared that the rest of his co-​religionists (homodoxois) would be killed by the sword and that the Bulgars would learn how Roman cities could be taken with siege instruments.”40 Both statements demonstrate that the Christian faith provided a shared religious-​ cultural background which served as a basis that underpinned the cultural image of Roman peoplehood of the participants in the civil war. The status of an imperial subject and a shared Christian identity were the main means to distinguish the Romans from the Bulgar allies of the emperor who had not yet been Christianized. 35 For the latter meaning, see Theodora Papadopoulou, Συλλογική ταυτότητα και αυτογνωσία στο Βυζάντιο. Συμβολή στον προσδιορισμό της αυτοαντίληψης των Βυζαντινών κατά την λόγια γραμματεία τους (11ος—​αρχές 13ου αι.) (Αthens: Σύλλογος προς διάδοσιν ωφελίμων βιβλίων, 2015), 314.

36 Genesios, Basileiai 2.1; in another part of the text (1.6) the author speaks of Thomas’s Armenian origin. This contradiction should rather be attributed to the fact that there were two different people named Thomas that rebelled, as mentioned above. 37 Genesios, Basileiai 2.2. 38 Genesios, Basileiai 2.7.

39 George the Monk, Chronicle, ed. Carl de Boor, Georgii monachi chronicon, 2 vols. (Leipzig: Teubner, 1904), 796. 40 Genesios, Basileiai 2.8.

110

110

Yannis Stouraitis

The use of language alluding to a notion of shared kinship in the representation of the common identity of the participants of this civil war is also attested in the mid-​ tenth-​century history of Theophanes Continuatus. The author reports that Emperor Michael II addressed the rebels who attacked the walls of Constantinople, “proclaiming and promising them amnesty for evils done if only they would change sides and refuse to stain themselves with the blood of their own people (homophylōn) and brothers (adelphōn).”41 In another part of the narration where he presents the same story as Genesios about the Bulgar ruler’s offer of aid, he observes that the emperor refused it, “either because he in truth felt shame and showed mercy on the army of his own people (homophylōn), or else because he was sparing money.”42 The content of the term homophylōn here seems to be informed once again by the vision of belonging to a political community of imperial subjects torn by an internal armed conflict (emphylios polemos) over the imperial throne. The term “brothers,” on the other hand, represents a more explicit effort to apply the language of metaphoric kinship to the Eastern Roman community. For the ideological background that facilitated this image of brotherhood, one needs to look at the shared religious-​cultural identity of the members of the two fighting sides in the civil war. The author of the text pointed to the role of the Christian faith as the main cultural marker and bond of Roman subjects already at the beginning of his account of the civil war when he depicted the rebel Thomas as a traitor of the empire and the Christian faith, arguing that he waged war against the Roman imperial power and the Christians (sic).43 In addition to this, one needs to consider that the image of the emperor’s subjects as brothers of the same faith was highlighted and propagated as the main incentive for Byzantine soldiers to stick together and fight on behalf of their fellow believers on the battlefield according to the testimony of contemporary military treatises.44 John Skylitzes, writing about the same events in the late eleventh century, reasserts the role of Christian identity in providing a shared religious-​cultural background for imperial subjects. Interestingly, the author’s allegation that the rebel invaded the empire acting in benefit of the Muslims did not prevent him from designating the conflict as a civil war (emphylios polemos).45 Examining Skylitzes’s identity discourse in this case it is particularly enlightening, if we consider that his history is the first of the high-​medieval period to identify subjects of the emperor as being Romans by birth.46 41 Theophanes Continuatus (libri I–​IV), 92. 42 Theophanes Continuatus (libri I–​IV), 96. 43 Theophanes Continuatus (libri I–​IV), 78.

44 Syrianos magistros, Rhetorica militaris 36.8–​10, ed. Immacolata Eramo, Siriano, Discorsi di Guerra (Bari: Edizioni Dedalo, 2010), 79–​81; Leonis VI Tactica 12.57, ed. Dennis, 248–​50; cf. the analysis of the latter passage in Stouraitis, “Reinventing Roman Ethnicity,” 76–​78. 45 John Skylitzes, Synopsis, 29.

46 Stouraitis, “Reinventing Roman Ethnicity,” 80.

111



Collective Identifications

111

The author reports that Thomas presented himself as the deceased Emperor Constantine VI because he did not want to be regarded as a foreigner (allophylos) of different faith (allopistos) by the Romans when he invaded the empire.47 To understand the content of the term allophylos in this case, one needs to consider that the author uses the terms apostasia (apostasy) and emphylios polemos (internal armed conflict) to describe Thomas’s rebellion and the war that ensued while highlighting that the rebel was a Roman subject of barbarian ethnic origin. This representation of the rebel’s identity indicates that the image of a foreigner (allophylos) invading from the Muslim empire seems once again to deviate from the ethnic connotations of the term (i.e., belonging to a different race), since Thomas’s Roman identity was not understood in terms of ethnic descent in the first place. What categorized him as an allophylos was his defection to and alliance (also in terms of religious conversion) with the archenemy of the empire, the Muslims. In general, the image of the Roman community in Skylitzes’s reports on Byzantine civil wars seems to follow the normative image of an imperial-​political order, the members of which were first and foremost bound together by their status as imperial subjects and their shared religious culture. For instance, when the author reported on the efforts of Emperor Michael I (r. 811–​813) to avoid internecine slaughtering (emphylia miaiophonia) during the usurpation against him by Leo the Armenian, he stated that the emperor preferred to surrender his throne to the usurper in order to avoid the spilling of Christian blood (haimatos christianikou).48 In the case of the usurpation of Bardas Phokas against Emperor John I Tzimiskes (r. 969–​976), he reported that the leader of the imperial forces, Bardas Skleros, was instructed to try to avoid the spilling of their own people’s blood (emphyliois haimasi)—​a phrase reminiscent of the aforementioned quote of the Taktika in which the notions of blood and peoplehood were explicitly related to the notion of a supra-​ethnic community of imperial subjects. Finally, about the rebellion of the Bulgar Samuel against Basil II, Skylitzes reported that this was an apostasy from imperial rule, thus applying to the rebel the status of a Roman subject, but then he went on to distinguish the ensuing war terminologically from the war caused by the apostasy of the general Bardas Skleros against the same emperor. The latter was designated as an internal armed conflict (emphylios polemos), because Skleros contrary to Samuel aimed at the imperial throne.49 A more explicit use of the metaphoric language of common kinship can be found in the discourse of the eleventh-​century chronicle of Michael Psellos regarding the civil war between General George Tornikes and Emperor Constantine IX in 1047. The author reports that the rebel ordered the captives from the emperor’s forces to stand in front of the walls of Constantinople and appeal to him to stop fighting so as not to spill the blood 47 John Skylitzes, Synopsis, 29. 48 John Skylitzes, Synopsis, 8. 49 See note 11 above.

112

112

Yannis Stouraitis

of his own people (homophylou) and kin (syggenikou).50 In the late-​twelfth-​century history of Niketas Choniates, ethnic discourse is articulated even more clearly when it comes to the participants of an internal armed conflict. Choniates’s historiographical work is extraordinary in that the author wrote an initial version of the text as a high-​ ranking imperial official in Constantinople in the 1190s, which he then reworked and expanded to include the events of the Fourth Crusade as a refugee in Nicaea after the fall of Constantinople in 1204.51 This means that Choniates wrote the text in a period of transition from a centralized Roman imperial community to a politically disintegrated Roman ethno-​cultural group, something which may explain the emphasis on the discourse of Roman ethnicity in his text. As already mentioned in the previous section of this chapter, the author does not deviate from the terminological pattern that designated as internal armed conflict (emphylios polemos) only those conflicts between subjects of the emperor, aiming at the imperial throne. Within this framework, however, Choniates presents the identity of the opposite sides in terms of common kinship. Regarding the rebellion of Andronikos Komnenos against Alexios II in 1182, he reported: As a result, the Asiatic cities were fraught with internal strife and civil wars. The actions now taken were more grievous than those undertaken by the neighbouring enemies; in other words, whatever the hand of those speaking a different language did not pluck, the right hand of the inhabitant reaped; the countrymen (to homophylon), ignoring the laws of kinship (syggeneias), went to war against one another as though they were barbarians.52

In this case, Choniates seems to have applied the term homophylon to allude to a common identity in terms of a shared ethnic descent of the participants of the civil war, as indicated by the allusion to ethno-​linguistic difference with the enemies of the empire and the explicit reference to the laws of common kinship. An ethno-​linguistic image of the Rhomaioi reappears in the author’s report on the civil war between General Branas and Emperor Isaak II (r. 1185–​1195). Choniates has the emperor addressing his army before battle and saying that it is preferable and non-​reprehensible in the eyes of God that the lawful ruler braves danger for his subjects rather than to yield to a revolutionary who foments civil war among those who speak the same tongue (homoglōtton).53

In another part of his account, however, Choniates presents the kaisar Renier of Montferat addressing his Latin mercenaries during an internal armed conflict within 50 Michael Psellos, Chronographia, 6.117, ed. Dieter R. Reinsch, Michaelis Pselli Chronographia, Band 1: Einleitung und Text (Berlin: De Gruyter, 2014), 159. 51 Alicia J. Simpson, “Before and after 1204: The Versions of Niketas Choniates’ “Historia”,” Dumbarton Oaks Paper 60 (2002): 189–​221, at 202–​5.

52 Choniates, Historia, 263.

53 Choniates, Historia, 384–​85.

113



Collective Identifications

113

Constantinople between his forces and the forces of the regency of Alexios II with the following words: It would have been better if we had put on our armours and taken up the swords against the enemies of the cross and not against compatriots (homophylōn) and coreligionists (homopístōn); it is because they have badly mismanaged the affairs of the Roman imperial power (basileias) that they must be deprived of it, and it is by necessity and not by choice that we have sharpened our lances against them. We shall oppose our assailants without lack of dignity, and we shall not consider that we are of the same faith (tautopistoi) and the same people (homogeneis) but look upon them first as enemies of God whose holy temple they enter without shame, and then defend ourselves against them as adversaries.54

This is probably one of the many invented harangues in Choniates’s Historia and it is, of course, hard to tell whether this way of thinking is representative of the participant’s way of thinking or not.55 The language of the passage, though, sheds light on the ways the author understood and used notions of metaphoric kinship. The terms homophyloi and homogeneis refer here to Roman subjects whose Roman identity was not considered to be determined by a shared Roman ethno-​cultural identity. Renier and his retinue were Latins whose status as Roman subjects was determined by their Christian faith and their loyalty to Roman imperial rule. Choniates’s adherence to the traditional image of a Roman political community in this case indicates that the image of a distinct Roman ethno-​cultural group within the empire, which is testified in other parts of his account, continued to play in fact little role in marking the boundaries of the Roman political community as well as the image of an internal armed conflict within that community. This is made further evident if we consider how he chose to represent the conflict between Emperor John II Komnenos (r. 1118–​1143) and a small group of native Greek-​ speaking Chalcedonian Christians, that is, Romans, who lived outside the boundaries of the imperial state in Asia Minor and refused to accept subjugation to his authority. Choniates was keen to allude to a shared ethnic background between those people and the Romans, highlighting their common descent and shared Christian religion. At the same time, he deprived them of the Roman label and refrained from defining their armed conflict with the emperor as an emphylios polemos, that is, as an internal armed conflict.56 The reason for the author’s stance was that these people were not subject and loyal to the centralized rule of Roman imperial power of Constantinople at the time, that is, they did not identify with it.

54 Choniates, Historia, 238–​39.

55 On invented speeches in Choniates’s work, see Herbert Hunger, Die hochsprachliche profane Literatur der Byzantiner, vol. 1 (Munich: Beck Verlag, 1978), 438–​39.

56 Choniates, Historia, 37; cf. Page, Being Byzantine, 83; Stouraitis, “Reinventing Roman Ethnicity,” 80.

114

114

Yannis Stouraitis

Choniates’s identity discourse, a product of the author’s effort to balance between the pre-​1204 dominant image of a Roman imperial-​political community and the post-​ 1204 reality of a politically fragmented Roman ethnic group, seems to anticipate the developments after 1261. The restoration of a Constantinopolitan imperial state fully reinstituted the normative vision of the Eastern Romans as an imperial-​political community, demarcated by the boundaries of centralized imperial rule and excluding all those ethno-​culturally Roman populations outside these boundaries. The account of the great civil wars within the Constantinopolitan imperial state (1321–​1357) in the history of Emperor John VI Kantakuzenos provides an insight into the development of the political discourse of Romanness in that period. The author’s frequent use of the term homophyloi (those of the same race or group) when referring to the participants in the internal armed conflict has been taken as an indication that shared ethnic identity played a central role in shaping the image of a civil war among the Romans in this period.57 This interpretation seems to be supported by the fact that an image of the Rhōmaioi as an ethnic group had been consolidated since 1204, as mentioned above. However, the interrelation of the term homophyloi with an ethnopolitical character of the Roman community under Constantinopolitan rule in this period needs to be examined against the absence of similar terminology in the wars fought between the Romans of the Constantinopolitan imperial state and those autonomous Roman polities whose members shared the same ethno-​cultural markers with the Constantinopolitan ruling elite.58 For instance, the conflict between Constantinople and Epirus is not defined as a civil war (emphylios polemos) in Kantakouzenos’s history and a representation of the two sides as homophyloi is absent from his discourse. In the negotiations between himself as a representative of the Constantinopolitan emperor Andronikos III and the people of the major city of Arta during the campaign of 1339–​1340, he did not use the term homophyloi to refer to the Arteans and did not raise a shared ethnic identity as the reason why they should submit to the emperor and become politically united with the Romans of Constantinople. Instead, his argument was that they should surrender because their forefathers had been ruled by the Roman emperors since the time of Caesar and their lands were the ancestral heritage of the emperors of Constantinople.59 Kantakuzenos’s discourse stems from the same traditional notion of an imperial-​ political Roman community, which Emperor Basil II had employed to justify the subjugation of the Bulgars back in the early eleventh century after the end of the long-​ drawn conflict which had started with the apostasy of Samuel in the 970s. In an imperial edict of the year 1020, the emperor stated that divine providence had allowed for the 57 Kyriakidis, “The Idea of Civil War in Thirteenth and Fourteenth-​Century Byzantium,” 246–​47.

58 See Kyriakidis, “The Idea of Civil War in Thirteenth and Fourteenth-​Century Byzantium,” 250, for the acknowledgment of the Roman ethno-​cultural identity of the Epirots by the Palaiologoi.

59 John Cantacuzenus, Historiae, ed. Ludwig Schopen, Ioannis Cantacuzeni eximperatoris historiarum libri iv, vol. 1, CSHB (Bonn, 1828), 1:520.1–​521.19.

115



Collective Identifications

115

Bulgars to be subjugated, thus bringing back under imperial rule those parts of the realm that had been separated in recent times.60 If Basil II’s statement was based on a view of the Bulgars as rebellious Roman subjects that were brought back under centralized rule along with their separated territories, the report of Kantakuzenos some centuries later shows that, despite the fact that the people of Epirus shared common markers of cultural Romanness with the Constantinopolitans, this did not make their status as outsiders to the Roman imperial-​political community any different from that of the Bulgars some centuries earlier. They were equally viewed as former Roman subjects of the empire who possessed territory which belonged to the Constantinopolitan emperors. The conflicts between the Romans of Constantinople and those Romans outside the current borders of imperial authority demonstrate that the Byzantine notion of a Roman internal armed conflict (emhylios polemos) remained diachronically bound to the image of a Roman political community circumscribed by the boundaries of imperial authority. To that, it should be added that the largest part of the armed forces in the civil wars within the post-​1261 Roman imperial state of Constantinople consisted of foreign mercenaries.61 The latter were viewed as subjects of the Roman political community, as the report of Pachymeres on the emphylios polemos between Andronikos II Palaiologos and his Alan mercenaries demonstrates.62 In light of this, the use of the term homophylos by authors like Kantakouzenos in the context of civil warfare within the boundaries of the Constantinopolitan emperor’s authority seems to have had less to do with a self-​perception of the Romans as an ethnonational group. It referred to those that were currently subjects of the Roman emperor, thus distinguishing them from those outside the boundaries of imperial authority. The latter’s hetero-​ or self-​categorization as Romans in ethno-​cultural terms did not alter the fact that they were excluded from the vision of a Roman political community whose boundaries were demarcated by subject status and loyalty to the Roman emperor’s centralized authority.

Conclusion

Throughout this chapter, I have sought to examine the development of identity discourse in civil wars in the medieval East Roman world between the seventh and fifteenth centuries. The Byzantine understanding of an internal armed conflict was dependent upon the image of a Roman political community, imagined in hierarchical and centripetal terms, whose boundaries were demarcated by the fluctuating limits of enforceable imperial authority. Even though the Greek term for internal armed conflict, emphylios polemos, bears connotations of common ethnic descent in etymological terms, 60 Heinirch Gelzer, “Ungedruckte und wenig bekannte Bistumsverzeichnisse der orientalischen Kirche II,” Byzantinische Zeitschrift 2 (1893): 40–​73, at 44. 61 Mark C. Bartusis, The Late Byzantine Army Arms and Society, 1204–​1453 (Philadelphia: University of Pennsylvania Press, 1992), 85–​102. 62 See note 22 above.

116

116

Yannis Stouraitis

in the case of Byzantine civil wars it was employed to denote an armed conflict within a supra-​ethnic community of imperial subjects whose aim was the usurpation of the imperial power. This conceptualization of civil war by Byzantine authors demonstrates the impact of the Roman imperial legacy and, in particular, the importance of the imperial office, that is, of centralized imperial rule in shaping the image of a Roman political community, as well as the practices of identification with it. The fact that the notion of a shared Roman ethnic origin played no role whatsoever in defining a war as internal armed conflict in medieval East Roman political thought relates to the operative ideology which underpinned the unity of this imperial community up to 1204, and which made political Romanness first and foremost a practice of ideological attachment to the divinely ordained monocracy of the Christian Roman emperor of Constantinople. Civil war could solely be understood as a practice intended to reproduce the system of centralized imperial rule through the change of ruler, thus reasserting belief in and identification with the perpetual political order of Constantinople—​New Rome and its emperor. Within this framework, the shared Christian faith of the emperor’s subjects was a main means to construct a common cultural background for all the participants in Byzantine civil wars, and became the main vehicle through which the language of metaphoric kinship was incrementally applied to an imperial community consisting of subjects of various ethnic backgrounds. The development of identity discourse in the historiographical reports on civil wars seems to follow the gradual demarcation of a dominant Roman ethno-​cultural category consisting of native Greek-​speaking Chalcedonian Christians within the empire in the High Middle Ages, the image of which was consolidated by the events of 1204. Despite these developments, the notion of civil war in Byzantine thought after 1261 continued to be applied to conflicts within the boundaries of the contracted imperial state of Constantinople, and not to conflicts among the members of the broader Roman ethno-​cultural community whose boundaries extended far beyond the boundaries of that imperial state. The consistent conceptualization of Byzantine civil war as a conflict among the emperor’s subjects demonstrates that the image and the boundaries of the Roman political community remained independent from and incongruent with those of an ethnic group.

117

Chapter 7

WARFARE AND PEOPLEHOOD: THE VIKINGS AND THE ENGLISH Clare Downham This chapter provides

a case study of conflict and identity in ninth-​ and tenth-​century England. It questions some of the common labels we use, starting with those we call vikings and arguing that their identities in English-​speaking lands were diverse rather than cohesive, and evolved over time. The analysis will briefly explore terminology in primary sources before looking at the role of battles and negotiations in the creation of a unified kingdom of England. Finally, consideration will be given to the St. Brice’s Day massacre in 1002 when, according to the Anglo-​Saxon Chronicle, the English king “ordered to be slain all the Danish men who were in England.”1 This would seem to be the starkest example of opposition between the English and vikings. The story of the English kingdom is often represented as a clash of cultures between Alfred the Great and his descendants against Scandinavian settlers. The case presented here is that integration and negotiation were more important in the making of England than warfare. It was only once the boundaries of the English kingdom were successfully expanded that there was growing intolerance to perceived enemies within, and a more rigid use of ethnic labels was applied. One essential problem we have in interpreting how (the people we call) vikings perceived their identity in the ninth century is the lack of contemporary written sources written by them. As Judith Jesch has pointed out in her recent book, there are three ways of defining the term “viking.” We can address its Old Norse etymology, historical usage, and current meaning. The etymology is debated but a víkingr was someone defined by their actions not their ethnicity; in historical usage Scandinavian víkingar “refers to people (always in groups) who were engaged in some sort of military activity, often but not always piratical”—​it tends to be pejoratively applied to opponents in one’s own country but could have more positive connotations referring to the activities of one’s own people elsewhere.2 It also turns up as a personal name in Norse-​speaking communities. However, the label is not used that frequently among Scandinavians or English speakers who used the cognate word wicing. The common use of the word “viking” is a modern phenomenon, when it was adapted and reintroduced into English in the nineteenth 1 The Anglo-​Saxon Chronicle: A Collaborative Edition. 6. MS. D: A Semi-​Diplomatic Edition with Introduction and Indices, ed. George P. Cubbin (Woodbridge: Brewer, 1996), s.a. 1002: “het ofsleagan ealla þa deniscan men þe on Angelcynne wæron”; Engl. transl. Dorothy Whitelock et al., The Anglo-​ Saxon Chronicle (London: Eyre and Spottiswoode, 1965), 86. 2 Judith Jesch, The Viking Diaspora (London: Routledge, 2015), 6.

118

118

Clare Downham

century to mean Scandinavian adventurer.3 Jesch defends this last usage as useful “to give a name to a phenomenon which previously had no name but needed one,” but this of course poses the risk of anachronism and imposing our categories of what needs to be named on the past.4 In separate publications, I have theorized the viking phenomenon as a militarized trading diaspora. It was not exclusively Scandinavian, although early raids were led from Scandinavia. Displaying elements of Scandinavian cultural affiliation was a means of engaging in this economic venture.5 The cultural identity of those we would call vikings was eclectic. The movement of people led to a plethora of new localized identities to be created. Scandinavian-​influenced “viking” cultures and Old Norse language spread and intermingled with the language and culture of different peoples across medieval Europe.6 I do not consider that the viking phenomenon represents a diaspora defined by Scandinavian biological descent. This matter has been problematized by Lesley Abrams.7 Her model of Viking Age society is one of elite courts “sharing a culture based not just on language and origins but on ongoing contacts and competitive interplay and imitation,” which then influenced cultural values in the immediate localities of these lordly centres that might allow individuals to identify themselves in different ways in different contexts.8 Drawing on a later range of evidence, Jesch refutes Abrams, asserting a Scandinavian dimension with “remarkable patterns of continuity.”9 However, this assumes that later evidence reflects continuity rather than recognizing dynamism in the identity of communities over time. Rather, the later evidence can show that Scandinavian identities sharpened over time where they did not disappear, and that later uses of Scandinavian identity were situational and opportunistic. Scandinavian-​ness could serve a political purpose, whether to claim a level of independence from other peoples, to affiliate oneself with Scandinavian royal power, or to promote pan-​Scandinavian links.10 This growth of ethnic assertiveness can be linked to the centralization of power 3 Jesch, Viking Diaspora, 7

4 Jesch, Viking Diaspora, 5.

5 Clare Downham, “Coastal Communities and Diaspora Identities in Viking Age Ireland,” in Maritime Societies of the Viking and Medieval World, ed. James Barratt and Sarah Jane Gibbon (Leeds: Maney, 2015), 369–​83.

6 Clare Downham, “Vikings Were Never the Pure-​Bred Master Race White Supremacists Like to Portray,” The Conversation, September 28, 2017 (Retrieved from https://​thec​onve​rsat​ion.com/​); cf. Catrine L. Jarman, Identities Home and Abroad: An Isotopic Study of Viking Age Norway and the British Isles (master’s thesis, University of Oslo, 2012), 69.

7 Lesley Abrams, “Diaspora and Identity in the Viking Age,” Early Medieval Europe 20 (2012): 17–​38, at 20. 8 Abrams, “Diaspora,” 21.

9 Jesch, Viking Diaspora, 70.

10 Jesch, Viking Diaspora, 71–​80. By way of comparison, Brittonic identity was strategically promoted in Strathclyde in the twelfth century to assert the independence of the diocese of Glasgow.

119



Warfare and Peoplehood

119

in the eleventh, twelfth, and thirteenth centuries in Europe.11 Myths were invented and labels were promoted at this time to justify a developing sense of pride in perceived ancestry, essentially rewriting the past to justify the present.

Viking Conquerors and Their Identities

To turn to the contemporary written evidence from England, we are largely dependent on literature composed in English court circles or major churches to provide insight into viking activities and identities.12 This provides an opportunity to see if and how ethnic identities were forged in adversity. The Anglo-​Saxon Chronicle is used by historians as the main source for the chronology of viking raids and settlement in England. However, the limitations of its record can sometimes be tested by comparison with other sources.13 It nevertheless remains the main record of raids that were led against English-​speaking peoples from the late eighth century. It records the achievement of different viking groups in taking over large areas of territory in the 860s and 870s, leaving only parts of northern Northumbria and Wessex in the hands of English rulers. After that time, the kingdom of Wessex under the rule of Alfred the Great and his successors rolled back the boundaries of viking rule to eventually forge a unified kingdom of England in the mid-​ tenth century. The core text of the Anglo-​Saxon Chronicle was put together during the reign of King Alfred the Great and it was continued to promote the interests of the West Saxon royal dynasty after his death. The Anglo-​Saxon Chronicle thus gives a rivals’ perspective on the political identities of vikings in England. A variety of labels is used within the text from 787 to 1000. The term wicing (viking) is employed in annals 879 and 885 and version A, in annal 917. The word hæðen (heathen) occurs frequently from 832 to 872. Norðmenn is recorded in annal 787, which purports to record the first viking raid, but this is unlikely to be a contemporary record. The term then reappears in 920 (version A), 937, and 942. It seems to be a catchall term for Scandinavian speakers until the eleventh century, when it might also have been used in specific contexts to define Norwegians once the Norwegian kingdom became a more territorially defined state.14 The term Dene is used sporadically in the tenth century. The most popular term to describe vikings in the Anglo-​Saxon Chronicle in the ninth and tenth centuries was Denisc. Like Norðmenn, the terms Dene and Denisc were used as general terms for Scandinavian speakers (or the 11 Robert Bartlett, The Making of Europe: Conquest, Colonization and Cultural Change 950–​1350 (Princeton: Princeton University Press, 1993), 236. 12 The use of material culture to interpret identity is beyond the scope of this analysis.

13 Clare Downham, “The Earliest Viking Activity in England?” English Historical Review 132 (2017): 1–​12. 14 John Bosworth and Thomas Northcote Toller, eds., An Anglo-​Saxon Dictionary: Based on the Manuscript Collections of the Late Joseph Bosworth (Oxford: Clarendon Press, 1908), 725–​76; Downham, “Hiberno-​Norwegians,” 142–​43.

120

120

Clare Downham

peoples under their rule) in the ninth and tenth centuries.15 There is little in the Anglo-​ Saxon Chronicle to support the notion that there was antagonism between vikings from Norway and Denmark who settled in England during the ninth and early tenth centuries, although this has been something of a historiographic commonplace in the twentieth century. It may reflect modern notions of national identity and rivalries being projected onto the past. Vikings were also referred to in military terms, so that in annals from 860 to 892 the word here (invading army) was the favoured usage. It is particularly interesting to note that English regional names Norþ(an)hymbre and Eastengle are used to describe viking armies from 893 to 919 and Norþ(an)hymbre was used again from 944 to 948. This implies that decades after vikings settled, they were identified and may have identified themselves by the English lands where they lived. It seems plausible that the armies that were put into the field by viking kings of East Anglia and Northumbria comprised people of local descent as well as settlers. The labels are a reminder that there were plural viking polities in England, but also that regional identities could be shared by people of different origins. An insight into the political organization of viking-​held territories may be provided by administrative divisions. Scandinavian influence on territorial organization in England was far from homogenous. This may reflect the decentralized nature of early settlements and adaptation to local circumstances. The reuse of an existing administrative framework has been posited in viking-​ruled Mercia, and this may have encouraged new settlers to identify themselves with existing communities. Units of government which appear in one area of Scandinavian influence such as the ridings of Yorkshire and Lincolnshire do not appear in other areas such as the East Midlands. This suggests different leaders applied different strategies in different areas at different times.16 There may have initially been a degree of independence and self-​government in some political units. One example is the Wirral peninsula (in northwest England). From the evidence of the Fragmentary Annals of Ireland, it has been deduced that this land was granted to a viking leader called Ingimund at the start of the tenth century.17 The place-​name Thingwall in the Wirral is derived from Old Norse þing-​völlr and denotes an assembly place. Roughly four miles away the place-​name Raby (ON Rá-​býr) denotes a boundary settlement, which may have marked the southern limit 15 Bibire, “North Sea Language Contacts,” 89; Clare Downham, “ ‘Hiberno-​Norwegians’ and ‘Anglo-​Danes’: Anachronistic Ethnicities in Viking Age England,” Mediaeval Scandinavia 19 (2009): 139–​69, at 139; Sara Pons-​Sanz, The Lexical Effects and Anglo-​Scandinavian Linguistic Contact on Old English (Turnhout: Brepols, 2013), 221–​22. 16 John Baker and Stuart Brookes, “Governance at the Anglo-​Scandinavian Interface: Hundredal Organization in the Southern Danelaw,” Journal of the North Atlantic 5 (2013): 76–​95; cf. Dawn M. Hadley, “Multiple Estates and the Origins of the Manorial Structure of the Northern Danelaw,” Journal of Historical Geography 22 (1996): 3–​15.

17 The Fragmentary Annals of Ireland, ed. and trans. Joan N. Radner (Dublin: Dublin Institute of Advanced Studies, 1978), § 429.

121



Warfare and Peoplehood

121

of this small viking enclave.18 The lack of administrative uniformity in areas under viking rule may support the hypothesis that viking-​held areas did not share a strong common identity, but that loyalties might be territorially defined as vikings reached accommodation with local powerbrokers (such as the archbishop of York in the kingdom of Northumbria). Although different viking armies united from time to time in opposition to a common foe in the ninth and early tenth centuries, such alliances tended to be temporary in nature. The label “Danelaw” to define Scandinavianized areas in England is first recorded in a law code of CE 1008, but it is often, anachronistically projected back to the late ninth century when the country was divided between different viking kingdoms.19 The unity of the Danelaw may be seen as a creation of royal advisors in the late tenth century, not a perceived reality of viking settlers in the ninth century. The concept of the Danelaw has been a staple of English historiography, represented on maps as a single block in the late ninth and early tenth centuries standing in opposition to the territories ruled by kings of Wessex. One might even challenge how Danish the Danelaw was. Katherine Cross has argued that some of the legal practices which were allowed to remain regionally distinct on the basis of Scandinavian inheritance may well cover for pre-​Viking-​Age regional differences.20 In other words, pleading Scandinavian exceptionalism was to become a way of maintaining local identity in the face of a centralizing state, but the roots of this division were regional not ethnic. The modern use of the terms “Danelaw” and “Anglo-​ Scandinavian” by scholars tends to obscure the lack of political and cultural unity within the viking polities of England.21 In contrast, the development of terminology to represent the unity of the English is well documented. In the eighth century, Bede had promoted the idea of a shared history of the English-​speaking peoples. This idea was used by later ecclesiastical writers. In his famous letter responding to the sack of Lindisfarne in 793, Alcuin wrote: Lo, it is nearly 350 years that we and our fathers have inhabited this most lovely land, and never before has such terror appeared in Britain as we have now suffered from a pagan race.22

Just as Bede referred to the arrival of Germanic people as a punishment to the Britons for their sins, Alcuin seems to have feared that the English-​speaking peoples 18 Paul Cavill, “Major Place-​Names of the Wirral: A Gazeteer,” in Wirral and its Viking Heritage, ed. Paul Cavill, Stephen Ernest Harding, and Judith Jesch (Nottingham: English Place Name Society, 2000), 125–​47 s.v. Raby, Thingwall.

19 Katherine Holman, “Defining the Danelaw,” in Vikings and the Danelaw: Selected Papers from the Proceedings of the Thirteenth Viking Congress, Nottingham and York, 21–​30 August 1997, ed. James Graham-​Campbell et al. (Oxford: Oxbow, 2001), 1–​11, at 2. 20 Katherine Cross, Heirs of the Vikings? History and Identity in Normandy and England c. 950-​c. 1015 (Woodbridge: Boydell & Brewer, 2018), 193–​98. 21 Cross, Heirs of the Vikings, 15, 155.

22 Two Alcuin Letter Books, ed. Colin Chase (Toronto: Pontifical Institute of Medieval Studies, 1975), 55; Whitelock, trans., English Historical Documents, 776.

122

122

Clare Downham

would be overwhelmed by a fresh migration of pagans if they did not mend their ways.23 In the mid-​ninth century, the word Angelcynn appears in charters as a vernacular equivalent to gens Anglorum.24 The label was promoted during the reign of Alfred the Great as an inclusive label for the peoples of Wessex and Mercia. The promotion of literacy and court-​sponsored literature was part of the centralizing program of Alfred’s reign. Among other texts, he sponsored a vernacular translation of Bede’s Historia and his court was closely linked with the creation of the Anglo-​Saxon Chronicle. Both texts promoted a unified view of English history, the latter with a pronounced Wessex spin.25 Alfred also issued a law book which cherry picked from earlier Wessex and Mercian codes and was prefaced by a translation of the Law of God as given to Moses.26 Thus, the Angelcynn were presented as one people with a single set of laws sanctioned by divine will. Ethnic labelling was important to legitimize new structures of power, and to generate a sense of loyalty among the elite. It was backed up by historical and biblical discourse during King Alfred’s campaign for vernacular literacy.27 Angelcynn and Englisc were useful terms and they were successful, according to Walter Pohl’s discussion of identities, in marking “a clear boundary among competing groups of warriors” and “at the same time open enough to accommodate those who had been recently won over, or even whose support might be desirable in future.”28 Englishness was an identity that people could opt into, even with an awareness of or affiliation to Scandinavian heritage. The Anglo-​Saxon Chronicle even gave the kings of Wessex a link to Scandinavian ancestry in presenting a lengthy genealogy of Alfred’s father Æthelwulf under the year 855. The inclusion of the name Scyld provided a link to the Danish royal line.29 As well as reflecting on Englishness this link may have also been seen by Alfred as a way to legitimize his rule over Scandinavian settlers, should the opportunity have arisen, or perhaps to find common ground in his negotiations with viking rulers. 23 Councils and Ecclesiastical Documents Relating to Great Britain and Ireland, ed. Arthur W. Haddan and William Stubbs, 3 vols. (Oxford: Clarendon Press, 1869–​71), 3:510. The theme is repeated in Archbishop Wulfstan’s “Sermon of the Wolf to the English” in 1014. 24 Patrick Wormald, “Engla lond: The Making of an Allegiance,” Journal of Historical Sociology 7 (1994): 1–​24, at 11.

25 Clare Downham, “Annals, Armies, and Artistry: ‘The Anglo-​Saxon Chronicle’, 865–​96,” in No Horns on Their Helmets? Essays on the Insular Viking Age, ed. Clare Downham (Aberdeen: University of Aberdeen, 2013), 9–​37. 26 Wormald, “Engla lond,” 15.

27 Walter Pohl, “Introduction: Strategies of Distinction,” in Strategies of Distinction: The Construction of Ethnic Communities, 300–​800, ed. Walter Pohl and Helmut Reimitz (Leiden: Brill, 2013), 1–​15, at 2. 28 Pohl, “Introduction,” 6–​7.

29 Alexander C. Murray, “Beowulf, the Danish Invasions, and Royal Genealogy,” in The Dating of Beowulf, ed. Colin Chase (Toronto: University of Toronto Press, 1997), 101–​12.

123



Warfare and Peoplehood

123

Warfare and Englishness Traditional pedagogy has fostered the idea that history is shaped by decisive battles.30 In relation to Alfred and his successors in creating an English kingdom, the most critical conflicts are identified as Edington (878), Tettenhall (910), and Brunanburh (937). Edington was a significant victory for Alfred as it brought the viking leader Guthrum into an agreement which included giving hostages, oaths, and the baptism. It seems Guthrum or his allies had previously broken an agreement bound by oaths and hostages in 876.31 The resulting conversion of Guthrum seems to have been effective. Rather than being forced, Guthrum’s baptism may have served an agenda to build links with the Church in his new kingdom. Guthrum continued to rule East Anglia until his death in 890.32 The battle of Tettenhall in 910 was fought as a result of a broken treaty between the army of Northumbria and Alfred’s son, Edward the Elder. The Northumbrians may have chosen to ravage Mercia because the lord of Mercia was an ally and brother-​in-​law of Edward the Elder, king of Wessex. He also seems to have been ill at the time and his wife Æthelflaed may have been in control of the government. An army of Wessex and Mercia intercepted the Northumbrian attackers on their way home and “many thousands” were killed. Brunanburh is sometimes hailed as one of the greatest English victories as Athelstan (son of Edward the Elder) led the men of Wessex and Mercia to victory against a coalition of vikings, Scots, and Britons. However, the battle was far from decisive as Athelstan died two years later and Northumbria came back under viking control. The significance of these three battles may be interpreted in the context of other events of this period. From 871 to 954, the Anglo-​Saxon Chronicle records many battles and skirmishes, the building and taking of fortresses to control areas of land, and a significant number of peace treaties and submissions. Peace treaties were often short-​lived but they were important in limiting bloodshed and maintaining lines of communication between rival groups. The Anglo-​Saxon Chronicle makes references to peace agreements in 871, 872, 873, 876, 877, 878, 885, 893, 906, 910, 918, 919, 920, 921, 926, 941, 943, 946, 947, and 948, as well as numerous acts of submission. The greatest gains for the Wessex dynasty were made not through battles but through oaths and agreements. Most decisively, the Northumbrians’ expulsion of their last viking king in 954 and acceptance of the Wessex king Eadred marked a final chapter in the unification of England. The kingdom may have been forged more at the negotiating table than on the battlefield. During this period, English identity does not seem to have been a barrier to leadership of groups identified as Danes and vice versa. This is illustrated after the death of Alfred the Great when the rule of Edward the Elder was disputed by his cousin Æthelwold. Æthelwold was recognized as king by the armies of Northumbria and East Anglia. At the 30 Edward S. Creasy, Fifteen Decisive Battles of World History from Marathon to Waterloo (London: R. Bentley, 1851), with countless reprints. 31 Whitelock et al., Anglo-​Saxon Chronicle, 48–​49.

32 Whitelock et al., Anglo-​Saxon Chronicle, 53.

124

124

Clare Downham

Battle of the Holme, Æthelwold and his English followers fought alongside the viking leaders and men of East Anglia against Edward’s Kentish allies. In 918, the Mercian Register reports that leaders of York were willing to pledge obedience to Æthelred’s widow Æthelflæd. She was recognized by her own people as holding the lordship of Mercia after her husband’s demise. After Æthelflæd’s death in 918, there is the extraordinary circumstance of medieval leadership passing from a mother to her daughter. The succession indicates determination among the Mercian elite to retain a level of independence from Wessex. Ælfwynn’s time in power was soon curtailed. She was “deprived of all authority in Mercia and taken into Wessex” by her uncle, Edward the Elder.33 Edward’s subsequent rule in Mercia may have been unpopular as later sources report a revolt in Chester at the end of his reign.34 After Edward’s death, Mercia would be ruled separately from Wessex for a short time by his younger son Athelstan in 924. It was significant that Athelstan had been fostered in the court of Æthelflæd and had an opportunity to build good relations with the local nobility. Mercia would be ruled separately on further occasions in 957–​959 and 1016. At least theoretically, on each occasion, the history of the Mercian kingdom could have followed a separate path from Wessex over a longer timeframe. These circumstances show that regional loyalties continued and that they could outweigh desire for a unitary ethnic kingdom of Angelcynn.35 Distinctions continued to be made between different peoples under the rule of English kings. In the poem which celebrates the Battle of Brunanburh in the Anglo-​Saxon Chronicle the enemies are defined as “Scots,” “Northmen,” and “pirates,” but the heroes are the people of “Wessex” accompanied by the “Mercians.” It suggests that regional labels continued to hold significance. At the end of the poem a rousing statement is made relating to the common glory of the English-​speaking peoples: Never yet in this island before this, by what books tell us and our ancient sages, was a greater slaughter of a host made by the edge of a sword, since the Angles and Saxons came hither from the east, invading Britain over the broad seas, and proud assailants, warriors eager for glory, overcame the Britons and won a country.36

This statement instructs or reminds its audience of the shared history of the English peoples, as if it was not self-​evident. In presenting the battle as a victory won primarily by the people of Wessex, but with the Mercians on board, the Chronicle continued to champion a vision of a united English realm with the people of Wessex at the helm. 33 Whitelock et al., Anglo-​Saxon Chronicle, 69.

34 David Griffiths, “The North-​West Frontier,” in Edward the Elder 899–​924, ed. Nick Higham and David Hill (Abingdon: Routledge, 2001), 167–​87. 35 For a contrary view, see Wormald, “Engla lond,” 8.

36 Anglo-​Saxon Chronicle (D) s.a. 937: “Ne wearð wæl mare on þisne iglande æfre gita folces gefylled beforan þyssum sweordes ecgum, þæs þe us secgað bec ealde uðwitan, siððan eastan hider Engle 7 Seaxe up becomon ofer brade brimu, Britene sohton wlance wigsmiðas, Wealas ofercomon earlas arhwæte, eard begeaton”; Whitelock et al, Anglo-​Saxon Chronicle, 69–​70.

125



Warfare and Peoplehood

125

It is interesting to note that when Northumbrians were subjected to Wessex (a non-​linear process which ran from 927 until 954), royal diplomas began to include the derogatory terms “barbarian,” “gentile,” or “pagan” to describe non-​English peoples which the Wessex dynasty claimed to rule. In other contexts, such terms are used to describe enemies.37 This suggests that while the process of othering might be a unifying tactic against an enemy in conflict, it became more pronounced once a people had been subjected.38 The connotations of this language are imperial and biblical, marking out English superiority. Imperial labels such as basileus and imperator were used for tenth-​ century English kings in diplomas and other texts. The multifaceted nature of their rule was recognized in coronation ordo, laws, and charters. An example is the Quadripartite rule attributed to them in a handful of charters issued from the 940s until 968, which identified the Wessex monarchs as rulers over “Anglo-​Saxons, Northumbrians and Pagans and Britons.”39 It is interesting in this context that being Northumbrian is presented as a separate category to Anglo-​Saxon and Pagan, implying a regional rather than a religious or ethnic allegiance. From 980, a wave of Scandinavian attacks commenced against England that would eventually culminate in the conquest of the kingdom by Cnut. This is often called the “Second Viking Age”; however, such a term obscures the continuity of viking activity outside England (for example, in Ireland), and it also draws what may be a false comparison between the raiders at the end of the eighth century and the more centralized structures of power which were behind the attacks at the end of the tenth century. Interestingly, the term Denisc was applied both to the raiders from Scandinavia and the native-​born Scandinavian population in England during the reign of Æthelred II. His kinsman and chronicler Æthelweard wrote an account of English history before the Scandinavian raids of the 980s. Æthelweard elaborated on the mythical Scandinavian element to Wessex royal history recorded in the Anglo-​Saxon Chronicle.40 This may show 37 Peter Sawyer, Anglo-​ Saxon Charters: An Annotated List and Bibliography (London: Royal Historical Society, 1968), S548: “Eadred rex rite Anglorum gloriosissimus rectorque Noþhanhimbra et paganorum imperator Brittonumque propugnator”; S549: “Ealdredus rex Ængulsæxna ond Norðhymbra imperator, paganorum gubernator. Brittonumque propugnator”; S632: “ego Adwig basileus Anglorum huiusque insule barbarorum”; S569: “Eadred rex Angulsæxna et Norþhimbra imperator paganorum gubernator. Brettonumque propugnator”; S725: “ego Eadgarus gentis Anglorum et barbarorum atque gentilium rex.” 38 Raymond I. Page, A Most Vile People: Early English Historians on the Vikings (London: Viking Society for Northern Research, 1987). Comparison may be made with Ireland when xenophobic attitudes towards those of viking descent emerged more strongly in literature of the eleventh and twelfth centuries after the viking towns of Ireland had lost much of their independence.

39 Sarah Foot, “The Making of Angelcynn: English Identity before the Norman Conquest,” Transactions of the Royal Historical Society 6 (1996): 25–​49, at 46; Simon Keynes, “The Henry Loyn Memorial Lecture for 2008: Welsh Kings at Anglo-​Saxon Royal Assemblies (928–​55),” Haskins Society Journal 26 (2015): 69–​122, at 98–​99. 40 Cross, Heirs of the Vikings, 37–​40.

126

126

Clare Downham

an interest in demonstrating the inclusivity of English rule and identity to all those who wanted to adopt it, even if they were aware of ancestral links to Scandinavian culture. However, this mood of inclusivity did not apply years later, when in 1002 “the king ordered to be slain all the Danish men who were in England … On St Brice’s Day” (November 13).41 The context of the event seems to be the desertion of a former Scandinavian ally of the king, Pallig, in 1001, and the hefty payment of 24,000 pounds of silver to make peace with the Danes in 1002. The Anglo-​Saxon Chronicle claims that the king heard of a Danish plot to overthrow Æthelred and his councillors that year. The suspicion may have fallen on troops from Scandinavia who were still in England, or other recent arrivals to the kingdom.42 It seems highly unlikely that a general populace who claimed Scandinavian ancestry would have been put to the sword. Areas linked to ninth-​century viking settlement were also under attack and supported the English war effort.43 Nevertheless, the order was intended to stir up ethnic division and brutality. The St. Brice’s Day massacre may bear some comparison with early eleventh-​century attacks on minorities elsewhere in Europe. While they did not pose a military threat, there were attacks on Jews in parts of France recorded from 1012–​1015 and heretics were burnt in Orleans in 1022. These events have been interpreted by R. I. Moore as the early stages of a new persecuting mentality.44 In his words, “persecution began as a weapon in the competition for political influence and was turned by the victors into an instrument for consolidating their power over society at large.”45 Æthelred may have hoped that the scapegoating of Danish residents in England could help strengthen his rule and act as punishment for suspected disloyalty among their ranks. It is hard to know whether the king mobilized a large number of his subjects, or if the order was met with apathy or resistance. A clear example of its implementation was in Oxford where a charter was issued to the church of St Frideswide, which was destroyed when a group of Danes sought to barricade themselves in: having broken by force the doors and bolts, and resolved to make refuge and defence for themselves therein against the people of the town and the suburbs; but when all the people in pursuit strove, forced by necessity, to drive them out, and could not, they set fire to the planks.46

41 Whitelock et al., Anglo-​Saxon Chronicle, 86. Anglo-​Saxon Chronicle (D) s.a. 1002: “Se cyng het ofsleagan ealla þa deniscan men þe on Angelcynne wæron, on Britius mæssedæg.”

42 Reynolds, “What Do We Mean by ‘Anglo-​Saxon’ and ‘Anglo-​Saxons’?” Journal of British Studies 24 (1985): 395–​414. 43 Holman, “Defining the Danelaw,” 8.

44 Robert I. Moore, The Formation of a Persecuting Society, 2nd ed. (Oxford: Blackwell, 2007), 135. 45 Moore, Formation, 138.

46 Sawyer, Anglo-​Saxon Charters, S909: “fractis per uim ualuis et pessulis, intrantes asylum sibi propugnaculumque contra urbanos suburbanosque inibi fieri decreuerunt; sed cum populus omnis insequens, necessitate compulsus, eos eiicere niteretur nec ualeret, igne tabulis iniecto.”

127



Warfare and Peoplehood

127

The charter also mentions the killings as “a most just extermination.”47 The accession of Cnut in England may have heralded a more positive view of Scandinavian identity. As ruler of Denmark, Norway, and England, it may have been in Cnut’s interests to promote Scandinavian identity linked through a common language, the dƍnsk tunga.48 Long after Danish rule in England had ceased, there were pronounced regional displays of Scandinavian identity in England. This may be exemplified in northwest England through the use of Scandinavian runes in twelfth-​century inscriptions.49 These displays of difference may relate to continued trading links within a Scandinavian-​speaking world and the desire for a distinctive regional identity. Notions of identity, inclusion, or exclusion were reinvented and manipulated through time.

Conclusion

While the narrative of the creation of the English kingdom in the ninth and tenth centuries is often presented as an ethnic battle between Anglo-​Saxons and vikings, it may provide a useful alternative to see this as a struggle between supra-​regional power and regional power, in which treaties and submissions played a significant role. In the course of the ninth and tenth centuries, power in England and Scandinavia became more centralized. However, in line with the ideas of Siniša Malešević, it does not seem that conflict sharpened ethnic identities; rather, a centralizing ideology was at play.50 The concept of Englishness was introduced well before viking raids began but it was promoted by Alfred the Great and his successors as a unifying factor in the areas they sought to rule. It was made flexible enough to absorb people with Scandinavian cultural affiliations who could opt in. However, by the turn of the eleventh century a process of othering toward subjected peoples (who are regarded as inferior) or those whose loyalties were suspect came sharply into view. It was among those furthest from the seat of power that a continuing Scandinavian regional identity can be observed. The changes which took place in the ninth and tenth centuries sowed the seeds for developing notions of Englishness which can be observed in the twelfth century. This concept of Englishness was permeable and it would later absorb the identity of Norman settlers. It was also antagonistic in fostering a negative view of non-​English peoples that the elites sought to control.51

47 “iustissima examinatione.”

48 Skaldic Poetry, ed. M. Clunies Ross et al., vol. 1 (Turnhout: Brepols, 2007), 555–​56.

49 With examples at Pennington, Conishead Priory, Bridekirk, and Carlisle Cathedral in Cumbria.

50 Sinsa Malešević, The Sociology of War and Violence (Cambridge: Cambridge University Press, 2010), 34, 65. 51 John Gillingham, “The Beginnings of English Imperialism,” Journal of Historical Sociology 5 (1992): 392–​409.

128

129

Chapter 8

MEDIEVAL EUROPEAN CIVIL WARS: LOCAL AND PROTO-​NATIONAL IDENTITIES OF TOULOUSAINS, PARISIANS, AND PRAGUE CZECHS Philippe Buc*

Looked at from the Far East, and specifically Japan, the civil wars or intercultural conflicts fought in High and Late medieval Catholic Europe look remarkably odd: coloured, to a lesser or greater degree, by an exclusivist religion or by semi-​secularized byproducts of this same religion, they veer easily to a Manichaean black-​and-​white tenor. Japan was at war for most of its medieval centuries, from the twelfth century that saw the founding of the Kamakura Shogunate (1192), all the way to Sekigahara, 1600, the battle that definitely put an end to the Sengoku jidai, the era of the contending principalities. In contrast with run-​of-​the mill western European history writing, Japanese chronicles either depict enmities soberly, led by power politics and thus rich in side-​switching, yet without much moralizing, or with an acute sense of the tragic, that is, by putting in dialogue contending norms or duties.1 Civil wars, understood here as armed conflicts sustained over time and embracing a majority of the politically active members of a polity (civitas), could be expressive or formative of identities. So could be religion, in its purer or secularized forms. While historiography in the 1970s foregrounded the issue of ethnicity, and began fruitfully to emphasize its fluidity,2 it now recognizes that identity, whatever it is, had in the medieval West, and already in the earlier Middle Ages, much to do with religion, as envisioned by * My thanks for input and ideas go to my Vienna University BA seminar (Winter Term 2017–​2018) students Joseph de Durfort, Christoph Haill, Karina Müller Mares, and Hannah Riedler. Also, thanks to my colleague Christina Lutter for a close reading of this text and her comments. This chapter was elaborated thanks to two fellowships, one at the Institute for Advanced Studies of the Central European University (October 2018 to January 2019), the other at Stanford University (Krater Visiting Professorship in the Department of History, February 2019). My gratitude to both institutions. 1 The idiosyncratic logic of medieval civil war, plus medieval implicit or explicit “emic” understandings of bellum civile, make all the more surprising its absence from a recent book. David Armitage’s Civil Wars: A History in Ideas (New Haven: Yale University Press, 2017) jumps happily from ancient Greece and Rome to early modern Europe over the medieval European millennium, claiming to provide a genealogy for our contemporary notions of intestine conflict and violence.

2 The earliest Viennese flagship book is Herwig Wolfram, Geschichte der Goten: von den Anfängen bis zur Mitte des sechsten Jahrhunderts: Entwurf einer historischen Ethnographie (Munich: Beck, 1979), American translation as History of the Goths (Berkeley: University of California Press, 1990). The German forerunner was Reinhard Wenskus, Stammesbildung und Verfassung: das Werden der frühmittelalterlichen gentes (Cologne: Böhlau, 1961).

130

130

Philippe Buc

local elites.3 The following considerations, therefore, cross-​fertilize civil war, religious conceptions, and the imagined belonging to a group by focusing on three dossiers. They are, namely, (1) the so-​called Albigensian Crusade (1208/​9–​1229); (2) the Franco-​ French civil war opposing the Armagnacs and the Bourguignons (1407–​1435); and (3) the Crusades waged by Emperor Sigismund against the dissident Bohemian Hussites and the latter’s equally holy resistance to papal Catholicism (1419–​1436). In all three cases, one identification at play was the city: Toulouse, Paris, and Prague.

Toulouse … when the news came … that Count Raymond [VII of Toulouse] had entered Toulouse to raise it up again, to destroy the French and to elevate Pretz (Prowess), all through the lands one found speech again, and one shouted out: “Toulouse! May God guide and save her, may He aid and help her, guard and defend her! May He give him [the Count] strength to amend what’s been lost, to rescue Paratge (Nobility) and fire up joy!”4

An intertextual reading of the narrative sources produced during the so-​ called Albigensian Crusade5 of 1208/​9–​1229 and in its aftermath allows one to compare a monastic and an aristocratic rendition of civil war, mediated by two source genres with which each outlook has elective affinities—​chronicle and epic. The Albigensian Crusade has been seen as a northern French invasion of the south, a master narrative highlighting the opposition between northern French and Occitan elites and their political cultures.6 But the crusade worked also on pre-​existing, internal cleavages and fault lines crisscrossing the Languedoc and Provence—​factions, rivalries, feuds even.7 Pope Innocent III preached the crusade owing to the 1208 murder of his legate, 3 Two publications among many, Walter Pohl and Gerda Heydemann, eds., Strategies of Identification: Ethnicity and Religion in Early Medieval Europe (Turnhout: Brepols, 2013); Eirik Hovden, Christina Lutter, and Walter Pohl, eds., Meanings of Community across Medieval Eurasia: Comparative Approaches (Leiden: Brill, 2016).

4 Canso, § 186, ed. Émile Martin-​Chabot, La chanson de la croisade albigeoise, 3 vols. (reprint Paris: Belles-​Lettres, 1957–​61), 2.306, trans. Janet Shirley, The Song of the Cathar Wars: A History of the Albigensian Crusade (Aldershot: Ashgate, 1996), 130 (modified).

5 For master narratives see the following: exalted but rich in insights, and not obviating recourse to earlier books, Mark Gregory Pegg, A Most Holy War: The Albigensian Crusade and the Battle for Christendom (Oxford: Oxford University Press, 2008); more sober, Jörg Oberste, Der Kreuzzug gegen die Albigenser: Ketzerei und Machtpolitik im Mittelalter (Darmstadt: Wissenschafliche Buchgesellschaft, 2003); and Christopher Tyerman, God’s War. A New History of the Crusades (Cambridge, MA: Harvard University Press, 2006), 563–​605. Recently published, Gregory. E. M. Lippiatt, Simon de Montfort and Baronial Government, 1195–​1218 (Oxford: Oxford University Press, 2017). 6 Fredrick L. Cheyette, Ermengard of Narbonne and the World of the Troubadours (Ithaca: Cornell University Press, 2001), describes, as it were, “the world before.” 7 See for a vivid portrait Cheyette, Ermengard, as well as Laurent Macé, Les comtes de Toulouse et leur entourage: Rivalités, alliances et jeux de pouvoir, XIIe–​XIIIe siècles (Toulouse: Privat, 2003).

131



Medieval European Civil Wars

131

Pierre de Castelnau, at the hand of a retainer of Raymond VI, count of Saint-​Gilles and Toulouse. Vehemently suspected of favouring the Cathar heretics,8 and inculpated for the assassination, Raymond was initially the crusade’s main target. He managed—​for a while—​to deflect it against his regional rival, Raymond-​Roger Trencavel, Viscount of Carcassone. Raymond-​Roger, defeated, soon died in prison. This did not save Count Raymond. After some hesitations, Innocent III stripped him of his honours in favour of Simon IV de Montfort, military leader of the expedition. Raymond VII’s own younger brother Baudoin, after having resisted the Crusaders, joined them. Captured, he was hung in 1214 for treason at Count Raymond’s order.9 Many southern lords changed sides, sometimes several times, which accounts for the multiple accusations of treason that dot the sources, and in part for the war’s savagery.10 Massacres, executions, and mutilations dot the record. This record is rich. There are, next to poems and letters, four narratives that allow entry into the relation between war and identity. The one farthest from the events was penned in 1275. It is a Latin chronicle, owed to William (Guillaume, Guilhem) of Puylaurens, a secular cleric who belonged to the entourages of the Cistercian Fulk, bishop of Toulouse (1205–​1231), and of the Count of Toulouse Raymond VII (d. 1249).11 By the 1270s, Raymond VII’s heiress, Jeanne, and her husband Alphonse, brother of King Louis IX, had died childless (1271), and the county had devolved to the Capetian crown. Guilhem penned his Chronicle, therefore, with the benefits of hindsight. Also clerical, but much more fervent in its crusading zeal, is a second Latin chronicle, the Historia Albigensis, by the young Cistercian monk Pierre des Vaux-​de-​Cernay.12 Pierre was one of the many Cistercians who joined the crusade as preachers and confessors; the Order also provided the cadre of prelates that led it spiritually. Cistercian involvement in the 8 Pegg, A Most Holy War, re-​iterates the position that he first expressed in The Corruption of Angels: The Great Inquisition of 1245–​1246 (Princeton: Princeton University Press, 2001) that there was no such thing as a unified Cathar heresy, and even no heresy at all. The Corruption was severely reviewed by Pete Biller, in Speculum 78 (2003): 1366–​70. See also Caterina Bruschi and Peter Biller, eds., Texts and the Repression of Medieval Heresy (Woodbridge: York Medieval Press, 2002), and now the debate in Antonio Sennis, ed., Cathars in Question (Woodbridge: York Medieval Press, 2016), with the clear if partisan review by Jan Rüdiger, “Rezension zu: Sennis, Antonio (Hrsg.): Cathars in Question”, H-​Soz-​Kult, July 5, 2017, www.hsozk​ult.de/​public​atio​nrev​iew/​id/​rez​buec​her-​26786. 9 Macé, Les comtes de Toulouse, 74–​86.

10 Macé, Les comtes de Toulouse, 268–​79, on fidelity and treason in the Toulousain.

11 Guillaume de Puylaurens, Chronique, prologue, ed. Jean Duvernoy (Paris: Éditions du CNRS, 1973); see The Chronicle of William of Puylaurens: The Albigensian Crusade and its Aftermath, trans. William Arnold Sibly and Michael D. Sibly (Woodbridge: Boydell, 2003).

12 Pierre des Vaux-​de-​Cernay, Hystoria Albigensis, ed. Pascal Guébin and Ernest Lyon, Petri Vallium Sarnaii Monachi Hystoria Albigensis, 3 vols. (Paris: Champion, 1926–​39); English trans. The History of the Albigensian Crusade: Peter of les Vaux-​de-​Cernay’s Historia Albigensis, trans. William Arnold Sibly and Michael D. Sibly (Woodbridge: Boydell, 1998). See Christopher M. Kurpiewski, “Writing beneath the Shadow of Heresy: The Historia Albigensis of Brother Pierre des Vaux-​de-​Cernay,” Journal of Medieval History 31 (2005): 1–​27.

132

132

Philippe Buc

fight against the dualist heresy known as “Catharism” had been traditional since Bernard of Clairvaux’s mid-​twelfth-​century preaching tours in the South of France.13 Reading the record, it becomes even clearer that the Cistercians trumped in zeal Pope Innocent III, and forced him, reluctantly, to take the hardest of possible lines against the Raymondine lineage.14 At least Pierre seems proud that his order’s leaders saw the truth—​that there should be no compromise with tepid or dissimulating southern lords—​more clearly and resolutely than the Pope (whose position, as documented in his letters and also self-​ servingly by the Canso de la Crozada’s second author, was far less black-​and-​white). Third and fourth, the bipartite Canso de la Crozada,15 an epic, which unlike Guilhem and Pierre’s Latin narratives was written down (and one must assume, sung) in Occitan.16 Two successive authors produced the Canso, an immigrant from Navarra, the cleric Guilhèm de Tudela, the other anonymous. This second, anonymous poet picks up where Guilhèm’s verses break off—​with the intervention north of the Pyrenees by King Pere II of Aragon, hero of the Christian victory against the Almohads at Las Navas de Tolosa (1212). The first poet, Guilhèm, favours the crusade, while frowning at its excesses. The second poet, the anonymous, is squarely on the side of the young Count Raymond VII, who led along with the Count of Foix resistance to Simon de Montfort and the Cistercians. With great craft, he exalts the Toulousains and delegitimizes the crusade.17 Consonant with his partisan position, one hears hardly anything about heresy in this second half of the Canso. 13 As explained, anew, by Simon de Montfort’s most recent biographer, Lippiatt, Simon de Montfort, 79–​98, in the wake of Beverly Mayne Kienzle, Cistercians, Heresy and Crusade in Occitania, 1145–​1229: Preaching in the Lord’s Vineyard (Woodbridge: Ashgate, 2001), esp. 109–​73.

14 Pierre des Vaux-​de-​Cernay, Hystoria Albigensis, ch. 14, § 438–​41, ed. Guébin and Lyon, 2.128–​34; trans. Sibly and Sibly, 199–​201. This is only hinted at; the Canso does make this reversal a product of Cistercian bullying, Canso, § 143–​52, ed. Martin-​Chabot, 2.39–​89, with Sharon Kinoshita, “Uncivil Wars: Imagining Community in La Chanson de la Croisade Albigeoise,” in Medieval Boundaries: Rethinking Difference in Old French Literature, ed. Sharon Kinoshita (Philadelphia: University of Pennsylvania Press, 2006), 200–​35 and 280–​6, at 212–​17.

15 Eugène Martin-​Chabot, La chanson de la croisade albigeoise, 3 vols. (reprint Paris: Belles-​Lettres, 1957–​61), English trans. Janet Shirley, The Song of the Cathar Wars: A History of the Albigensian Crusade (Aldershot: Ashgate, 1996).

16 For Occitan poetry and politics, see Martin Aurell, La vielle et l’épée: Troubadours et politique en Provence au XIIIe siècle (Paris: Aubier, 1989). See as well Barbara Rosenwein, “Poetic Dissent: The Troubadours at Toulouse,” in Disciplined Dissent: Strategies of Non-​Confrontational Protest, ed. Fabrizio Titone (Rome: Viella, 2016), 23–​40, with recent bibliography; Rosenwein, Generations of Feeling: A History of Emotions, 600–​1700 (Cambridge: Cambridge University Press, 2016), 113–​43, proposes a rich panorama of emotion-​words linked to treason and fidelity. Regrettably, however, limited credit is given to Cheyette, who first underlined how troubadour love language and political language (including fidelity and treason) were one and the same, see Ermengarde, chap. 13, “Love and Fidelity,” 233–​47 (the only references to Cheyette in Generations are 125n78 and 128n180).

17 See most recently Marjolaine Raguin, Lorsque la poésie fait le souverain: étude sur la ‘Chanson de la croisade albigeoise’ (Paris: Champion: 2015); Raguin, “The Words To Say It: The Crusading Rhetoric of the Troubadours and Trouvères,” in Singing the Crusades: French and Occitan Lyric

133



Medieval European Civil Wars

133

Pierre des Vaux-​de-​Cernay dreamt that the heretical south would be swallowed up, and digested, in Catholic universalism. A digestion with refuse: the religious dissenters were unlikely to convert sincerely, and so had to be destroyed (their lay friends, however, being devious, might fake return to orthodox Christendom, and therefore were to be tricked into unmasking themselves). One was duty bound to be uncompromising. Pierre even lauds and promotes pious deceit to bring the perfecti and perfectae, the heretics’ leaders, to the stake. He recounts with stunning approbation what took place at the siege of Minerve (1210), a fortified village surrounded by deep canyons, where a sizable number of the Cathar “perfects” were holed up. There, the Cistercians convinced the crusade’s lay leadership to give quarter to the heretics on the condition that they would convert. De facto, the abbot of Cîteaux “wanted the enemies of Christ to die but as a monk and priest, he dared not condemn them to death.” Hence this proposal, to offer the alternatives of conversion or death. A “noble and dedicated Catholic,” a warrior as unforgiving as the monks, objected that the Cathar religious elite, the “good men,” would dissimulate and pretend to return to orthodoxy; the Cistercian abbot answered that there was not any reason to worry; “I believe that very few of them will accept conversion.” And indeed it so came to pass; as deserved, all went to the pyre.18 Catholic universalism: there is little trace in Peter’s exalted (some would say fanatical) writing of a predilection for Frenchness; there is, however, a strong indictment of the southern region. Its central city, Toulouse, is also the centre of heretical evil, not simply because of its heretical count, but also because of its citizens. Toulouse since its very foundation “had hardly ever or rarely been free from” heresy; this virus had passed from father to son (a patribus in filios successive veneno superstitiosae infidelitatis diffuso); it had infected the whole region. The Cistercian mentions as divine vengeance that one of its kings, Alaric, was hung from the gallows before the city’s gates, an extreme dishonour. Tolosa, dolosa, wordplaying on Toulouse and deceit, dolus, appears in the very first folia of his Historia, and never leaves it.19 Pierre harangues the city, in the negative: O nidus hereticorum Tolosa! o taberna predonum! (O Toulouse, nest of heretics; o den of bandits).20 For Pierre, its count, Raymond VI, is, to cite Laurent Macé, “the archetype of the perjurer.”21 Responses to the Crusading Movements, 1137–​1336, ed. Linda M. Paterson (Woodbridge: Boydell & Brewer, 2018), 259–​85.

18 Pierre des Vaux-​de-​Cernay, Hystoria Albigensis, ch. 7, § 154–​56, ed. Guébin and Lyon, 157–​61, trans. Sibly and Sibly, 83–​85; see Kienzle, Cistercians, 157.

19 Pierre des Vaux-​de-​Cernay, Hystoria Albigensis, § 8, ed. Guébin and Lyon, 1.7, trans. Sibly and Sibly, 9; later the citizens of Toulouse (Tolosani) are dolosani, dealers in deceit, and Count Raymond, comes Tholosanus, immo dicamus melius “dolosanus” (§ 68).

20 Pierre des Vaux-​de-​Cernay, Hystoria Albigensis, § 359, ed. Guébin and Lyon, 2.57. The city of Toulouse que ab antiquo fuit et erat receptaculum hereticorum et sentina, Pierre des Vaux-​de-​ Cernay, Hystoria Albigensis, § 420, ed. Guébin and Lyon, 2.112. 21 Macé, Les comtes de Toulouse, 277.

134

134

Philippe Buc

Writing more than half a century after the conflict, right after the Toulousain had been incorporated into the French crown, Guilhem of Puylaurens had less reasons to tar the region, and to essentialize it into a unitary entity. A good Catholic, he sought to explain the spread of Catharism but also the Crusaders’ inability to conquer the south easily. More even-​handedly than the Cistercian Pierre, Guilhem expresses at the very beginning of his Chronicle the causes of the former: the “unhappy lands” were punished “for the sins of their people … but I do not exclude the negligence of the prelates and princes.”22 Consistent with this nuanced sense of human failings and responsibilities, Guilhem of Puylaurens expressed compassion for the Toulousains after their bloody defeat at the decisive open-​field battle before the walls of Muret (1213), where Pere of Aragon also met death. He mused: “It was pitiful to see and hear the laments of the people of Toulouse as they wept for their dead”; indeed, there was hardly a single house that did not have someone dead to mourn, or a prisoner they believed to be dead. The blame he laid on the King of Aragon.23 Guilhem attributed the difficulty to subject the south to a greed that had made the Crusaders deviate from their religious mission. His verdict on many lords was not tender. Some had deservedly found death because they had neglected the fight against heresy, preferring to go to war for domains and possessions. They “fell from their status, and God, offended, was angry at them.” Of two Crusader lords, who ended up beheaded, and these heads displayed on stakes, Guilhem comments, in the words of John’s Revelation (cf. Apoc. 3.16): “They were neither hot nor cold, and so since they were lukewarm, the Lord began to vomit them out of His mouth, and expel them for the land that they had acquired through His help.”24 Finally, for Guilhem, unlike for the Cistercian Pierre, Simon de Montfort’s death was not a martyrdom whose responsibility fell on the shoulders of the evil Toulousain heretics; rather, this just man had fallen to punish the less than perfect Crusaders, his followers. Nuanced, Guilhem of Puylaurens refused to pit narratively two groups against one another, one essentially sinful (and resisting both conversion and just punishment), and one essentially saintly (and fighting God’s war in the negotium pacis et fidis).25 However Guilhem was still a clerk, believing in good and evil. In the early years of his mission, Fulk, bishop of Toulouse, had established a so-​called “white” confraternity in the city, to combat heresy and usury.26 In reaction, a second, “black” confraternity had come into being. Violent affrays, with banners, weapons, and horses in plate armour, between “whites” and “blacks” split the town. Citing (but elaborating with the authoritative Gloss on) Christ’s obscure saying in Matt. 10.34 (“I did not come to bring peace but a sword”), 22 Guillaume de Puylaurens, Chronique, prologue, ed. Duvernoy, 23, trans. Sibly and Sibly, 7. 23 Guillaume de Puylaurens, Chronique 21, ed. Duvernoy, 84–​86, trans. Sibly and Sibly, 49.

24 Guillaume de Puylaurens, Chronique 31, ed. Duvernoy, 110.

25 This comparison is subtly developed in Karina Müller Mares, “Das tückische Böse: Gottes Hand & die Konstruktion des inneren Feindes im Albigenserkreuzzug,” Universität Wien seminar paper (Winter Term 2017–​2018). 26 Kienzle, Cistercians, 167–​69.

135



Medieval European Civil Wars

135

Guilhem of Puylaurens explained that the Lord had come via His servant, Bishop Fulk, to bring not an evil peace but a good sword to split them (non pacem malam, sed gladium bonum mittere inter eos).27 There was bad peace uniting evil, sinful men; against it, a cleaving sword was good. It is the anonymous poet responsible for the second half of the Canso de la Crozada who intertextually, as it were, answers the Cistercian Pierre’s attacks against Toulouse. In this part of the Canso, which opposes the crusade while avoiding the question of heresy,28 pride in resistance to the invaders is fundamentally associated with the city. As Sharon Kinoshita notes, the poet has the Count of Toulouse use the nos, “we,” when speaking of the shared enterprise of resistance against the French, and he brings together great nobles, Catalans, and citizens of Toulouse of all classes, “knights, bourgeois, and commoners.” And in the work of rebuilding Toulouse’s fortifications, all classes, as well, labour with building tools.29 The city’s patron saints alongside the Virgin Mary protect it. “Toulouse” is the battle cry of those men and women who resist the misguided crusade. Furthermore, the French identity of the Crusader majority,30 positive in the Canso’s first part, inverts its valence. In the mouth of a valiant warrior in the Crusader camp, Robert of Picquigny, the poet puts forward an ethnic self-​critique:31 a Frenchman must by his very nature first conquer; and he goes on conquering till he soars higher than a hawk. And when he is [high] on the wheel [of Fortune], he behaves

27 Guillaume de Puylaurens, Chronique 15, ed. Duvernoy, 66 (my translation); trans. Sibly and Sibly, 35–​37. Duvernoy amusingly states (66n1) that “evil” and “good” were added owing to “le souci de rendre moins choquante la citation [du Nouveau Testament] pour le lecteur non averti.” It comes in fact from the Gloss ad loc., which states: “Pacem mittere: Missum est bellum bonum ut rumperetur pax mala.” The interlinear gloss states “Non ad hoc inter homines veni, ut carnales affectus confirmem, sed spirituali gladio dissociem” (the inter comes from here). Toulouse did not remain long split. The “white” confraternity did participate in the crusade’s siege of Lavaur, but when the Northerners turned against Toulouse, it dissolved itself, and the citizenry stood in the main united against several sieges. See Michel Roquebert, L’épopée cathare, vol. 1 (Toulouse: Privat, 1970), 415, 422–​25, with the Histoire générale de Languedoc, ed. Claude Devic, Jean Vaissette, et al., vol. 5 (Toulouse: Privat, 1892), 584–​86; Pegg, A Most Holy War, 115: In 1211 the citizens wrote to Pere II of Aragon, expressing their fear that they might perpetrate a crimen proditionis, and refusing to reject the count and withdraw from his dominium and fidelitas. They explained their reconciliation: “… super hoc discordias et dissentiones que in civitate nostra et suburbio diu fuerant pacificamus, et divina cooperante gratia totam villam nostram ad unitatem, ita quod numquam fuit melius, reformavimus.” 28 With one exception, after the defeat at Muret, the Toulousains swear oaths to Simon de Montfort and “on proper conditions returned into the Church,” Canso, § 141, ed. Martin-​Chabot, 2.34, trans. Shirley, 71. This hapax could indicate that the anonymous author rewrote a pre-​existing, pro-​ crusade source (perhaps now lost stanzas by Guilhèm of Tudela). 29 Kinoshita, “Uncivil Wars,” 210, 227, 231.

30 See Daniel Power, “Who Went on the Albigensian Crusade?” The English Historical Review 128 (2013): 1047–​85, for the beginnings of a prosopographical investigation based on charters. 31 Here I disagree with the otherwise fine piece by Kinoshita, “Uncivil Wars,” 228–​29, that such internal debates among the “French” signify an erosion of morale among the Crusaders. The Canso merely makes use of enemy self-​critique to validate critique.

136

136

Philippe Buc

with such arrogance that his pride smashes the ladder, breaks it, and tosses it down. And the man himself falls, tumbles and lies flat on the ground. He loses all that he has gained, for he is not a good lord of the land. It was for French pride and for petty deeds that Roland and Oliver died in Spain.32

Intimately identified with its young Count Raymond VII, the city of Toulouse is courageous and pious. The traitors are in the other camp. Those Cistercians who, with Pierre des Vaux-​de-​Cernay and in his chronicle, accuse the southerners of deceit and treason, are the betrayers par excellence. Mirrored, two moments threaten Toulouse. In the first, Simon de Montfort marches against the city (which having surrendered to him is now in full rebellion and has sided with its legitimate count); in the second, which closes the Canso on a cliffhanger, it is France’s crown prince whose arrival the Toulousain resistance expects. For both, the poet musters the image of an evil and bloodthirsty clergy, distorting the Church’s teachings to call for merciless extermination. For both, he invokes the protection of the Lord Jesus and of the saints. The first episode, placed in the fall of 1217, enlists the hated Bertrand, cardinal priest of San Giovanni e Paolo: Count [Simon], said the Cardinal, … as soon as you enter it, have the men hanged and the counts put to torture. Make sure that no one escapes … Never fear that God will require them of you or exact amends for them. But the Cardinal was wrong. For the [God the] King, Who steers [the All] and surveys with a sharp and clear eye, and gave His precious blood to cure from sin, wants to defend Toulouse! He wants to defend Toulouse, the heavenly King, Judge, ruler who sees right and wrong.33

The second episode, announcing yet another siege, in 1219, also involves Cardinal Bertrand: Now may the Virgin’s Son, Who is light and glory, Who gave His precious blood for mercy’s victory, defend reason and justice (Dreitura); may He cause the guilt and faults to fall back on those who plan evil! For now the King of France’s son comes in pride bringing thirty or forty counts and so many troops that no man in this world, no matter his science, can evaluate their thousands and hundreds. For the Cardinal from Rome too, preaching and lecturing, wants that death and slaughter must lead the way, so that in Toulouse and its appurtenances, there shall remain no man, nor living being, [nay,] neither noble lady, noble girl nor pregnant woman, and no other creature, not [even a] suckling child, but all must suffer martyrdom in burning flames. But the Virgin Mary will preserve them from this, she who rights wrongs according to justice, so that innocent blood will not be shed. Let them not be afraid, Saint Sernin [patron of Toulouse] guides them, and God; justice, strength, the young count and the saints will defend Toulouse for them.34 32 Canso, § 192, ed. Martin-​Chabot, 3.68, trans. Shirley, 143 (modified).

33 Canso, § 186–​87, ed. Martin-​Chabot, 2.306–​9, trans. Shirley, 130 (modified).

34 Canso, § 214, ed. Martin-​Chabot, 3.318–​20, trans. Shirley, 193–​94 (modified).

137



Medieval European Civil Wars

137

Clerical hatred for the Toulousains and clerical deceit are a leitmotiv. As one can see in these two passages, cruelty, appanage of the truly valiant but truly wicked Simon de Montfort, characterizes also the crusade’s prelates. In the first of these two episodes, Cardinal Bertrand is painted as addressing the army in terms that unveil him as a hypocrite: My lords, the Spiritual King admonishes all you that the fires of hell are in this town. It is fully a haven of crimes and sins, for there inside it is Raymond, their prince and lord. Whoever attacks this place will be saved before God. Recapture the town, seize every house! Let neither living man nor woman escape alive; let no church, no relics or hospital protect them, for in holy Rome sentence has been given: the sharp sword of death shall touch them. If it is true that I am a good and holy man, worthy and loyal, and that they are guilty, wicked and forsworn, let the killing sword come down on all of them!35

Evidently, Bertrand is by no means “a good and holy man, worthy and loyal,” and the Toulousains are by no means “guilty, wicked and forsworn.” Ultimately, the city will defeat the Crusaders, and Simon de Montfort will be killed by a mangonel’s stone. Perhaps to counter what is hardly ever evoked in the Canso,36 the charge that the southerners are heretics or defend heretics—​an attested fact—​but perhaps as well because many of the Toulousains were Catholics and did not accept the crusade as a legitimate instrument for conversion and correction, the poet thus underlines constantly Toulouse’s heavenly defenders, in an implicit contrast with the falsely pious clerics and monks in Simon de Monfort’s ranks.37 At one point he even puts in the citizens’ mouth a form of layman’s creed (a type of prayer attested already in the popular Chanson d’Antioche, an epic recounting the high point of the First Crusade, whose Occitan version Guilhèm of Tudela, author of the Canso’s first part, mentions as a model).38 Right before our cliffhanger second passage, the Canso lists at length by name the nobles who stand ready to defend Toulouse against France’s “pride.”39 The men so named were possibly the expected audience, virtual or in presence, of the anonymous poet’s sung performance. In the moment, they were the Toulousain—​a tightly knit unit that intertextually but positively inverts Pierre des Vaux-​de-​Cernay’s suspicion that the 35 Canso, § 187, ed. Martin-​Chabot, 3.10–​12, trans. Shirley, 131 (modified). 36 With the exception noted above, note 28.

37 See Aurell, La vielle et l’épée, 51–​55, for false clerics; Aurell, “Foi et perfidie à la croisade albigeoise selon les troubadours,” in Confiance, bonne foi, fidélité: La notion de Fides dans la vie des sociétés médiévales (VIe–​XVe siècles), ed. Wojciech Falkowski and Yves Sassier (Paris: Garnier, 2018), 239–​56. 38 Canso, § 196, ed. Martin-​Chabot, 3.102–​4, trans. Shirley, 150; see Canso § 2, ed. Martin-​Chabot, 1.8. The anonymous author of the second part also follows in this epic tradition. For the layman’s creed, see Philippe Buc, “Evangelical Fundamentalist Fiction and Medieval Crusade Epics,” Cahiers de Recherches Médiévales et Humanistes 37 (2019): 189–​209, with further bibliography. 39 Canso, § 214, ed. Martin-​Chabot, 3.191–​93.

138

138

Philippe Buc

southern heretics constituted a single heretical people.40 But a decade later, Raymond VII would have to consent to a peace agreement, signed in Paris in 1229, that in effect meant the surrender of some of his territorial claims, and that came with the affiance of his daughter and sole heir to the king’s brother. He would rebel again, briefly, in 1241–​1242. In 1271, Raymond’s daughter and her Capetian husband died childless, and their honours fell to the French crown. Lacking a dynasty with local roots, the memory of resistance around Toulouse failed to bear the fruits of a vibrant local identity.41 If there was a “Languedoc” in the Later Middle Ages, argued the regretted Philippe Contamine, it was because the monarchy administered it as a separate unit.42 The crusade and Simon de Montfort were commemorated, but in the official Grandes Chroniques de France—​a source copied and reworked many times over the centuries—​in a section placed right before the Capetian victories at La Roche aux Moines against John Lackland of England and at Bouvines against the German emperor Otto IV and rebellious barons (1214). It thus was a core part of French Capetian memory.43 And when memories of the crusade were revived in the sixteenth century, it was in the context of the French Wars of Religion, and for the ultra-​Catholic cause. It mustered Bishop Fulk’s “white” confraternity as a model for a militant anti-​Protestant crusade, along with the memory of Muret, where the “heretics” (as the anti-​Montfortian coalition was styled) had been defeated alongside King Pere II of Aragon.44 In the actual moment of the crusade, a Toulousain centred on the city of Toulouse had existed, vibrantly. The fate of its counts and of the county, however, meant that for centuries this intensely felt identification went dormant.

Paris

The religious dimension played out differently in the case of Paris during the Armagnac–​ Bourguignon civil war (1407–​1435).45 While this Franco-​French conflict was by no 40 For Pierre, Kurpiewski, “Writing,” 3. See Hystoria Albigensis, ch. 2, § 9, ed. Guébin and Lyon, 8 (with Sibly and Sibly, 12n22): “the people of Toulouse, truly the progeny (genimen) of vipers” (cf. Matt. 23.33).

41 See Jean-​Marie Moeglin, “L’histoire des princes et la cristallisation du sentiment d’identité régionale: du Comté de Toulouse au Landgraviat de Hesse (XIIIe–​XVe siècles), ” in Les princes et l’histoire du XIVe au XVIIIe siècle, ed. Chantal Grell, Werner Paravicini, and Jürgen Voss (Bonn: Bouvier, 1998), 13–​42. 42 Philippe Contamine, “Qu’est-​ce qu’un étranger pour un Français de la fin du Moyen Âge,” in Peuples du Moyen Âge: Problèmes d’identification, ed. Claude Carozzi and Huguette Taviani-​Carozzi (Aix-​en-​Provence: Publications de l’Université de Provence, 1996), 27–​43. 43 Grandes Chroniques, in Recueil des historiens des Gaules et de la France, ed. Léopold Delisle et al., vol. 17 (Paris: Victor Palmé, 1878), 402–​3.

44 “Publication de la croisade faite à Toulouse contre ceux de la nouvelle religion,” in Histoire générale de Languedoc, ed. Claude Devic et al., vol. 12 (1890; rev. ed. Paris: 2004), 888–​89, with Denis Crouzet, Les Guerriers de Dieu: la violence au temps des troubles de religion, vers 1525-​vers 1610, 2 vols. (Seysell: Champ Vallon, 1990), 1:387.

45 Bertrand Schnerb, Les Armagnacs et les Bourguignons: La maudite guerre (Paris: Perrin, 1988), unfortunately without notes; Bernard Guenée, Un meurtre, une société: L’assassinat du duc d’Orléans,

139



Medieval European Civil Wars

139

means a war of religion, the lexicon of religious war coloured partisan chronicling and, likely, the actual fighting. Its prequel had opposed initially the Orléans and Bourgogne (Burgundy) princes of the blood. But in the aftermath of the Duke of Orléans’ assassination in 1407, one faction’s military leadership passed to Bernard d’Armagnac, and after 1418, when the Count of Armagnac was mowed down, to the future Charles VII, last surviving son of the mentally disabled king of France, Charles VI (r. 1380–​1422). The civil war was complicated by the existence of English claims to the crown of France. The line represented by Isabelle of France, wife of England’s Edward II (r. 1307–​1327), and daughter of France’s Philippe le Bel (d. 1314), had been passed over in favour of the Valois cousin (Philippe VI, r. 1328–​1350) of Philippe the Fair’s last son, Charles IV (d. 1328). This second issue was constantly latent, and explains the presence of English troops on the continent even during the long truces that dotted the so-​called Hundred Years War. Even before the renewal of royal English campaigns in 1415, the two French factions, Orléans and Burgundy, had sought English help, and employed English mercenaries. Finally, reformist Parisian groups allied with Burgundy perpetrated two sets of massacres in the city, in 1413 and 1418. Civil war borrowed some of its lexicon from the Crusades. The expectation that the enemy kills “en église et dehors église,”46 without respect for holy places, belonged to the repertoire of evils perpetrated by pagans and in particular Muslims. The anonymous chronicler known as the Bourgeois of Paris indeed constantly assimilates the Armagnacs to the Saracens.47 One example among many: in 1411, north of Paris, the Armagnacs “perpetrated the sort of evils that the Saracens would have committed; they hanged people up by the thumbs or by the feet, others they killed or took for ransom, they raped women (efforçaient femmes), and started fires.”48 The Armagnacs were thus associated with the Muslim enemy, and conversely, they were not truly French. They were “those who called themselves French.”49 The Bourgeois de Paris’s assertion that in the aftermath of the Armagnac faction’s return to power in the capital (1413), the wives 23 novembre 1407 (Paris: Gallimard, 1992). Also a classic, Richard Vaughan, John the Fearless: The Growth of Burgundian Power (Woodbridge: Boydell, 1966).

46 Anonymous, Journal d’un Bourgeois de Paris, § 11 (1410), modernized French ed. Colette Beaune (Paris: Livre de Poche, 1990), 35. For obvious reasons (including this chapter being in English), while also giving the library-​accessible pages in Beaune, I shall privilege the older edition by Alexandre Tuetey, Journal d’un bourgeois de Paris 1405–​1449 (Paris: Champion, 1881), here at 7. 47 The bourgeois was not alone; see inter alia the so-​called Religieux de Saint-​Denis, Michel Pintoin, Chronique XXXV.xi, XXXV.xviii, Chronique du religieux de Saint-​Denys, contenant le règne de Charles VI, de 1380 à 1422, ed. Louis François Bellaguet, 6 vols. (Paris: Imprimerie Royale, 1839–​52), 3.326 (foreign troops), 3.356 (Parisian rebels).

48 Anonymous, Journal (1411), § 16, ed. Tuetey, 11; ed. Beaune, 39; see Shirley, 55. Cf. inter alia Anonymous, Journal (1414), § 100–​2, Tuetey, 54, Beaune, 77–​79; Anonymous, Journal (1417), § 172, Tuetey, 83, Beaune 103; Anonymous, Journal (1422), § 328, Tuetey, 163, Beaune, 177–​78.

49 Anonymous, Journal (1434), § 644, Tuetey, 299; see Beaune, 331.

140

140

Philippe Buc

of the pro-​Burgundian Cabochien leaders were sent to bordellos is also on a continuum with crusade fantasies—​and realities—​about the fate of Christian women at the hand of Muslim victors.50 Civil war writing also borrowed from the religious lexicon in the dimension of martyrdom and retribution. The starved poor cursed openly or in secret nobles and rulers with the words that John’s Revelation attributes to the martyrs: “ ‘Alas, alas, alas, dear sweet God, when shall we have an end to these bitter troubles, of this intolerable life, this accursed war?’ They would often exclaim, O God, vindica sanguinem sanctorum (Apoc. 6.10)! Avenge the blood of the good creatures who are dying undeservedly by [the hand of] these false traitors Armagnacs.’ ”51 Parisian insurrections, spectacular episodes in the civil war, are coloured now and then by Scripture. At the apex of the pro-​Burgundian Parisian uprising known as the révolte cabochienne (1413), its leadership called for a final purge, including of people close to King Charles and the Queen Isabeau of Bavaria.52 On May 12, a delegation forcibly broke into the Hotel Saint-​Pol, residence of the king, all the way to his rooms.53 There, Charles VI and the princes of blood, including the royal crown prince, were forced to listen to the Carmelite friar Eustache de Pavilly, “at the request 50 Anonymous, Journal (1414), § 115, ed. Tuetey, 58, with n1. Also Beaune, 83. For western fantasies, see most recently Simon Barton, Conquerors, Brides, and Concubines: Interfaith Relations and Social Power in Medieval Iberia (Philadelphia: University of Pennsylvania Press, 2015). One troubling sign that sexual violence could occur is in the description of the conquest of Bethlehem and Jerusalem by Saladin, penned by his secretary ʻImād al-​Dīn al-​Iṣfahānī (d. 1201); see the French translation by Henri Massé, `Imâd ad-​Dîn al-​Isfahânî, Conquête de la Syrie et de la Palestine par Saladin—​al-​Fath al-​qussi fî l-​fath al-​qudsî (Paris: Paul Geuthner, 1972), 34, 50. Also Yvonne Friedman, Encounter between Enemies: Captivity and Ransom in the Latin Kingdom of Jerusalem (Leiden: Brill, 2002), 169. 51 Anonymous, Journal (1421), § 325, ed. Tuetey, 163, ed. Beaune, 176, trans. Shirley, 166 (modified).

52 Narrative in Francoise Autrand, Charles VI: La folie du roi (Paris: Fayard, 1986), 470–​94. See Philippe Buc, “Christianity’s stamp: Of hybrids, traitors, false peace, massacres and other horrors,” History and Anthropology 34 (2023): 123–44, which analyzes the fearful rumours propagated in both factions, which bona parte secularized the positive elimination of evil in Revelation and the prophetic books, and mala parte attributed a perverse inversion of the same (a purge of the good) to the adversary. This belief contributed to the remarkable violence in Paris in 1413 and 1418. See also Michael Sizer, “The Calamity of Violence: Reading the Paris Massacres of 1418,” Journal of the Western Society for French History 35 (2007): 19–​39, a fine article, and Emily J. Hutchison, “Knowing One’s Place: Space, Violence and Legitimacy in Early Fifteenth-​Century Paris,” The Medieval History Journal 20 (2017): 38–​88. 53 King Charles VI later (September 18, 1413) protested vigorously about forcible entry, imprisonments, and massacres in town, Lettre Patente now in the Archives Départementales de l’Hérault, Série A1, fols. 334v–​339v. A transcription by Jean-​Claude Toureille is available on the web, at https://​sour​cebo​oks.ford​ham.edu/​fre​nch/​cab​och.asp. The insurgents justified their actions in a letter to Noyon (likely also directed to other towns, May 2, 1413), edited by Félix Bourquelot,

141



Medieval European Civil Wars

141

of the leaders of these civil rebellions (sediciones civiles).”54 Eustache propounded the “fiction of a garden,” where nettles and other useless weeds prevented the good plants from bearing fruits. These—​servants of the king and queen—​were to be weeded out.55 Maître Jean de Troyes then went to the king’s son and designated successor, the Duke of Guyenne, and repeated the fable, explaining that the noxious weeds were “traitors who dwelled in the royal house and had led him by their perverse advice into multiple vices.” These advisors and servants were to be handed over and jailed. We have here straightforwardly a translation, into the sphere of civil conflict, of the Gospel pericope of the “tares” (zizania, Matt. 13.24–​43): Christ sowed his field with good wheat seed; then came Satan who sowed bad weeds; the two grow together. The servants asked their master: Shall we uproot the bad? And the answer was, wait until all is grown. Until the late twelfth century, the pericope had forbidden the execution of heretics, reserving the weeding to Christ at the Last Judgement. Was it not impossible to have clear knowledge in this present age of who was truly guilty, and of who might not repent if spared? But as one can see with the reworking of Peter the Chanter’s still traditional position by the Dominican Postilla (ca. 1230), the prohibition had lapsed, and the execution of evident (identified, refusing to convert, therefore contumacious) heretics was permitted. It was now figured by the End of Times weeding, which Christ and the angels would execute in full knowledge of individuals’ guilt.56 The pericope was well known and much used. “Correspondance entre le corps municipal de la ville de Paris et celui de la ville de Noyon en 1413,” Bibliothèque de l’École des Chartes 7 (1846): 52–​69.

54 I follow here the French vernacular in Jean Juvénal des Ursins, Histoire de Charles VI (s.a. 1413), in Choix de chroniques et mémoires de l’Histoire de France, ed. Jean-​Alexandre Buchon (Paris: Bureau du Panthéon littéraire, 1861), 479b. He draws here on Michel Pintoin, Chronique XXXIV.xii, ed. Bellaguet, 3.40–​42: Ad propositum inter cetera exempla, ineptum reputans hortulanum in horto odorifereo herbas non eradicantem nocivas, que non sinunt odoriferas incrementum suscipere, sic et impunes relinqui, impedientes ne a floribus liliorum prodeuntes virtutum summum culmen optineant, indecentissimum reputavit, addens regem debere optare ut tales velud radices inutiles evellantur. Pintoin is the one who speaks of sediciones civiles.

55 The proceedings were documented immediately by the Paris Parliament Clerk Nicolas de Baye, from direct hearsay, Journal de Nicolas de Baye, Greffier du Parlement de Paris, 1400–​1417, vol. 2, ed. Alexandre Tuetey (Paris: Renouard, 1988), 112–​13: the intruders “after a sermon (collation) made by Master Eustache of Pavilly, a master in theology, from the Order of Our Lady of the Carmel, arguing that one should to take away the good from the bad, and narrating the great problems that the kingdom has since long suffered, requested that there should be handed over to them several lords and ladies and other servants, both from the King and the Queen’s households, and from Mylord the Crown Prince.” 56 Compare Peter the Chanter’s exegesis, edited in Philippe Buc, “Exégèse et violence dans la tradition occidentale,” Annali di Storia moderna e contemporanea 16 (2010): 131–44n3 with the Dominican Postilla directed by Hugh of Saint-​Cher. The latter recycles the Chanter’s distinction between three types of weeding (eradicatio), that is, festinata, dampnosa, suspiciosa: “Nota autem quod prohibet his festinata, dampnosa, [et] festinata eradicatio. Festinata: quia prius debet corrigi. Damnosa: ut cum multitudo sit in causa vel princeps, nisi fuerit manifesta et redundans in iniuriam ecclesie … Suspiciosa: quando latent hęretici, vel de eorum infidelitate dubium est, excludendi non sunt, sed cum contumaces sunt, in cuius figura iudicii non propter suspitionem, sed propter causam

142

142

Philippe Buc

Master Jean Gerson, otherwise a partisan of Orléans and an enemy of disorder in Paris, used it in the traditional way.57 While the nobles seized in the royal suite were merely jailed for a while (and then freed by a counter-​revolution on August 2–​4), bloody weeding there was in the streets and houses of Paris (April and May 1413). “Peace,” for the sake of which one wars, is (so to speak) on the side of one’s side.58 One’s Burgundian friends will make an orderly entry into Paris to chase out the wicked Armagnacs, to the shout of “Notre Dame! Vive le roi et le dauphin et la paix” (Our Lady! Long live the King, the Crown Prince, and Peace).59 The Armagnac enemies, however, will burst in, and here the slogan is “à mort! ville gagnée! vive le roi et le dauphin et le roi d’Angleterre! Tuez tout! Tuez tout!” (Death! City won! Long live the King, the Crown Prince, and the King of England! Kill all! Kill all!).60 The slogan proclaims that the city is to be sacked, with massive massacres. And thus it is not all too surprising that a second recourse to secularized theological tropes was the image of the “paix fourrée.”61 The Cabochiens rejected violently in 1413 attempts to reconcile the feuding princes of the blood, in what would be evidently profitable to the Armagnacs, fearing that they would be sacrificed to that peace. They called it a “pacem … ovinis pellibus circumtectam,” a peace wrapped fully in sheep’s skin.62 The allusion to Matt. 7.15, the arrival of false manifestam [Christus] damnabit.” I use here the incunabula, Biblia latina cum postillis Hugonis de Sancto Caro, 7 vols. (Basel: 1498–​1502), vol. 6, accessed at https://​digi​tal.ub.uni-​dues​seld​orf.de/​ink/​ cont​ent/​pagev​iew/​7869​625. The Postilla subverts Augustinian scepticism as to the possibility of unequivocal knowledge. It provides as such a fantastic fantasy of transparent and certain knowledge, which fits well the ambitions of the Dominicans, the religious order that staffed inquisitions. For early Dominican ideology, see Christine Caldwell Ames, Righteous Persecution: Inquisition, Dominicans, and Christianity in the Middle Ages (Philadelphia: Pennsylvania University Press, 2011). 57 Jean Gerson, “In festo sancti Michaelis Archangeli,” in Oeuvres complètes, ed. Palémon Glorieux, vol. 5 (Paris: Desclée, 1963), 308.

58 For the contemporary debates among the learned on “peace,” triggered by the murder of Louis d’Orléans, see Guenée, Un meurtre, 232–​64.

59 Anonymous, Journal (1418), § 190, ed. Tuetey, 89, ed. Beaune, 109; Enguerrand de Monstrelet, Chronique 189, ed. Louis Claude Douët-​D’Arcq, La Chronique d’Enguerrand de Monstrelet: En deux livres, avec pièces justificatives, 1400–​1444, 6 vols. (Paris: J. Renouard, 1857–​62), 3.259–​62.

60 Anonymous, Journal (1418), § 192, ed. Tuetey, 90, ed. Beaune, 110. The pairing of the king, his crown prince, and the hereditary English enemy is an evident slander; see Beaune’s note 47.

61 I have been unable to consult Denis Clauzel, “La paix fourrée à l’automne du Moyen Age,” in Il n’est trésor au monde que de paix, ed. Isabelle Clauzel (Saint-​Martin-​Boulogne: Cercle d’études en Pays Boulonnais II, 2007), 57–​78. Examples of the expression are in Frédéric Godefroy, Dictionnaire de l’ancienne langue française, et de tous ses dialectes du IXe au XVe siècle, 10 vols. (Paris: F. Vieweg, 1881–​1902), 9:652.

62 Chronique du religieux de Saint-​Denys 34.24, 3.82. See the odd reversible fur in Chronique du religieux de Saint-​Denys 34.29 (to reveal the true nature of the fabric), 3.122: pacemque hanc vulpinis pellibus involutam asserens. See for the context Alfred Coville, Les Cabochiens et l’ordonnance de 1413 (Paris: Hachette, 1888), 340–​42. The expression was likely known in Paris since the dubious peace of Chartres imposed by the murderous Duke of Burgundy on his fearing rivals, sealed by

143



Medieval European Civil Wars

143

prophets, wolves clad in sheep’s clothing, is clear. The biblical verse reads: “Beware of false prophets, who come to you in the clothing of sheep (in vestimentis ovium), but inwardly (intrinsecus) they are ravening wolves.” By the fourteenth century, these wolves were either heretics or falsi fratres, treacherous brethren (or worse, Antichrist himself).63 Saint Jerome had recycled Cicero to criticize a false peace within which war was wrapped up, in a reference to a verse exegetically related, the classic jeremiad by the Prophet Jeremias, “they clamor, peace, peace, and there is no peace” (Ier. 6.14): It is one thing to seek, and another to show off; one thing to trumpet with words concord, and in fact to exact servitude. We too want peace, and not only do we want it, but we call for it. But [we call for] Christ’s peace, the true peace, the peace without enmities; the peace within which no war is wrapped up (involutum); the peace that is not in order to put in one’s subjection enemies, but to join them as friends.64

Yet what did one do with a false peace? One met it with violence, if needed with force and weapons. This was the theological position, mobilized countless times in European conflicts.65 Thus the good called for real concord and friendship within the polis, but could also mandate stasis and civil war.66 Christ had said, “Blessed be the peace-​makers, since they shall be called sons of God” (Matt. 5.9). Charlemagne’s advisor, Alcuin of York (ca. 735–​804), commented as follows: However, this peace is to be preserved with good men who obey God’s commands, not with criminals and unjust people who have a peace among themselves in their [shared] sins. Christ’s peace serves the attainment of eternal salvation; the peace which is in the Devil has as its endpoint eternal damnation.67

much pageant in the capital; see Pierre Cochon, Chronique Normande, ed. Auguste Vallet de Viriville, Chronique de la Pucelle ou Chronique de Cousinot suivie de la Chronique normande de P. Cochon (Paris: Delahays, 1859), 401.

63 See the Dominican commentary on the Bible, under the direction of Hugh of Saint-​Cher, Biblia latina cum postillis Hugonis de Sancto Caro (as note 56 above), ad loc., from https://​digi​tal.ub.uni-​ dues​seld​orf.de/​ink/​cont​ent/​pagev​iew/​7869​580&-​81. 64 I owe this reference to Luc Ferrier. Jerome, Epist. 82, 2, PL 22, col. 736–​37, and Cicero, Orationes 2.6.19: “Nec ego pacem nolo, sed pacis nomine bellum involutum reformido. Qua re si pace frui volumus, bellum gerendum est; si bellum omittimus, pace numquam fruemur.”

65 Jehangir Yezdi Malegam, The Sleep of Behemoth: Disputing Peace and Violence in Medieval Europe, 1000–​1200 (Ithaca: Cornell University Press, 2013), 4–​12, and passim; Philippe Buc, Holy War, Martyrdom, and Terror: Christianity, Violence, and the West (Philadelphia: University of Pennsylvania Press, 2015), 194, 197–​98; earlier, Philippe Buc, The Dangers of Ritual: Between Medieval Texts and Social Scientific Theory (Princeton: Princeton University Press, 2001), 150, 257, and passim. 66 See Philippe Contamine, “Charles VII, les français et la paix, 1420–​1445,” Comptes-​rendus des séances de l’Académie des Inscriptions et Belles-​Lettres 137 (1993): 9–​23, at 11, for the polemics against the treaty of Troyes (1420), citing “L’honneur de la couronne de France”: Quatre libelles contre les Anglais (vers 1413-​vers 1429), ed. N. Pons (Paris: Klincksieck, 1990), 132.

67 Alcuin, Liber de uirtutibus et uitiis ad Widonem comitem vi, PL 101, 617b: “… Sed haec pax cum bonis et Dei praecepta servantibus custodienda est, non cum iniquis et sceleratis, qui pacem inter

144

144

Philippe Buc

The position would be taken up again and again, in the case of civil wars, when one side accused the other of rupturing the peace, an accusation that had to be countered. The Gregorian Herrad could thus answer Bishop Walram, a partisan of Henry IV (r. 1165–​1104), by citing Matthew 10, and adding: “What is this? Why did Christ appoint the sword to peace, declare war to peace? Forsooth, [He did so] in order to destroy the Devil’s peace.”68 In conformity with this principle, the Chancellor of the University of Paris, Jean Gerson, who belonged rather to the anti-​Burgundian camp, could also argue against an earlier peace, concluded in Chartres between the partisans of the murdered Louis d’Orléans and those of his murderer, John, Duke of Burgundy (1409). He wrote: “If the only way to obtain peace is by giving support to falsehood (fausseté) against the faith or good doctrine, one should rather suffer some war or scandal. … And on this Christ said, Non veni pacem mittere sed gladium, understand, in order to support the faith and truth.”69 One of the very foremost experts on late-​medieval France’s political culture, Bernard Guenée (d. 2010), has suggested that on this issue the theologian Jean Gerson reasoned on the basis of axioms that were essentially different from those of his opponents, at their core jurists.70 Still, both Gerson and the defenders of Jean Petit’s Justification of the Duc de Bourgogne (a tract arguing that Louis d’Orléans had been a tyrant and so legally murdered) built on scriptural arguments and on authoritative exegesis of the Bible. One wished for urban concord in Paris, but the counter-​argument was, therefore, that there existed bad civic concord. The Bourgeois de Paris and Jean Juvénal des Ursins, one writing contemporaneously and the other recounting and likely embellishing his father’s deeds, interpreted the Parisian insurrection of 1413, led by the butchers, through different lenses. For one, the proposed peace was a trick; for the other, something that had to be achieved.71 The so-​called Cabochiens (named after the butchers’ leader, Caboche) were willing to fight to prevent the implementation of the peace; the opposite faction was symmetrically ready to use force to impose it. In the final confrontation, dramatically recounted by the then young Jean Juvénal, the various sections of Paris face se habent in peccatis suis. Pax Christi ad salutem proficit sempiternam. Pax quae in diabolo est, ad perpetuam pervenit perditionem.” 68 See Gerd Althoff, “Selig sind, die Verfolgung ausüben”: Päpste und Gewalt im Hochmittelalter (Darmstadt: Wissenschaftliche Buchgesellschaft, 2013), 182, explaining the so-​called “Epistola de causa Heinrici regis”, ed. Ernst Dümmler, Libelli de Lite, vol. 2 (Hannover: Hahn, 1892), 285–​91, at 289. See as well Malegam, Sleep of Behemoth, chap. 2. 69 Cited by Guenée, Un meurtre, 263–​64, from Jean Gerson, Oeuvres complètes, ed. Palémon Glorieux, 10 vols. (Paris: Desclée, 1960–​73), 7:216. 70 Guenée, Un meurtre, 263–​64.

71 Jean Juvénal des Ursins, Histoire de Charles VI, 481–​84; Anonymous, Journal (1413), § 71, ed. Tuetey, 38, ed. Beaune, 62–​64: here it is the commons (le menu commun) in arms who shouts out for peace, despite Caboche’s demonstration that the peace agreed upon in Pontoise was neither to the king’s honor, nor to the Duke of Burgundy’s, not to the advantage of Paris and its inhabitants, but honored instead the Armagnacs, who had so often lied and betrayed their sworn word.

145



Medieval European Civil Wars

145

off near Saint-​Germain l’Auxerrois.72 Spooked by an odd confrontation—​one peacenik artisan, Gervaisot Dyonnis, came up to his personal enemy, the Cabochien leader Jean de Troyes, and said, drawing his sword, “treacherous wretch, this time I shall get you”—​ the Cabochiens suddenly fled. The hero in this episode of Juvénal’s Chronicle was (surprisingly?) his own father, Jean Juvénal the elder, Lord of Traignel (1360–​1431). The older Juvénal made it be known that the gates of Paris were to remain open, so that “all those who want to leave [Paris] leave, and who wants to stay, stay; one sought only peace and [to be in] good love together” (bon amour ensemble).73 An invocation of peace in a moment of violence. We can well imagine that, had the two sides being forced by closed city doors to remain together, the peacenik artisan Gervaisot Dyonnis would have ended up running through Jean de Troyes. What therefore made Parisian identity during this civil war? Was there, indeed, a single identity? The complaints of the anonymous Bourgeois de Paris show well enough that he considered that violence impacted Paris as a whole; it accounted for high taxation, scarcity in food supply, and higher prices. Initially chalked to the Armagnacs, so to insiders and thus traitors to the monarchy, these issues were later and progressively put on the account of the English (and certainly, by 1435, when the Treaty of Arras, 1435, reconciled the Valois King Charles VII and Philippe, Duke of Burgundy, son of the murdered John). In the first phase, the English were present on the continent, but initially seldom blamed for exactions and difficulties; Parisian militias might even ally with them, and Paris supplied the English army during its siege of Orléans in 1429, fighting also to protect English provisioning lines.74 In 1420, the Bourgeois even listed among “our French lords” the king of England, next to the king of France.75 However, one sees a negative discourse emerge; in 1419, the English get the normal label of the Armagnacs, worse than Saracens.76 Progressively as well, recriminations against the initially loved House of Bourgogne increased, since it was felt it no longer worked to ensure peace in and around Paris. In 1422, exasperated, the anonymous chronicler even inveighed against both John the Fearless and his successor and son Philippe that they were slow and negligent in helping Paris, “a city which had loved him [Philippe] so well 72 Jean Juvénal des Ursins, Histoire de Charles VI, 84b: “les prevost des marchands et eschevins, les archers, et arbalestriers de la ville, et tous les cabochiens, estoient assemblés en Greve, de mille à douze cens bien ordonnés, se doutans qu’on ne leur courust sus, prest de se defendre”. 73 Jean Juvénal des Ursins, Histoire de Charles VI, 485a.

74 Anonymous, Journal (1429), § 494–​95, ed. Tuetey, 230–​33, ed. Beaune, 249–​52, with Joseph de Dufort, “The Political Identity in Paris during the Burgundian-​Armagnac Civil War in Le Bourgeois de Paris,” Universität Wien Seminar Paper (Winter Term 2017–​2018), 9. 75 Anonymous, Journal (1420), § 291, ed. Tuetey, 144 (see Beaune, 162, amused by the expression): “… noz signeurs de France, c’est assavoir, [le roy de France], le roy d’Engleterre, les deux roynes, le duc de Bourgongne, le duc Rouge et plusieurs autres signeurs, tant de France que d’ailleurs);” see Shirley, 153. 76 Anonymous, Journal (1419), § 256, ed. Tuetey, 127, ed. Beaune, 144; see Shirley, 140.

146

146

Philippe Buc

and which had suffered so much and still was suffering for him and for his father.”77 His Paris was royalist, but not fundamentally French. It turned French when it became clear, with the Treaty of Arras, that peace would be ensured not by an alliance of Burgundy and England, but by a reconciliation of Burgundy and the erstwhile Armagnac party, now led by Charles VII of France.78 Such was the Bourgeois, likely representative but not fully representative of Paris. He, like for that matter those of his contemporaries who described the massacres of 1413 and 1418, liked to portray unanimity. Initially, the Bourgeois had imagined Paris as a unity despite, but more tellingly also because of a minority of traitors within the city, allies of the Armagnacs. The enemy within testified to the urban group’s identity, since Paris faced this enemy so well. Yet in rare moments the quill had to yield to the reality of discord. When the Bourgeois had to admit to schisms or Parisian excesses in violence, his reflex was to chalk them up to the “petit people,” the lower commons (this despite his real sympathy for the lower classes).79 The written myth of unanimity broke down. There was thus a majority and a minority.80 For all these divisions, and perhaps thanks to them, however, a Parisian tradition of armed resistance, the sort historians associate with the Low Countries’ urban culture, lasted into the following century. Likely as in northern cities borne by memories and trade associations,81 but also by the university, it was activated anew during the ferocious and bloody French Wars of Religion (1562–​1598). Then, ultra-​Catholic Paris was to be defended at all costs,82 even to the point of rebelling against a king not too keen on persecuting the Protestants and rejoicing at his assassination (1589).83 77 Anonymous, Journal (1422), § 331, ed. Tuetey, 165, ed. Beaune, 179; trad. Shirley, 168. The Bourgeois had begun to criticize John’s slowness in acting in 1418; his refusal to confront the Armagnacs had led to a spike in the price of necessities. 78 I draw in this paragraph on de Durfort, “The Political Identity in Paris,” but modulate its chronology.

79 Characteristic is the Clément de Fauquembergue’s report under August 30, 1418, Journal de Clément de Fauquemberge, Greffier du Parlement de Paris, 1417–​1435, ed. Alexandre Tuetey and Henri Lacaille, vol. 1 (Paris: Renouard, 1903), 158–​59, that the June and August 1418 atrocities had been perpetrated by men belonging to the “smaller populace of Paris, [men] of small status.” In reverse, the Bourgeois attributed the forced “bad peace” of 1413 to the same small people (le menu commun), even though it is clear that the bourgeois led the fight to roll back the Cabochien revolution; see the Journal (1413), § 71, ed. Tuetey, 38, ed. Beaune, 64 (as above, note 71). 80 All the more, Paris did not speak for all its near environment. A score of miles away from the capital, the royal abbey of Saint-​Denys’s chronicler, Michel Pintoin, had, through this civil war, staunchly written for royal France, and berated foremost the Burgundian faction, if also occasionally the Armagnacs.

81 Marc Boone, À la recherche d’une modernité civique: La société urbaine des anciens Pays-​Bas au bas Moyen Age (Brussels: Éditions de l’Université de Bruxelles, 2010). 82 Cf. Barbara Diefendorf, Beneath the Cross: Catholics and Huguenots in Sixteenth-​Century Paris (Oxford: Oxford University Press, 1991); Buc, Holy War, 226, 250–​51. 83 Crouzet, Les guerriers de Dieu.

147



Medieval European Civil Wars

147

Prague The so-​called Hussite Revolution witnessed the rebellion of parts of Bohemia and Moravia against the pretender king, Sigismund of Luxemburg, Holy Roman Emperor and King of Hungary, culminating in a series of unsuccessful crusades led by the same ruler against the dissenting Hussites.84 The motivations to fight against Sigismund, heir to his half-​brother Wenzel IV, King of Bohemia (d. 1419), entangled religion and national honour.85 Since the 1380s, an ardent movement of religious reform, initially fostered by Emperor Charles IV of Luxemburg, had led to factionalism within the University of Prague and the clergy of the Czech lands (the Kingdom of Bohemia and the Margraviate of Moravia).86 Even Charles, despite massive investments in making Bohemia’s capital, Prague, a holy city, had not escaped criticism.87 One of the fault lines running through both clerics and university masters split Germans, often more conservative, and Bohemians, usually reform friendly (although some Germans did side with the reformers). The two camps began trading accusations of heresy. This was easier for one side since some reformers drew on the excommunicate John Wyclif’s writings. Wyclif, rabidly hostile to the papacy, called for a poor clergy deprived of all forms of lordship, and opposed indulgences, in particular for the crusade. He even exhorted the English king and the insular nobility to coerce the greedy clerics.88 To these Wyclifite ideas, a circle of priests and university masters added the call to allow the laity to take the Eucharist in utraque specie, that is with both bread and the chalice’s wine (hence the labels “Utraquist” and “Calixtine”). Two of these masters, Jan Hus and Jerome of Prague, condemned as heretics by the Council of Constance, would be burnt at the stake in respectively 1415 and 1416. Reaction led to escalation among the reformers. It began in Prague, when radicals from

84 František Šmahel, Die hussitische Revolution, German tr., 3 vols., MGH Schriften 43:1–​3 (Hannover: Hahn, 2002); Thomas A. Fudge, The Magnificent Ride: The First Reformation in Hussite Bohemia (Aldershot: Ashgate, 1998); earlier works include Frederick Heymann, John Žižka and the Hussite Revolution (Princeton: Princeton University Press, 1955); H. Howard Kaminsky, A History of the Hussite Revolution (Berkeley: University of California Press, 1965).

85 Analysis of the complicated and not constant relationship between Czech “nationalism” and Hussite religious aspirations in František Šmahel, “The Idea of the ‘Nation’ in Hussite Bohemia,” Historica 16 (1969): 143–​247, and Historica 17 (1969): 93–​197. The better term is likely “patriotism,” given the local connotation of the term patria and the potential anachronisms liked to “nationalism.” 86 Olivier Marin, L’archevêque, le maître et le dévôt: Genèses du mouvement réformateur pragois, années 1360–​1419 (Paris: Honoré Champion, 2005).

87 Philippe Buc, “Religion, violence, pouvoir, vers 1050—​vers 1500: doute et contrainte,” in Formes de convivència e la baixa edad mitjana, ed. Flocel Sabaté Curull (Lerida: 2015), 47–​59.

88 The best study of Wyclif’s political impact is Anne Hudson, The Premature Reformation: Wycliffite Ideas and Lollard History (Oxford: Clarendon Press, 1988). See, as well, Michael van Dussen, From England to Bohemia: Heresy and Communication in the Later Middle Ages (Cambridge: Cambridge University Press: 2012).

148

148

Philippe Buc

the “New Town,” the Neustadt, defenestrated magistrates in City Hall (July 30, 1419);89 it was joined by the mobilization of other radicals who streamed toward the capital in armed bands. In 1420, the papacy called for a first crusade against the Hussites—​itself a factor playing for the self-​identification of Czech and good Christian: “the Church raised the cruel cross against all of the faithful in our kingdom with bloody hands.”90 As is well known, the decade-​long wave of crusades, which did divide the Czech lands since they were also inhabited by Christians loyal to Rome, fed the pre-​existing Bohemian identity all the way to the Thirty Years’ War. For Hus and Jerome’s disciple Lawrence of Brezowá (Vavřinec z Březové), author of the key chronicle of the conflict’s first phase,91 the death of the two masters was a product of German hatred for the Czechs (as he names in the vernacular the ethnic group he calls in Latin boemos). The bonfires dishonoured the Czech nation in tarring it, through Jan and Jerome, with heresy. This position was not Vavřinec’s alone. About ten letters of protests, some already before the execution, came from Czech noblemen, condemning the proceedings on these same grounds. To one parchment letter, dated May 12, 1415, were affixed 250 seals.92 The Czech partisans of Jan Hus felt themselves persecuted as if by a vast conspiracy of enemies. My Vienna student Christoph Haill has pointed out how the plurality in ethnic origins of the prison wardens guarding Hus in Constance is echoed by the same Vavřinec’s description of Sigismund’s army: this vast host too is multi-​ethnic. On June 30, 1420, “the king of Hungary, Sigismund, neared Prague’s castle with a strong army, consisting both of Bohemians and of many and diverse nations,” setting up a camp into which “daily there flowed together men from various kingdoms of the world, duchies, provinces and regions to assault the famous and magnificent city of Prague …” The clergy “had promised” indulgences “to destroy the faithful Bohemians of both sexes.”93 The enumeration of “people, tribes and languages belonging to varied nations” then follows—​more than forty are named. One is reminded of the vast multi-​ethnic Muslim army brought by Baligant to relieve Spain against the 89 Howard Kaminsky, “The Prague Insurrection of 30 July 1419,” Medievalia et Humanistica 17 (1966): 106–​26. 90 Manifesto dated April 3, 1420, trans. Fudge, 59, from Archiv Český, ed. František Palacký, vol. 3 (Prague: Kronberger, 1844), 212–​13.

91 Kronika Husitsá, ed. Jaroslav Goll (Fontes Rerum Boemicarum 5; Prague: Knihkupectvi Edvard Valečka, 1893), 327–​534. Bibliographic entry “Vavřinec z Březové,” in Husitské Století, ed. Pavlína Cermanová, Robert Novotný, and Pavel Soukup (Prague: Nakaladatelství Lidové Noviny, 2014), 121 (my thanks to Dr. Alexandra Kaar for this and other references). See Lisa Maria Langmayr, Die Chronik des Laurentius von Březová (master’s thesis, Universität Wien, 2015).

92 See Fudge, Magnificent Ride, 88; Kaminsky, History, 138–​40; Šmahel, Hussitische Revolution, 2:926–​36 (and the reproduction of the sealed parchment letter now in Edinburgh in vol. 1, 13); Šmahel, “Nation [part 1]”, 188–​91, looks at the references to Czech “language” (jazyk, lingua & linqwagium) and Czech “crown” (koruna, corona) in these letters of protest.

93 Kronika Husitsá, 332, 339, 383 (citation), with Christoph Haill, “Notions of Nationalism in Laurentius de Brezowa’s Historia Hussitica,” BA Seminar paper, Universität Wien (Winter Term 2017–​2018), 8.

149



Medieval European Civil Wars

149

crusader-​emperor Charlemagne in the Song of Roland. And no wonder there was a latent reference to the vast world of heathendom in coalition against Christianity: later in his Kronika Husitsá, Vavřinec brings out the point with a biblical allusion, in a paraphrase of Psalm 33: “Wait and see, for the Lord Himself will grant help to those who fear him, and will put you, who trust in a multitude of people (multitudo gencium), to flight, brought low by a smaller army.”94 Unchristian bloodthirstiness Vavřinec attributed to the other, Roman Catholic, imperial side. The Hungarians, especially, raped; the Germans killed and tortured women and children; they burnt alive the true Christians.95 The Hussites did massacre and burn prisoners, but in the Kronika Husitsá at least, only the radical Utraquists (the Taborites) perpetrate such excesses. Some of this recourse to execution by fire was retaliation. For instance, after the Germans had burnt Czech Utraquists outside Prague, the radicals burst into Prague’s town hall, and forced the officials to hand over German prisoners. And so, “sixteen among them were burnt in barrels before the town, in plain sight of the Germans.”96 But burning alive heretics was also the Taborites’ felt religious duty, consonant with the apocalyptic articles the Chronicle reports: the elect were to execute the prophesized retribution “by fire and by the sword and by the seven End of Times calamities described in Ezechiel 39 [39.33–​36; cf. Is. 66.16].”97 Vavřinec considered Taborite excesses a blot on the Czech name, and came close to denouncing them as heretics. Czech violence, though, was fine in his eyes when it came to the extremists’ own extremists. His Kronika Husitsá recounted how the Taborite captain Jan Žižka’s extermination of the antinomian Adamites (the antinomian “Picards”), an episode which Vavřinec closes with a “Glory to You, O Lord.” The chronicler also uses this exclamation to praise steadfast Utraquist martyrs, two priests burnt alive.98 This said in passing, we have here a fine testimony to the medieval belief of the equivalence in merits of, on the one hand, passive martyrdom at the hand of pagans or religious deviants and, on the other hand, the extermination on God’s behalf of the wicked. For Vavřinec, in its majority, Prague was much, much sounder.99 People like him, university masters or capital patrician elites, identified the city with “the head of the 94 Kronika Husitsá, 532. The threat does not stop here, and remains biblical: “and instead of the most authentic chalice of Christ’s blood, which you seek to destroy by persecuting His faithful, soon enough rabid dogs will lick the blood of you cruel beasts.”

95 For the differences between these two groups in Vavřinec, see Haill, “Notions of Nationalism,” 15–​18.

96 Kronika Husitsá, 395–​96 (as noticed by Haill, “Notions of Nationalism,” 12–​13; see before him Šmahel, “Nation [1]‌,” 235 and nn. 130–​31). 97 See Buc, Holy War, 86–​88, with further historiography.

98 Kronika Husitsá, 519–​20, with Langmayr, “Chronik des Laurentius von Březová,” 77–​78, 94, on gloria tibi praising the death of Hussite martyrs and of perverted Adamite heretics. See Šmahel, Hussitische Revolution, 2:1148–​58 for Picards and Adamites. 99 See the fine discussion in Šmahel, “Nation,” of the place of Prague and proto-​nationalism, notably part I, 159–​68, and part II, 102–​8.

150

150

Philippe Buc

Czech Kingdom” from which religious Truth flowed, and politically as the communitas primogenita.100 It could not be denied that there dwelled also in the Bohemian capital a number of radicals. These allies of Tabor were geographically located not in the more uppity town of Altstadt, but in the Neustadt. The presence of radicals could not be denied, but could be papered over. The worse that Vavřinec attributed explicitly to those Neustadt townsmen was the despoiling and expulsion of Germans during Sigismund’s siege of Prague in 1420. In other cases, Vavřinec chalked their participation in misdeeds narratively to the Taborites. The Bourgeois de Paris was not much different, when he minimized Parisian affinities for the House of Orléans, and attributed anti-​Burgundian actions within Paris to Armagnac initiatives. Historically seen, the siege of Prague constituted a highpoint for civic identity. Its successful lifting in late 1420 with the battle before the fortress of the Vyžehrad granted the city a military aura it cashed in at the Diet of Časláv (1421), where it positioned itself at a rank rivalling that of the Hussite higher aristocracy in defying Sigismund.101 Vavřinec’s rendition of the battle before Vyžherad further allows one to see how religious and ethnic identities crossed. Allegedly, the wicked ruler had sent those Catholic Moravian nobles who belonged to his army to certain death, by assigning them the most difficult position on the battlefield. Vavřinec noted how the noble Hussites had tried to save the Moravian nobles from extermination at the hand of the fanatical peasants, and bemoaned the many fallen, mourning the beautiful younger nobles, all the same concluding that “thus few of the barons from the Moravian land, who had attacked the communion under both species, remained alive.”102 Czech national sentiment and Hussite certitude of election and religious orthodoxy co-​existed, but did not mix.103 The Kronika Husitsá breaks off in the year 1421. Thus, Vavřinec could not anticipate either the length of the wars, or their resolution. Yet the resolution corresponded well with his Czech patriotism alloyed with moderate Utraquist fervour. Defeated crusade followed defeated crusade; the moderate Hussites and their radical allies, led until his death by the blind military genius Jan Žižka, kept winning on the battlefield. Raids into the empire turned the target of the Catholic crusade into feared aggressors. Resolution came via the battle of Lipany (1434), which saw the defeat of the radical Hussites at the hand of an alliance of moderate Calixtines and Czech Catholics. The defeated included the Taborites and the Orebites, but also men from Prague’s New City; among the victors were the troops of Prague’s Altstadt. Lipany opened the way to the Hussite acceptance of a compromise negotiated with the Catholic Council of Basel (between 1431 and 1436). Prague was already from 1420 the hegemonic leader of an alliance of cities, in rivalry 100 Šmahel, “Nation [1]‌,” 230–​31; Ferdinand Seibt, “Communitas primogenita. Zur Prager Hegemonialpolitik in der hussitischen Revolution,”Historisches Jahrbuch 81 (1962): 80–​100. 101 Šmahel, Hussitische Revolution, 2:1177–​80, with a discussion of the historiography.

102 Kronika Husitsá, 440, as already noticed along so much else by Šmahel, “Nation [1]‌,” 230.

103 As explained for all segments of the Hussite movement, and all Czechs visible in the sources, by Šmahel, “Nation.”

151



Medieval European Civil Wars

151

with the Taborite and Orebite confederations, and quasi a city-​state itself—​but under the leadership of the Old Town’s upper bourgeoisie. At Lipany, Prague’s moderate faction triumphed, the Altstadt against Tabor but also the Neustadt.104 So reconfigured, Prague would in the years around Sigismund’s death (1437) be able to play kingmaker and participate in policy-​making alongside the traditional elite, the Bohemian and Moravian higher nobility.105

Conclusion

In a chapter devoted to cities in late-​medieval Flanders and what is now the Netherlands, Wim Blockmans proposed that an urbanite could identify himself or herself on three levels. A first level of identification was social with the craft guild; the second was local with the urban community, “experienced in innumerable acts”; the third was territorial or national. But, cautions Blockmans, “nationality was a more distant and occasional feeling.” Occasional: “circumstances sometimes dictated that the most obvious identifications with guild and citizenship were overruled by national identity or even a coalition like that of peasants and craftsmen within a part of the principality.”106 War was, one may elaborate without betraying Blockman’s thrust, an event which restructured for the occasion existing structures and their tendential operation. In Toulouse, Bishop Fulk’s white confraternity pitted craft guild interest against craft interest, since the crusade’s reform face included, next to the fight against heresy, struggling against vicious usury. The city united against the crusade’s exterminatory face, and united with neighbouring regions and lordships. This became, for a while, the Toulousain. Prague was the major military and intellectual power in the Hussite wars, and competed with Tabor for the status of the most essentially Czech and apostolic community. Notions of Czech election pre-​existed the Crusades and counter-​crusade, but were reinforced by Catholic persecution. This had been since earliest Christianity the normal reaction to the use of coercion in matters of religion.107 Yet the many instances of convergence between “nation” and religious “truth” were met with many, if not more, disjunctions. František Šmahel’s verdict is elegant and eloquent: “[the most varied groups] all wished to unify the ‘nation’ through their faith and no other. It was, however, just at this level [true faith], which was of the greatest importance for contemporaries, that the ‘nation’ was deeply disintegrated.”108 If there was integration, it was in and for Prague. Paris, if the Bourgeois is representative, pretended to be united despite the constant existence 104 See Seibt, “Communitas,”, 84.

105 Šmahel, Hussitische Revolution, 3:1739–​44.

106 Wim Blockmans, “The Impact of Cities in State Formation: Three Contrasting Territories in the Low Countries, 1300–​1500,” in Resistance, Representation, and Community, ed. Peter Blickle (Oxford: Clarendon Press, 1997), 256–​71, at 265.

107 Wilhelm Kamlah, Christentum und Geschichtlichkeit: Untersuchungen zur Entstehung des Christentums und zu Augustins “Bürgerschaft Gottes” (Stuttgart: Kohlhammar, 1951), 76. 108 Šmahel, “Nation [2]‌,” 96–​97.

152

152

Philippe Buc

in its midst of factions considered treacherous, and therefore denied status as insiders. Its identification with France was limited, despite its great affection for the disabled King Charles VI. All in all, Paris’s love and loyalties went to those political actors best able, at a low cost in taxation, to ensure the security and provisioning of the city and its contado. The so-​called Journal is a long litany of laments over exactions and price hikes. Rather than making Paris more French, therefore, the civil wars—​those of the fifteenth century’s first half as well as those of the sixteenth-​century Wars of Religion—​made Paris more … Parisian!

153

Chapter 9

THE CRUSADES AND FRENCH POLITICAL IDENTITY IN THE THIRTEENTH-​CENTURY MEDITERRANEAN Gregory Lippiatt The association between

the French and the Crusades is one familiar to modern historiography. Chateaubriand and Michaud epitomize the nineteenth-​century romantic appropriation of the crusading past, anachronistically casting the medieval movement as an early example (and justification) of contemporary French colonialism. The Salle des croisades at Versailles, the invasion of Algeria in 1830, even Sykes-​ Picot and the Syrian Mandate: all drew on the allegedly unique French genius of the Crusades.1 In our postcolonial age, this connection remains strong in the popular mind, both within and without France, although it obviously ignores the pan-​European nature of the expeditions, which drew—​among others—​Italian, English, German, Magyar, and Scandinavian participants. But this perception of the French as integral to the Crusades is not simply a modern invention. Since the eleventh century, these “Franks” or “French”—​Franci or Francigene in Latin, from the northern regions of the modern hexagon (especially around the city of Paris), united by a shared langue d’oïl with a high degree of mutual intelligibility—​ presented themselves as specially linked to the defence of Christendom, particularly in the semi-​legendary Matter of France.2 In the Chanson de Roland and its associated epic songs, Charlemagne is represented as an emperor, but one who has much more in common with a Capetian than a Carolingian. He is the emperor of “France dulce,” 1 Christopher Tyerman, The Debate on the Crusades (Manchester: Manchester University Press, 2011), 101–​16, 144–​46, 148–​50. For more recent—​and less nationalistic—​studies on Latin Christian identity in Outremer, see Alan V. Murray, “Ethnic Identity in the Crusader States: The Frankish Race and the Settlement of Outremer,” in Concepts of National Identity in the Middle Ages, ed. Simon N. Forde, Leslie Johnson, and Alan V. Murray (Leeds: School of English, University of Leeds, 1995), 59–​73; Murray, “How Norman Was the Principality of Antioch? Prolegomena to a Study of the Origins of the Nobility of a Crusaders State,” in Family Trees and the Roots of Politics: The Prosopography of Britain and France from the Tenth to the Twelfth Century, ed. Katharine S. B. Keats-​ Rohan (Woodbridge: Boydell Press, 1997), 1–​14; Kevin J. Lewis, The Counts of Tripoli and Lebanon in the Twelfth Century: Sons of Saint-​Gilles (London: Routledge, 2017), 89–​91. 2 For a discussion of the term Franci in eleventh-​ and early twelfth-​century crusade sources, see Marcus G. Bull, “Overlapping and Competing Identities in the Frankish First Crusade,” in Le concile de Clermont et l’appel à la croisade, ed. André Vauchez (Rome: École Française de Rome, 1997), 195–​211. Pace Matthew Gabriele, An Empire of Memory: The Legend of Charlemagne, the Franks, and Jerusalem before the First Crusade (Oxford: Oxford University Press, 2011), 136–​37, Franci and Francigene were not neatly distinct terms for “Franks” and “French,” respectively: Bull, “Overlapping and Competing Identities,” 197n6, 199, 207–​8; and note 7 below.

154

154 Gregory Lippiatt not of Rome or Germany; his men the “Francs de France,” rather than the historical Charlemagne’s Teutonic Franks.3 The opening of the later Voyage de Charlemagne shows the emperor, a resident of Paris and Chartres, signing himself with the cross at Saint-​ Denis before setting out for Jerusalem: the very image of a Capetian king preparing for crusade. Indeed, the Voyage may have been composed in the shadow of the Second Crusade (1145–​1149).4 But even before the call for the First Crusade in 1095, the connection between the French and combat on behalf of Christendom was clear. After his rearguard action against the Saracens of Saragossa, the dying Roland thinks about “dulce France” before committing his soul to God.5 Despite the offensive nature of Charlemagne’s expedition into Spain, Roland views his duty as both the spiritual fight against the infidel and the protection of the honour of his French homeland. France and Christendom may not have been synonymous, but the former embodied, for the anonymous poet at least, the best of the latter and provided the warriors who fought on behalf of her spiritual integrity. The example of Roland’s death in Spain foreshadows a further important dimension of this French identity that is often overshadowed by historiographical interest in the twelfth-​century Holy Land: the diaspora created by crusading encompassed a much wider geographical space than Syria. In fact, initiatives under the banner of the cross established a cultural France d’outre-​mer across the northern Mediterranean from Narbonne to Ascalon in the thirteenth century.

Frenchness and Crusader Culture

The image of the French as standard bearers of holy war was immediately taken up by contemporaries writing about the Crusades. Nearly a decade after the First Crusade, Robert of Rheims would echo—​in Latin—​the tropes of the vernacular chansons de geste in his Historia iherosolimitana, including the special destiny of the French in achieving God’s work on earth: in this case, liberating Jerusalem from the Muslims.6 The anonymous Gesta Francorum and Raymond of Aguilers’s Historia Francorum, among the earliest eyewitness sources for the First Crusade, enshrine this association in their titles, though these are unlikely to be contemporary with their contents.7 The Gesta in 3 La Chanson de Roland, ii, xii, ed. Frederick Whitehead (Oxford: Blackwell, 1957), 1, 6; Stefan Vander Elst, The Knight, the Cross, and the Song: Crusade Propaganda and Chivalric Literature, 1100–​1400 (Philadelphia: University of Pennsylvania Press, 2017), 21. Charlemagne’s French identity would likewise be stressed by the thirteenth-​century French translator of William of Tyre: “L’Estoire de Eracles empereur et la conqueste de la terre d’Outremer,” in Recueil des historiens des croisades: Historiens occidentaux, vol. 1 (Paris: Imprimerie royale, 1844), 14; below, pp. 157–58.

4 Il Voyage de Charlemagne, ed. Guido Favati (Bologna, 1965), 132, 198; Theodor Heinermann, “Zeit und Sinn der Karlsreise,” Zeitschrift für romanische Philologie 56 (1936): 550–​59. 5 Chanson de Roland, clxxvi, p. 70.

6 Vander Elst, Knight, Cross, and Song, 55–​56, 63–​64, 68–​70. Cf. Gabriele, Empire of Memory.

7 Samu Niskanen, “The Origins of the Gesta Francorum and Two Related Texts: Their Textual and Literary Character,” Sacris erudiri 51 (2012): 301–​2. Raymond seems to distinguish between

155

The Crusades and French Political Identity

155

particular draws on the chansons de geste for its models of the Christian—​and French—​ warrior.8 Neither of the authors of these works were themselves “French”: the author of the Gesta was most likely an Apulian follower of Bohemond of Taranto; he may have known French from the Siculo-​Norman court, but he seems to have been a member of the petty aristocracy and struggles to translate certain French words into Latin correctly.9 Raymond of Aguilers was the chaplain to the count of Saint-​Gilles and took the cross as a canon of the cathedral of Le Puy in the county of Velay; he could probably speak French, as he was educated in northern Burgundy, but his culture was more likely Auvergnat, his daily speech a dialect of langue d’oc.10 Nevertheless, these writers saw “Frankishness” or French identity—​employed in the chansons de geste with a more limited value than had been the case in the historical period they describe—​as categories paradoxically wider than their classical or Carolingian senses, capable of encompassing Christian warriors like themselves and their compatriots. Such a liberal interpretation of Frenchness is less surprising when one considers the culture of European courts. Throughout the twelfth and thirteenth centuries, French was spoken by elites from Scotland to Germany to Lombardy.11 Like English today—​ though with a more pronounced class element—​French was the literal lingua franca used by lay elites not only to communicate across national boundaries, but also within individual curial cultures. For example, when the Ibelin barons of Jerusalem wrote to Emperor Frederick II in June 1241, proposing Simon of Montfort, sixth earl of Leicester, as imperial bailiff for the minority of Frederick’s son, King Conrad II, they wrote in French, presumably expecting this shared medium to better facilitate comprehension at the imperial court than Latin.12 Those participants in the First and subsequent Crusades who describe their fellows as French are simultaneously echoing the chansons de geste while reflecting the linguistic reality among the royal and baronial leadership. However, in many places this Francophonia went beyond mere court language. Just as Finnish nobles spoke Swedish or the British royal family spoke German into the twentieth century, many elite societies outside of France considered themselves linguistically—​and therefore, to some extent, culturally—​French, not by affection but by birth. Rather than the German of their distant descendants, English royals spoke French Francigene (French) and Franci (in his usage, Franks or the Crusaders generally), but also admits that among the host, the former was applied to the French and the Provençals together: Raymond of Aguilers, Liber, ed. John H. Hill and Laurita L. Hill (Paris: Geuthner, 1969), 52. 8 Vander Elst, Knight, Cross, and Song, 44–​45, 48–​49.

9 Gesta Francorum et aliorum Hierosolimitanorum, ed. Rosalind Hill (Oxford: Oxford University Press, 1962), xi–​xii. Cf. Murray, “How Norman Was Antioch?” 14. 10 Bull, “Overlapping and Competing Identities,” 198.

11 Significantly for our purposes, however, this was not the case in the Midi, or south of modern France.

12 Manners and Household Expenses of England in the Thirteenth and Fifteenth Centuries, ed. Thomas H. Turner (London: Shakespeare Press, 1841), xixn1. I am grateful to Prof. Nicholas Vincent for bringing this to my attention.

156

156 Gregory Lippiatt as their mother tongue through the mid-​thirteenth century. French culture became more pronounced as one approached the Capetian orbit: in Flanders, Flemish—​like English across the Channel—​remained the language of the majority of the population, but the nobility—​not only the comital court—​spoke French from the cradle.13 The extension of French identity beyond the Seine valley was not only an invention of the historians of the First Crusade. Nor was this linguistic affinity unique to western Europe. The Crusades that attacked Syria (which became known under the Franks as Outremer, “oversea”), Romania (modern Greece and Turkey), the Midi (modern southern France), and Sicily (both the island and modern southern Italy) were not transitory phenomena. When the armies returned home, there were always some few who—​zealous for the Lord’s work, their own gain, or both—​stayed on as rulers in new principalities; their numbers were often later increased by further settlers from Europe. Despite the polyglot constitution of the Crusades themselves, those who remained in the conquered territories and their descendants formed ruling elites that once again spoke French and sometimes used it in administrative documents. This is not to say, as Michaud might have done, that these societies were proto-​colonies; they clearly had their own political, social, and economic structures and traditions (save in the Midi, where the Crusader conquests were in fact incorporated into the royal domain over the course of the thirteenth century). However, they were part of a wider Francophone community, culturally linked with France and each other. In the early fourteenth century, the author of the Chronique de Morée claims that Prince William II of Morea, hearing of the victory of Charles of Anjou over King Manfred of Sicily at Benevento, “demena grant joie pour ce que la langue et la seignorie estoit de France auces aprochié vers lui et vers son pays.”14 Around the same time, the Catalan Raymond Muntaner notes that the knights of the Morea, before their extermination in 1311 at the hands of his fellow mercenaries, spoke French to rival that of Acre or, according to a version edited in the sixteenth century, of Paris itself.15 This common means of communication reinforced the identification found in their shared literature, in which they imagined themselves as heirs of Charlemagne. And not only of him. William the Breton, court historian to the French king Philip II Augustus, would imagine Simon V of Montfort, a veteran of the Fourth and leader of the Albigensian Crusades, exhorting his French troops at the desperate battle of 13 Michael T. Clanchy, From Memory to Written Record: England 1066–​1307, 3rd ed. (London: Wiley, 2013), 215; Thérèse de Hemptinne and Walter Prevenier, “La Flandre au Moyen Âge: Un pays de trilinguisme administratif,” in La langue des actes (2005), http://​elec.enc.sorbo​nne.fr/​CID2​003/​ de-​hemp​tinn​e_​pr​even​ier. 14 Chronique de Morée (1204–​1305), ed. Jean Longnon (Paris: Librairie Renouard, 1911), 170; cf. The Chronicle of Morea, ed. John Schmitt (London: Methuen & Company, 1904), 406–​7.

15 David Jacoby, “Knightly Values and Class Consciousness in the Crusader States of the Eastern Mediterranean,” Mediterranean Historical Review 1 (1986): 162 and note 25.

157

The Crusades and French Political Identity

157

Muret in 1213 with reminders of their alleged Trojan ancestry.16 But William, though contemporary, viewed the Albigensian Crusade from a distance, as the king kept the entire affair at arm’s length. Closer sources, such as Peter of Vaux-​de-​Cernay’s Hystoria albigensis, do not make much of the Crusaders’ French identity, despite (or perhaps because of) the fact that Frenchmen dominated this crusade more than any other, save the Angevin conquest of Sicily.17 This silence suggests that such dominance was taken for granted. Certainly the southerners saw the Albigensian expedition as a particularly French undertaking: in addition to the invective of the troubadours against the frances, William of Puylaurens, a contemporary southerner sympathetic to the crusade, begins his later history by noting that the Crusaders were recruited especially from France, “que semper consuevit gerere bella Dominica.”18 As with Raymond of Aguilers a century earlier, even observers who were not strictly French themselves imagined a particular connection between that nation and holy war. In the distant eastern Mediterranean, however, this connection with a mythic French past needed to be stressed by the Crusaders themselves. The First Crusade Cycle, a series of late twelfth-​century chansons de geste celebrating the Crusaders who liberated Jerusalem, is a prominent example of this framing of crusade history into a French vernacular context.19 But while these songs likely circulated in Outremer, they were clearly Western creations. The greatest historian of Outremer, Archbishop William II of Tyre, who wrote his Historia rerum in partibus transmarinis gestarum in the late twelfth century, makes little mention of France or the French, despite having studied in Paris and Orléans in his youth. A much more popular French translation and multiple continuations of William’s work into the thirteenth century, however, emphasize the specifically French foundation of the kingdom of Jerusalem by supplying details absent from William’s account.20 This translation was almost certainly composed in France or possibly Flanders, but all of the continuations seem to have been written in Outremer.21 Thus, while the francizing revisions of William’s history are of European origin, by the 16 William the Breton, “Philippidos,” in Œuvres de rigord et de Guillaume le Breton, historiens de Philippe Auguste, ed. Henri-​Francois Delaborde, vol. 2 (Paris: Société de l’Histoire de France, 1885), 234.

17 Philippe Contamine, “Le jeudi de Muret (12 septembre 1213), le dimanche de Bouvines (27 juillet 1214): Deux “journées” qui ont “fait la France”?,” in La croisade albigeoise, ed. Michel Roquebert (Carcassonne: Nouvelles Editions Loubatières, 2004), 117–​18. 18 William of Puylaurens, Chronica, ed. and trans. Jean Duvernoy (Paris: Éditions du Centre National de la Recherche Scientifique, 1966), 54.

19 The Old French Crusade Cycle, ed. Jan A. Nelson and Emanuel J. Mickel, vols. 4–6 (London: Tuscaloosa, 1977–​2003).

20 William II of Tyre, “Historia rerum in partibus transmarinis gestarum,” in Historiens des croisades, vol. 1; Peter W. Edbury and John G. Rowe, William of Tyre: Historian of the Latin East (Cambridge: Cambridge University Press, 1988), 15; “Estoire de Eracles,” in Historiens des croisades, i–​ii; Philip Handyside, The Old French William of Tyre (Leiden: Brill, 2015), 55–​58, 62–​67, 120. 21 Handyside, Old French William, 55, 58, 73–​75; Peter W. Edbury, “The French Translation of William of Tyre’s Historia: The Manuscript Tradition,” Crusades 6 (2007): 94.

158

158 Gregory Lippiatt thirteenth century they were circulating among the Franks of the Levant, who embraced this French account of their origins. Further evidence for this adoption of French identity may be found in legal treatises. Legal writing—​a genre that came into vogue in both Outremer and France in the thirteenth century—​often provided its own histories in order to justify its authority. This was especially true in Outremer, where the memory of the fall of Jerusalem in 1187 led jurists like Philip of Novara in the 1250s to construct a vision of prelapsarian legal purity that had been lost with the Holy City. In his Livre de forme de plait, Philip—​ as his name suggests, a Lombard, but one writing in French and living in French-​ speaking Outremer—​describes the “Letres dou Saint Sepultre,” a collection of legal customs and precedents that formed an evolving body of statutes, stored in the Holy Sepulchre and consulted at need. Peter Edbury believes that this archive never existed; it was invented, by Philip or by someone else. This fiction was, Edbury argues, likely a response to pressure from French-​born outsiders who insisted on the validity of their native customs within the kingdom, which certainly makes sense at the time of Philip’s writing.22 However, it may be that Philip was simply recording received wisdom about the “Letres” that had been created decades earlier. In the second quarter of the thirteenth century, the bailiffs of Frederick II, custodian of the kingdom of Jerusalem in right of his son, had attempted to govern in accordance with the assumptions of imperial Roman law, and stories about the “Letres” may have begun circulating even then to give weight to an independent, customary legal tradition in Outremer. To oversimplify the issue, custom, with all its regional variety, emphasis on precedent, and dependence on local memory, had been characteristic of France, while the prescriptive dictates of Roman law were warmly embraced by ecclesiastical canonists, Italian communes, Provençal nobles, and imperial jurists. All of this went alongside a growing preference across Christendom in the thirteenth century for written statutes over subjective memory.23 Despite their later possible use as a bulwark against French encroachment, the “Letres” may initially have allowed the barons of Outremer to preserve their “French” legal traditions in the face of imperial pressure while sanctifying them with the authority of anachronistic codification. The story of the “Letres” was not limited to the Livre de forme de plait. Another jurist of the kingdom of Jerusalem, John of Ibelin, composed his own treatise, the Livre des assises, about a decade after Philip. Sometime between the completion of John’s work and 1290, a recension was added to the prologue of one of the manuscript traditions, which repeated Philip’s notice concerning the “Letres,” but now integrated it with 22 Philip of Novara, Le livre de forme de plait, ed. Peter W. Edbury (Nicosia: Cyprus Research Centre, 2009), 119; Peter W. Edbury, “Law and Custom in the Latin East: Les letres dou Sepulcre,” Mediterranean History Review 10 (1995): 71–​79. 23 Edbury, “Law and Custom,” 76–​77; Gregory Lippiatt, “Reform and Custom: The Statutes of Pamiers in Early Thirteenth-​Century Christendom,” in Simon de Montfort (c. 1170–​1218): Le croisé, son lignage et son temps, ed. Martin Aurell, Gregory Lippiatt, and Laurent Macé (Turnhout: Brepols, 2020), 65–​66.

159

The Crusades and French Political Identity

159

John’s foundation narrative centred on the veteran of the First Crusade and advocate of the Holy Sepulchre, Godfrey of Bouillon.24 Thus were the “Letres” made not just a legal tradition autonomous from that of the empire but directly associated with one of the greatest knights of the medieval pantheon and—​as imagined in Outremer since at least the 1170s—​a Frenchman.25 This French tradition provided a powerful means of identification with the great heroes of the past and dignified the independent political society of Outremer. As we have seen, that tradition need not be tied directly to France herself. In the fourteenth century, a collection known as the Uxanze de Romania, setting out the legal customs of the Frankish principality of Morea and likely based on an earlier collection, included its own foundation myth. This was an abridged version of the prologue from the Livre des assises (but without the interpolation from the Livre de forme de plait), recounting the establishment by Godfrey of Bouillon and the Crusaders of separate courts for barons and burghers; the provision of courts for Suriens—​Syrians, meaning Semitic-​speaking Christians, a section of John’s prologue obviously irrelevant to the situation in Romania—​is omitted. The rest of the prologue contains a précis of the Fourth Crusade, the conquest of Constantinople, the Crusaders’ request for legal customs from the kingdom of Jerusalem, and finally a confused account of how the Uxanze came to govern the principality of Morea.26 Leaving to one side the fact that almost everything about this account is utter fantasy, one can see that as time went on, such genealogies yielded multiple benefits. For one, the princes of Morea, like John of Ibelin and his later interpolator, could claim the prestige not only of French but of saintly Crusader origin for their legal system. Second, like Philip of Novara, they could point to its independent provenance when under pressure to conform with external structures: no small gain for the principality in the fourteenth century, caught between Siculo-​Angevins, Venetians, Greeks, and the Catalan Company.

Law and French Identity in the Areas of Conquest

Legal texts did not only advertise a common French culture through narrative prologues. Their content also signalled their kinship. Identity might be constructed negatively, by placing the French elite in opposition to the indigenous population. Fulcher of Chartres, 24 John of Ibelin, Le livre des assises, ed. Peter W. Edbury (Leiden: Brill, 2003), 51–​54, 629.

25 Simon John, The Creation of a First Crusade Hero: Godfrey of Bouillon in History, Literature, and Memory, c.1100–​c.1300 (PhD diss., Swansea University, 2012), 166. The German pilgrim John of Würzburg complains that Godfrey’s memory in the kingdom had been appropriated by the French: Saewulf, John of Würzburg, and Theoderic, Peregrinationes tres, ed. Robert B. C. Huygens (Turnhout: Brepols, 1994), 124–​25.

26 John of Ibelin, Livre des assises, 51–​53; Les assises de Romanie, ed. Georges Recoura (Paris: Librairie ancienne H. Champion, 1930), 146–​53; Jean Longnon, “Les assises de Romanie,” Journal des Savants (1953): 22–​23. For dating and lost French precedents to the Uxanze, see David Jacoby, La féodalité en Grèce médiévale: Les ‘Assises de Romanie’, sources, application et diffusion (Paris: Mouton & Co, 1971), 13, 63–​65, 82–​88.

160

160 Gregory Lippiatt writing shortly after the foundation of the kingdom of Jerusalem, famously paints a rosy picture of Frankish integration in their new Levantine environment, and Christopher MacEvitt has shown that there is a degree of truth in his account, at least in the twelfth century.27 However, even before the composition of the thirteenth-​century assizes, the parliament of Nablus in 1120 included miscegenation and sumptuary clauses intended to distinguish Franks from Muslims.28 Over a century later, local Christians were excluded from the government of the kingdom and segregated to separate courts (though the latter is presented by John of Ibelin as a favour granted by one of the kings at the petition of his indigenous subjects). Only Franks might bear witness against other Franks in the High Court. When establishing property boundaries, Philip of Novara records a hierarchy of authority privileging the testimony of Franks, then Syrians, Greeks, any other Christians, and finally Muslims. The penalty for a blow struck against a Syrian or Greek was half that for the same offence against a Frank. At the turn of the century, the Livre au roi, a customary commissioned by King Aimeric to recover pre-​1187 legal traditions and preserve royal power in the kingdom, had already forbidden the alienation of a fief to Syrians.29 In the Midi under Simon of Montfort, linguistic—​and therefore national—​ difference was less pronounced, and therefore less consequential. Many local knights were members of Simon’s court, though offices such as marshal or seneschal were bestowed exclusively upon Frenchmen. The Statutes of Pamiers, a constitution drafted in 1212 by Simon of Montfort to govern his conquests during the Albigensian Crusade, make distinctions with regard to inheritance and marriage discussed below, but those indigenous knights who had always been Catholic were allowed to retain their traditional terms of service; those who were believers in the heretics, however, were held to serve according to the will of the count and his barons.30 Sicily provided a similar story: an imported French aristocracy existed alongside and to some extent integrated with the native elite, Italian nobles performed important roles in the royal court, and there is no evidence of formal discrimination. After the battle of Tagliacozzo, Crusaders received lands confiscated from enemies of the Angevin government, but the key distinction was loyalty to the new (French) regime, not nationality itself.31 No doubt due to greater 27 Fulcher of Chartres, Historia hierosolymitana (1095–​ 1127), ed. Heinrich Hagenmeyer (Heidelberg, 1913), 748; Christopher MacEvitt, The Crusades and the Christian World of the East: Rough Tolerance (Philadelphia: Pennsylvania University Press, 2008). 28 Benjamin Z. Kedar, “On the Origins of the Earliest Laws of Frankish Jerusalem: The Canons of the Council of Nablus, 1120,” Speculum 74 (1999): 333–​34.

29 Philip of Novara, Livre de forme de plait, 80, 132, 148; John of Ibelin, Livre des assises, 54–​55, 167, 671; MacEvitt, Rough Tolerance, 102–​4, 138–​39; Livre au roi, ed. Myriam Greilsammer (Paris: L’Académie des Inscriptions et Belles-​Lettres, 1995), 220. 30 Gregory Lippiatt, Simon V of Montfort and Baronial Government, 1195–​1218 (Oxford: Oxford University Press, 2017), 151–​53, 193–​95; Histoire générale de Languedoc, ed. C. Devic and J. Vaissète, 2nd ed., viii (Toulouse: Édouard Privat, 1879), column 630; below, pp. 161–62. 31 Jean Dunbabin, The French in the Kingdom of Sicily (Cambridge: Cambridge University Press, 2011), 162–​67, 175–​77, 179.

161

The Crusades and French Political Identity

161

confessional and linguistic divides, the situation in the Morea mirrors much more closely that of Outremer than of the Montfortine Midi or Angevin Sicily in according a separate legal status to Greek nobles, or ἄρχοντεϛ. A grant of land made to an ἄρχων was irrevocable, even by the heir of the donor. However, this privilege was included in the same assize recognizing the permanence of gifts to villeins, though only for the donor’s lifetime.32 Thus the Uxanze distinguishes the unfree (who, in the Morea, were all Greek) from Greek lords, while simultaneously associating them in an assize dealing with donations to non-​Franks. The linkage is made explicit in the Uxanze’s treatment of mixed marriages, where a woman who ‘contraza matrimonio cum villan, o cum arcundo’ has no right to her dowry upon the death of her husband.33 Another concession to Greek custom, the equal partition of “fiefs” predating the Frankish conquest among sons and daughters, was even more insidious: such division would likely have disadvantaged the fiefholders in the face of Franco-​Moreot absolute primogeniture.34 Even while recognizing the irrefutable fact of Greek nobility, the Franks were able to proclaim French supremacy in the conquered principality. For positive indicators of French identity across the diaspora, attitudes to marriage and inheritance present perhaps the most accessible examples. While far from uniform, notable similarities emerge when these texts are placed alongside each other. In a compromise between traditional royal prerogatives over the choice of husbands for fief-​holding ladies, growing baronial concerns about disparagement, and ecclesiastical emphasis on the importance of consent in marriage, the Livre au roi stipulates that an enfeoffed widow must choose from three suitors presented by the king. Philip of Novara would extend this procedure to heiresses.35 On the other hand, the Statutes of Pamiers require a comital licence only for those heiresses or widows seeking to marry indigenous (and therefore untrustworthy) lords. They might marry Frenchmen freely, and after ten years they were free to marry without restrictions.36 These both, in different ways, attempt to deal with the problems of an embattled and isolated elite on a permanent war footing, but also reflect a strong concession toward the principle of consent in marriage. A salient point of comparison comes from contemporary England, where Magna Carta promises only that the king will not disparage heiresses: no provision is made for the choice of the actual bride.37 These French examples, on the frontiers—​ as they imagined it—​of Christendom, fighting infidels and heretics, sought to come to terms with baronial independence in marital matters while still ensuring the effective defence of their principalities. 32 Assises de Romanie, 271–​72. 33 Assises de Romanie, 280.

34 Assises de Romanie, 246–​47.

35 Livre au roi, 220; Philip of Novara, Livre de forme de plait, 161.

36 Histoire de Languedoc, vol. 8, column 634; Lippiatt, “Reform and Custom,” 45.

37 David Carpenter, Magna Carta (London: Penguin, 2015), 40.

162

162 Gregory Lippiatt Closely associated with marriage was the issue of inheritance, where broad agreement was visible throughout the French Mediterranean. Usually this was not a case of direct influence, but of shared culture and circumstance. In Outremer, a liegeman or -​woman could not alienate a fief “a nule religion, ni a yglise, ni as gens de coumune, ni a nul Surien, ni a nul home rendable, ni a home qui ne soit chevaler ou d’estraite a chevaler ou a dame.”38 The Statutes of Pamiers restrict the alms from an estate to a fifth “ad consuetudinem et usum francie circa parisius,” and totally forbid them from baronies and fortifications.39 In Romania, the Uxanze does in fact copy the prohibition of the Livre au roi, excising the references—​such as to Syrians—​specific to the kingdom of Jerusalem.40 But most similarities were simply the result of common assumptions: the faithful wife of a knight who forfeits his fief—​for heresy or treason, for example—​would not lose her dowry, whether she lived in Outremer, the Montfortine Midi, Romania, or Champagne.41 This custom may seem an obvious point of justice, but the fact that it was being codified, especially in the vernacular (true of all these cases save the Statutes of Pamiers), was an important trend in the legal culture of the thirteenth century. Primogeniture was a feature of French inheritance custom that was especially well-​suited to the needs of the crusading diaspora. In the Midi before the Albigensian Crusade, fiefs tended to be transmitted in accordance with written, Roman law: though circumstances might vary, the inheritance would be divided equally if the succession were ab intestat. In the Statutes of Pamiers, Simon of Montfort abolished this tradition wholesale, replacing it for both noble and commoner with a system of succession “secundum morem et usum francie circa parisius.” Among other regulations that must be inferred from contemporary references, this “custom and use” reserved the majority—​ probably four-​fifths, as in the statute concerning ecclesiastical alienations mentioned above—​of the estate for the eldest son.42 The case was much the same in Outremer, where the Livre au roi retains a fief for the eldest son, with any younger sons holding smaller appanages from him; this applies, however, only to the fief-​holding elite, not—​ as in the Statutes of Pamiers—​to the population as a whole.43 At first, the kingdom of Jerusalem would also preserve the fief in default of male heirs, transmitting it to the eldest daughter. But around 1171, at the urging of the Crusader Count Stephen I of Sancerre, this was brought into line with French practice by dividing a fief owing the service 38 Livre au roi, 271.

39 Histoire de Languedoc, viii c. 628.

40 Assises de Romanie, 223; Jacoby, Féodalité en Grèce, 43.

41 Livre au roi, 197; Histoire de Languedoc, viii cc. 633–​64; Assises de Romanie, 223; L’Ancien coutumier de Champagne (XIIIe siècle), ed. Paulette Portejoie (Poitiers: P. Oudi, 1956), 110n4.

42 Hélène Débax, La seigneurie collective: Pairs, pariers, paratge: Les coseigneurs du XIe au XIIIe siècle (Rennes: Presses Universitaires de Rennes, 2012), 81–​82; Olivier Martin, Histoire de la coutume de la prévôté et vicomté de Paris, vol. 1 (Paris: E. Leroux, 1922), 28; Pierre Timbal, Un conflit d’annexation au Moyen Âge: L’Application de la coutume de Paris au pays d’Albigeois (Toulouse: Édouard Privat, 1949), 25–​26; Histoire de Languedoc, viii cc. 633, 634.

43 Livre au roi, 231n141.

163

The Crusades and French Political Identity

163

of several knights or sergeants equally among daughters, the younger ones holding from the eldest. This solution for daughters would also be adopted in the principality of Antioch, where it would be described in assizes composed in the first third of the thirteenth century.44 However, at Antioch, male primogeniture was even more strictly interpreted: the entire fief descended to the eldest son, whose only responsibility was to care for the marriage of his sisters. The jurists of the Morea would adhere so closely to the idea of primogeniture as to grant an entire fief to the eldest child, regardless of sex. As a result, the idea that these customs were directly imported from France must be complicated: neither Norman Sicily nor Champagne—​sources of these principalities’ respective founding dynasties—​have records of such stringent primogeniture.45 Instead, primogeniture was variously adapted according to local circumstances and traditions in order to ensure—​at least in theory—​that fiefs would be maintained by men capable of performing personal military service with sufficient concentrated wealth to render it, a necessity for an alien nobility surrounded by potential enemies. The codification of inheritance procedure was important, suggesting as it did an interest in guaranteeing rights and customs. These guarantees were underpinned by the principle of objective justice. The phrase confessus aut conuictus is repeated throughout the Statutes of Pamiers with reference to judgement, not to introduce a doctrine of due process previously alien to the Midi (local charters demonstrate similar concerns), but to reassure Simon of Montfort’s new subjects that these customs will be enforced justly, not abused through arbitrary will. In fact, the statutes went further, promising free justice and the provision of advocates for those too poor to afford them.46 Though less concerned with the practical details of court provision, the Livre au roi similarly justifies its procedures and penalties with the refrain of la raison—​rather than the king or lord—​ juge et comande and refers judgement to the High Court, made up of the entire knightly class.47 In Angevin Sicily, the parliament of San Martino in 1283 likewise recognized the right of fiefholders to trial by peers.48 Again, France (linguistically interpreted) affords parallels: the so-​called “Penal Charter” of Hainaut—​issued by Count Baldwin VI (IX 44 Livre au roi, 231; Philip of Novara, Livre de forme de plait, 141; John of Ibelin, Livre des assises, 763, 766; Martin, Coutume de Paris, vol. 1, 302; Assises d’Antioche, ed. and trans. L. M. Alishan (Venice: Imprimerie arménienne médaillée, 1876), 20.

45 Assises d’Antioche, 18; Assises de Romanie, 178–​79, 222; Jacoby, Féodalité en Grèce, 32; Jacoby, Coutumier de Champagne, 42–​46. Primogeniture in the Morea was, as we have seen, limited to the Franco-​Moreot aristocracy: see above, p. 161.

46 Histoire de Languedoc, viii cc. 628, 629, 630, 632. For southern precedents, see Gregory Lippiatt, The Statutes of Pamiers: Crusading, Power, and Reform in Thirteenth-​Century Christendom (Oxford, forthcoming). 47 Livre au roi, 208–​10 and passim. Usāma ibn Munqidh confirms the principle of judgement by peers in the twelfth-​century kingdom of Jerusalem: Usāma ibn Munqidh, The Book of Contemplation: Islam and the Crusades, trans. Paul M. Cobb (London: Penguin, 2008), 76–​77.

48 Jean Dunbabin, Charles I of Anjou: Power, Kingship and State-​Making in Thirteenth-​Century Europe (Cambridge: Longman, 1998), 110.

164

164 Gregory Lippiatt of Flanders, later Eastern Roman emperor) in July 1200 as he prepared to set out on the Fourth Crusade—​concludes its list of crimes and penalties with the provision that “hec omnia per bonam veritatem comprobanda sunt.”49 By the early thirteenth century, the writing down of traditional customs was seen as an important dimension of good government, especially among the wider French cultural community. A baronial argument one hundred years later illuminates this unified authority of custom and codification. In 1303, the foreign prince of the Morea, Philip I of Piedmont, arrested his chancellor, Benjamin of Kalamata, without due process: as a liegeman, Benjamin’s fief ought to have served as his pledge. According to the French version of the Chronique de Morée, when confronted by his marshal, Nicholas III of Saint-​Omer, over this arbitrary violation of les usances et les coustoumes dou pays, the prince demanded to see these customs. Nicholas responded by drawing his sword and replying, “Veez ci nos coustumes! car par ceste espée conquesterent nostre anciseur cest pays, et par ceste espée deffendons nos franchises et nos usances contre ceaux qui le nous voudront rompre ne amenrier.”50 At first glance this episode seems to undermine the argument of the previous paragraphs: the marshal’s naked sword is a far cry from the written customs requested by Philip and examined elsewhere. Georges Recoura took it as evidence that no written customs existed in the principality at this time, though David Jacoby points to evidence in the Chronique for assizes having been composed by 1276.51 In fact, the existence of assizes is irrelevant to the point of this story. Nicholas’s brandishing of his sword—​like that of John de Warenne before Edward I of England in the 1280s—​is not an admission that no written customs exist, but rather a revelation of the truth that lies behind the written expression of those customs.52 The violence represented by the marshal’s blade is the guarantee of the order of the principality, established by his French crusading ancestors to include the privileges he shares with the chancellor. The assizes are not themselves the law: they—​like the sword—​are a means of communicating a pre-​existing reality that the Franco-​Moreot barons understood to exist outside of the prince’s pleasure. Such ideas about rational lordly restraint were not new. They were deeply conservative in theory: the texts discussed above attempted to preserve traditional legal practices and procedures, whether authentic or imagined, and never claimed novelty. Nor were they exclusive to the French Mediterranean. Though later than most of the examples treated here, customaries were compiled in various regions of France and 49 Charles Faider, Coutumes du pays et comté de Hainaut, vol. 1 (Brussels: Fr. Gobaerts, 1871), 9.

50 Chronique de Morée (1204–​1305), ed. Jean Longnon (Paris: Librairie Renouard, 1911), 339–​41; Assises de Romanie, 160–​61, 179–​80. 51 Assises de Romanie, 39; Jacoby, Féodaltié en Grèce, 63–​65.

52 Walter of Guisborough, The Chronicle of Walter of Guisborough, ed. Harry Rothwell (London: Royal Historical Society, 1957), 216n. d. The uncertainties surrounding the composition of the Chronique de Morée and the fact that the English account of the quo warranto proceedings may be an interpolation make it difficult to establish which story influenced the other. Cf. Clanchy, Memory to Written Record, 41–​44.

165

The Crusades and French Political Identity

165

elsewhere: written law and due process had spread across Europe by the late thirteenth century.53 This was not the product of some French genius: the 1266 crusading conquest of the kingdom of Sicily, with its tradition of highly centralized royal administration, had more of an effect on the legal traditions of France than vice versa.54 It would be difficult even to argue that these trends were wholly “positive” for the history of legal development. Sophisticated legal processes in fact facilitated condemnations for treason in Sicily under King Charles I, and even if one escaped, a determined administration might still convict on a trumped-​up charge in a different jurisdiction, as happened to Count Adolph IV of Acerra in 1293.55 More generally, the emphasis placed on testimony or confession—​and the rejection of trial by ordeal following the Fourth Lateran Council of 1215—​contributed to the emergence of torture in cases where proof was doubtful.56 Nevertheless, the customaries of the French Mediterranean provide a conscious testimony to a common cultural heritage and a shared vision of righteous rulership. Despite the whiggish image of unaccountable, selfish, and anarchic barons, government was as much a concern for the French nobility as it was for the king. In the late twelfth century, a “neogregorian” attitude to reform had emerged, most especially from the alliance between the Cistercians and the schools of Paris, seeking not only the liberty of the Church but also the moral correction of lay behaviour. The election of Pope Innocent III in 1198 placed a neogregorian alumnus of Paris on the throne of Saint Peter and prepared the ground for the Crusades which would settle French barons, recruited by neogregorian preachers, around the Mediterranean.57 The close relationship between Crusader and Church did not end with the fulfillment of the former’s vows: the French translator of William of Tyre notes that after the conquest of Jerusalem, Godfrey of Bouillon established the Holy Sepulchre and the Temple of the Lord “selonc les establissemenz et les us des granz eglises de France.”58 Presumably the translator means that these churches were granted the same liberty and resources as those of his contemporary thirteenth-​century kingdom. But as early as the parliament of Nablus, King Baldwin II and his barons granted, in a classically 53 Dunbabin, French in Sicily, 244–​45.

54 Dunbabin, French in Sicily, 235, 238–​48.

55 Jean Dunbabin, “Treason, Sodomy, and the Fate of Adenolfo IV of Acerra,” Journal of Medieval History 34 (2008): 424; and below, p. 167.

56 Robert I. Moore, The War on Heresy: Faith and Power in Medieval Europe (London: Profile Books, 2012), 301–​2. For the significance of the disappearance of the ordeal, see Robert I. Moore, The Origins of European Descent (London: University of Toronto Press, 1977), 259–​61. 57 These ideas have been explored by John W. Baldwin, Masters, Princes, and Merchants: The Social Views of Peter the Chanter and His Circle, 2 vols. (Princeton: Princeton University Press, 1970); and Philippe Buc, L’Ambiguïté du livre: Prince, pouvoir, et peuple dans les commentaires de la Bible au Moyen Âge (Paris: Beauchesne, 1994). For a case study of baronial reception of the neogregorian programme, see Gregory Lippiatt, “Simon de Montfort, les cisterciens et les écoles: Le contexte intellectuel d’un seigneur croisé, 1187–​1218,” Cahiers de civilisation médiévale 61 (2018): 269–​88. 58 “Estoire de Eracles,” vol. 1, 376.

166

166 Gregory Lippiatt Gregorian manner, tithes throughout the kingdom to the jurisdiction of the Church.59 This was especially remarkable since in these conquered lands, unlike in the West, no real precedents for tithes existed. In Romania, the new Emperor Baldwin I instituted tithes owed to the Roman Church in 1206, an example followed by princes and barons throughout the eastern empire over the next decade.60 In the Midi west of the Rhône, which had previously shown such little enthusiasm for the Gregorian revolution, the Statutes of Pamiers guaranteed the liberty of the Church, demanded the restitution of tithes, and imposed a novel hearth census of 3d. annually, paid to the Holy See.61 Though French barons did not universally subscribe to the neogregorian program, the independence of ecclesiastical revenue was, by the thirteenth century, almost unquestioned. Lay dedication to the neogregorian vision of righteousness also went beyond the endowment of churches and restitution of tithes. For example, Christianity had sought to regulate sex since its earliest days, reserving it within the boundaries of lifelong marriage. At the turn of the thirteenth century, sexual morality among the laity was a prominent theme of neogregorian preachers like Fulk of Neuilly and Robert of Courson. They translated their words into actions: Fulk founded a convent for reformed prostitutes at Saint-​Antoine-​des-​Champs, while Robert, at the council of Paris in 1213, expelled prostitutes from within the city walls.62 The condemnations of prostitution by Fulk and Robert coincided with their preaching and support of the Fourth and Albigensian Crusades, respectively. It is therefore unsurprising to see leaders of these expeditions, such as Baldwin of Flanders or Simon of Montfort, placing an emphasis on marital chastity in their own lives. Baldwin, after his election as emperor in 1204, forbad his barons from spending the night in the imperial palace in Constantinople with anyone other than their wives, while Simon supported Fulk’s foundation at Saint-​ Antoine, anticipated Robert’s measures against prostitutes in the Statutes of Pamiers, 59 Kedar, “Laws of Frankish Jerusalem,” 332; Hans E. Mayer, “The Concordat of Nablus,” Journal of Ecclesiastical History 33 (1982): 539–​41. 60 Jean Richard, “Le paiement des dîmes dans les états des croisés,” Bibliothèque de l’École des chartes 150 (1992): 81–​82.

61 Histoire de Languedoc, viii cc. 626, 627. For enforcement of these provisions, see Lippiatt, Simon of Montfort, 32, 168.

62 James of Vitry, The Historia occidentalis of Jacques de Vitry, ed. John F. Hinnebusch (Fribourg: Saint-​Paul, 1972), 99–​100; Roger of Howden, Chronica, ed. William Stubbs, vol. 4 (London, 1871), 76; Ralph of Coggeshall, Chronicon anglicanum, ed. Joseph Stevenson (London: Longman & co, 1875), 82; Rainier, “Annales,” in MGH SS, vol. 16, 654; “Ex chronico Cluniacensis cœnobii,” in Recueil des historiens des Gaules et de la France, ed. Martin Bouquet and Léopold Delisle, rev. ed., vol. 18 (Paris: Imprimerie royale, 1879), 742; Baldwin, Masters, Princes, and Merchants, 1:37, 2:95n141; Sacrorum conciliorum nova et amplissima collectio, ed. Giovani D. Mansi et al., vol. 22 (Venice: Antonius Zatta, 1778), c. 854.

167

The Crusades and French Political Identity

167

and was reportedly disgusted by the infidelity of his opponent, King Peter II of Aragon, during the Muret campaign of 1213.63 At least some important French Crusaders of the early thirteenth century mirrored the growing ecclesiastical interest in sexual probity. That is obviously not to say that the French—​Crusaders or otherwise—​were more righteous in the bedroom than any other group of men throughout history. In the Voyage de Charlemagne mentioned above, Count Oliver, in fulfillment of a drunken boast, demonstrates French prowess one hundred times in one night on the virgin daughter of the king of Constantinople; his stamina is attributed to divine aid.64 Even where sexual misconduct was condemned, it was often a mask for more mundane motives. The gruesome deaths of the brothers Adolph of Acerra and Henry of Aquino, mentioned above, were brought about not by the official charge of sodomy—​of which their guilt is doubtful—​but the insecurity of King Charles II of Sicily and a persistent curial belief in Adolph’s treachery.65 Similarly, the Livre au roi’s harsh penalties for heiresses who had illicitly lost their virginity are clearly less concerned with neogregorian pastoral priorities and more anxious about the stability of the kingdom’s fiefs: an heiress who had not saved herself for the marriage bed might cast doubt on the legitimacy of her future issue and struggle to find a husband who can render the service owed by her holding.66 Elsewhere, the Livre au roi makes explicit its independence from the ecclesiastical vision of marriage. Should a liegeman abandon his lands and become a Muslim, his marriage would be considered dissolved after a year and a day, and his wife may marry another baron who can provide service for her fief. Though expedient for the defence of the kingdom, this is a contradiction of not only neogregorian but any Christian understanding of the permanence of the married state.67 While the Livre au roi dates from the same generation as Baldwin and Simon, their priorities are markedly different. These Crusaders and the barons of Outremer held a broad conception of a shared French identity, but its expression through neogregorian reform was characteristic of those who set out at the turn of the thirteenth century, distinguishing the foundation of Frankish Romania and the Montfortine Midi from the classical French settlement in the Latin East. 63 Niketas Choniates, Historia, ed. Jan-​Louis van Dieten, vol. 1 (Berlin: De Gruyter, 1975), 597; Histoire de Languedoc, viii, c. 633; Lippiatt, “Simon de Montfort, les cisterciens et les écoles,” 278–​79. 64 Voyage de Charlemagne, 202–​4.

65 Dunbabin, “Treason, Sodomy, and Adenolfo of Acerra,” 427–​32.

66 Livre au roi, 228–​30. The harsh punishments for sexual misconduct found in the much earlier parliament of Nablus, though perhaps less politically minded, are reflections of Byzantine penal codes rather than precursors of the neogregorian movement: Kedar, “Laws of Frankish Jerusalem,” 313–​24, 335. 67 Livre au roi, 200, 203–​4. Cf. John A. Brundage, “Marriage Law in the Latin Kingdom of Jerusalem,” in Outremer: Studies in the History of the Crusading Kingdom of Jerusalem, ed. Benjamin Z. Kedar, Hans E. Mayer, and Raimund C. Smail (Jerusalem: Yad Izhak Ben-​Zvi Institute, 1982), 266–​67.

168

168 Gregory Lippiatt

Conclusion Outremer, the first and pre-​eminent destination of crusading, seems to have adopted a strong French identity only in the thirteenth century, no doubt due in part to the increased insecurity and dependence on constant European intervention for survival in the period after the loss of Jerusalem and most of its territory in 1187. But the networks formed with other French principalities emerging “oversea” as a result of the Fourth, Albigensian, and Sicilian Crusades likely also facilitated the francization of the Latin East. It became easier and more advantageous to be a French—​rather than generally Frankish—​elite abroad as similar aristocracies established themselves around the Mediterranean coastline. But this diaspora was not monolithic, nor was French influence, though predominant, isolated. Each society encountered particular problems and developed appropriate, though divergent, solutions. Settlement in these territories continued to draw immigration from across western Europe. Crusading dynasties adopted local traditions and practices and, despite legal attempts to avoid absorption, intermarried with native Christians. This did not necessarily diminish the elite’s understanding of itself as French. Cultural superiority was still assumed, supported by the dominant position held by France on the European stage. Indeed, “Frenchness” and its associations with the defence of Christendom became a quality to be envied and aspired to. It is significant that of the legal treatises discussed in this chapter, only one group—​that from the thirteenth-​century kingdom of Jerusalem—​ survives in French. The Statutes of Pamiers were composed in Latin, reflective of their early date and perhaps their ecclesiastically influenced context. The Assises d’Antioche were translated into Armenian by Sempad the Constable from a lost French original in the mid-​thirteenth century, a result of the principality’s brief dynastic union and continuing ties with the kingdom of Cilicia.68 The Uxanze de Romania are so called as a result of their exclusive survival in Venetian manuscripts. Jacoby argues for a French original or at least predecessor, but the French element of the Morea declined over the course of the fourteenth century. After the Turkish conquest of the Peloponnese, the Uxanze survived only in the islands and Greek territories where they had been promulgated by the Republic of Saint Mark, thereby preserving and even expanding the French customary tradition.69 And at the very beginning of the assizes remains the foundation myth of Godfrey of Bouillon, that “French” crusading hero and (imagined) lawgiver. This odd juxtaposition—​the Venetian account of an imperial vassal establishing the legal foundation of the kingdom of Jerusalem later transmitted to Constantinople and the Peloponnese; all invented, interpreted, and transmitted through a French tradition—​ demonstrates the power and flexibility of the medieval myth of French identity in the Mediterranean, a myth made reality by the efforts of the Crusader elite to survive and to justify their survival in their new conquests. 68 Assises d’Antioche, xx, 2.

69 Jacoby, Féodaltié en Grèce, 82–​88.

169

Chapter 10

THE SONG–​JURCHEN CONFLICT IN CHINESE INTELLECTUAL HISTORY Shao-​yun Yang In 1120, the Chinese Song dynasty negotiated an alliance with the Jurchen people

of Manchuria, who had lately achieved remarkable military success in a revolt against their overlords, the Inner Mongolian Kitans (or Khitans).* The Song had enjoyed more than a century of peace with the Kitan Liao empire since 1005. But that peace rested on a treaty that required the Song to pay an annual indemnity of silk and silver to the Liao and renounce its claim to the strategic Yan-​Yun region on its northern border, which a previous dynasty had ceded to the Liao in exchange for military aid in 938. A hawkish, irredentist segment of the Song literati elite had always resented the loss of Yan-​Yun. Although a majority of ministers at the Song imperial court felt that war with the Liao was too risky, the chief minister Wang Fu (1079–​1126) and the eunuch general Tong Guan (1054–​1126) finally convinced emperor Zhao Ji (posthumously Huizong, r. 1100–​ 1126) that the Jurchen revolt was a heaven-​sent opportunity to recover Yan-​Yun.1 Under the Song–​Jurchen alliance’s terms, the Song undertook to capture the Liao empire’s Southern capital (Yan, modern Beijing) while the Jurchens overran the Kitan heartland in eastern Inner Mongolia and took the Western capital (Yun, modern Datong). Once the war was over, the Jurchens’ new Jin empire would recognize Song possession of the Yan region (northern Hebei) in return for receiving the same annual indemnity once made to the Liao: 300,000 bolts of silk and 200,000 ounces of silver. The question of who would get the Yun region (northern Shanxi) was left somewhat open, to be subject to further negotiation. It seemed like a relatively good deal for the Song, but things started going wrong almost immediately. A large rebellion broke out in southeast China in late 1120, delaying the planned Song invasion of Yan until late 1122. By that time, the Jurchens had already conquered all four of the Liao empire’s other capitals, including Yun. Next, the Song army badly bungled its assault on the Southern capital, only to see its defenders surrender to a Jurchen army shortly afterwards.2 The Song had clearly failed to fulfill its part of the alliance agreement. Nonetheless, under a new Song–​Jin treaty concluded in 1123, the Jurchens grudgingly agreed to hand over the entire Yan-​Yun region in exchange for financial compensation. Only months * I would like to thank Julia Schneider, Anthony Kaldellis, and Benjamin Ridgway for their helpful comments on earlier versions of this chapter. 1 This account of the origins of the alliance follows Patricia Buckley Ebrey, Emperor Huizong (Cambridge, MA: Harvard University Press, 2014), 373–​91. 2 Ebrey, Emperor Huizong, 395, 408.

170

170 Shao-yun Yang later, however, the Song pushed its luck too far and broke one of the treaty’s terms by giving asylum to a defector: a former Liao official who had rebelled against the Jin and offered to surrender another strategically valuable border prefecture to the Song. Emboldened by growing awareness that the massive and outwardly formidable Song army had serious tactical and organizational weaknesses, the Jurchens used the issue of the defection as a pretext to invade their erstwhile ally in 1125.3 At this point, Huizong’s impulsive and imprudent decision to abdicate the throne to his inexperienced heir apparent, Zhao Huan (Qinzong, r. 1126–​1127), fatally undermined the already ineffectual Song defence. The Jurchens easily took Yan and Yun and carried their assault all the way to the Song capital, Kaifeng. At the beginning of 1127, Qinzong surrendered after the Jurchen siege broke through Kaifeng’s outer walls. The Jurchens looted Kaifeng for three months and then departed with over 14,000 captives, including Huizong and Qinzong and nearly the entire Song imperial clan. Many of the female captives were soon handed out to Jurchen nobles and officers as concubines or slaves. The other captives were taken north into harsh lives of permanent exile in the Jurchens’ Manchurian homeland. By 1131, only some 10 percent of the imperial clan was still living; the rest had died from hunger, exhaustion, disease, abuse, or suicide. Huizong and Qinzong were relatively fortunate, as the former lived on for another four years and the latter for another thirty, but it is doubtful that they found much meaning or dignity in their existence.4 In the years following the fall of Kaifeng, the Jurchen Jin empire conquered most of north China, employing its superiority in cavalry warfare and growing competence in siegecraft to overcome numerous pockets of Song loyalist resistance. But despite mounting a raid across the Yangzi River in 1129–​1130, the Jin lacked both the means and the ambition to extend its territorial control into south China. In the Yangzi delta, therefore, a new Song court established itself under Zhao Gou (Gaozong, r. 1127–​1162), the only one of Huizong’s twenty-​six sons who had escaped the siege of Kaifeng. This “Southern Song” state gradually reasserted authority over the rest of the south, putting down an assortment of rebels and warlords. In 1141, following numerous military and diplomatic twists and turns, the Jin and Southern Song courts came to a peace agreement that fixed their border along the Huai River in the east and the Qin Mountains in the west. The Song–​Jin treaty of 1141 (subsequently ratified in 1142) is infamous in Chinese historical memory for involving formal acceptance of foreign domination. On the advice of his chief minister Qin Hui (or Qin Gui, 1090–​1155), but over the heated objection of many other ministers, Gaozong submitted to the Jin emperor as a vassal and committed to making annual indemnity payments of silk and silver that would officially be termed as tribute. Just as infamously, Gaozong imprisoned and later executed the fervently pro-​ war general Yue Fei (1103–​1142), whom he and Qin Hui accused of using the prestige gained from military successes against the Jurchens to build up an independent power 3 Ebrey, Emperor Huizong, 409–​20.

4 Ebrey, Emperor Huizong, 421–​503.

171

The Song–Jurchen Conflict

171

base that posed a threat to the Song court. In 1140, Yue and his army had purportedly come close to recapturing Kaifeng before being ordered to withdraw so that peace negotiations could begin.

Filial Revanchism as a Pillar of Southern Song Irredentism

The treaty of 1141 maintained peace for twenty years until 1161, when an overambitious Jin emperor personally led an invasion of the Southern Song. Upon suffering defeat by Song naval forces on the Yangzi River, he was promptly assassinated by his own army and replaced with a more sensible rival. Gaozong, who had grown weary of governing after thirty-​five years as emperor, now chose to retire and hand the throne to his adopted heir apparent Zhao Shen (Xiaozong, r. 1162–​1189). The peace party’s dominance at the Southern Song court had waned since Qin Hui’s death in 1155, and Xiaozong himself had irredentist ambitions. The Southern Song therefore launched a counter-​offensive across the Huai River in 1163, but it, too, failed to hold on to its initial gains. Stalemated, the two sides agreed in 1164–​1165 to revised peace terms that favoured the Song, upgrading Xiaozong’s status from a vassal to a junior kinsman and reducing the size of the annual indemnity (henceforth called a gift and not tribute) from 250,000 units to 200,000 units each of silk and silver. Although the most humiliating aspect of the 1141 treaty was thus eliminated, many leading Southern Song thinkers and men of letters were not satisfied and continued, over the next four decades, to demand the reconquest (huifu 恢復) of north China. Modern historians in both China and the Western world have tended to interpret the strength of irredentist sentiment in the Southern Song through the lens of ethnic nationalism and patriotism.5 In fact, it has recently been argued that in the eleventh century, the Song developed a mode of premodern elite nationalism centred on the irredentist issue of Yan-​Yun and the idea that all ethnically Chinese (Hua 華, Xia 夏, or Han 漢) people belonged under Song rule. Nicolas Tackett suggests that both the ill-​ fated Song–​Jurchen alliance and Southern Song irredentism were natural outgrowths of this new national consciousness.6 My reading of the evidence suggests that irredentist discourse in the Southern Song is better understood as driven by an ideological imperative of “filial revanchism,” as distinct from the modern ethnic nationalist version of revanchism based on recovery of territory seen as rightfully belonging to one’s nation. This ideology was complemented by a discourse of Chinese identity that I think is better described as ethnic supremacism than nationalism. The “Confucian” classics studied by the Southern Song literati contained a principle dictating that when a man is unjustly killed, his sons have the filial duty to avenge him 5 See for example Rolf Trauzettel, “Sung Patriotism as a First Step toward Chinese Nationalism,” in Crisis and Prosperity in Sung China, ed. John Winthrop Haeger (Tucson: University of Arizona Press, 1975), 199–​213. 6 Nicolas Tackett, The Origins of the Chinese Nation: Song China and the Forging of an East Asian World Order (Cambridge: Cambridge University Press, 2017).

172

172 Shao-yun Yang by killing those responsible, as “one cannot live under the same sky with the killers of one’s father.” If his sons fail to take revenge for any reason, the duty passes down to their descendants, who can also legitimately take revenge on the killers’ descendants within up to five generations. The influential Gongyang Commentary to the Annals popularized the idea that since a ruler was like a father to his subjects, they, too, had the moral obligation to avenge him; it also claimed that in the case of avenging a ruler, even a hundred generations were not too long—​in other words, the moral duty required of his subjects’ descendants had no expiration date.7 Under the influence of such ideas, many Southern Song literati embraced irredentist warfare rather than peaceful coexistence with the Jurchen Jin empire because their understanding of their moral duty, as loyal subjects of the Song, compelled them to avenge the humiliation and death in captivity of the emperors Huizong and Qinzong.8 In a set of policy essays written in 1175–​1177, for example, Ye Shi (1150–​1223) argued that opportunistically breaking peace treaties with “barbarians” (Yi-​Di 夷狄, a pejorative Chinese term for foreign peoples) was normally not a morally acceptable option for the Chinese, who by relying on “deceit and brute force” risked becoming just as barbaric as their enemies.9 In fact, Ye pinned the blame for the loss of north China on Huizong’s bad faith in allying with the Jurchens, when the Song–​Liao peace treaty instead obliged him to render military aid to the Liao or at least offer sanctuary to the Liao emperor. But, Ye Shi also claimed, the Song dynasty’s duty to avenge Huizong and Qinzong meant that a different set of rules now applied to relations with the Jurchens. The moral duty to “avenge the humiliation of one’s father and elder brother” took precedence over the moral value of trustworthiness and rendered any peace agreement with the Jurchens both “illegitimate” (wuming 無名) and “immoral” (feiyi 非義). It followed that because the Song was duty bound to take revenge on the Jurchens, “the people in charge of making policy have miscalculated by conciliating them on the pretext of their being [mere] barbarians.”10 Ye Shi did not specify whom he meant by “the people in charge of making policy.” It would have been impolitic and dangerous for him to be specific, since the retired Emperor Gaozong (then still living) was the chief offender in this regard. Gaozong is known to have justified his choice of peaceful coexistence with the Jurchen Jin by making one version or another of the following argument: 7 Li Longxian 李隆獻, Fuchou guan de xingcha yu quanshi: Xian Qin liang Han Wei Jin Nanbeichao Sui Tang bian 復仇觀的省察與詮釋:先秦兩漢魏晉南北朝隋唐編 (Taipei: Taida chuban zhongxin, 2012), 26–​34. 8 Fang Cheng-​hua 方震華, “Fuchou dayi yu Nansong houqi duiwai zhengce de zhuanbian” 復仇大 義與南宋後期對外政策的轉變, Zhongyang yanjiuyuan lishi yuyan yanjiusuo jikan 中央研究院歷史 語言研究所集刊 88 (2017): 311–​12. 9 On the idea of Chinese moral superiority and barbarian moral inferiority that served as a foundation of imperial Chinese identity, see Shao-​yun Yang, The Way of the Barbarians: Redrawing Ethnic Boundaries in Tang and Song China (Seattle: University of Washington Press, 2019), 7–​15. 10 Ye Shi 葉適, Ye Shi ji 葉適集 (Beijing: Zhonghua shuju, 1961), 687.

173

The Song–Jurchen Conflict

173

The Central Lands (Zhongguo 中國) [of the Chinese] and the barbarians are like yin and yang. Neither can exist without the other. There has never been a possibility of exterminating the barbarians. If it were possible to exterminate them, the emperors of the Qin, Han, and Tang dynasties would have done so long ago.11

This view of the classical Chinese–​barbarian dichotomy as complementary rather than oppositional may have been derived from arguments made in the 1080s by conservative Song officials who opposed the military expansionism of Huizong’s father, Zhao Xu (Shenzong, r. 1067–​1085).12 Ye Shi believed that it was wrong-​headed to apply it to Song–​Jin relations, since the Jurchens were not simply barbarian neighbours of the Song. Instead, they had committed a great and unforgivable wrong against the Song dynasty that must be avenged. One of Ye Shi’s intellectual rivals, the “Neo-​Confucian” philosopher Zhu Xi (1130–​ 1200), had a more complicated attitude toward filial revanchism. In memorials addressed to Xiaozong in 1162–​1163, Zhu Xi forcefully argued against a return to peace negotiations and reminded the new emperor of his duty to seek filial revenge.13 In 1191, a disciple asked Zhu why past attempts at reconquest had all started strong but then fizzled out. He replied that it was because those in charge had been driven by selfish ambition and not a sincere desire for filial revenge.14 During the last year of his life, however, Zhu Xi expressed the opinion that filial revanchism was now an inadequate basis for rallying subjects of the Southern Song against the Jurchens, and in fact had been inadequate for quite a long time. He believed that the cause of avenging Gaozong’s father and elder brother had still held raw emotional power in the 1130s and could have fuelled enough popular rage to drive a successful reconquest of the north, since the Jurchens responsible for the fall of Kaifeng were still alive. By the time Xiaozong became emperor in 1162, however, the guilty parties were long dead and it was impossible to “drum up the people’s enthusiasm” with calls for revenge. That was why the 1163 counter-​offensive had failed. Zhu Xi felt that even capturing and killing the Jin emperor would now be meaningless as an act of filial revenge, since the current emperor was not responsible for his ancestors’ deeds.15 It is clear from Zhu Xi’s other recorded sayings that he never stopped seeing the reconquest of north China as a legitimate cause, even if he came to believe that revenge 11 Lü Zhong 呂中, Leibian huangchao dashiji jiangyi/​Leibian huangchao zhongxing dashiji jiangyi 類編皇朝大事記講義/​類遍皇朝中興大事記講義 (Shanghai: Shanghai renmin chubanshe, 2014), 683, 688. Note that in this chapter, I have translated Zhongguo as “the Central Lands” rather than as “China” to avoid anachronistic conflation with the modern nation-​state of China (which also uses Zhongguo as its unofficial name). On the origins of the Central Lands concept and its significance to imperial Chinese identity, see Yang, The Way of the Barbarians, 8. 12 See e.g., Fan Zuyu 范祖禹, Tangjian 唐鑒 (Shanghai: Shanghai guji chubanshe, 1984), 79–​82.

13 Zhu Jieren 朱傑人 et al., Zhuzi quanshu 朱子全書 (Shanghai and Hefei: Shanghai guji chubanshe/​ Anhui jiaoyu chubanshe, 2002), vol. 20, 573, 633–​34. 14 Li Jingde 黎靖德, Zhuzi yulei 朱子語類 (Beijing: Zhonghua shuju, 1994), 3196. 15 Li, Zhuzi yulei, 3197–​99.

174

174 Shao-yun Yang no longer had much to do with it.16 In other words, he remained an irredentist despite ceasing to advocate filial revanchism. In the 1180s, Zhu still believed that the Southern Song should renounce its treaty with the Jin and stop sending annual indemnities, but he also advocated maintaining a defensive posture while using reforms to achieve good governance and military strength before mounting an offensive. In the 1190s, he argued that a successful reconquest would take ten to thirty years of preparation and complained that the imperial court was not making the necessary sacrifices. If the court would only divert money from luxuries to the army, he claimed, “then we’d be able to cut off the [Jurchen] caitiffs’ heads and put them on display.”17 In 1198, he lamented that he was likely to die without seeing the reconquest of north China. But he evidently still believed that a reconquest would happen sometime after his death.18 In a conversation recorded in either 1190 or 1199, Zhu Xi revealed that his confidence in a future reconquest was based on a cyclical theory of history. When a disciple asked whether Heaven (tian 天) was a god-​like being who controlled the universe or just a set of impersonal cosmic principles, Zhu opted for the latter and equated Heaven with the core Neo-​Confucian cosmological concepts of li 理 (cosmic principle) and qi 氣 (material force): This is simply how li works. The movement of qi has always gone from flourishing to decline, then from decline back to flourishing. It just keeps going in a cycle like that, so there is never [permanent] decline without [subsequent] flourishing. That is why [it is said that] when [Heaven] has visited extraordinary disaster on the world, it will surely give birth to an extraordinary man [to reverse it].

After quoting a poem about the vicissitudes of history by the eleventh-​ century philosopher Shao Yong (1011–​1077), Zhu Xi made what appears to be a boldly optimistic prediction about the future: A well-​ordered age will surely be followed by an age of disorder, and an age of disorder will surely be followed by a well-​ordered age. The barbarians are [in the end] only barbarians; they will have to return the Central Plains (i.e., north China) to us.19

Six years after Zhu Xi’s death, the powerful Song chief minister Han Tuozhou (1152–​1207) tried and failed to prove himself the “extraordinary man” who would reconquer the north and inaugurate a new “well-​ordered age.” Ironically, Zhu Xi had opposed Han Tuozhou and supported his political rival Zhao Ruyu (1140–​1196). Han retaliated in 1196 with an official proscription of Neo-​Confucian teachings and blacklisting of their adherents. The proscription was eventually lifted in 1202, apparently to regain support from Neo-​Confucian irredentists in preparation for a military expedition against 16 On this I agree with a recent analysis in Fang Cheng-​hua 方震華, “Zhu Xi yu huifu lun—​‘Lichang gaibian shuo de jiantao” 朱熹與恢復論—​“立場改變說”的檢討, Tang Song lishi pinglun 唐宋歷史評 論 5 (2018): 187–​200. 17 Li, Zhuzi yulei, 3200. 18 Li, Zhuzi yulei, 3196. 19 Li, Zhuzi yulei, 5.

175

The Song–Jurchen Conflict

175

the Jin. Other moves to rally irredentist sentiment followed in 1204–​1206, as Han Tuozhou had Yue Fei posthumously honoured with the title of “prince” and conversely stripped Qin Hui of all posthumous honours. In a recent study, the Taiwanese historian Fang Cheng-​ hua has shown that both the honouring of Yue Fei and the vilification of Qin Hui featured rhetoric emphasizing the principle of filial revanchism. As if to prove Zhu Xi’s pessimism about filial revanchism wrong, the propaganda for Han Tuozhou’s northern expedition of 1206 also relied heavily on filial revanchist rhetoric.20 Unfortunately, despite the support of Ye Shi and several other prominent irredentist literati, the expedition failed miserably due to the military unpreparedness about which Zhu Xi had long warned. Han Tuozhou’s credibility was so badly damaged that his political enemies assassinated him and sent his severed head to the Jin emperor in a box as a peace offering. In 1208, the Song and Jin agreed to new peace terms that raised the Jin emperor’s kinship status slightly (from junior uncle to senior uncle) and increased the annual indemnity to an unprecedentedly high 300,000 bolts of silk and 300,000 ounces of silver. As Fang Cheng-​hua has also shown, the failure of the 1206 expedition discredited Han Tuozhou but did not diminish filial revanchism’s appeal to the Southern Song literati, especially the Neo-​Confucians—​although Zhu Xi may well have been right that it had limited resonance with the common people. In 1214, the Southern Song court received word that the Jin emperor, under attack by Chinggis Qan’s (r. 1206–​1227) new Mongol khanate, had fled his capital at Zhongdu (modern Beijing) and moved his court to Kaifeng. Immediately, the calls for filial revenge resumed. The Song court faced strong pressure to weaken the Jin by discontinuing the annual indemnity and to prepare for military action, possibly in coordination with the Mongols. When one official dared suggest that it was wiser to support the faltering Jin and use it as a buffer against the rising Mongols, a group of Imperial University students prostrated themselves outside the imperial palace and indignantly (though unsuccessfully) demanded his execution.21 The Mongols completed their destruction of the Jin in 1234, with some help from a Southern Song expeditionary force. Almost immediately, the short-​lived alliance with the new “barbarian” power broke down and the Mongols began raiding Song territory.22 Nonetheless, literati in the south continued to maintain that the Song had been right to seek the Jin’s downfall at any cost. One can see this from a lecture that the historian Lü Zhong (fl. 1234–​1264) wrote between 1234 and 1247.23 In it, Lü argued that Qin Hui and 20 Fang, “Fuchou dayi,” 313–​17. 21 Fang, “Fuchou dayi,” 321–​23.

22 On the origins and collapse of the Song–​Mongol alliance, see Charles A. Peterson, “Old Illusions and New Realities: Sung Foreign Policy, 1217–​1234,” in China among Equals: The Middle Kingdom and Its Neighbors, 10th-​14th Centuries (Berkeley: University of California Press, 1983), 218–​30. 23 After passing the jinshi civil service examinations at the top of his cohort in 1247, Lü Zhong worked as an instructor at a prefectural school. Soon afterwards, he published his past lectures in two volumes, marketed commercially as reference works for teachers and prospective examination candidates.

176

176 Shao-yun Yang other pro-​peace ministers of the past were no different from barbarians and animals because they failed to see filial revenge as a moral imperative: As long as heaven and earth exist, the three bonds [of ruler–​subject, father–​son, and husband–​wife] will exist … Human beings who do not observe them may wear robes and caps but are merely animals. A state that does not observe them may be the Central Lands of the Chinese but is merely a land of foreign barbarians. That is why “one cannot live under the same sky with the murderers of one’s ruler or father.” That is the supreme foundation of human ethics and the li (cosmic principle) ordained by Heaven … The calamity that our dynasty suffered during [Qinzong’s reign] should have been avenged even if it took ten thousand generations. If not, how could the Way of humanity rest secure for even a day?24

Indeed, Lü Zhong later claimed that his reason for publishing his lectures on Song history was to “explicate the Way of humaneness and moral duty, so that [the people of this age] will understand what makes the Central Lands different from the barbarians, human beings different from animals, and morally superior men (literally “noble men,” junzi 君子) different from morally inferior men (xiaoren 小人).”25 In Lü’s understanding of revanchism, then, the ethnic difference between Chinese and Jurchens was ultimately rooted in moral difference, was equivalent to the difference between human beings and animals, and could only be maintained by exacting filial revenge.

Ethnocentric Moralism and Chinese Supremacism

In a history lecture analyzing the fall of Kaifeng to the Jurchens, Lü Zhong argued that traditional interpretations faulting Huizong’s court for breaking the Song–​Liao peace agreement, establishing diplomatic relations with the Jurchens, and antagonizing the Jurchens by accepting a defector had all failed to identify the root of the problem. Lü’s own interpretation was that the Song proved so vulnerable to a Jurchen invasion because it was already corrupted by moral barbarism on the inside: “Since antiquity, there has never been a state that had no inner barbarians and yet suffered calamity from outer barbarians.” Huizong’s court was dominated by “barbaric morally inferior men,” relations with the Jurchens were handled by Tong Guan, a “barbaric eunuch,” the army was full of cowardly “barbaric soldiers and generals,” and there was a rebellion by “barbaric bandits” in the south. Under such circumstances, even without a barbarian invasion from outside, the Song would have collapsed from the influence of the “barbarians” within.26 Lü Zhong’s interpretation of the Jurchen invasion was not wholly original. Already in 1135, the Neo-​Confucian philosopher Hu Hong (1105–​1161) had sought to convince Gaozong that all barbarian invasions in history stemmed from morally barbaric behaviour by the Chinese government, since “all things in the world respond to their 24 Lü, Leibian huangchao, 432. 25 Lü, Leibian huangchao, 436.

26 Lü, Leibian huangchao, 377–​78 (cf. the reiterations of this argument at 392, 497).

177

The Song–Jurchen Conflict

177

own kind.” In Hu Hong’s case, he laid the blame for the recent disaster on a favourite target of Neo-​Confucian opprobrium: the controversial reform program initiated by chief minister Wang Anshi (1021–​1086) and favoured by emperors from Shenzong to Huizong. Hu argued that if Gaozong could return to traditional Confucian values and reverse the moral decline caused by the reforms, “then the Way of the Central Lands will be restored and the rebellion of the barbarians can be defeated.”27 This belief in a direct causative relationship between moral decline (or barbarism) and barbarian invasion predated the Jurchen invasion and originated with Cheng Yi (1033–​1107), the effective founder of Neo-​Confucian thought.28 It was, in fact, one of the most notable Neo-​Confucian contributions to a distinct discourse that I call “ethnocentric moralism.” For most of imperial Chinese history, the Chinese assumed themselves to be morally superior to all other peoples (“the barbarians”) because of their unique schooling in principles of “ritual propriety and moral duty” (liyi 禮義). This ethnocentric idea can be traced back to Confucian thinkers of the Han dynasty, who typically reinterpreted older passages from the Confucian classics to justify it. Ethnocentric moralist discourse, which first appeared in the ninth century and gained popularity in the late eleventh century, subverted Confucianism’s ethnocentric interpretation of morality (“the Chinese are moral and the barbarians are immoral”) for rhetorical purposes by balancing it with a moralistic interpretation of ethnicity (“to be immoral is to become a barbarian”). For example, Cheng Yi claimed hyperbolically that a single violation of ritual propriety could turn a Chinese person into a barbarian, and a second violation could turn that barbarian into an animal.29 Ethnocentric moralist interpretations of the difference between Chinese and barbarians can be found in the writings of most prominent Neo-​Confucian philosophers in the Southern Song. Intriguingly, Zhu Xi was an exception who seems to have found the idea of Chinese people suddenly turning into barbarians both logically and cosmologically unsound. Zhu’s preference for systematic theorizing about the workings of li and qi may have led him to see the differences between barbarians and Chinese as too significant to be negated by changes in behaviour. Neo-​Confucian thinkers since Cheng Yi and his brother Cheng Hao (1032–​1085) had asserted that all living beings possess the same innately moral nature rooted in li and the universal Way (Dao 道), but human beings are the only living things that have balanced and pure qi. Cheng Hao also implied, however, that barbarians have imbalanced qi and are thus more like animals and plants than human beings.30 A conversation between Zhu Xi and his students, recorded in 1188, shows him building on these theories. Zhu argues that even humans 27 Hu Hong 胡宏, Hu Hong ji 胡宏集 (Beijing: Zhonghua shuju, 1987), 87–​89. See also Yang, The Way of the Barbarians, 131–​32, 150. 28 Yang, The Way of the Barbarians, 130–​31. 29 Yang, The Way of the Barbarians, 119–​22.

30 For the Cheng brothers’ views on human qi, see Cheng Hao and Cheng Yi, Ercheng ji 二程集 (Beijing: Zhonghua shuju, 2004), 4, 54, 122, 198–​99, 211.

178

178 Shao-yun Yang have inborn concentrations of murky or turbid qi that cloud their moral awareness, like a thatched roof blocking out most of the light of the sun or moon. Fortunately, humans have enough intelligence, originating from their unique ability to stand fully upright, to “penetrate” their qi obstruction, achieve a full understanding of li, and thus become fully moral. In contrast, animals are limited by their physical forms to displaying only rudimentary and partial moral qualities at best. Then, almost as an afterthought, Zhu Xi adds, “When it comes to barbarians, they are between human beings and animals, so it will always be difficult for them to change [their immoral behaviour].”31 This statement indicates that Zhu Xi, like Cheng Hao, did not see non-​Chinese people as fully human, even though their physical forms were no different from those of the Chinese. Perhaps he believed that all barbarians had impenetrably murky or turbid qi that could not be overcome even by human-​level intelligence, although he never states this explicitly in any extant source. It seems clear, in any case, that he did not see barbarians as capable of any significant moral improvement, whereas to his mind all Chinese people, no matter how morally benighted, had the human potential to recover their original moral natures through self-​cultivation and attain the perfect moral understanding that Neo-​Confucians called “sagehood” (sheng 聖). It stood to reason, then, that Chinese people could never descend into barbarism simply by behaving immorally, as that would make their moral redemption almost impossible. In 1198, right after Zhu Xi sighed and lamented that he would not live to see the reconquest of north China, one of his students noted provocatively that the now-​deceased Jin emperor Wanyan Yong (Shizong, r. 1161–​1189) was known to have been “committed to practicing humane governance” and had earned the nickname “Little Yao and Shun” from the north Chinese literati who now served the Jin. Yao and Shun were the legendary sage kings of antiquity whom Confucians most revered for their selfless moral character and good governance. Zhu Xi’s first response was to say, “If he could honor and practice the Way of Yao and Shun, then even if he was called a Big Yao and Shun, that would be fine.” But he then gave the matter more thought and said, “How can he have changed his barbarian ways? I suspect he was just talented enough to govern humanely by chance.”32 Zhu Xi’s skeptical comment that Shizong cannot have “changed from his barbarian ways” was essentially correct, but not for the reasons assumed. Rather than being unable to change his ways completely, Shizong was unwilling to do so because of skepticism about the supposed superiority of Chinese civilization. Despite making efforts at encouraging literacy and Confucian learning among the Jurchens, he was also prone to idealizing the Jurchen people’s “old customs” as “pure and honest” and fretting that they were being corrupted morally by Chinese culture in a way that would eventually spell the dynasty’s doom.33 Zhu Xi could not have known about or empathized with Shizong’s concerns; to Zhu’s mind, the only thing that could prevent a barbarian from adopting 31 Li, Zhuzi yulei, 58.

32 Li, Zhuzi yulei, 3196.

33 Toqto’a 脫脫 et al., Jinshi 金史 (Beijing: Zhonghua shuju, 1975), 7:158–​59, 164, 8:191, 89:1980, 1989. Taiwanese historian Hsu Ping-​yu has linked Shizong’s anxiety to a crisis of ethnic solidarity

179

The Song–Jurchen Conflict

179

Chinese norms of morality was an innate inability to do so, not pride in or attachment to his own “uncivilized” traditions. Zhu Xi and other Southern Song Neo-​Confucian thinkers held firmly to a belief in Chinese moral superiority, in part because it constituted the foundation for a relatively new ideology that is best termed “Chinese supremacism.” Extant sources suggest that this ideology first gained prominence in the twelfth century, as a reaction to a world order in which the Chinese Empire could no longer claim supremacy and instead was subject to the supremacy of the Jurchens. To put it simply, Chinese supremacists held that the Chinese, being morally superior by nature, were entitled to dominate the world, whereas barbarians, being immoral by nature, were meant to be dominated by the Chinese for their own good. If the positions were reversed, with morally inferior barbarians holding geopolitical superiority, then civilization would collapse, and the world would descend into anarchy, chaos, and catastrophe. These ideas were legitimated with claims, based on dubious readings of the classics, that Confucius himself had espoused them. They were also something of a departure from earlier Chinese imperial ideology, which had anchored the ruling dynasty’s right to supremacy in the moral superiority of its emperors both past and present (the so-​called Mandate of Heaven), rather than the moral superiority of the Chinese people as a whole. Chinese supremacism was especially important to the Song Neo-​Confucians, who saw familial, social, political, gender, and ethnic hierarchies as integral to cosmic order and civilization. For Southern Song thinkers who had only known a world of Jurchen military dominance, Chinese supremacy over the barbarians seemed like a thing of the remote past, but many believed that civilization’s best hope lay in affirming this ideal’s validity and one day restoring it to reality. As an anonymous Southern Song commentator, writing ca. 1196, put it: Among all things created by heaven and earth, human beings are superior to all others by virtue of their intelligence. Among human beings, however, the Central Lands are superior to the barbarians because they have ritual propriety and moral duty. That is why it is right for the superior to rule over the inferior, while it is wrong for the inferior to infringe on the superior.34

The commentator went on to acknowledge that “when barbarians admired the Central Lands, [Confucius] would promote (jin 進) them [to the level of the Central Lands].” This idea of “promotion” originated from classical exegesis on a historical text attributed to Confucius, the Annals (Chunqiu 春秋), and so did a corresponding and hitherto more influential idea—​often used as a basis for ethnocentric moralist arguments—​that when among the Jurchens and the beginnings of Mongol power on the steppe: Hsu Ping-​yu 徐秉愉, “Jin Shizong shiqi Nüzhen minzu de weiji” 金世宗時期女真民族的危機, Hanxue yanjiu 漢學研究 19 (2001): 249–​79.

34 Hong Benjian 洪本健, Ouyang Xiu shiwen ji jiaojian 歐陽修詩文集校箋 (Shanghai: Shanghai guji chubanshe, 2009), 1564. On the authorship and dating of this comment, see Yang, The Way of the Barbarians, 200n29.

180

180 Shao-yun Yang Chinese rulers behaved like barbarians, Confucius regarded them as barbarians.35 But the commentator then posed a rhetorical question: How can it be acceptable for barbarians to “infringe on the legitimate succession of [Chinese] rulers” even when they have been promoted for admiring and imitating the Chinese? In other words, inferior barbarians might possibly be granted equal status with the Chinese, but it would never be legitimate for them to seek political supremacy by invading the “Central Lands” and setting themselves up as a Chinese-​style imperial dynasty, as the Jurchens had done. The discourse of ethnocentric moralism was useful for making rhetorical assertions that Chinese identity could not exist without ritual propriety and moral duty. But universalizing barbarism into an abstract moral problem, unlimited by ethnic and cultural boundaries, could also open something of a can of worms for adherents to Chinese supremacism: if it was so easy for the Chinese to become barbarians by behaving badly, then on what was their absolute right to supremacy really based? Could the reverse not also be true, with “promoted” barbarians becoming Chinese by acquiring morality? If, as Neo-​Confucianism taught, all human beings had an equal potential to attain moral perfection, then was Confucius merely being bigoted when he (purportedly) regarded barbarians as inferior? Was even the great sage not morally perfect enough to see all human beings as equal? One could solve this problem via Zhu Xi’s approach of ignoring the ethnocentric moralist discourse completely and interpreting barbarians as subhuman, thus excluding them from the Neo-​Confucian teaching of universal human moral potential. Other Neo-​ Confucians did not deny that barbarians were human but did attempt to make their supremacism morally legitimate by grounding Confucius’s ethnocentrism in the concept of qi. In a historical commentary completed around 1155, Hu Hong’s brother Hu Yin (1098–​1156) used a kind of imaginary dialogue to present such an argument: Q: When Heaven created human beings, were they divided into Chinese and barbarians? A: No. Q: In that case, when the sage [Confucius] regarded the Chinese as his own people and the barbarians as outsiders, and when he regarded the barbarians as inferior and the Central Lands as superior, wasn’t he being contrary to Heaven? A: If the barbarians could behave like [people of] the Central Lands, then they too would be [people of] the Central Lands. It is just that they are inhumane and immoral, avaricious and murderous, which is contrary to the principles of humanity. That is why [Confucius] regarded them as inferior outsiders, strongly disapproving of their throwing the Central Lands into disorder with their conquests. Q: If, when Heaven created human beings, there was no division between Chinese and barbarians, then why are the barbarians inhumane and immoral, avaricious and murderous, contrary to the principles of humanity?

35 Yang, The Way of the Barbarians.

181

The Song–Jurchen Conflict

181

A: All [human beings] are endowed with the same qi, yet some are wise and some foolish. This does not mean that Heaven is unduly partial toward the wise and miserly toward the foolish. All [human beings live] within the space between [heaven and earth], yet some are in the Central Lands and some are barbarians. This does not mean that Heaven makes the Central Lands good and the barbarians bad. Rather, their concentrations [of qi] differ in purity and balance, which makes their apportioned capacities differ as well.36

This still does not answer the question of why all barbarians would have more impure or imbalanced qi than all Chinese. For an Imperial University lecture on the Annals in 1182, the Neo-​Confucian Lu Jiuyuan (1139–​1192) came up with a slightly more sophisticated model based on geographical determinism: When the sage [Confucius] regarded the Central Lands as superior and the barbarians as inferior, he was not being unduly partial to the Central Lands. The Central Lands have received balanced qi from heaven and earth, and so that is where ritual propriety and moral duty are found. Regarding the Central Lands as superior is not really regarding the Central Lands as superior but rather regarding ritual propriety and moral duty as superior. Even though [the Central Lands] had gone through decline and disorder [in Confucius’s day], the sage-​kings’ institutions and laws still survived, and the remnants of their customs had not been fully extinguished. When the barbarians became militarily strong and began swallowing up smaller states, they were about to take advantage of their strength to invade and conquer the Chinese. If this happened, there would be no place left for ritual propriety and moral duty.37

Lu Jiuyuan drew on a longstanding notion, again going back to the Han dynasty, that the “Central Lands,” the Chinese homeland and civilizational core in north China, had exceptionally balanced qi that allowed civilization and morality to flourish, whereas peripheral lands had inferior qi that produced barbarism.38 In a recent paper, Hans Van Ess suggested that Hu Yin and Zhu Xi also subscribed to a geographical determinist or environmental determinist understanding of barbarians’ purportedly inferior qi, but he acknowledged that this is merely speculative since neither man’s extant works or recorded teachings mention this idea.39 Possibly, Hu Yin and Zhu Xi avoided endorsing such a geographical interpretation of Chinese superiority because it would raise a question of obvious relevance to the Southern Song situation: if the barbarians 36 Hu Yin 胡寅, Dushi guanjian 讀史管見 (Changsha: Yuelu shushe, 2011), 245–​46.

37 Lu Jiuyuan 陸九淵, Lu Jiuyuan ji 陸九淵集 (Beijing: Zhonghua shuju, 1980), 277.

38 Shao-​yun Yang, “ ‘Their Lands are Peripheral and Their Qi is Blocked Up’: The Uses of Environmental Determinism in Han (206 BCE–​220 CE) and Tang (618–​907 CE) Chinese Interpretations of the ‘Barbarians,’ ” in The Routledge Handbook of Identity and the Environment in the Classical and Medieval Worlds, ed. Rebecca Futo Kennedy and Molly Jones-​Lewis (Abingdon: Routledge, 2016), 390–​412. 39 Hans Van Ess, “What is Chinese? Culturalism or Ethnicism in Song China? With an Appendix on Siku quanshu Censorship of Ethnic Prejudice,” Asia Major 34 (2021): 77–​80.

182

182 Shao-yun Yang occupied the “Central Lands” of the north and gained access to its qi, would they not eventually also develop ritual propriety and moral duty and learn to use the “sage-​kings’ institutions and laws”? Lu Jiuyuan seems to have ignored that troubling question, but in 1178 Chen Liang (1143–​1194), who like Ye Shi was a “statecraft” thinker critical of Neo-​Confucian moralism, had already submitted a memorial to Xiaozong that explored and answered it, in a fashion that would have shocked many Chinese supremacists. Chen, a staunch irredentist, argued that the balanced qi unique to the “Central Lands” (north China) had been “blocked up” for over fifty years by the intrusion of “the deviant qi of barbarians from beyond [the boundaries of] heaven and earth”—​that is, the Jurchens. The Southern Song had taken the “robes and caps and rites and music of the Central Lands” (i.e., Chinese civilization) and relocated them to the “peripheral region” of the southeast. Although the Mandate of Heaven (i.e., political legitimacy) and “the people’s hearts” (i.e., the loyalty of Chinese people in the north) had remained attached to the Song court in its new location, this situation could not continue indefinitely. If the Southern Song remained confined to the peripheral south, cut off from the qi of the Central Lands, that qi would eventually build up to a point where it had to “vent itself” and fill the “Central Lands” again, enabling the Jurchens to overcome their inferior qi endowment and acquire civilized ways. At that point, the Southern Song would lose the Mandate of Heaven and the people’s hearts to the Jurchens, who could then legitimately conquer the south.40 Chen Liang’s qi theory suggests that he had changed his mind since 1160 when, at the age of seventeen, he wrote an essay on the “barbarian” northern ruler Fu Jian’s (r. 357–​385) famously disastrous failure to conquer the southern dynasty of Eastern Jin (317–​420) in 383. This historical event’s contemporary resonance was obvious: in 316–​581, a number of “barbarian” states or empires ruled north China and a succession of “Chinese” dynasties ruled the south after the collapse of the Western Jin dynasty (266–​316), a geopolitical configuration very similar to that of the Jurchen Jin and Southern Song.41 Moreover, the Eastern Jin had claimed continuity with the Western Jin just as the Southern Song claimed to be a legitimate continuation of the Song dynasty. Fu Jian’s huge army had suffered an unexpected rout at the Fei River (a tributary of the Huai) and his Former Qin empire had then collapsed in a wave of rebellions. Chen explained this outcome as a manifestation of the Chinese people’s sacrosanct right to superiority over “barbarians,” even in times of political and military weakness: “Although Eastern Jin was weak, it was the Central Lands; although [Former] Qin was strong, it was a barbarian [state]. Since antiquity, when have the barbarians ever been able to 40 Chen Liang 陳亮, Chen Liang ji 陳亮集 (Beijing: Zhonghua shuju, 1987), 1–​2.

41 Andrew Chittick has recently argued against interpreting the southern dynasties as unproblematically “Chinese” in an ethnocultural sense, due to their cultural diversity. Song-​era historians did, however, tend to view them as such. See Andrew Chittick, The Jiankang Empire in Chinese and World History (Oxford: Oxford University Press, 2020).

183

The Song–Jurchen Conflict

183

swallow the Central Lands whole?”42 This argument assumed any Chinese dynasty to represent “the Central Lands” even if it lost control of north China. In contrast, Chen Liang’s 1178 memorial implied that no dynasty, no matter how Chinese ethnically, could continue calling itself “the Central Lands” without controlling the geographical space of the northern civilizational core. In fact, Chen Liang’s memorial argued that the Southern Song capital at Lin’an (Hangzhou) was highly vulnerable to conquest from the north because fifty years as the seat of the imperial court had exhausted all its regional qi, which in any case was “the peripheral qi of heaven and earth.” He went on to offer a solution to this problem: the Southern Song should abandon its peace agreement with the Jin and reconquer the “Central Lands” of the north as soon as possible by launching an offensive from the middle Yangzi region. That region’s relegation to a strategically peripheral status since the Tang dynasty (618–​907) had left it with a large store of unused peripheral qi, making it ideal as a staging point.43 Chen also warned that if, on the other hand, no such action was taken, then the Southern Song’s fate could be foretold based on historical precedent. The Eastern Jin and its successor dynasty in the south, coincidentally also named Song (420–​479), wasted multiple opportunities to reconquer the north. As a result, Chen claimed, the qi of the “Central Lands” finally vented itself on the Northern Wei dynasty (399–​534), ruled by the “barbarian” Tuoba Xianbi (or Xianbei, *Särbi) people. Receiving that superior qi made it possible for the Northern Wei emperor Tuoba Hong (Xiaowendi, r. 471–​499) to move his court to the old Chinese capital of Luoyang and “adopt the robes and caps and rites and music of the Central Lands”: a reference to a series of reforms in the 490s aimed at “Sinicizing” numerous aspects of Tuoba Xianbi elite culture, including language, clothing, and ritual. The Northern Wei thus gained the Mandate of Heaven and the people’s hearts from “the old robes and caps and rites and music to the [south]east of the [Yangzi] River.” Chinese civilization returned to the north, and the southern dynasties lost their legitimacy by deferring reconquest for too long. Although the Northern Wei eventually collapsed without achieving reunification, its legitimacy passed down to the Northern Zhou (557–​581) and then the Sui (581–​ 618), which conquered the last southern dynasty in 589. “Because of this,” Chen writes, “the reunification of all under heaven finally came from the northwest and not the southeast —​how can one not be terrified by this conjunction of cosmic forces and human will?”44

42 Chen, Chen Liang ji, 71. For the date of this essay see Hoyt Cleveland Tillman, Utilitarian Confucianism: Ch’en Liang’s Challenge to Chu Hsi (Cambridge, MA: Council on East Asian Studies, Harvard University, 1982), 72. 43 Chen, Chen Liang ji, 7–​8. 44 Chen, Chen Liang ji, 1–​2.

184

184 Shao-yun Yang

The Northern Wei Question and the Idea of “Sinicization” In the 1970s, the historians Rolf Trauzettel and Hoyt Tillman characterized Chen Liang as a proto-​nationalist whose irredentism was rooted in a new concept of an ethnically defined Chinese nation-​state.45 This interpretation seems unpersuasive, considering Chen’s opinion that the “barbarian” Northern Wei became a legitimate Chinese dynasty during the reign of Xiaowendi. Chen Liang was not an outlier in this regard: although numerous recent treatments of this subject have emphasized “hard supremacist” attitudes that disqualified any barbarian dynasty from legitimacy on ethnic grounds alone,46 Southern Song historians’ assessments of the Northern Wei were actually diverse and often surprisingly ambivalent, even when influenced by revanchist and supremacist agendas of denying legitimacy to the Jurchen Jin. In this they were not entirely different from nuanced eleventh-​century arguments acknowledging either that the Northern Wei did achieve legitimacy via “Sinicization” (i.e., the adoption of classical Chinese norms of governance), or that it could have done so under the right circumstances (for example, successful reunification).47 For example, the ardently revanchist Hu Yin believed that the Southern dynasties inherited the Mandate of Heaven from the Eastern Jin and remained legitimate until the last of them was conquered by the Sui, widely regarded as the first ethnically Chinese dynasty to rule the north since the Western Jin.48 But he also felt that Xiaowendi, despite being a barbarian, was so “morally worthy” on account of his “Sinicizing” reforms that he could well have achieved reunification and thus earned legitimacy had he not died relatively young in 499. To an imaginary critic’s objection that Heaven would never let a barbarian dynasty enjoy legitimacy, Hu Yin merely offered a skeptical and cryptic question: “Alas! But is that really so?”49 Evidently, Hu Yin was not merely being facetious when he claimed that “[i]‌f the barbarians could behave like [people of] the Central Lands, then they too would be [people of] the Central Lands,” although he (unlike Chen Liang) made no effort at explaining how Xiaowendi had overcome his innate disadvantage in qi. 45 Trauzettel, “Sung Patriotism as a First Step toward Chinese Nationalism,” 212; Hoyt Cleveland Tillman, “Proto-​Nationalism in Twelfth-​Century China? The Case of Ch’en Liang,” Harvard Journal of Asiatic Studies 39 (1979): 403–​28, esp. 425–​26. 46 For some recent examples see Tackett, The Origins of the Chinese Nation, 190–​93; Liu Pujiang, 劉浦江, Zhengtong yu Hua Yi: Zhongguo chuantong zhengzhi wenhua yanjiu 正統與華夷:中國傳 統政治文化研究 (Beijing: Zhonghua shuju, 2017), 31–​32; Puning Liu, “Song Scholars’ Views on the Northern Wei Legitimacy Dispute,” Archiv Orientální 86 (2018): 17.

47 For these see Liu, “Song Scholars’ Views on the Northern Wei Legitimacy Dispute,” 10–​23; also, Yang, The Way of the Barbarians, 147–​48.

48 In fact, the first Sui emperor’s ethnic identity is ambiguous. His clan had intermarried with prominent Xianbi families, and the veracity of its own claim to descent from an eminent Chinese clan has been questioned by modern historians. The same could be said of the first Tang emperor and his clan. For Hu Yin’s filial revanchist views, see Toqto’a 脫脫 et al., Songshi 宋史 (Beijing: Zhonghua shuju, 1977), 435.12920–​21. 49 Hu, Dushi guanjian, 432; see also Yang, The Way of the Barbarians, 150–​52.

185

The Song–Jurchen Conflict

185

Consider also Zhu Fu (1140–​1215), a student of the statecraft thinker Chen Fuliang (1137–​1203). Zhu published, ca. 1205, a monumental hundred-​scroll commentary on Chinese history from Yao and Shun to the founding of the Song, a preface to which states that its central principle is to distinguish between legitimate and illegitimate dynasties in history, and especially to deny legitimacy to the rulers whom its author considered the worst usurpers in history. The preface’s author, none other than Ye Shi, claims: “There is no historiographical principle more important than this [in the book]; second to it are [the principles of] avenging wrongs and humiliations and of maintaining Chinese superiority and barbarian inferiority”—​in other words, filial revanchism and Chinese supremacism.50 The fact that Zhu Fu placed filial revenge at the same level of importance as ethnic difference, and moral condemnation of usurpers even higher, is most evident in his treatment of the Later Tang dynasty (923–​937). Although Zhu saw the Later Tang emperors’ “barbarian” Shatuo Türk origin as an imperfection, he also recognized the Later Tang as a legitimate continuation of the Tang dynasty. This was because he respected the Shatuo Türks’ commitment to destroying the Later Liang dynasty (907–​923), whose rulers Zhu condemned as illegitimate usurpers responsible for the last Tang emperor’s dethronement and murder.51 In contrast, when a barbarian dynasty was not avenging a legitimate Chinese dynasty but threatening it, then Zhu Fu could be unequivocal in his Chinese supremacism. For example, he saw Fu Jian’s invasion of the Eastern Jin as nothing more than “a barbarian caitiff bringing disorder to the Central Lands”—​an act equal in immorality to regicide and usurpation. Zhu argued that Fu Jian’s empire collapsed following the failure of his invasion because “heaven and earth and the spirits and gods would surely not tolerate” a barbarian state trying to destroy a legitimate Chinese dynasty.52 When it came to assessing the Northern Wei, however, Zhu Fu was strikingly prone to self-​contradiction. His commentary on the Northern Wei’s unification of north China in 439 begins with yet another discourse explaining why Confucius’s bias against barbarians was justified; indeed, Zhu claims that without such a bias, civilization would collapse and the whole universe would descend into chaos: The sage [Confucius] respected the Central Lands and saw the barbarians as inferior; he regarded the Chinese states as his own people and the barbarians as outsiders. He did not draw this distinction for superficial or arbitrary reasons. Rather, he did so because the Central Lands are the source of rites and music, civilizing teaching, humaneness, moral duty, and the moral power of the Way. They have distinctions between rulers and subjects, superiors and inferiors; affection between fathers and sons and brothers; and moral duty between husbands and wives and friends. The balance between heaven and earth, the light of the sun, moon, and stars, and the survival of all living things depend on this. As for

50 Ye, Ye Shi ji, 208.

51 Siku quanshu cunmu congshu 四庫全書存目叢書, vol. 280 (Jinan: Qilu shushe, 1997), 644–​46 (cf. 660, 666, 677 for Zhu’s view of the Shatuo as inferior to the Chinese). 52 Siku quanshu cunmu congshu, 280: 531.

186

186 Shao-yun Yang the barbarians, they are like animals attacking each other and tearing each other apart in their fights for dominance, biting and devouring each other in their pursuit of food and warmth. One cannot instruct them with reason or regulate them with laws. That is why the sage deeply abhorred them and distanced himself from them.53

On the other hand, Zhu Fu found it difficult to dismiss the Northern Wei as just another “barbarian” state, largely because of his relatively positive opinion regarding Xiaowendi’s reforms. Immediately after attributing “the disasters brought by barbarians” to their “addiction” to warfare and killing, he changed tack and began affirming the Northern Wei’s contributions in laying the foundations for the Sui reunification and the Tang dynasty’s core institutions: In the north, the [Northern] Wei absorbed the territory of fifteen states and was succeeded by the [Northern] Zhou and the Sui, which then conquered the Chen dynasty (557–​589) and reunified [north and south]. This was not due to the merits of the original Tuoba [Xianbi], that vile breed from north of the [Gobi] desert. [The Northern Wei’s] laws and regulations became a model for the age, and its ritual institutions emulated the legacy of the sage-​kings. Its policies for appointing officials, distributing farmland, recruiting personnel, and organizing the army were essentially the foundation for those of the [early Tang dynasty].54

Zhu then provided an explanation for why the Northern Wei was unusually long-​ lived and successful compared to earlier barbarian dynasties: Was this not also due to the benefits of using Chinese [ways] to change the barbarians? The method of the Annals is that when the Central Lands [states] use barbarian [ways], then they are regarded as barbarians, and when barbarians are promoted to the level of the Central Lands, then they are regarded as [people of] the Central Lands. If the sage [Confucius] had taken up his writing brush to write the Annals between the Sui and Tang dynasties, even he would have been inclined to favour the [Northern] Wei.55

Here, Zhu Fu’s language alluded to a famous summation by the late Tang writer Han Yu (768–​824) of the principle of Annals exegesis already mentioned earlier, as well as a phrase, “using Chinese [ways] to change the barbarians,” that first appeared in the Mencius and later became a conventional way of describing acculturative “Sinicization.”56 If Zhu Fu was prepared to concede that the Northern Wei’s “Sinicization” could get it “promoted to the level of the Central Lands,” he was still not ready to recognize it as legitimate from Xiaowendi’s reign on. Instead, he later claimed that the Southern Qi dynasty (479–​502) still held legitimacy, despite being inferior to the Northern Wei in governance, moral virtue, and military strength, simply because the Southern Qi 53 Siku quanshu cunmu congshu, 280:548.

54 Siku quanshu cunmu congshu, 280:548.

55 Siku quanshu cunmu congshu, 280:548.

56 On Han Yu’s summation and its influence, see Yang, The Way of the Barbarians. In Mencius 3A.4, Mencius claims that he has heard of “using Chinese [ways] to change the barbarians” but not of the Chinese “being changed by barbarian [ways],” implying that the former is desirable and the latter is not.

187

The Song–Jurchen Conflict

187

emperors were Chinese and not barbarians.57 He also argued that Xiaowendi’s reforms were motivated by an ambition to conquer the south, rather than genuine admiration for Chinese civilization. Even if Xiaowendi sincerely desired to “emulate antiquity,” he had also “indulged his greedy, insatiable heart and presumptuously schemed to imitate his betters, not knowing his limits,” thus bringing about his dynasty’s fall.58 Zhu even negated his earlier statements about the Northern Wei being an example of “the benefits of using Chinese ways to change the barbarians,” now claiming instead that barbarians could only bring disaster upon themselves by emulating Chinese ways. Xiaowendi’s premature death and the Northern Wei’s subsequent political decline were proof of this: I have previously been so bold as to suggest that the Central Lands and the barbarians each have their proper place. Changing the Chinese to be like barbarians will certainly lead to chaos, but there have never been barbarians who imitated the Central Lands and were not destroyed as a result.59 It seems likely that by this point in the text, Zhu Fu’s attitude toward the Northern Wei was torn between admiration for Xiaowendi’s reforms and apprehension that they would serve as a precedent for the Jurchen Jin. He reconciled this tension by claiming that “Sinicization” might be good for barbarian dynasties in the short term but would be fatal to them in the long term, since it transgressed the proper boundaries between Chineseness and barbarism. An even more skeptical take on Xiaowendi is found in a collection of iconoclastic and independent-​minded reading notes on the Confucian classics and dynastic histories that Ye Shi produced in 1208–​1223, while in retirement following his dismissal from office in the wake of his patron Han Tuozhou’s assassination. It should be noted first that Ye Shi was a revanchist but not a Chinese supremacist. To his pragmatic mind, whether “barbarian” dynasties prevailed over their Chinese opponents, or vice versa, was a matter of statecraft and strategy and had nothing to do with any inherent Chinese superiority in morality, legitimacy, or civilization. He gave no credence to the idea, for example, that the Eastern Jin defeated Fu Jian because it was the legitimate dynasty, or that Fu Jian’s empire fell because his barbarian origin made him inherently illegitimate: “If we see things from the perspective of Fu Jian’s interests, he was just seeking to unify all under heaven, so what harm was there in having designs on the [Eastern] Jin?” In Ye’s opinion, Fu Jian was a ruler of impressive vision who had failed only because his army was inadequately trained and lost its cohesion at the Fei River, while the Eastern Jin army was well trained and kept its nerve despite being outnumbered. Likewise, Fu had lost his empire in the aftermath of that defeat only because it was overextended and insufficiently consolidated.60 57 Siku quanshu cunmu congshu, 280:565.

58 Siku quanshu cunmu congshu, 280:572.

59 Siku quanshu cunmu congshu, 280:571–​72.

60 Ye Shi 葉適, Xixue jiyan xumu 習學記言序目 (Beijing: Zhonghua shuju, 1977), 434–​35, 437–​38.

188

188 Shao-yun Yang Ye Shi viewed Xiaowendi’s reforms, especially his efforts at reviving classical Chinese ritual and his moving of the capital, as misguided policies that undermined the Northern Wei’s strong foundations for the sake of acquiring the mere trappings of a civilized Chinese state. He argued that the Northern Wei differed from the fourth-​ century “barbarian” states, whose founders had lived in north China for generations and were similar to the Chinese in values and tastes. Instead, the Tuoba “were roughly similar to today’s Jurchens” in invading north China from outside and initially “using purely steppe-​nomadic (Hu 胡) customs to change the Chinese people by force.”61 Unlike Zhu Fu, Ye Shi credited this use of “purely steppe-​nomadic” methods, not “Sinicization,” for the Northern Wei’s relative longevity as a conquest dynasty. As a result, Ye argued that Xiaowendi’s attempt to “Sinicize” his dynasty initiated the Northern Wei’s decline not because barbarians were meant to be inferior, but because their ways weren’t objectively inferior at all: That being so, using Chinese [ways] to change the barbarians is the Way of the sage [Confucius], but using barbarian [ways] to control the Chinese is beneficial to the barbarians. When they lose that benefit, they decline, and when they go against what is normal, they fall.62

Ye’s opinions may have been influenced by observing that the once mighty Jurchens were now being overrun by the Mongols after having adopted many of the political and ritual institutions that the Chinese identified with civilization. In their case, too, “Sinicization” appeared to have made them weaker rather than stronger.

Conclusion

Looking back on the Jin empire’s destruction by the Mongols in 1234, Lü Zhong was inclined to see Heaven itself intervening to set the world right: Surely Heaven desires to give prosperity to the Chinese and calamity to the barbarians! … In the past, people used to say, “Barbarians are not fated to rule for a hundred years.” How true these words are! Now the Jin caitiffs, those little upstarts from the desert, filled our Central Lands with the reek of mutton, defiling our [civilization of] robes and caps. It seemed as if it would not be easy to destroy them. But just as we approached the hundredth year since their rise, their own kinsmen rebelled against them from within, and the Tatar (i.e., Mongol) cavalry attacked them from without. Their domain gradually shrank until finally they were utterly wiped out. Meanwhile our dynasty, with its majestic legitimacy, remains calm and at ease. Is it not clear as day that Heaven gives prosperity to the Chinese and calamity to the barbarians?63

Lü Zhong thus claimed that Heaven would never permit a barbarian dynasty to last for more than a hundred years, and he spoke as if the notion was a longstanding one. Recent research by Fang Cheng-​hua suggests, however, that this idea first emerged 61 Ye, Xixue jiyan xumu, 468. 62 Ye, Xixue jiyan xumu, 491.

63 Lü, Leibian huangchao, 829–​30.

189

The Song–Jurchen Conflict

189

among Southern Song literati, like Lü Zhong himself, in the aftermath of the Jin empire’s fall.64 These literati interpreted the Jin’s demise as evidence that Heaven was protecting the Chinese after all, making the Song dynasty invincible in spite of temporary threats and setbacks. Fang has also argued that the Southern Song court’s refusal to negotiate peace with the Mongols after 1234, despite multiple opportunities to do so, was driven by the elevation of filial revanchism to ideological dogma.65 But since filial revenge was not a relevant factor in the Song–​Mongol conflict, I am inclined to attribute the hardline attitude to restored Chinese supremacist faith that the Song would outlast its barbarian enemies. No barbarian power had ever conquered the entire Chinese world, and none ever would. Heaven would never allow it. We know now, of course, that such faith was misplaced, as the Southern Song finally fell to the Mongol Yuan empire in 1276–​1279. When Yuan armies began their advance into the Yangzi delta in late 1274, the Southern Song’s Grand Empress Dowager Xie (1210–​ 1283) issued an edict on behalf of her newly enthroned grandson, three-​year-​old Zhao Xian (1271–​1323), to exhort officials and commoners throughout the realm to form new volunteer armies and come to the capital’s defence, thus “embodying high Heaven’s desire to give prosperity to the Chinese.” The edict also expressed indignation and confusion that these “vile caitiffs,” the Mongols, would “destroy the proper order of heaven and earth” by trying to conquer the whole world: after all, “since antiquity there has never been an age when the barbarian caitiffs ruled everywhere.”66 Several months before this, Khubilai Khan (r. 1260–​1294) had himself issued a declaration of war in Chinese to “officials, clerks, generals, military officers, literati, and commoners in Song prefectures and commanderies,” promising them generous treatment if they laid down their arms peacefully. Intriguingly, this document—​surely composed by one or more of Khubilai’s north Chinese literati advisors—​begins with the following lines: Dwelling in the centre, we have made grand plans, and Heaven has ordained this as the hour of unification. Moving from north to south, our great army marches like peals of thunder. We will first punish the guilty and condole with the people,67 then use Chinese [ways] to change the barbarians. We wish to bring the Yangzi and Qiantang rivers under our sway and extend our unification [of the world] to the shores of the southern sea.68

64 Fang Cheng-​hua 方震華, “Yi-​Di wu bainian zhi yun —​yunshu lun yu Yi-​Xia guan de fenxi” 夷狄 無百年之運—​運數論與夷夏觀的分析, Taida lishi xuebao 臺大歷史學報 60 (2017): 159–​91. 65 Fang, “Fuchou dayi,” 330–​38.

66 Quoted in Fang, “Yi-​Di wu bainian zhi yun,” 175. For further analysis of the edict and its effects, see Richard L. Davis, Wind against the Mountain: The Crisis of Politics and Culture in Thirteenth-​ Century China (Cambridge, MA: Council on East Asian Studies, Harvard University, 1996), 61–​63. 67 On this expression, which conveys the meaning that the Mongols’ quarrel is only with the corrupt Song court and that its innocent subjects will not be harmed, see Shao-​yun Yang, “Letting the Troops Loose: Pillage, Massacres, and Enslavement in Early Tang Warfare,” Journal of Chinese Military History 6 (2017): 7.

68 Tao Zongyi 陶宗儀, Nancun chuogeng lu 南村輟耕錄 (Beijing: Zhonghua shuju, 1959), 16–​17 (emphasis added).

190

190 Shao-yun Yang It seems almost certain that the Chinese ways mentioned here are not those of the Song, and the barbarians are not the Mongols. Instead, it is the Yuan that is now Chinese and the Southern Song that is barbaric, presumably because of their respective geographical locations: the Yuan occupying the “Central Lands” of the north, and the Song left with only the “peripheral” south. One might read this piece of war propaganda as the Mongol Empire’s answer to Southern Song pretensions to superiority vis-​à-​vis the “barbarians” of the north. As for whether the Mongols really identified with “Chinese ways” when ruling China, that is a complicated question best left for a different discussion.

191

Chapter 11

FAITHFUL TO A VANISHING PAST: NARRATING WARFARE AND PEOPLEHOOD IN YUAN CHINA Francesca Fiaschetti The Mongol campaigns in East Asia in the thirteenth and fourteenth centuries

affected perceptions of ethnicity and identification in many ways: they reshaped territorial boundaries, led to the destruction of existing kingdoms, and to the creation of new ones.* In terms of demography, the medieval Mongols triggered the movement of large groups from one corner of Eurasia to the other through the mobility of military personnel and troops, such as Central Asian troops engaged in the conquest of Yunnan, but also through the migration of refugees into Southeast Asia. This movement involved both prominent individuals, as part of the Mongols’ administrative policies and the social mobility resulting from them, and mass migration, radically changing the ethnic landscape of East Asia.1 Warfare, alongside other dynamics, played a major role in enabling the creation of new ways of identification, as the Mongol worldview was famously built around the dichotomy of “enemies and allies” (il bulγa irgen), a military-​based narrative which then applied, alongside the expansion of the empire, to all aspects of society. From social groups to administrative offices and geographical perceptions, the medieval Mongols developed a new language to refer to the non-​Mongol identities that they conquered, and new rules to administer them.2 Alliances or repeated displays of loyalty during crucial moments of Mongol expansion became the basis on which many non-​Mongol citizens of the empire constructed a new career and a new identity for themselves, as well as for their lineages. In the available sources, this is often symbolically connected with recurring ritual elements associated with Inner Asian traditions, such as the granting of a new name or title by the emperor. Thus, the perception of belonging to a certain * A first draft of this chapter was presented during my cooperation with the ERC project “Mobility, Empire and Cross-​Cultural Contacts in Mongol Eurasia” at the Hebrew University of Jerusalem. I am grateful to my colleagues in the project, Geoffrey Humble and Florence Hodous, for sharing some relevant translations. I am also grateful to Márton Vér for his comments and suggestions. All shortcomings remain my own. 1 On this see, among others: Thomas T. Allsen, “Population Movements in Mongol Eurasia,” in Eurasian Nomads as Agents of Cultural Change, ed. Reuven Amitai and Michal Biran (Honolulu: Hawaii University Press, 2015), 119–​51.

2 As analyzed in: Francesca Fiaschetti, “The Borders of Rebellion: The Yuan Dynasty and the Rhetoric of Empire,” in Political Strategies of Identity Building in Non-​Han Empires in China, ed. Francesca Fiaschetti and Julia Schneider, Asiatische Forschungen 154 (Wiesbaden: Harrassowitz, 2014), 127–​46.

192

192 Francesca Fiaschetti community was achieved, in Mongol Eurasia, by a constant renarration of the construction of the empire through its crucial moments, and through the ritualized construction of a collective memory. This impacted not only the identity of individuals and elites, but also the sense of belonging of communities and political groups in Mongol-​ruled East Asia. This becomes evident if we look at the frequent reference in Yuan imperial documents to the long-​lasting confrontation of the Yuan dynasty (1260–​1368) with the Song dynasty (960–​1279) and how it changed the political landscape in China and Southeast Asia. This chapter will present an overview of some of the main theories showing the linkage between military rhetoric and the representation of various elites and communities in Yuan China. Further, by reviewing some notable examples from Chinese sources of the period, the chapter will offer some thoughts for a comparative analysis of this period.

Mongols, New Mongols, Foreign Mongols

As the Mongols proceeded eastwards and arrived to conquer East Asia, the representation of non-​Mongol identities was adapted to the new context, and a new terminology was created by merging Inner Asian and Chinese geographical, political, and military concepts. Borrowing some labels established under previous dynasties, the population of the Yuan empire was famously divided into four groups, which were granted different status, positions, and privileges.3 The principle behind this division was the display of loyalty, as these four categories corresponded to the progressive expansion of the Mongol Empire. Thus, the division of the ruled population into the four groups reflects the chronological aspect of the Mongol worldview, as it built a hierarchy of identities mirroring the expansion of their empire—​from the Mongol core (mengguren 蒙古人), that is, the original nucleus of tribes from which the Mongol Empire originated, to the southerners (nanren 南人), that is, the subjects of the Southern Song (1127–​1279), who were the last to be subjugated. As famously shown by Yoshiyuki Funada, the construction of social groups and group identities was not as rigid as the terminology of the Chinese sources seems to imply.4 Instead, a relative flexibility was in place, as the creation of these categories was the result of negotiation between rulers and subjects: it was not uncommon, in fact, for people outside of the original nucleus of the Mongolic tribes to be granted the label 3 The four groups were: the Mongols (mengguren 蒙古人), the semuren 色目人 (Central Asians), hanren 漢人 (“Northeners”), and nanren 南人 (“Southerners”), as described in Tao Zongyi’s 陶宗儀 Nancun chuogenglu 南村輟耕綠 (Repr. Beijing: Zhonghua shuji, 1959), juan 1:14. Among the many studies on the topic, see: Yoshiyuki Funada, 舩田善之. “Semuren yu Yuandai zhidu, shehui: chongxin tantao Menggu, semu, hanren, hanren huafen de weizhi 色目人与元代制度・​社会-​重新探討蒙古・​ 色目・​漢人・​南人劃分的位置”, Yuanshi Luncong 元史論叢 9 (2004): 162–​74. For a recent analysis (focusing especially on the concept of hanren) see: Mark Elliott, “Hushuo: The Northern Other and the Naming of the Han Chinese,” in Critical Han Studies, ed. Thomas Mullaney, James Patrick Leibold, Stéphane Gros, and Eric Armand Vanden Bussche (Berkeley: University of California Press, 2011), 173–​90. 4 Funada, “Semuren yu Yuandai zhidu.”

193

Faithful to a Vanishing Past

193

“Mongol” after some worthy deeds toward the empire. These individuals, which Michael Weiers labels as “foreign Mongols,” and Lhamsuren Munkh-​Erdene calls “new Mongols,” were granted all the privileges that the Mongols reserved for their own.5 According to Weiers, a determining factor in becoming a “foreign Mongol” was participation in the Western Campaigns (in Eastern Europe, 1236–​1241).6 However, as the Chinese sources show, non-​Mongol military personnel fighting in the campaign against Khwarezm (1219–​1221) were granted a similar status and label to that of native-​born Mongols. One of the most famous recorded examples of a “foreign Mongol” is the well-​studied case of the Kitan Yelü Xuedu 耶律薛阇 (r. 1226–​1238), son of the founder of the Liao dynasty, Yelü Liuge 耶律留哥 (1165–​1220): The Emperor said: “[The Khitan] Xuedu is a Mongol now. He followed Us to the Western Regions, and when the Muslims surrounded the heir apparent in the city of Khwarazm, Xuedu whisked him to safety with a thousand men, though he himself was wounded by a lance. He also fought with Us against the Muslims in Bukhara and Samarkand and was struck by an arrow. Because he repeatedly rendered such services, he was given the title baγatur 拔都魯 [Mong. “hero”].”7

The association of a new label with the mention of a specific military event has two meanings here. First, it shows how shared military experiences created a sense of belonging to a specific group, and legitimized membership in it.8 Second, what the label “Mongol” here defines is not a process of “Mongolization” in cultural terms, but the acknowledgement of the specific political status of subjects of the Mongol Empire, due to their military contribution in creating this empire. Such a shift of ethnic labels into ethnopolitical identifications is not exclusive to the Mongol Empire and in Central Asia it can be seen also in the case of the label “Türks” applied to Hungarians and Mongols.9 In the case of the medieval Mongols, this flexible ascription to the Mongolian core group of the empire challenges us to reflect on the meaning of the term “Mongol,” 5 Michael Weiers, Geschichte Der Mongolen (Stuttgart: Kohlhammer, 2004), 13–​15, 35–​38, and Lhamsuren Munkh-​Erdene, “Where Did the Mongol Empire Come From? Medieval Mongol Ideas of People, State and Empire,” Inner Asia 13 (2011): 211–​37. 6 Weiers, Geschichte, 67–​68.

7 帝曰:「薛闍(舍)〔今〕為蒙古人矣,其從朕之征西域也,回回圍太子於合迷城,薛闍引千 軍救出之 […] 以是積功為拔都魯 […] . Song Lian 宋濂, Yuanshi 元史 (repr. Beijing: Zhonghua shuju, 1976), juan 149:3154, quoted after Michal Biran, “The Mongols and Nomadic Identity: The Case of the Kitans of China,” in Eurasian Nomads as Agents of Cultural Change, ed. Reuven Amitai and Michal Biran (Honolulu: Hawaii University Press, 2015), 164.

8 The role of shared experiences in creating a sense of belonging has been analyzed among others by George C. Bentley, “Ethnicity and Practice,” Comparative Studies in Society and History 29 (1987), 27, who states that: “consciousness of affinities of interest and experience embodies subliminal awareness of objective commonalties in practice.” 9 As analyzed by István Vásáry, “Hungarians and Mongols as ‘Turks’: On the Applicability of Ethnic Names,” in Between Byzantium and the Steppe. Institute of Archaeology, ed. Ádám Bollók, Gergely Csiky, and Tivadar Vida (Budapest: Hungarian Academy of Sciences, 2016), 537–​43.

194

194 Francesca Fiaschetti which, in the thirteenth and fourteenth centuries, was obviously different from the ethnonym we currently associate with it. In the Secret History of the Mongols (thirteenth century), the appellative Mongqol first appears to designate the Mongqol ulus (“Mongol confederation”)10 in the description of the assembly (quriltai) of 1206, in which Chinggis Qan (r. 1206–1227) was elected emperor.11 Similarly, as Igor de Rachewiltz has shown, the idea of an institutionalized yeke Mongqol ulus (a “State of the Great Mongols”) was coined only later, after 1221, as Chinggis Qan’s response to the parallel denomination of the “Great Jin state” (Da Jin guo 大金國), thus signifying equal status vis-​à-​vis the Jurchen polity.12 Also in the Chinese sources the translation of yeke Mongqol ulus (Da menggu guo 大蒙古國) appears rarely in the Yuanshi 元史 (History of the Yuan Dynasty, 1370), starting with the section on the rule of Ögödei (r. 1229–​1241), the heir of Chinggis Qan (r. 1206–​1227).13 On the one hand, the term “Mongol” (and its flexibility) represents the formalization of Inner Asian political dynamics of rewarding shared military experiences and displays of loyalty, which were at the base of the creation of the Mongol Empire. On the other hand, it answers the medieval Mongols’ necessity to communicate with the neighbouring polities on equal terms, in order to legitimize their imperial project. This is an issue that other empires previous to the Mongol yeke ulus faced too, as seen for example in the Orkhon inscriptions and the particular usage of el and bodun there.14 It was only in a later phase, with the further expansion of the Mongol Empire, that a process of Mongolization in cultural terms could take place among the intellectual and administrative elites of the empire. In Yuan China for example, those subjects who changed not only their political, but also their cultural identity into a Mongol one, expressed it not by being named “Mongols,” but by acquiring a Mongolian name. These names could be granted by Mongol emperors, as well as by their most prominent generals,15 and the Chinese sources testify to this practice at least up to the mid-​Yuan

10 There is a vast literature on the meaning and proper rendering of the term ulus as an organized political entity. See, among others: Munkh-​Erdene, “Mongol Empire” and Pavel Rykin, “The Social Group and Its Designation in Middle Mongolian: The Concepts Irgen and Oboq,” trans. Sarah Turner. Antropologicheskii Forum /​Forum for Anthropology and Culture 1 (2004): 183–​221.

11 Secret History, tr. de Rachewiltz, § 202: 133–​34. See the analysis of Munkh-​Erdene, “Mongol Empire,” 215–​16, who points out that the term “Mongol” here has a broader connotation, whereas the same source uses the term Mongoljin to refer to ethnic Mongols. See, also, Igor de Rachewiltz, “The Genesis of the Name ‘Yeke Mongγol Ulus’,” East Asian History 31 (2006): 53–​56. Weiers, Geschichte, 15–​18, too questions the application of this term as a later phenomenon. 12 de Rachewiltz, “The Genesis of the Name,” 54.

13 See the three mentions in Yuanshi 124:3050, 149:3512, and 208:4625.

14 See on this: István Zimonyi, “Bodun und El im Frühmittelalter,” Acta Orientalia Scientiarum Hungaricae 56 (2003): 57–​79.

15 The Yuanshi in fact records the case of Tian Xiong 田雄 (1189–​1246), an official who was granted his new name per request of the general Muqali (Yuanshi 151:3579).

195

Faithful to a Vanishing Past

195

period.16 In the process of name granting, Mongol practices encountered Chinese ones, as it is clearly stated in the Yuanshi, in the case of the biography of Shi Tianlin 石天麟 (1217–​1309): [When] the emperor [Ögödei] ordered the Secretary-​General (of the Central Secretariat) Yelü Chucai to deal with general affairs, and to choose a virtuous and able person as Assisting Official (or Service Official), Tianlin was chosen, and [the emperor] granted him the name of Manggutai. When the imperial prince(s) went on campaign to the Western regions, Tianlin was made duanshiguan 斷事官 [judge].17

It has also been shown that the “name granting” (ciming 賜名) of the Chinese sources sometimes simply indicated that an official title was being conferred such as the military title baγatur (“hero”) or the civil one of uran (“artisan”). One notable example is the Yuanshi biography of the Tangut artisan and official Xiaochou 小丑 (fl. 1220s), which reads as follows: Durutai was a Tangut. His grandfather was called Xiaochou [“little clown”]. When Chinggis had pacified Xi Xia, he took all kinds of craftsmen [with him], and Xiaochou, because his occupation was bow[-​making], entered [the royal presence], and he was granted the name of Queyan wulan 怯延兀蘭 (üčigen uran?), and he was ordered to be centurion over bow[-​maker] households at the camp of the qieliankou 怯憐口 [guards].18 He accumulated positions until he became Colonel of the Second Class, and moved to live at Helin, where he died.19

Recent scholarship has analyzed in detail some notable cases of non-​Mongol subjects, whose cultural assimilation to the Mongols was expressed in many aspects of their lives, from their knowledge of the Mongolian language to marriage relationships, to the choice of Mongolian names for their offspring. All these elements moulded a new, mixed identity for entire lineages over the course of many generations.20 Yet, changing one’s name was not only a prerogative of the ruling elite, but also the choice of many individuals. So, for example, the Yuanshi records the case of some 16 The Yuanshi in fact records an episode of name granting by the emperor Tuq Temür (r. 1328–​1332). See: Yuanshi 143:3423.

17 帝命中書令耶律楚材釐正庶務,選賢能為參佐,天麟在選,賜名蒙古台。宗王征西域,以 天麟為斷事官. Yuanshi 153: 3619. Translated and analyzed in: Florence Hodous, “A Judge at the Crossroads of Cultures: Shi Tianlin,” Asiatische Studien—​Études Asiatiques 71 (2017): 1147.

18 Note here that qielienkou 怯憐口 is the Chinese rendering of boghol. See David M. Farquhar, The Government of China under Mongolian Rule: A Reference Guide (Stuttgart: Franz Steiner Verlag, 1990), 372.

19 朵羅台,唐兀氏。祖小丑,太祖既定西夏,括諸色人匠,小丑以業弓進,賜名怯延兀蘭,命 為怯憐口 行營弓匠百戶,徙居和林,卒。Yuanshi 34: 3264, trans. Florence Hodous. The lineage of this individual is also analyzed by Tomoyasu Iiyama, “ ‘Steles and Status’: Evidence for the Emergence of a New Elite in Yuan North-​China,” Journal of Chinese History 1 (2017): 3–​26. 20 See Geoffrey Humble, “ ‘Han’ Cultural Mobility under Mongol Rule: Biographies of the Jia 賈 Family,” Asiatische Studien—​Études Asiatiques 71 (2017): 1153–​67; Hodous, “Shi Tianlin,” and Dandan Zhang, Menggu zaoqi Menggu hua Hanren jinchen qunti yanjiu 蒙 元 早 期 蒙 古 化 汉 人 近 臣 群 体 研 究 民族学 (unpublished master’s thesis, Nanjing University, 2014).

196

196 Francesca Fiaschetti individuals who changed their family name after escaping from a defeated state or dynasty.21 It is also noteworthy that Han-​Chinese subjects also gave Mongolian names to their children for various reasons: some of them hoped for better career chances for their offspring, others did it out of genuine loyalty toward the dynasty.22 The choice made by these individuals to change their names or those of their children is interesting as it reveals the various intentions behind the dynamics of identity-​building under the Mongols: shifts in one’s own identity were dictated by interest or necessity. In this case, they were a clear byproduct of the multicultural society created by the Yuan rulers.23 Testimony to this is the fact that acculturation went both ways, with some non-​Han people taking Chinese names.24 These examples show how an important outcome of the Mongol conquest of China was the creation of an environment where multiple identities could coexist without contradictions in an individual’s or in a family’s personal history. New identities were attributed not only to military elites but also to civil ones, and they were ritually sanctioned by the granting of a new name, something which again is reminiscent of the Secret History and Inner Asian tradition. In 1189, when Temüjin was first made Qan, the Secret History reports how his peers gave him the name of Chinggis Qan. In the same passage, rewards in terms of official positions were granted to those who had displayed loyalty and maintained an alliance with him.25 The personal connection with Chinggis Qan has been understood as the main element in the biographies of the boghol (bondsman) category in the history of the medieval Mongols, as this was another label used to distinguish identities and create a hierarchy in the society of the Mongol Empire. The Mongolian term boghol bears the literal meaning of “slave.”26 However, several studies on the subject have shown that more than the literal idea of “slavery,” this term 21 For example, the Yuanshi records the case of Li Ting 李庭, a former subject of the Jin dynasty (1125–​1234), who changed his surname after fleeing from the Jin (Yuanshi 162:3795).

22 See Elizabeth Endicott-​West, Mongolian Rule in China: Local Administration in the Yuan Dynasty (Cambridge, MA: Harvard University Asia Center, 1989), 79, 81–​82.

23 Among the various studies on the subject, see Brose’s analysis of semuren elite members in Michael C. Brose, Subjects and Masters: Uyghurs in the Mongol Empire (Bellingham, 2007); Michael C. Brose, “Central Asians in Mongol China: Experiencing the ‘Other’ from Two Perspectives,” The Medieval History Journal 5 (2002): 267–​89; Michael C. Brose, “Uyghur Technologists of Writing and Literacy in Mongol China,” T’oung Pao, Second Series 91, no. Fasc. 4/​5 (2005): 396–​435. 24 See the case of Amdula 暗都剌 and his brothers adopting their grandmother’s family name Xun 荀 in: Xu Youren 許有壬, Xiyu shizhi Haji Haxin bei 西域使者哈只哈心碑 (Epitaph dedicated to the Messenger from the Western Lands, Hajj Ḥasim), in: Li Xiusheng, ed., Quanyuanwen 38: 389. 25 Secret History, tr. de Rachewitlz, § 123: 49–​50 (for the name giving) and § 124–​25: 50–​52 (for the granting of official positions).

26 Among the several studies on the boghol see Shiraiwa Katzuiko 白岩一彦, “On the Ötegü Bogol in the Jämi’ Al-​Tavärikh of Rashid Al-​Dïn,” Acta Orientalia 47 (1986): 27–​31; Tat’jana D. Skrynnikova, “Boghol, a Category of Submission at the Mongols,” Acta Orientalia 58 (2005): 313–​19; Skrynnikova,

197

Faithful to a Vanishing Past

197

describes a relationship of submission and a strategy of inclusion of groups of non-​ethnic Mongols into the system of the empire.27 Boghol were bondsmen: individuals or entire communities of subjects of the Mongols, who were defined by their function of rendering service (voluntarily or by force) to the imperial elite. In this sense, it is particularly the Persian rendering of the term boghol that explains its value in terms of community construction. In Rashīd al-​Dīn’s Jāmiʿ al-tawārīkh (Compendium of Chronicles, fourteenth century) for example, the translation of the term as banda ‫( هدنب‬lit. “slave,” but also “bound, included,” or someone “subject to orders”)28 expresses both the idea of the service of an inferior toward a superior, as well as the idea of a bond or relationship of subordination, both of which characterized this institution.29 Thus, the boghol was not a slave proper, but someone who fulfilled a bond of submission by providing a service to the ruler. This corroborates the view of Yekemingghadai Irinchin, who traces the origin of this institution to the Khitan period (eleventh–thirteenth century) and shows how the term boghol designated a sort of “labour force,” entitled to have properties, families, and even some personal freedom.30 The term ötegü boghol (“old servant”) on the other hand is reported by Rashīd al-​Dīn in transcription and not in translation,31 signifying that this should be treated as “Relations of Domination and Submission: Political Practice in the Mongol Empire of Chinggis Khan,” in Imperial Statecraft: Political Forms and Techniques of Governance in Inner Asia, Sixth–​ Twentieth Centuries, ed. David Sneath (Bellingham, 2006), 85–​116; Yekemingghadai Irinchin, “Regarding the Mongol Bo ‘ol in the 11th and 12th Centuries,” in Chinese Scholars on Inner Asia, ed. Xin Luo and Roger Covey (Bloomington: Indiana University Press, 2012), 315–​30; Christopher P. Atwood, “Mongols, Arabs, Kurds, and Franks: Rashīd Al-​Dīn’s Comparative Ethnography of Tribal Society,” in Rashīd Al-​Dīn as an Agent and Mediator of Cultural Exchanges in Ilkhanid Iran, ed. Anna Akasoy, Ronit Yoeli-​Tlalim, and Charles Burnett (London: The Warburg Institute, 2013), 223–​50. The issue is also briefly reviewed in: Michal Biran, “Forced Migrations and Slavery in the Mongol Empire (1206–​1368),” in The Cambridge World History of Slavery, ed. Craig Perry et al. (Cambridge: Cambridge University Press, 2001), 76–​99. 27 Analyzed in Skrynnikova, “Boghol”, esp. 315–17; Skrynnikova, “Relations of Domination and Submission”; Irinchin, “Mongol Bo’ol,” esp. 320–​21.

28 Francis J. Steingass, A Comprehensive Persian-​English Dictionary, Including the Arabic Words and Phrases to Be Met with in Persian Literature (London: Routledge & K. Paul, 1892), 202. An example of banda is the case of Arghun Aqa becoming subject to the orders (lit. “slave and servant” banda wa khidmatgār) of Ilüge, a major commander of Ögödei (Rashīd al-​Dīn, Compendium of Chronicles, tr. Thackston, 1998, 1:33 and 51). 29 As analyzed by Skrynnikova, “Boghol,” 317. 30 Irinchin, “Mongol bo’ol,” 321.

31 “The people [qawmī] of the Mongols that they currently call ötegü boghol received this name during Chinggis Khan’s time. Ötegü boghol means that they are hereditary slaves of Chinggis Khan’s fathers and forefathers. Some of them [i.e., of such hereditary slaves] are those who performed pleasing services during Chinggis Khan’s time and established their rights and that is why they are called ötegü boghol. Those who maintain their status as ötegü boghol will be mentioned individually in the proper places” (Rashīd al-​ Dīn, Compendium of Chronicles, trans. Thackston, 1998, 1:117). This passage has been at the origin of a long debate on the correct understanding of the term ötegü boghol, on which see Skrynnikova, “Relations of Domination and Submission”; Irinchin, “Mongol bo’ol”; Atwood, “Mongols, Arabs, Kurds,” esp. 239.

198

198 Francesca Fiaschetti a different institution. The term ötegü boghol (Ch. lao nübi 老奴婢), in fact, indicated a high rank or official position, obtained after having rendered meritorious service to the imperial elite.32 Thus, these were not labourers like the boghol,33 but constituted a kind of elite due to the longstanding loyal relationship of their lineage to the imperial lineage of the Borjigid. In the Chinese sources, this is expressed, for example, by the translation of the Mongolian term as “meritorious descendants” (xunzhou 勳冑).34 Furthermore, the construction of the social class of the ötegü boghol was highly ritualized: it implied a story (real or fictive) of meritorious service or submission directly to the founder of the Mongol Empire, Chinggis Qan.35 Chinese sources stress further the new social identity expressed by this bond, reporting that the ötegü boghol were granted a new name, a device often used by Mongol rulers toward prominent non-​ Mongol elite members. This is illustrated by the Yuanshi biography of Zhang Rong 張榮 (1159–​1231), a former subject of the Jurchen Jin dynasty and, possibly, an ötegü boghol: Zhang Rong was a man of Qingzhou, but later he moved to Yanling. In the year jiaxu, he followed the Jin ruler Taobaoji in peaceful submission [to the Mongols]. … In the 15th year he led the garrison of the military artisans and followed Taizu (Chinggis Qan) in the campaign against the Western regions. In the eighth month of the 17th year, as they arrived at the Molan River, Taizu convened [his advisors] to find a plan to cross the river and Zhang Rong suggested building ships. Taizu said, “Building ships is difficult, how do you intend to do it?” Zhang Rong asked for a month’s time; he was the overseer of the works and built 100 ships with which they were able to cross the river. Taizu highly praised his skills and granted him the name of Usunči [“One who can manage the water”].36

Both the bond of submission and the high official position related to the status of ötegü boghol were conferred upon Zhang Rong’s offspring, as testified by the choice of their personal names, as well as by their careers. The Yuanshi, in fact, further reports that: 32 See the example in Irinchin, “Mongol bo’ol,” 326.

33 Irinchin, “Mongol bo’ol”, 326–​27 shows that although some of the boghol could become ötegü boghol, there was no direct relationship between the two categories. 34 Atwood, “Mongols, Arabs, Kurds,” 238–​39.

35 Skrynnikova, “Boghol,” 317. It is to note that Irinchin, “Mongol bo’ol,” 326, however, found one case in which the Chinese sources mention a ötegü boghol who didn’t have a direct connection with Chinggis Qan.

36 張榮,清州人,後徙鄢陵。歲甲戌,從(金)太保明安降. […] 戊寅,領軍匠,從太祖征西域 諸國。庚辰八月,至西域莫蘭河,不能涉。太祖召問濟河之策,榮請造舟。太祖復問:「舟卒 難成,濟師當在何時?」榮請以一月為期,乃督工匠,造船百艘,遂濟河。太祖嘉其能,而賞 其功,賜名兀速赤。Yuanshi 151:3581, trans. Geoffrey Humble. See also Qiu Yihao’s work, dealing in detail with another notable ötegü boghol, the commander Dordoqa: Qiu Yihao, “The Scenario of a Conflict between Qaidu and Qubilai in Mamluk Chronicles,” paper presented at the international workshop: Mongol Warfare between Steppe and Sown: Military, Social and Cultural Perspectives, Sofia, May 26, 2017.

199

Faithful to a Vanishing Past

199

His son was Nubi 奴婢 (lit. ‘slave’, but also ‘boghol’), was granted a tiger tally and held the office of Supreme Commander of the maritime artillery. He was the overseer of all military artisans. Similarly, Nubi’s son Junyou 君佐 (“one who assists the ruler”), was granted a tiger tally and held the office of Supreme Commander of the maritime artillery.37

It is interesting to note that these examples of ötegü boghol subjects are connected to the ritual of name-​giving, a rite of passage shared by many communities as a device to include new members.38

Facing the Song

When expanding their power in China and after the establishment of the Yuan dynasty in the 1260s, a major challenge for the Mongols was their relationship with the Song dynasty. Decades of military confrontations between the two polities deeply affected the sense of loyalty and belonging of communities in South China and Southeast Asia in general. Participation in major battles leading to the conquest of the Song became fundamental events on which to build the collective memory of the Yuan subjects. In these battles, Mongol military elites were supported by new ethnic groups who saw in their defection to the Yuan a better chance for their career and private interests. Notable examples include the Persian merchant Pu Shougeng 蒲壽庚(fl. ca. 1250–​1281), who was instrumental in the Mongol conquest of Quanzhou.39 Similarly, the defection to the Mongols of the Han Chinese general Liu Zheng 劉整 (1211–​1275) played a determining role in the Mongols’ implementation of naval warfare in the battle of Xiangyang (1268–​ 1273).40 Because these battles against the Southern Song constituted the reference point around which to build and legitimize membership in the Yuan political and administrative elites, they performed the same function in the East as the campaigns against Khwarezm and Europe did in the West. Moreover, Mongol warfare impacted maritime Asia through the spreading and strengthening of the Han communities, who remained loyal to the Song, along the shores of Southeast Asia. Military refugees were particularly relevant, as Song military leaders fled together with their troops, introducing into Southeast Asia new technologies of 37 子奴婢,襲佩虎符、砲水手元帥,領諸色軍匠。[…] 子君佐,襲佩虎符、砲水手元帥,戍蔡 州。五年,都元帥阿朮,命將砲手兵攻襄陽。至元八年,調守襄陽一字城、槖駝嶺,攻南門牛 角堡,破之。攻樊城,親立砲摧其角樓,樊城破。Yuanshi 151:3582–​83, trans. Geoffrey Humble.

38 See, for example, in the study by Arnold van Gennep, The Rites of Passage (Chicago: University of Chicago Press, 1960), 67f. 39 As recorded in Yuanshi 9:191. For a recent analysis of his career and legacy see John W. Chaffee, “Pu Shougeng Reconsidered: Pu, His Family, and Their Role in the Maritime Trade of Quanzhou,” in Beyond the Silk Roads: New Discourses on China’s Role in East Asian Maritime History, ed. Angela Schottenhammer and Robert J. Antony (Wiesbaden: Harrassowitz, 2017), 63–​76.

40 See his biography in: Yuanshi 161:3785–​88. The battle is analyzed in Lo Jung-​pang, China as a Sea Power, 1127–​1368: A Preliminary Survey of the Maritime Expansion and Naval Exploits of the Chinese People during the Southern Song and Yuan Periods, ed. and annotated by Bruce A. Elleman (Singapore: NUS Press, 2012), 218–​46.

200

200 Francesca Fiaschetti warfare, and playing a major role in resisting the Mongol expansion into Southeast Asia. So, for example, the Chinese sources report that, in the campaign of 1285 against Đại Việt (1054–1804), the Mongols encountered so many Song troops as part of the Vietnamese army that they were discouraged from fighting by the thought that the Song had arrived all the way there.41 Even though we do not have precise figures regarding Han migration to Southeast Asia, we can infer its scale from the fact that, from the Yuan period onward, Han communities appear in the sources more and more frequently in the territories of Vietnam, Java, Malaysia, and South India.42 In response to these dynamics, Yuan emperors and diplomats frequently mentioned their military success against the Song in their correspondence with their neighbours in the East, as a way of legitimating the Mongol imperial project in maritime Asia. This was the case, for example, in Qubilai’s (r. 1260–​1294) diplomatic correspondence with the Kingdom of Koryŏ (918–​1392): At the moment in the entire world only you and the Song haven’t submitted. The Song relied on the Changjiang, but the Changjiang is not an obstacle anymore. They relied on the provinces Chuan (Sichuan) and Guang (Guangzhou), but Chuan and Guang are no pillars anymore. The border guards spontaneously retreated from their posts. The imperial army lies already on their chest; like a fish in the offering bowl, or a swallow in a tent, they will be prostrate in the time between sunrise and sunset.43

Similarly, in an edict to the ruler of the Trần dynasty (1226–​1400) in Vietnam, the Yuan ruler Qubilai (r. 1260–1294) states: You were the loyal subjects of the Song. How do you value the strength [of the Song] now that they have been defeated?44

Even when dealing with Liuqiu 瑠求 (present-​day Taiwan), Qubilai’s edict states:

It has been already seventeen years since we annexed and pacified Jiangnan, and among all the foreign countries overseas there is none that hasn’t pledged alliance to Us. Only Liuqiu, near the border of Min [Fujian], has never turned to Us and submitted.45

Whereas this kind of rhetoric was a necessary, common type of language in the maritime Asia shaped by Song networks, it should be also mentioned that for various 41 See on this: Hok-​lam Chan, “Chinese Refugees in Annam and Champa at the End of the Sung Dynasty,” Journal of Southeast Asian History 7 (1966): 1–​10.

42 The sources and some examples of famous refugees are surveyed in: Lo Jung-​pang, China as a Sea Power, 326–​29 and Hok-​lam Chan, “Chinese Refugees.”

43 今也,普天之下未臣服者,惟爾國與宋耳。宋所恃者長江,而長江失險;所藉者川、廣,而 川、廣不支。邊戍自徹其藩籬,大軍已駐乎心腹,鼎魚幕燕, 亡在旦夕。Yuanshi 208:4610–​11. 44 爾嚐臣事亡宋,自揆氣力何如?Zhiyuan ershiwunian ershiyue yu Annan shizi zhao 至元二十五 年十二月諭安南世子詔(Lê Tắc, Annan zhilüe 2:51–​52).

45 詔曰:「收撫江南已十七年,海外諸蕃罔不臣 210:4667.

屬。惟瑠求邇閩境,未曾歸附。Yuanshi

201

Faithful to a Vanishing Past

201

countries in maritime Asia the Mongols also represented a new player, whose disruptive military power set the conditions for a shift in the local political dynamics. Effective resistance to the Mongols has in fact been found by scholarship to be one of the foundational elements of many existing and new political identities and states in Southeast Asia.46 As such, it became a fundamental moment in the construction of the historical memory of these countries, as seen for example in their works of literature and epics. The most famous examples in this regard are present-​day Myanmar and Java. In Myanmar, the confrontation with the Mongols exposed the internal weakness of the Pagan dynasty (849–​1297) in North Myanmar. It ended in fact with the assassination of the ruler Narathipate (r. 1238–​1287) immediately after his submission to the Mongols. The turmoil that followed led to the rise of the short-​lived Myinsaing Kingdom (1297–​ 1313), which, for the first time, managed to shift the power southwards and unify the territory of Central Myanmar.47 In Java, the Mongol force was instrumentalized to settle the internal dispute between the kingdom of Kediri (which took power through a coup in 1292) in the East, and Raden Wijaya (r. 1293–​1309), heir of the kingdom of Singhasari (1222–​1292), in Southeast Java, who had been overthrown by the same coup. Because the Mongols were unaware of the local context, they were tricked by Raden Wijaya into supporting him against Kediri, a fact that contributed to the creation of the Majapahit Empire (1293–​ 1527), which soon became one of the most powerful polities in Southeast Asia. As these examples show, alliance with and rebellion against the Mongols played a role also in thirteenth-​ and fourteenth-​century maritime Asia, as the response to Mongol military actions deeply shaped the formation of later, early modern dynamics in this macro-​ region. The confrontation with the Mongols became such a foundational chapter in the history of Java that the account of it was revisited and fictive elements were added in various Javanese literary and historiographical works.48 The case of Vietnam is similar. Its successful resistance to the Mongol attacks was widely celebrated in later historical accounts as an important moment of state formation.49 46 This has been analyzed by James A. Anderson, “Man and Mongols: The Dali and Đại Việt Kingdoms in the Face of the Northern Invasion,” in China’s Encounters on the South and Southwest: Reforging the Fiery Frontier over Two Millennia, ed. James A. Anderson and John K. Whitmore (Leiden: Brill, 2014), 106–​34.

47 For an overview of Mongol–​Burmese relations comparing Chinese and local sources, see Geoff Wade, “An Annotated Translation of the Yuan Shi Account of Mian (Burma),” in The Scholar’s Mind: Essays in Honor of Frederick W. Mote, ed. Eugene Perry Link (Hong Kong: Chinese University Press, 2009), 17–​49. 48 As analyzed in David Bade, Khubilai Khan and the Beautiful Princess of Tumapel: The Mongols between History and Literature in Java (Ulaanbaatar: A. Chuluunbat, 2002).

49 For an overview of this issue and related sources: Vu Hong Lien and Peter Sharrock, Descending Dragon, Rising Tiger: A History of Vietnam (London: Reaktion Books, 2014).

202

202 Francesca Fiaschetti

Conclusion In East Asia under Mongol rule, the construction of identities paralleled the construction of the empire. In shaping their own and foreign identities, the Mongols were, in fact, at the same time constructing and continuously narrating their own history, from the Western Campaigns to the defeat of the Song dynasty. Labelling and the creation of identities was the product of a synergy of concepts and structures: Inner Asian strategies of empire building were adopted by the Mongols, together with Chinese concepts and traditions. However, the variety of terms used by sources of the period to label the multi-​ethnic groups and communities of the Mongol Empire all point toward a similar criterion: the preference for political and military loyalty over ethnicity, as elements to create membership and cohesion within a group. In connection with questions of identity and peoplehood, it is exactly in the flexibility resulting from a such dynamic that we can find the main contribution of the medieval Mongols. By stressing loyalty over ethnicity or culture, they allowed individuals and groups to find their place within or outside the imperial project. For those who chose to belong to the imperial project, on the one hand, individual and collective identifications were constructed through the combined efforts, across time, cultures, and languages, of many actors, all of whom looked at military rhetoric and shared military experiences as fundamental elements of their collective memory. As a process imposed from above by the ruling elite, for example, labelling individuals as Mongols or “new Mongols” served the administrative purposes of the empire, and was mainly centred in the distinction between elites (or aristocracy) and commoners. As a result of the needs, initiative, and agency of the subjects of the Mongols, their display of cultural elements pertaining to various communities instead reflected the multiculturalism and mobility which characterized their career and personal paths. On the other hand, for those who remained outside the Yuan imperial project—​ either by fleeing or offering resistance—​the military confrontation with the Mongols became, too, a constituting moment in their historical consciousness, and something which shaped their way of representing themselves for centuries to come.

203

Chapter 12

WAR AND COLLECTIVE IDENTIFICATIONS IN MEDIEVAL SOCIETIES: DRAWING COMPARISONS Yannis Stouraitis As I have already stated in the introduction of the current book, our cross-​cultural

approach to the relationship between war and collective identities in the period between the sixth and the fifteenth centuries does not claim to offer a wholistic coverage of the topic. In this light, my effort in the current chapter to draw some comparative conclusions from the presented case studies cannot do justice to many aspects of that relationship in the period under examination; especially if we consider the significant number of cultures that were inevitably left out of the picture. The goal when we started this dialogue was, of course, not to state the obvious, namely warfare’s diachronically important role in the processes of configuration or renegotiation of different types and visions of community. It was rather to problematize the multifaceted character of war’s impact on high-​level identifications. The individual chapters speak to each other in many ways when it comes to the question of when and how warfare in the form of interstate conflict, conquest, or civil war converged with institutions, political ideologies, religion, and ethnicity, or informed practices of memorialization in order to shape collective attachments, enhance or weaken groupness, and construct otherness. The role of religious identifications and religious–​political proto-​ideologies looms large in the two case studies on civil wars. The well-​studied events of the Albigensian Crusade testify to civil war in medieval western Europe as a phenomenon that was circumscribed by the imagined community of western-​European Christendom, attributing a secondary role to the state and/​or shared ethnicity.1 Even though the nominal kingdom of France provided the political-​territorial scene of the conflict that has been interpreted as an attack of the northern French against the Occitan elites, the southerners distinguished themselves from the French ethnicity of the northerners. Philippe Buc highlights the two main visions of identity that emerge from the narrative sources: the first, a product of Crusading ideology and practice, was a vision employed to demarcate the opposite parties in terms of Catholic versus Cathar and pious versus sinful in order to justify the actions of war; even though such distinctions were by far not clear-​cut on the ground. The other referred to local identification with the city 1 On the events of the Albigensian Crusade see: Michael D. Costen, The Cathars and the Albigensian Crusade (Manchester: Manchester University Press, 1997); Jörg Oberste, Der Kreuzzug gegen die Albigenser: Ketzerei und Machtpolitik im Mittelalter (Darmstadt: Primus, 2003); Elaine Graham-​ Leigh, The Southern French Nobility and the Albigensian Crusade (Woodbridge: Boydell & Brewer, 2005); Laurence W. Marvin, The Occitan War: A Military and Political History of the Albigensian Crusade, 1209–​1218 (Cambridge: Cambridge University Press, 2009).

204

204

Yannis Stouraitis

of Toulouse, the very target of the crusade and centre of the Languedoc. The conflict reinforced attachment to the city, demarcating its pious defenders, the Toulousain, from the impious Crusaders and the northerners. Due to its situational character as a direct consequence of the war, this vibrant identification regressed eventually in its aftermath due to the lack of a dynasty with local roots that would use the memory of resistance in order to capitalize on local attachments. The Armagnac–​Bourguignon civil war was a rather different affair. The framework of internal armed conflict was evidently set by the state insofar as this was a conflict between two powerhouses, the Valois and the Orléans, for control over the policies of the kingdom of France.2 As Buc shows, religious identity discourse was not employed to demarcate the otherness of the two sides in the conflict, but rather to delegitimize the opposite party. Frenchness was thus connected with proper Christian behaviour during the war, as—​for instance—​the discourse that paralleled the Armagnac with the infidel Saracens due to their war deeds exemplifies. In the context of the religiously charged representation of the actions of the rival parties, Buc zooms in on pro-​Burgundian insurrections in Paris, in particular the Cabochien uprising in 1413. He shows how the war contributed to a local Parisian identity becoming salient. This identity was by no means coherent and monolithic but eventually came to supplant allegiances to the two major parties of the civil war in favour of the interests of different social groups within the city that interpreted the goal of local peace differently. In the Hussite civil war, religious controversy and the supra-​ethnic political framework of western European Christendom and the Holy Roman Empire set the tone again.3 Historians have focused on the role of the Crusades against the religiously dissident Hussites as a factor that reinforced Czech ethnic or national identity.4 In fact, the case of the Hussites pops up in the aforementioned ethno-​symbolist study of John Hutchinson as an example of premodern nationalism.5 Buc argues for a more cautious approach to the relationship between nationalism and the Hussites’ religious aspirations, declaring his preference for the term patriotism, instead, as a notion stemming from the local patria and, hence, less amenable to anachronism. 2 Bertrand Schnerb, Les Armagnacs et les Bourguignons: La maudite guerre (Paris: Perrin, 2009); Simona Slanička, Krieg der Zeichen: Die visuelle Politik Johanns ohne Furcht und der armagnakisch-​ burgundische Bürgerkrieg (Göttingen: Vandenhoeck & Ruprecht, 2002). 3 Frantisek Šmahel, Die Hussitische Revolution, 3 vols. (Hannover: Hahnsche Buchhandlung, 2002); Thomas A. Fudge, The Magnificent Ride: The First Reformation in Hussite Bohemia (Aldershot: Ashgate, 1998); Frederik G. Heymann, John Žižka and the Hussite Revolution (Princeton: Princeton University Press, 1955); Howard Kaminsky, A History of the Hussite Revolution (Berkeley: University of California Press, 1965).

4 Frantisek Šmahel, “The Idea of the ‘Nation’ in Hussite Bohemia,” Historica 16 (1969): 143–​247; Šmahel, “The Idea of the ‘Nation’ in Hussite Bohemia,” Historica 17 (1969): 93–​197; Noman J. Housley, “Pro Deo et Patria Mori: Sanctified Patriotism in Europe, 1400–​1600,” in War and Competition between States, ed. Philippe Contamine (Oxford: Clarendon Press, 2000), 221–​48. 5 Hutchinson, Nationalism and War, 24.

205



War and Collective Identifications

205

The Crusades against the Hussites contributed to the reinforcement of pre-​existing notions of Czech ethnicity based on beliefs of election. In the course of the conflict, the latter were articulated in terms of good versus bad Christians, with the papacy and the holy Roman emperor being cast as bloodthirsty and unchristian. However, Hussite patriotism promoted integration mainly in and for Prague, the centre of intellectual power of the Hussite movement, whereas its religious incentive evidently excluded the Catholic part of the “nation.” The civil war was not between the Czechs and the non-​ Czech Crusaders; it was equally a conflict between Hussite Czechs and Catholic Czechs that sided with the outsiders of the crusade. Hence, religious beliefs were a means for accentuating a Czech identity, but also the cause of its disintegration. Hussite religiously inspired patriotism had little to do with nationalism as a unifying factor, that is, as a dominant operative ideology fostering national cohesion. Compared to Buc’s examples from late-​medieval western Europe, in the Byzantine case study it was not religious controversy but the need for the reproduction of the centralized imperial state that overdetermined the image of civil war and the discourses of collective identification emerging in its context.6 The period under examination is divided in two main phases: the time up to 1204 during which the East Roman imperial order remained highly centralized under the sole rule of the emperor in Constantinople, and the time between 1204 and the fifteenth century when political fragmentation dominated in the territories of the former empire.7 The first part of the chapter zooms in on the terminology of internal armed conflict in Byzantine sources. The terms used to denote a rebellion, whether the latter concerned an attempt to usurp the imperial rule or an attempt to secede from it, testify to the centrality of the notion of loyalty and subordination to the imperial power for membership in a political community circumscribed by the limits of enforceable imperial authority. Usurpers and secessionists were equally viewed and identified as apostates from imperial rule irrespective of their ethno-​cultural categorization or identification. The goal of the rebellion (usurpation or secession), however, was an indicator of the rebels’ intention to continue to identify with Roman centralized imperial authority or not. In this context, the practice of Byzantine authors to confine the term denoting war within the community (emphylios polemos) exclusively to conflicts resulting from rebellions of usurpation testifies to a strategy of collective identification that 6 On civil war in Byzantium: Walter E. Kaegi, Byzantine Military Unrest, 471–​843: An Interpretation (Amsterdam: Hackett, 1981); Jean-​Claude Cheynet, Pouvoir et Contestation à Byzance, 963–​1210 (Paris: Publications de la Sorbonne, 1990); Yannis Stouraitis, “Civil War in the Christian Empire,” in A Companion to the Byzantine Culture of War, ca. 300–​1204, ed. Yannis Stouraitis (Leiden: Brill, 2018), 92–​123.

7 On the post-​1204 Byzantine world see Michael Angold, “The Greek Rump States and the Recovery of Byzantium,” in The Cambridge History of the Byzantine Empire, c. 500–​1492, ed. Jonathan Shepard (Cambridge: Cambridge University Press, 2008), 731–​58; Angeliki Laiou, “The Palaiologoi and the World Around Them (1261–​1400),” in The Cambridge History of the Byzantine Empire, c. 500–​1492, ed. Jonathan Shepard (Cambridge: Cambridge University Press, 2008), 803–​33.

206

206

Yannis Stouraitis

defined political Romanness as an ideological practice of attachment to a centralized imperial order and of commitment to its reproduction. War with rebels that aimed at secession acquired the image of interstate warfare, even when the seceding party was ethno-​culturally categorized as Roman. This is exemplified by the way the concept of emphylios polemos was employed in the period after 1261, when populations ethno-​ culturally categorized as Romans had organized themselves in separate regional polities and were not willing to submit to the centralized authority of the Roman emperor in Constantinople in order to be reintegrated into the Roman political community. War between those regional polities and the Constantinopolitan imperial state was not conceptualized as Roman civil war, but as interstate warfare insofar as anyone who did not identify with the notion of centralized Roman imperial rule and was not a subject of the emperor was practically and ideologically excluded from the image of Roman political peoplehood. Within this framework, the identity discourse of the accounts of Byzantine civil wars, explored in the second part of the chapter, demonstrates the central role of a shared Christian faith in providing a common religious-​cultural background that underpinned political attachment well into the High Middle Ages.8 The language of micro-​solidarity that accentuated family bonds and local attachments in order to highlight the horror of internal armed conflict translated into notions of Christian brotherhood at the broader level of the empire. Christian discourse in plain form, free of any doctrinal-​theological concerns of orthodoxy and heresy, was a main ideological means for propagating an image of homogeneity, supplanting the different regional and ethnocultural backgrounds, and the various doctrinal affiliations of the imperial subjects that formed the opposite parties in the civil wars. The Christian faith was, also, a main means to demarcate the members of the Roman political community from the major non-​Christian neighbouring rivals of the empire, the Bulgars (before they were Christianized in the 860s) and the Muslims—​both often involved as allied forces in Byzantine civil wars. Given that Christian identity discourse had adopted the language of ethnicity at its early stages,9 the ethnoreligious image of the imperial community, which the interchangeable use of the labels Roman and Christian by Byzantine authors for the collective designation of imperial subjects testifies to, facilitated the translation of a discourse of metaphoric kinship to a supra-​ethnic political community, in which the imperial subjects’ ethnocultural categorizations did not inform their political attachment. In this context, the discourse of Byzantine authors that demarcated a Roman ethno-​ cultural category within the empire had very little to do with the transformation of the 8 On the role of Christian identity in Byzantium: Hans-​Georg Beck, Die byzantinische Jahrtausend (Munich, 1978: reprint 1994), 11–​29, 34–​45, 87–​108; Yannis Stouratis, “Romanness and Chalcedonian Christianity in Byzantium,” in Ethnicity and Religion in the Middle Ages, ed. Walter Pohl, Ingrid Hartl, and Gerda Heydemann (Leiden: Brill, forthcoming).

9 Denise K. Buell, Why This New Race: Ethnic Reasoning in Early Christianity (New York, 2005); Todd Berzon, “Ethnicity and Early Christianity: New Approaches to Religious Kinship and Community,” Currents in Biblical Research 16 (2018): 191–​227.

207



War and Collective Identifications

207

image of Byzantine civil war into that of a conflict within an ethnic or ethnopolitical community. Internal conflict remained bound to the vision of a political community of imperial subjects, whose political Romanness was not a matter of shared ethnicity but of a centralizing ideology promoting identification with the Constantinopolitan ruler’s imperial monocracy. The presented case studies of civil war in late-​ medieval western Europe and Byzantium, respectively, draw our attention to the different function of the phenomenon in different sociopolitical contexts. In western Europe, it was a shared religious identity transcending political and cultural boundaries which, more often than not, set the general framework of internal armed conflict, with the state having a secondary role. In Byzantium, instead, it was the strong centralized state that determined the framework. The main result of this difference was that civil war reinforced different kinds of identification. These were primarily religious-​doctrinal and regional-​local in the west, whereas in Byzantium internal armed conflict over the throne reinforced a dominant Roman political discourse of ideological attachment and loyalty to centralized imperial rule, which was underpinned by a shared religious identity. The chapter on the development of Capetian Frankish or French identity in the High Middle Ages into a diaspora identity in the areas occupied by the Crusaders across the Mediterranean provides an alternative insight into the relationship between war, religion, and ethnicity. The intertwining of French identity and the Crusading movement—​a popular modern theme—​goes back to the eleventh century when the “Franks” or “French” from the northern regions, especially around the city of Paris, viewed themselves as bound by a shared, mutually intelligible langue d’oïl and as particularly linked to the defence of Christendom. In this context, French epic culture deprived the historical figure of Charlemagne of any Germanic background and appropriated him as the emperor of the Franks of “France dulce,” attributing a Capetian identity to a Carolingian ruler.10 Moreover, it attributed to the French a distinct role in the defence of Christendom already before the Crusades, depicting the protagonist of the Chanson de Roland as a fighter on behalf of his French homeland and Christianity. The sources written in the aftermath of the First Crusade testify that during the movement, “Frankishness” or French identity had become inclusive categories that integrated Christian warriors of various ethnic backgrounds.11 Given that in twelfth-​ and thirteenth-​ century European courts French was the main medium of elite communication, Crusading sources seem to reflect the linguistic reality among the royal and baronial leadership of the movement. The polities that emerged in Syria (Outremer) 10 Matthew Gabriele, An Empire of Memory: The Legend of Charlemagne, the Franks, and Jerusalem before the First Crusade (Oxford: Oxford University Press, 2011).

11 On French identity in the Crusading movement: Marcus G. Bull, “Overlapping and Competing Identities in the Frankish First Crusade,” in Le concile de Clermont et l’appel à la croisade, ed. André Vauchez (Rome: École Française de Rome, 1997), 195–​211; Sh. Kinoshita, “Crusades and Identity,” in The Cambridge History of French Literature, ed. William Burgwinkle, Nicholas Hammond, and Emma Wilson (Cambridge: Cambridge University Press, 2011), 93–​101.

208

208

Yannis Stouraitis

after the First Crusade,12 in the fragmented territories of the Byzantine Empire after the Fourth Crusade,13 in the Midi (south France) after the Albigensian Crusade,14 and in Sicily after 1266,15 were governed by ruling elites that spoke French. These polities formed part of a broader Francophone community that was culturally linked with France through language but also through various notions of a shared heritage. Collections of legal customs and precedents contributed to that end. The case of the “Letres dou Saint Sepultre” from the kingdom of Jerusalem in the thirteenth century and the Uxanze de Romania from the Frankish principality of Achaea in the fourteenth century exemplify how such collections preserved a “French” legal tradition while serving as a means of identification with the glorious Crusading past, thus dignifying their respective independent political societies. Most importantly though, law was a basic means to draw boundaries in terms of rights and privileges between the French ruling elites and the indigenous populations in the conquered areas. The indigenous Muslim and Christian populations in the Outremer, the Greek-​speaking Christians of medieval Greece, as well as the Cathar heretics of the Midi were legally demarcated from their French masters due to their linguistic and/​or confessional diversity, whereas Catholics in the Midi or in Sicily enjoyed a more integrated status.16 Lippiat shows that, besides demarcating themselves from the local populations, the conquerors circumscribed their “Frenchness” through common attitudes toward marriage and inheritance. Legal collections testify to the effort to maintain a balance 12 Malcolm Barber, The Crusader States (New Haven: Yale University Press, 2012); Charles MacEvitt, The Crusades and the Christian World of the East: Rough Tolerance (Philadelphia: University of Pennsylvania Press, 2008); Ralph-​Johannes Lilie, Byzanz und die Kreuzfahrerstaaten: Studien zur Politik des Byzantinischen Reiches gegenüber den Staaten der Kreuzfahrer in Syrien und Palästina bis zum vierten Kreuzzug (1096–​1204) (Munich: Fink, 1981); Engl. transl. Jean E. Ridings, Byzantium and the Crusader States 1096–​1204 (Oxford: Oxford University Press, 2004); Hans E. Mayer, ed., Die Kreuzfahrerstaaten als multikulturelle Gesellschaft: Einwanderer und Minderheiten im 12. und 13. Jahrhundert (Munich: Oldenbourg Wissenschaftsverlag, 1997).

13 Peter Lock, The Franks in the Aegean, 1204–​1500 (London: Routledge, 1995); David Jacoby, “The Latin Empire of Constantinople and the Frankish States in Greece,” in The New Cambridge Medieval History, vol. 5: c.1198–​c.1300, ed. David Abulafia (Cambridge: Cambridge University Press, 1999), 525–​42; Filip van Tricht, The Latin Renovatio of Byzantium: The Empire of Constantinople (1204–​1228) (Leiden: Brill, 2011); Benjamin Hendrickx, “Les duchés de l’Empire latin de Constantinople après 1204: origine, structures et statuts,” Revue belge de philologie et d’histoire 93 (2015): 303–​28. 14 Gregory E. M. Lippiatt, Simon V of Montfort and Baronial Government, 1195–​1218 (Oxford: Oxford University Press, 2017). 15 Jean Dunbabin, The French in the Kingdom of Sicily (Cambridge: Cambridge University Press, 2011).

16 On the Outremer: Benjamin Z. Kedar, “On the Origins of the Earliest Laws of Frankish Jerusalem: The Canons of the Council of Nablus, 1120,” Speculum 74 (1999): 310–​35. On Latin Greece: David Jacoby, “The Encounter of Two Societies: Western Conquerors and Byzantines in the Peloponnesus after the Fourth Crusade,” The American Historical Review 78 (1973): 873–​906, at 885f.; and Lock, Franks, esp. chap. 11. On the Midi: Lippiatt, Simon V of Montfort and Baronial Government. On Sicily: Dunbabin, The French in the Kingdom of Sicily.

209



War and Collective Identifications

209

between the principle of consent in marriage and the need of the conquerors to defend their status and their lands as an embattled social minority when it came to the choice of husbands for fief-​holding ladies, either heiresses or widows. In a similar vein, the issue of inheritance was controlled through restrictions concerning the alienation of fiefs or the provision of alms from an estate, whereas primogeniture was a major common feature of inheritance custom in the French diaspora. Its adaptation to local circumstances and traditions aimed primarily at securing that those best fitted for war would remain in control of the fiefs and of sufficient financial power in order to defend the interests of the rulers against the ruled. If the customaries of the French Mediterranean testify to a common cultural heritage and a shared vision of righteous rulership, these were reinforced by the relationship of the French diaspora with the Church. The independence of ecclesiastical revenue as reflected in the endowment of churches and the restitution of the tithes became common practice in the conquered lands. The “neogregorian” reform of the late twelfth century, promoting the liberty of the churches and the moral correction of lay behaviour, informed the French identity of the communities of Latin Greece and the Midi in the thirteenth century. The latter represented a development of the shared Frankish identity that had emerged out of the First Crusade. The Outremer’s French identity took full shape only in the thirteenth century as a result of the networks formed with the French principalities that emerged from the Crusades of this period: the Fourth Crusade (1204), the Albigensian Crusades (1209–​1229), and the Sicilian Crusade (1266). In light of the above, the French identity of an embattled diaspora in the late-​ medieval Mediterranean appears to have emerged as a byproduct of warfare, in particular Crusading warfare. Frenchness, as an inclusive identity that developed out of the lingua franca of west European elites and an intimate association with the defence of Christendom, had been transformed by the thirteenth century into a Mediterranean-​ wide shared ethnocultural heritage of ruling elites in Crusader principalities that needed to demarcate themselves from the conquered populations while defending their possessions. If the previous three case studies point to the convergence of war and religion, in particular Christianity, in the formation of different collective identities in different geopolitical contexts, the Song–​Jurchen conflict (1125–​1234) provides a different perspective on war and identity in the Southern Song Empire.17 Shao-​yun Yang begins his argument by pointing to “filial revanchism” as the main notion employed by Song intellectuals to propagate the reconquest of the northern lands from the Jin dynasty of the Jurchen. This was a notion that went back to the Confucian classics’ principle that the sons of an unjustly killed man have the filial duty to avenge him by killing those responsible. This idea was adapted to a patrimonial image of the Song imperial state 17 On the conflict see Peter Lorge, War, Politics and Society in Early Modern China, 900–​1795 (London: Routledge, 2005), 50–​57; Jing-​shen Tao, “The Move to the South and the Reign of Kao-​ tsung,” in The Cambridge History of China: Volume 5, The Sung Dynasty and Its Precursors, 907–​1279, ed. Paul J. Smith and Dennis C. Twitchett (Cambridge: Cambridge University Press, 2009), 556–​643.

210

210

Yannis Stouraitis

based on the notion that the emperor was a father to his subjects who had the moral obligation to avenge him, even if a hundred generations had passed. Hence, the death of the emperors Huizong and Qinzong in captivity alongside nearly the entire Song imperial clan after the capture of Kaifeng by the Jurchen in 1127 provided indefinite ideological incentive and justification for the reconquest of the north. The notion of avenging the dynasty fuelled irredentism while overshadowing contemporary views of the traditional Chinese–​barbarian dichotomy as being complementary instead of oppositional. “Filial revanchism” remained an active ideological incentive for reconquest even after the peace treaty of 1142 and—​in particular—​the revised treaty of 1164–​1165 whose terms were more favourable for the Song, despite the fact that by that time the concept’s pretensions had been relatively compromised by the death of the guilty parties on the Jin side. The Mongol offensive against the Jin under the leadership of Chinggis Qan in 1214 triggered new appeals for revenge among the Southern Song literati. Even though the destruction of the Jin empire by the Mongols with Song support in 1234 opened the way for Mongol incursions into Song territory, the Southern Song literati continued to maintain that the decision to back up the downfall of the Jurchen had been correct because “filial revanchism” was the means to exemplify the moral difference and superiority of the Chinese over the barbarians. The latter notions are related to the interpretation of the Jurchen invasion as the result of moral barbarism within the Song empire, an interpretation based on the Neo-​ Confucian thought that all barbarian invasions in the history of the Chinese Empire had been begotten by the Chinese government’s morally barbaric behaviour. The claims that there was a causative relationship between moral decline and barbarian invasion, and that for the defeat of the barbarian enemies a reversal of the moral decline through a return to Confucian values was needed, point to the role of the conflict in reinforcing notions of Song ethnocentric moralism.18 By the late eleventh century, Neo-​Confucians entertained a notion of ethnocentric moralism according to which becoming immoral turned someone into a barbarian. Such an abstract moral approach to barbarism which disassociated it from ethnic descent and ethno-​cultural boundaries posed an evident problem for the demarcation of the Chinese from their barbarian enemies. The Song–​Jurchen conflict in the new geopolitical context of the twelfth century, in which the Jurchen as barbarian enemies had come to dominate the Chinese, exemplified this. The response to the geopolitical domination of the barbarians came in the form of an ideology of Chinese supremacism whose content was nuanced. One line of thought opted to exclude the barbarians from Neo-​Confucian universalism, presenting them as subhuman, that is, not simply as immoral by nature but also as incapable of changing their immoral barbarian ways. Another line of Neo-​Confucian thought refrained from dehumanizing the barbarians, and sought to ground Chinese supremacism in the notion 18 On the concept of “ethnocentric moralism,” see Shao-​yun Yang, The Way of the Barbarians: Redrawing Ethnic Boundaries in Tang and Song China (Seattle: Washington University Press, 2019), 21–​23.

211



War and Collective Identifications

211

of superior qi. The latter was bound to the Chinese homeland in north China, the “Central Lands,” with their exceptionally balanced qi that had made the emergence of civilization and morality possible in the first place. Both approaches complemented the concept of “filial revanchism” regarding the pursuit of the reconquest of the north from a different angle. The first posited that, despite the geopolitical regression of Chinese supremacy, the Song maintained a right to world domination due to their moral superiority vis-​à-​vis the subhuman barbarian Jurchen. It was for their own good that the barbarians should be dominated by the morally superior Chinese,19 a way of thinking common to other imperial orders such as ancient Rome20 and its continuation, the medieval East Roman Empire.21 If the morally inferior barbarians held geopolitical superiority, this threatened to throw the world order into a state of anarchy, chaos, and catastrophe. The second approach made the reconquest of the “Central Lands” an existential matter for the morally superior identity of the Southern Song. If the latter remained away from the qi of the “Central Lands” for a longer time, that would be eventually claimed by those lands’ new occupiers, the Jurchen, who would thus be able to overcome their barbarian inferiority and become civilized. Hence, the Southern Song would be deprived of both the Mandate of Heaven and the northern Chinese people’s hearts, that is, they would lose their morally superior Chinese identity. Yang pinpoints the moralistic aspect of Chinese supremacism as the main reason that irredentist notions fuelled by the latter cannot be considered as grounded on the conception of an ethnic Chinese state. Ethnocentric moralism and even qi determinism both left the door open for barbarian dynasties to have a chance at “Sinicization” and thus legitimacy. In this context, the destruction of the Jurchen was the main means to both prove and maintain the claim of Song Chinese supremacy. The latter also determined the uncompromising attitude of the Song toward the Mongols after 1234, based on a supremacist belief that they would outlast all their barbarian enemies. Ironically, the very barbarians that overpowered the Song in 1276–​1279, the Mongol Yuan empire, would propagate a civilized, but de-​ethnicized, “Chinese” identity for themselves as rulers of the “Central Lands,” casting the Song as barbarians. The spread of Mongol identity through conquest in the thirteenth century gave birth to four basic categorizations as a result of the establishment of a Mongol Empire—​the Mongols (mengguren 蒙古人), the “Central Asians” (semuren 色目人), the “Northerners” (hanren 漢人), and the “southerners” (nanren 南人)—​these represent a reflection of the process of expansion of Mongol rule that determined a hierarchy of identities 19 This notion is reminiscent of the Aristotelian idea that civilized conquerors benefitted the uncivilized conquered; Aristotle, Politics 7.1333b.38–​1334a2, ed. W. D. Ross, Aristotelis politica (Oxford: Oxford University Press, 1964). 20 Benjamin Isaac, The Invention of Racism in Classical Antiquity (Princeton: Princeton University Press, 2004).

21 Yannis Stouraitis, “Conceptions of War and Peace in Anna Comnena’s Alexiad,” in Byzantine War Ideology between Roman Imperial Concept and Christian Religion, ed. Johannes Koder and Yannis Stouraitis, Veröffentlichungen zu Byzanzforschung 30 (Vienna: ÖAW Verlag, 2012), 74–​75.

212

212

Yannis Stouraitis

interrelated with different status, positions, and privileges.22 The dominant Mongol identity was not rigid and exclusivist, as Francesca Fiaschetti argues. The Mongol label and the privileges accompanying it could be acquired by imperial subjects as a reward for important deeds in the service of the empire. In this regard, “Mongolization” at the initial stage of expansion did not involve a process of acculturation but rather the acquirement of full-​blown political rights and privileges which were not bound to ethnic descent or acculturation but to participation in military activity on behalf of the Mongol imperial rule. The inclusive nature of Mongol rule is exemplified by the boghol category.23 In the hierarchical Mongol society, boghols were ethnically non-​Mongol persons or groups offering their service to the Mongol ruling elite as a labour force, either voluntarily or by force. The ötegü boghol, on the other hand, were a distinct category of persons that had received a high-​ranking position as a reward for their service to the imperial elite. The existence of such a social class whose members’ bond with the Mongol community could be mythologized by claiming a history of family service going back to Chinggis Khan facilitated political inclusion in the Yuan empire that was underpinned by name giving. The latter was one part of the process of cultural “Mongolization” that was attested after the Mongol conquest of China and which was achieved in various ways such as learning the Mongolian language, marriage relationships, being given a Mongolian name by the emperor or one of his generals, or choosing to give a Mongol name to one’s own offspring. In the Yuan Empire, however, the practices of renaming did not only concern Han-​Chinese adopting a Mongol name for better chances of social advancement. Non-​Han people also adopted Chinese names, a process pointing to acculturation going both ways and creating an imperial framework where multiple identifications were possible. The Mongol conquest caused loyalties and collective attachments in South China and Southeast Asia to be renegotiated. The major battles that the Mongols won in order to subjugate the Southern Song were a means of integration of members from other ethnic groups, Han and non-​Han, that participated in the process, thus legitimizing the Yuan 22 On the formation of the Mongols under Chinggis Khan and their expansion: Peter Jackson, “The Mongol Age in Eastern Inner Asia,” in The Cambridge History of Inner Asia: The Chinggisid Age, ed. Nicola Di Cosmo, Allen J. Frank, and Peter B. Golden (Cambridge: Cambridge University Press, 2009), 26–​45; Timothy M. May, The Mechanics of Conquest and Governance: The Rise and Expansion of the Mongol Empire, 1185–​1265 (Madison: University of Wisconsin Press, 2004); Lorge, War, Politics, and Society, chap. 3 and 4.

23 Yekemingghadai Irinchin, “Regarding the Mongol Bo’ol in the 11th and 12th Centuries,” in Chinese Scholars on Inner Asia, ed. Xin Luo and Roger Covey (Bloomington: Indiana University Press, 2012), 315–​30; Tat’jana D. Skrynnikova, “Boghol, a Category of Submission at the Mongols,” Acta Orientalia 58 (2005): 313–​19; Skrynnikova, “Relations of Domination and Submission: Political Practice in the Mongol Empire of Chinggis Khan,” in Imperial Statecraft: Political Forms and Techniques of Governance in Inner Asia, Sixth–​Twentieth Centuries, ed. David Sneath (Bellingham: Center for East Asian Studies, 2006), 85–​116.

213



War and Collective Identifications

213

Empire and contributing to the configuration of its administrative elites.24 On the other hand, Mongol expansion triggered the movement of members of the Song elite along with Han populations toward the shores of Southeast Asia where Han communities are attested in Vietnam, Java, Malaysia, and South India. These migrants brought with them military resources in the form of manpower and technology that the local polities used to repulse Mongol incursions. In the diplomatic exchanges with these areas, the Yuan Empire presented their military successes against the Song as a proof of legitimacy of their imperial project. As we have seen in the case of the Song–​Jurchen conflict, the defeat of the Song enabled Khubilai Khan to propagate that the Mongols were bearers of a civilized Chinese identity insofar as they held the “Central Lands” and had gained the Mandate of Heaven. Being the new dominant imperial-​military power in this macro-​region, the Mongols played an important role in the renegotiation of ethnic and political boundaries in the southeast. This is exemplified by the emergence of the short-​lived Myinsaing Kingdom (1297–​ 1313) in Myanmar as a result of the conflict with the Mongols or the creation of the Majapahit Empire (1293–​1527) due to the Mongol intervention in the internal armed conflict in Java.25 The ideological mechanisms of collective identification in the Mongol Empire demonstrate similarities with those in the Byzantine Empire insofar as a shared ethno-​ cultural identity was not propagated as the main means for membership and social advancement in the political community. In both instances, a centralizing political ideology was at play as the main means that promoted collective attachment to imperial rule. In the case of the Byzantines, confessional identity supplanted ethnicity as the principal criterion for political integration that translated into ideological attachment to the rule of the emperor in Constantinople and the privileges stemming from it. In the Mongol case, political integration and ideological attachment were the result of successful military service and the collective memory of shared military experiences, respectively. Contrary to the Byzantine, Mongol, and Chinese political orders that were conceptualized as imperial political orders, the cases of the Bulgar conquest of the northern Balkans in the late seventh century and the viking conquest of parts of England in the ninth century redirect our attention to the relationship between war and identity in the context of emerging kingdoms. The nomadic background of the Bulgar elite favoured a militarized social organization as evidenced by the fact that, following in the footsteps of other nomadic qaghanates, military and political administration were inseparable in the early Bulgar state, with high-​ranking offices combining military and civilian duties. Militarization was complemented by an increasing degree of centralization of political power after the late eighth century. A main trigger for the latter development 24 On the war, see Lorge, War, Politics and Society in Early Modern China, 78–​79.

25 James A. Anderson, “Man and Mongols: The Dali and Đại Việt Kingdoms in the Face of the Northern Invasion,” in China’s Encounters on the South and Southwest: Reforging the Fiery Frontier over Two Millennia, ed. James A. Anderson and John K. Whitmore (Leiden: Brill, 2014), 106–​34.

214

214

Yannis Stouraitis

was Byzantine aggression under Emperor Constantine V (741–​775), who conducted a series of nine campaigns against the Bulgars.26 In this period, the charismatic Dulo clan that had traditionally held power in the Bulgar realm died out, something which caused political instability marked by opposition between a pro-​Byzantine and an anti-​ Byzantine faction within the Bulgar elite.27 By the end of the eighth century, a new power elite was in place whose khans affirmed their control over the military aristocracy and restructured the power relations between the leading Bulgar tribes and lineages. One of the most successful rulers of that time, Krum (r. 803–​815), was able to solidify his power due to his successes in war in the territories of the disintegrating Avar qaghanate in the west as well as in the conflict with the Byzantines in the south.28 The campaign of the Byzantine emperor Nikephoros I (802–​811), which aimed to annihilate the Bulgar kingdom, resulted in a Bulgar triumph and the emperor’s death on the battlefield, thus reinforcing the anti-​Byzantine attitudes and setting in motion a Bulgar policy of aggression. Krum’s charisma was underpinned by the prestige of victory on the battlefield and the wealth amassed through booty which enabled him to secure the loyalty of the elite through the distribution of rewards. Hence, the gains of successful warfare transformed the relationship between the ruler and the members of elite clans from quasi egalitarian and horizontal in the previous century into hierarchical and vertical. The challenges posed to Bulgar rule by the majority of their Slavic subjects as well as by the Byzantine threat led the rulers of this period to reassert their cultural links with the world of the steppes. This is exemplified by the adoption of the notion of sacral rulership and their adherence to the cult of Tangra/​Tängri, the sky-​god and supreme divine being of many Central and Inner Asian nomads. On the other hand, while the tolerant treatment toward the Christian subjects of the realm seems to have regressed during the early ninth century, the large number of Christian captives that were deported to Bulgaria as a result of the Bulgar campaigns on Byzantine soil contributed to the spread of Christianity in the realm from below before the official Christianization under Tsar Boris took place in the 860s.29 In this context, the adoption of Christian symbolism and the imitation of Byzantine rituals and public expressions of power since the early ninth century contributed considerably to the political integration of the realm’s Christian populations. The best proof of the changed nature of the Bulgar polity in the ninth century is that no khan was killed or deposed after 815. The expansion of the “khan-​centred” state in the reign of Krum’s son, Ormutag, whose elite network of power included people 26 Dennis P. Hupchick, The Bulgarian-​Byzantine Wars for Early Medieval Balkan Hegemony: Silver-​ Lined Skulls and Blinded Armies (Cham: Palgrave Macmillan, 2017), 55–​60. 27 On the various interpretations of the social and ethnic background of this controversy, Hupchick, Bulgarian-​Byzantine Wars, 61–​62. 28 Panos Sophoulis, Byzantium and Bulgaria, 775–​831 (Leiden: Brill, 2012), chap. 5.

29 Robert Browning, “Byzantines in Bulgaria—​Late 8th–​Early 9th Centuries,” in Studia Slavico-​Byzantina et Mediaevalia Europensia: In Memoriam I. Dujchev, I, ed. Petar Dinekov et al. (Sofia: Dr. Peter Beron, 1988), 32–​36.

215



War and Collective Identifications

215

from different tribes or clans, indicates an effort to shape a new group identity that transcended older social boundaries. High-​ranking positions in state administration had become open now to “outsiders” beyond the traditional nobility, including Slavs, Christians, and provincial elites. Within this framework, the increased awareness of the Byzantine Empire as the “other” was conducive to strengthening the political ties between the ethnically heterogeneous components of the Bulgar realm and reinforcing a sense of collective identification with Bulgar rule. The case study of the early medieval Bulgars provides, thus, an example of war and conquest as factors that promoted the centralization of political power among a group of nomadic clans, turning them into a sedentary society and reinforcing the stability of their state. The latter became the framework for the configuration of a shared ethnicity at least among the members of the ruling elite. A different example of the role of conquest and protracted conflict emerges from Clare Downham’s chapter on the vikings and the English in the transition from the Early to the High Middle Ages. Focusing on the conquest of parts of England in the ninth century,30 Downham zooms in on the terminology of the Anglo-​Saxon Chronicle about the vikings which recounts the events from an English perspective.31 The variety of attested labels for their categorization such as wicing (viking), hæðen (heathen), Norðmenn (Norsemen), and Denisc, the latter being the most popular term in the ninth and tenth centuries, alludes to the lack of a coherent ethnic identity by the newcomers. This calls for caution toward the tendency of modern scholarship to do little justice to the absence of political and cultural unity between the various viking polities, as exemplified by the anachronistic back-​projection of the concept of “Danelaw” to the ninth century.32 In fact, the attested use of English regional names such as Norþ(an) hymbre (Northumbrian) and Eastengle (East Anglian) in the late ninth and mid-​tenth centuries indicates that the settlers had come to identify with and through the regions they had conquered instead of fostering an attachment to a putative shared ethnic origin. The armies of the various regional polities under viking rulers consisted of a mixture of indigenous people and settlers—​a result of various strategies of integration of the indigenous populations by the viking rulers such as the continuation of the older 30 Julian D. Richards, Viking Age England (London: Tempus, 1991).

31 On the debated content of the term “Viking,” see: Judith Jesch, The Viking Diaspora (London: Routledge, 2015), 6; Lesley Abrams, “Diaspora and Identity in the Viking Age,” Early Medieval Europe 20 (2012): 17–​38; Clare Downham, “Coastal Communities and Diaspora Identities in Viking Age Ireland,” in Maritime Societies of the Viking and Medieval World, ed. James H. Barratt and Sarah J. Gibbon (Leeds: Maney, 2015), 369–​83; Downham, “Vikings Were Never the Pure-​Bred Master Race White Supremacists Like to Portray,” The Conversation, September 28, 2017, https://​the c​onve​rsat​ion.com/​.

32 On “Danelaw” see the papers in James Graham-​Campbell et al., eds., Vikings and the Danelaw: Selected Papers from the Proceedings of the Thirteenth Viking Congress, Nottingham and York, 21–​30 August 1997 (Oxford: Oxbow, 2001).

216

216

Yannis Stouraitis

administrative framework, or some degree of independence and self-​government at the initial stage. The presence of the Scandinavian settlers and their polities reinforced a notion of shared English identity as reflected in the use of the term gens Anglorum since the eighth century.33 By the ninth century the equivalent vernacular term Angelcynn was in use, a label that was promoted during the reign of Alfred the Great as a common designation of the peoples of Wessex and Mercia. Alfred’s program of centralization included court-​ sponsored historiography, the promotion of literacy, and a law book, an effort to present the Angelcynn as a people with unified history and their own divinely sanctioned law.34 However, if ethnic labels were a useful means to underpin elite loyalty to a single king’s rule, the concept of Englishness remained open enough to accommodate those that were willing to submit or cooperate, even the ones with a Scandinavian background. In this light, at the epicentre of the process of formation of a centralized kingdom was rather the integration of regional identifications instead of ethnic ones, since regional loyalties represented the main factor working against the unification process of the kingdom of the Angelcynn. This is exemplified by the relationship between Wessex and Mercia during the tenth century, which was marked by tendencies of separatism and regional autonomy in the latter. Within this framework, the major gains in the process of centralization seem to have come more out of the large number of agreements and oaths than out of the great battles of the period, as Downham argues. The best example of this was the last major step toward the unification of England in 954 when the Northumbrians expulsed their last viking king and submitted to the king of Wessex, Eadred. Regional identities continued to hold significance, however, as reflected in Wessex’s superior position within the realm and the discourse that demarcated the Northumbrians from the Anglo-​Saxons. An accentuation of ethnic divisions is testified after the late tenth century. Æthelred II’s order on St Brice’s Day (November 13, 1002) for all Danish men of his kingdom to be slain represented an act of violence that de facto drew a line of ethnic division between Angelcynn and Denisc. By the time England came under attack by Cnut the Great, the term Denisc was employed to designate both the raiders from Scandinavia and the indigenous population of England with a Scandinavian background. The latter’s demarcation as a group was further reinforced through the promotion of a common language, the dƍnsk tunga, by Cnut after the conquest of England.35 33 Patrick Wormald, “Englalond: The Making of an Allegiance,” Journal of Historical Sociology 7 (1994): 1–​24. 34 On the reign of Alfred the Great see: Richard Abels, Alfred the Great: War, Kingship and Culture in Anglo-​Saxon England (London: Routledge, 1998); Timothy Reuter, ed., Alfred the Great (Aldershot: Ashgate, 2003); David Horspool, Why Alfred Burned the Cakes (London, 2006); Nicole G. Discenza and Paul E. Szarmach, eds., A Companion to Alfred the Great (Leiden: Brill, 2015).

35 Timothy Bolton, The Empire of Cnut the Great: Conquest and the Consolidation of Power in Northern Europe in the Early Eleventh Century, The Northern World 40 (Leiden: Brill, 2009); Alexander

217



War and Collective Identifications

217

The case studies of the viking settlement of England and the Bulgar settlement of the northeastern Balkans offer some interesting comparative insights. The nomad Bulgars’ conquest of the northeastern Balkans brought them in contact with the highly centralized political culture of the East Roman, so-​called Byzantine, empire. The protracted conflict next to the economic and cultural interactions with this powerful neighbour during the eighth century contributed to the centralization of the political culture of the nomadic newcomers. The outcome of this process was a centralized Bulgar kingdom by the early ninth century, which accommodated ethnic diversity while giving shape to elite Bulgar ethnicity. In contrast, the viking conquerors with their various identities settled on a politically divided island during the ninth century, in which regional-​political identifications mattered more than a unifying English ethnic discourse. As a result, the conflict did not lead to the formation of two ethnically demarcated, rival state cultures. Even though the presence of the vikings seems to have contributed to the accentuation of ethnic Englishness in the longue durée, the formation of a unitary kingdom was rather a result of a centralizing ideology aiming to respond to the newcomers’ integration into and appropriation of the various regional-​political identities on the island. The memorialization of war as a means to shape or renegotiate the content of identity discourses is the topic of both Peter Webb’s chapter on the pre-​Islamic battle of Dhū Qār between the Bakr tribe and the Sasanian Empire, and Sergio La Porta’s chapter on a story of Armenian captives in the History attributed to Sebēos. Webb begins his analysis by pointing to the absence of any references to the battle in near-​contemporary poetry from outside the Bakr macro-​tribal group, especially from those poets connected with politics and the rise of Islam who would have had an interest in glorifying a great Arab victory. This indicates that the battle was not perceived as an event of major significance for a supra-​tribal community in its own time. In this context, the testimony of early poetry on the battle regarding the identity of the participants shows that this had nothing to do with Arab peoplehood, in particular Arab ethnicity. The combatants were predominantly identified through their tribal subgroups, the Shaybān and ʿIjl, and secondarily through their macro-​tribal identity, Bakr. Inter-​Bakr poetry suggests that the victory of the tribal subgroups that participated in the battle did not resonate beyond the boundaries of a loose micro-​tribal entity whose members recognized one shared kinship. Moreover, it indicates that the conflict had not even united the whole macro-​tribal entity in action and that its victorious outcome did not promote a more cohesive corporate Bakr identity in practice.36 The absence of a notion of a broader Arab–​Persian ethnic dichotomy in these early sources is an indication that such a notion was absent on the battlefield as well. R. Rumble, ed., The Reign of Cnut: King of England, Denmark and Norway (London: Leicester University Press, 1994).

36 On the issue of the unity of the Bakr micro-​tribal unit, see Fred Donner, “The Bakr b. Wā’il Tribes and Politics in Northeastern Arabia on the Eve of Islam,” Studia Islamica 51 (1980): 5–​38, at 28–​36.

218

218

Yannis Stouraitis

Central Arabian nomads in the pre-​Islamic period used the label Maʿadd, not Arab, to conceptualize their belonging to an imagined supra-​tribal community,37 something which is testified by the identity discourse of central Arabian poetry at the time of the battle. However, the poetical praise of victory on the battlefield was not projected upon the broader Maʿaddite identity either. Nor was the opposite side presented as bearers of a Persian ethnic identity, but as the subjects of the Sasanian shah, an imperial-​political description of the enemy that did justice to the mixed identities of its forces that included some Arabian tribes as well. The latter’s siding with the enemy was not understood in terms of ethnic treason. The poetry that emerged after the Islamic conquest of Persia also did not pay particular attention to Dhū Qār as a precursor of the conquest. Throughout the early Umayyad period, the battle remained of interest only to poets from the Bark subgroups. It was only with the establishment of the Marwanid dynasty in the late seventh century that this started to change. The event stopped being viewed as an intra-​Bakr affair and became incrementally of interest to poets of non-​Bakr origin. With the benefit of hindsight, the battle could now be reinterpreted as the beginning of a process that marked the downfall of the Sasanian Empire. In this context, the opposite parties were now more clearly demarcated in terms of cultural otherness, with Muslim poets downplaying the siding of Arabian tribes with the Sasanians. That effort to propagate an invented image of unity between Arabian tribes in the past was clearly intended to serve current political needs insofar as these tribes were now united through a shared political-​confessional identity under the authority of the Islamic Caliphate. However, their current supra-​tribal identity in whose name Dhū Qār was to be celebrated was not yet conceptualized as Arab but as Maʿaddite. Marwanid poets referred to the supra-​tribal group that had claimed victory at Dhū Qār as Maʿadd. Arabness was still not the dominant sense of an ethnic identity in the Marwanid era, as Webb has extensively argued elsewhere as well.38 The Abbasid era testifies to a decisive turn regarding the representation of collective identifications in the memorialization of Dhū Qār. The association of the battle with the Prophet and the rise of Islam was made possible through the emergence of a late-​eighth-​ century hadith that should be seen almost certainly as an invention of the Abbasid era due to the lack of an isnād (chain of authorities). The hadith established the image of the battle as an ethnic conflict between Arabs and Persians, an approach adopted by later Muslim historiographers. The retrospective Arab ethnicization of the conflict facilitated its reinterpretation as a manifestation of Arab anger against Persian aggression in pre-​Islamic times. Moreover, it allowed for its integration into the most important developments of the early history of Islam and the triumphal rise of the Prophet. 37 Peter Webb, Imagining the Arabs: Arab Identity and the Rise of Islam (Edinburgh: Edinburgh University Press, 2016), 70–​77. 38 Peter Webb, “Ethnicity and Power in the Umayyad Era: The Case of Maʿadd,” in The Umayyad World, ed. Andrew Marsham (London: Routledge, 2020), 65–​102.

219



War and Collective Identifications

219

Thus, a typical frontier skirmish39 between the Persian Empire and Arabian nomads had been transformed into a cornerstone of the effort to project backwards an image of a cohesive Arab ethnic identity.40 Within this framework, the memory of the battle was also used to underpin the image of the supra-​ethnic confessional community of the Abbasid Empire. The tenth-​ century Iranian historian Muḥammad Balʿamī, a representative of a re-​emerging Persian identity in the Muslim world, offered a new version of the events of Dhū Qār as part of the process of the Sasanian Empire’s collapse in his New Persian translation of al-​Ṭabarī’s Arabic Tārīkh al-​Rusul wa-​l-​Mulūk. Balʿamī attributed the victory of the non-​Muslim Arabian nomads to their divine inspiration to proclaim Muhammad’s name as they engaged the Persians on the battlefield. The Prophet received news of the victory via the archangel Gabriel and sent a letter to the Persian shah asking him to embrace Islam. The historian’s unparalleled effort to Islamize the battle reflects the need of a Persian Muslim writing a historical account for a Persian audience to downplay notions of Arab ethnic superiority and Persian ethnic inferiority, respectively. By stressing the role of Islam in the outcome of the battle, he suggested that the latter should not be perceived as a humiliating defeat of the Persians by the Arabs that had led to the collapse of the Sasanian Empire. It should be instead understood as a result of the Persians fighting against divine will, and as the moment that opened the way to their Islamization, thus bringing them eventually to the right side of history, that of the victors. Webb’s case study shows that military conflict in itself was less a factor that generated or reinforced ethnic or macro-​tribal loyalties in the pre-​Islamic Arabian peninsula. Instead, it was the memorialization of the battle that was employed to reinforce malleable discourses of dominant ethnicity within the Umayyad and Abbasid empires by back-​projecting a coherent Maʿadd or Arab identity upon pre-​Islamic times. In this imperial context, it was also applied to reinforce supra-​ethnic collective attachment to centralized Muslim imperial rule. La Porta provides a similar insight into the role of the memorialization of war deeds in renegotiating the content of Armenian identity. The History of Armenia attributed to Sebēos recounts a story of prisoners of war that takes place in the political framework of the Sasanian Empire. The Armenian magnate Smbat Bagratuni was sent to the province of Vrkan, east of the Caspian Sea, to function as governor on behalf of the Sasanian king around 600.41 There he found a group of prisoners of war that had been deported from 39 On the nature of the clash see Clifford E. Bosworth, “Iran and the Arabs Before Islam,” in The Cambridge History of Iran Volume 3(1): The Seleucid, Parthian and Sasanian Periods, ed. Ehsan Yarshater (Cambridge: Cambridge University Press, 1983), 593–​612, at 608. 40 On the gradual configuration of ethnic Arabness as a consequence of the early Islamic conquest, see Webb, Imagining the Arabs, esp. 110–​73.

41 On the status of Armenian elites between the Byzantine and the Sasanian Empire see, Nina Garsoïan, “Frontier—​Frontiers? Transcaucasia and Eastern Anatolia in the Pre-​Islamic Period,” in La Persia e Bisanzio: Atti dei convegni Lincei (Rome: Accademia Nazionale dei Lincei, 2004), 327–​52; Christian Settipani, Continuité des élites à Byzance durant les siècles obscurs: Les princes caucasiens

220

220

Yannis Stouraitis

their native lands in Armenia, the Byzantine Empire, and Syria, including both Christians and non-​Christians. Sebēos constructs the Armenian identity of those prisoners, stating that these people had forgotten their own language, were deficient in literacy, and lacked the order of priesthood.42 If we leave aside the evident problems of what did constitute the Armenianness of those captives, and how this could have been inferred by Smbat Bagratuni at the time, Sebēos’s account could be read as a testimony to the destructive effect of warfare on identities. Even though we are not informed about when these war prisoners had been deported from their homelands, their deportation to a foreign land due to war appears as the primary cause of the loss of the basic markers of their Armenian identity, as highlighted by the author. Irrespective of whether the captives wished to recover their alleged lost identity or not, their re-​Armenization was set in motion as a top-​down process by the protagonist of the narrative, Smbat Bagratuni. Given that this process seems to have included all Christian captives, that is, also those that had not come from Armenian lands, the reconstruction of a putative Armenian identity points to “colonization,” which at least for a part of the captives, if not for all of them, meant a process of further distancing from their putative original identities, as well as the deconstruction of their current ones. Sebēos’s story of re-​Armenization offers a perspective on Armenian identity from the second half of the seventh century when the Sasanian Empire had ceased to exist, the Byzantine Empire had been considerably contracted, and the Islamic Empire was on the rise. The next two histories recounting the story of the captives were written at the time of the independent Armenian kingdoms that had been established by Prince Ašot Bagratuni in 884 and by the Arcruni princes of Vaspurakan around Lake Van in 908.43 The tenth-​century History of Yovhannēs V Drasxanakertc‘i, the catholicos of Armenia between 897 and 925–​929, provided a new version of the story which seems to have been informed by current ecclesiastical and political concerns. In contrast to Sebēos, he presented the captives as exclusively Armenian, highlighting an ethnic instead of a territorial Armenian identity, which was attested by their reduced linguistic and literary ability as well as by the presence of a priest in the community.44 Drasxanakertc‘i’s authorial agenda, as a member of the Armenian Church, is made explicit when he reports that the community became an extraterritorial diocese through the appointment of a bishop by the Armenian catholicos, Movsēs. The accentuation of et l’Empire du vie au IXe siècle (Paris: De Boccard, 2006); Christian Settipani, “The Seventh-​Century Bagratids between Armenian and Byzantium,” Travaux et mémoires 17 (2013): 559–​78. 42 Sebēos, The Armenian History Attributed to Sebeos, translated with commentary by R. W. Thomson and J. Howard-​Johnston, Parts 1–​2 (Liverpool: Liverpool University Press, 1999), 1:44.

43 Ninan Garsoïan, “The Independent Kingdoms of Medieval Armenia,” in The Armenian People from Ancient to Modern Times, Vol. I: The Dynastic Periods: From Antiquity to the Fourteenth Century (New York: St. Martin’s Press, 1997), 143–​85. 44 Yovhannēs Drasxanakertc‘i, History of Armenia, trans. Krikor Maksoudian (Atlanta: Scholars, 1987), 95.

221



War and Collective Identifications

221

ethnic identity served the Catholicosate’s need to highlight its historical authority and jurisdiction through a story testifying to the extension of its influence beyond Armenian lands in cooperation with the political authority. The latter’s central role, expressed through the person of Smbat Bagratuni, was connected with Drasxanakertc‘i’s intention to bolster the cohesion of the Bagratini Kingdom and legitimize Bagratuni rule at current time. Roughly one century later, the Universal History of Step‘anos Tarōnec‘i was little interested in pressing a similar agenda. His version of the story presented the captives as having fully lost their Armenian cultural markers such as language, script, and religion, while the Armenian Church played little role in their ethnocultural reintegration which took place through the re-​acquirement of literacy.45 The author’s position as a scholar in the Armenian Church seems to have been the main determinant of his approach to Armenian identity, which was not connected to the jurisdictional framework of the Church. The memory of the captives served in this case as a means to propagate an image of the Armenians as an intellectual community of shared literary heritage. The impact of a changing sociopolitical context on authorial agendas is made equally evident in the versions of the story from the late twelfth and thirteenth centuries. Two priests of the Cathedral of Ani, Samuēl and Mxit‘ar Anec‘i, writing at the same place roughly two decades apart from each other, the former between 1173 and 1176 and the latter around 1193, provided nuanced versions of the story. Considering that Ani was contested ground between the Georgian Bagrationi dynasty and the Kurdish Shaddādid emirs at the time,46 their differentiated versions rather point to the different impact that the absence of a dominant Armenian secular authority had on their identity discourses. Samuēl chose to downplay the cultural markers of language and literacy, while blurring the lines of political and religious authority by making Smbat responsible for the emergence of the bishopric. In contrast, Mxit‘ar accentuated the captives’ Armenian cultural markers and was keen to project an image of them as a single group of Armenian ethnicity, while distinguishing more clearly between the role of Smbat and that of the Church in the process of their re-​Armenization. The Historical Compilation of the monastic scholar and teacher Vardan Arewelc‘i (ca. 1200–​1271), dated soon after 1267, and the Chronicle of Mxit‘ar Ayrivanec‘i (1222–​ca. 1300), written after 1289, testify to their authors’ choice to propagate the ethnic instead of the territorial aspect of Armenian identity in their retellings of the story. This choice seems to have stemmed from a current need to come to terms with the dispersed and de-​territorialized nature of Armenian communities in that thirteenth century. The role of the authors’ personal ideologies of identity in stressing different aspects of Armenianness in different times and contexts warns against any simplistic use of 45 Step‘anos Taronac‘i, The Universal History of Step‘anos Tarōnec‘i. Introduction, Translation, and Commentary by Timothy Greenwood (Oxford: Oxford University Press, 2017), 154.

46 Sergio La Porta, “ ‘The Kingdom and the Sultanate Were Conjoined’: Legitimizing Land and Power in Armenia During the 12th and Early 13th Centuries,” Revue des études arméniennes 34 (2012): 73–​118, at 77–​86.

222

222

Yannis Stouraitis

their “evidence” for projecting a collective identity upon a medieval society, as La Porta cautions.47 Both his close reading of the Armenian sources and Webb’s close reading of the Islamic sources are, in fact, a good reminder that medieval authors more often than not lacked a sense of national history writing which seeks to unify the nation’s past through a static conceptualization of a coherent collective identity based on selected true constants. It is, more often than not, modern historians that undertake for themselves the task to “nationalize” medieval history writing by looking to uncover true historical constants in medieval texts in order to produce a coherent narrative of identity out of the incoherent and often contradictory ideologies of identity of those texts’ authors.

Conclusion

That war made an impact on the collective identities of medieval societies is a truism that needs no emphasizing. Instead, the here presented cross-​cultural approach focuses on the when and how of the convergence of war with various types of community in order to explore the kind of impact warfare had on identity formation in different cultures in different geopolitical and chronological settings. The emergence of the early Bulgar kingdom in the Balkans represents a case where conquest and the protracted conflict that followed it played a central role in the centralization of rule and the configuration of an ethnic regnal culture. The English–​Viking encounter, on the other hand, shows that when the indigenous groups confronted the conquerors from the north, conflict did not strengthen ethnic bonds on both sides, nor did it draw political boundaries along ethnic lines. Moreover, England’s eventual political unification in the High Middle Ages was rather the result of an integrative centralizing political ideology instead of shared ethnic identity forged through war. The Song, Byzantine, and Mongol case studies also highlight the role of war as a means of reinforcing or asserting centralizing imperial-​ political ideologies that promoted supra-​ethnic collective attachments. The latter could be based either on a religious operative proto-​ideology, a tradition of military service and virtue, or notions of cultural moralism and world order. The Crusades to the Holy Land set in motion the transformation of Frankish identity into the French ethnocultural identity of a late-​medieval diaspora of ruling elites. This was a shared cultural heritage employed to underpin the ruling culture of a number of distinct polities in the eastern Mediterranean. In late-​medieval Europe, religious discourse was an important means for shaping the image of the “other” between the opposite parties in internal armed conflicts such as the Albigensian Crusade, the Armagnac–​Bourguignon civil war, and the Crusades against the Hussites. The primary outcome of all three conflicts, however, was the strengthening of regional or local/​ 47 On various approaches to Armenian identity in the Middle Ages, see Nina Garsoïan, Interregnum: Introduction to a Study on the Formation of Armenian Identity (ca. 600–​750) (Louvain: Peeters, 2012); Theo Maarten van Lint, “The Formation of Armenian Identity in the First Millenium,” Church History and Religious Culture 89 (2009): 251–​78; Anne E. Redgate, “Myth and Reality: Armenian Identity in the Early Middle Ages,” National Identities 9 (2007): 281–​306.

223



War and Collective Identifications

223

city attachments. The Arab and Armenian case studies show that the retrospective narrative ethnicization of warfare, or its victims, can tell us very little about the collective identities that existed at the time of the conflict. It served current needs for the renegotiation of the content of ethnic discourses at the time of writing; a renegotiation bound to authorial ideologies of identity. The memorialization of war could thus be used to strengthen current notions of ethnic identity, but that identity’s content remained malleable and situational over time. That cautions against modern efforts to pin down the true constants of medieval people’s perennial ethnicity. Nations and nationalisms do not loom large in the aforementioned case studies, neither as agents nor as products of medieval warfare. Ethnicity, to the extent that it is attested as an outcome of the war conflicts under examination, appears rather as a distinct phenomenon from modern nationhood as defined by Siniša Malešević, namely an attachment generated and reproduced by nationalism as a societally dominant operative ideology. In fact, by showing how war could influence different types of collective identification and visions of community in the Middle Ages, the chapters of this book invite us to look beyond the debate on the perenniality of the phenomena of the nation-​state and nationhood.

224

225

SELECTED BIBLIOGRAPHY

Abrams, Lesley. “Diaspora and Identity in the Viking Age.” Early Medieval Europe 20 (2012): 17–​38. Bartlett, Robert. The Making of Europe: Conquest, Colonization and Cultural Change 950–​1350. Princeton: Princeton University Press, 1993. Berberian, Houri, and Touraj Daryaee, eds. Reflections of Armenian Identity in History and Historiography. Irvine: UCI Jordan Center for Persian Studies, 2018. Bibire, Paul. “North Sea Language Contacts in the Early Middle Ages: English and Norse” in The North Sea World in the Middle Ages, ed. Thomas Lizka and Lorna Walker, 88–107 (Dublin: Four Courts Press, 2001). Biran, Michal. “The Mongols and Nomadic Identity: The Case of the Kitans of China.” In Eurasian Nomads as Agents of Cultural Change: The Mongols and Their Eurasian Predecessors, edited by Reuven Amitai, Michal Biran, and Anand A. Yang, 152–​81. Honolulu: Hawaii University Press, 2015. Brubaker, Rogers. “Ethnicity without Groups.” European Journal of Sociology 43 (2002): 163–​89. Brubaker, Rogers, and Frank Cooper. “Beyond ‘Identity’.” Theory and Society 29 (2000): 1–​47. Buchberger, Erica. Shifting Ethnic Identities in Spain and Gaul, 500–​700: From Romans to Goths and Franks. Amsterdam: Amsterdam University Press, 2017. Bull, Markus G. “Overlapping and Competing Identities in the Frankish First Crusade.” In Le concile de Clermont et l’appel à la croisade, edited by André Vauchez, 195–​211. Rome: École Française de Rome, 1997. Burbank, Jane, and Frank Cooper. Empires in World History: Power and Politics of Difference. Princeton: Princeton University Press, 2010. Carozzi, Claude, and Huguette Taviani-​Carozzi, eds. Peuples du Moyen Âge: Problèmes d’identification. Aix-​en-​Provence: Publications de l’Université de Provence, 1996. Cross, Katherine. Heirs of the Vikings? History and Identity in Normandy and England c. 950–​ c. 1015. Woodbridge: Boydell & Brewer, 2018. Curta, Florin. The Making of the Slavs: History and Archaeology of the Lower Danube Region, c. 500–​700. Cambridge: Cambridge University Press, 2001. de Rachewiltz, Igor. “The Genesis of the Name ‘Yeke Mongγol Ulus’.” East Asian History 31 (2006): 53–​56. Fiaschetti, Francesca, and Julia Schneider, eds. Political Strategies of Identity Building in Non-​Han Empires in China. Asiatische Forschungen 154. Wiesbaden: Harrassowitz, 2014. Foot, Sarah. “The Making of Angelcynn: English Identity before the Norman Conquest.” Transactions of the Royal Historical Society 6 (1996): 25–​49. Forde, Simon N., Leslie Johnson, and Alan V. Murray, eds. Concepts of National Identity in the Middle Ages. Leeds: School of English, University of Leeds, 1995.

226

226 Selected Bibliography Gabriele, Matthew. An Empire of Memory: The Legend of Charlemagne, the Franks, and Jerusalem before the First Crusade. Oxford: Oxford University Press, 2011. Garipzanov, Ildar H., Patrick J. Geary, and Przemysław Urbańczyk, eds. Franks, Northmen, and Slavs. Turnhout: Brepols, 2008. Garsoïan, Nina. Interregnum: Introduction to a Study on the Formation of Armenian Identity (ca. 600–​750) (CSCO 640, subs. 127). Leuven: Peeters, 2012. Gat, Azar (with Alexander Yakobson). Nations: The Long History and Deep Roots of Political Ethnicity and Nationalism. Cambridge: Cambridge University Press, 2013. Geary, Patrick. The Myth of Nations: The Medieval Origins of Europe. Princeton: Princeton University Press, 2002. ——​. “Ethnicity as a Situational Construct in the Early Middle Ages.” Mitteilungen der anthropologischen Gesellschaft in Wien 113 (1983): 15–​26. Gillett, Andrew, ed. On Barbarian Identity: Critical Approaches to Ethnogenesis Theory. Turnhout: Brepols, 2002. ——​. “Ethnogenesis: A Contested Model of Early Medieval Europe.” History Compass 4 (2006): 241–​60 Goffart, Walter. Barbarian Tides: The Migration Age and the Later Roman Empire. Philadelphia: University of Pennsylvania Press, 2006. ——​. Barbarians and Romans, A.D. 418–​584: The Techniques of Accommodation. Princeton: Princeton University Press, 1980. ——​. “Rome’s Final Conquest: The Barbarians.” History Compass 6 (2008): 855–​83. Golden, Peter B. An Introduction to the History of the Turkic Peoples: Ethnogenesis and State Formation in Medieval and Early Modern Eurasia and the Middle East. Wiesbaden: Otto Harrassowitz, 1992. Haldon, John. The Empire that Would Not Die: The Paradox of Eastern Roman Survival, 640–​740. Cambridge, MA: Harvard University Press, 2016. Haldon, John and Yannis Stouraitis. “The Ideology of Identities and the Identity of Ideologies.” In Identities and Ideologies in the Medieval East Roman World, edited by Yannis Stouraitis,1–​18. Edinburgh: Edinburgh University Press, 2022. Halsall, Guy. Barbarian Migrations and the Roman West, 376–​568. Cambridge: Cambridge University Press, 2008. Hastings, Andrian. The Construction of Nationhood. Cambridge: Cambridge University Press, 1997. Heather, Peter. Empires and Barbarians: The Fall of Rome and the Birth of Europe. Oxford: Oxford University Press, 2010. Hovden, Eirik, Christina Lutter, and Walter Pohl, eds. Meanings of Community across Medieval Eurasia: Comparative Approaches. Leiden: Brill, 2016. Hutchinson, John. “Bringing the Study of Warfare into Theories of Nationalism.” Nations and Nationalism 24 (2017): 1–​16. ——​. Nationalism and War. Oxford: Oxford University Press, 2017. Hutchinson, John, Chris Wickham, Bo Stråth, and Azar Gat. “Debate on Azar Gat’s Nations: The Long History and Deep Roots of Political Ethnicity and Nationalism,” Nations and Nationalism 21 (2015): 383–​402. Ibn Nubāta, Jamāl al-Dīn Muḥammad, Sarḥ al-ʿuyūn fī sharḥ Risālat Ibn Zaydūn, ed. Muḥammad Abū al-Faḍl Ibrāhīm. Cairo: Dar al-Fikr al-ʿArabī, 1964.

227

Selected Bibliography

227

Ichijo, Atsuko, and Gordana Uzelac, eds. When is the Nation? Towards an Understanding of Theories of Nationalism. London: Routledge, 2005. Jesch, Judith. The Viking Diaspora. London: Routledge, 2015. Kaldellis, Anthony. Romanland: Ethnicity and Empire in Byzantium. Cambridge: Cambridge University Press, 2019. Kinoshita, Sharon. Medieval Boundaries: Rethinking Difference in Old French Literature. Philadelphia: University of Pennsylvania Press, 2006. Kumar, Krishan. The Making of English National Identity. Cambridge: Cambridge University Press, 2003. Lippiatt, Gregory E. M. Simon V of Montfort and Baronial Government, 1195–​1218. Oxford: Oxford University Press, 2017. Malešević, Sinisa. “ ‘Divine Ethnies’ and ‘Sacred Nations’: Anthony D. Smith and the Neo-​ Durkhemian Theory of Nationalism.” Nationalism and Ethnic Politics 10 (2004): 561–​93. ——​. Identity as Ideology: Understanding Ethnicity and Nationalism. Basingstoke: Palgrave Macmillan, 2006. ——​. “Nationalism and the Longue Durée.” Nations and Nationalism 24 (2018): 292–​99. ——​. Nation-​ States and Nationalisms: Organization, Ideology and Solidarity. Cambridge: Cambridge University Press, 2013. ——​. The Rise of Organised Brutality: A Historical Sociology of Violence. Cambridge: Cambridge University Press, 2017. ——​. The Sociology of War and Violence. Cambridge: Cambridge University Press, 2010. Mayer, Hans E., ed. Die Kreuzfahrerstaaten als multikulturelle Gesellschaft: Einwanderer und Minderheiten im 12. und 13. Jahrhundert. Munich, 1997. Munkh-​Erdene, Lhamsuren. “Where Did the Mongol Empire Come From? Medieval Mongol Ideas of People, State and Empire.” Inner Asia 13 (2011): 211–​37. Niethammer, Lutz. Kollektive Identität: Heimliche Quellen einer unheimlichen Konjunktur. Reinbek: Rowohlt, 2000. Noble, Thomas F.X., ed. From Roman Provinces to Medieval Kingdoms. New York: Routledge, 2006. Page, Gill. Being Byzantine: Greek Identity before the Ottomans. Cambridge: Cambridge University Press. Pohl, Walter. “Conceptions of Ethnicity in Early Medieval Studies.” In Debating the Middle Ages, edited by Lester K. Little and Barbara H. Rosenwein, 15–​24. Oxford: Blackwell, 1998. ——​. “Von der Ethnogenese zur Identitätsforschung.” In Neue Wege der Frühmittelalterforschung. Bilanz und Perspektiven, edited by Walter Pohl, Martin Diesenberger, and Bernhard Zeller, 9–​34. Vienna: ÖAW Verlag, 2018. Pohl, Walter, Clemens Gantner, and Richard Payne, eds. Transformations of Romanness in the Early Middle Ages: Early Medieval Regions and Identities. Berlin: De Gruyter, 2018. Pohl, Walter, Clemens Gantner, and Richard Payne, eds. Visions of Community in the Post-​Roman World: The West, Byzantium and the Islamic world, 300–​1100. Farnham: Ashgate 2012. Pohl, Walter, Cinzia Grifoni, and Sophie Gruber, eds., Ethnic Terminologies in the Early Middle Ages. Leiden: Brill, forthcoming. Pohl, Walter, Ingrid Hartl, and Gerda Heydemann, eds. Ethnicity and Religion in the Middle Ages. Leiden: Brill, forthcoming. Pohl, Walter, and Gerda Heydemann, eds. Post-​Roman Transitions: Christian and Barbarian Identities in the Early Medieval West. Turnhout: Brepols, 2013.

228

228 Selected Bibliography Pohl, Walter, and Gerda Heydemann, eds. Strategies of Identification: Ethnicity and Religion in Early Medieval Europe. Turnhout: Brepols, 2013. Pohl, Walter, and Rutger Kramer, eds. Empires and Communities in the Post-​Roman and Islamic World, c. 400–​1000. Oxford: Oxford University Press, 2021. Pohl, Walter, and Daniel Mahoney, eds. Historiography and Identity IV: Writing History across Medieval Eurasia. Turnhout: Brepols, 2021. Pohl, Walter, and Mathias Mehofer, eds. Archaeology of Identity—​Archäologie der Identität. Vienna: ÖAW Verlag, 2010. Pohl, Walter, and Helmut Reimitz, eds. Strategies of Distinction: The Construction of the Ethnic Communities, 300–​800. Leiden: Brill, 1998. Pohl, Walter, and Bernhard Zeller, eds. Sprache und Identität im frühen Mittelalter. Vienna: ÖAW Verlag, 2012. Reimitz, Helmut. History, Frankish Identity, and the Framing of Western Ethnicity, 550–​850. Cambridge: Cambridge University Press, 2015. 1300. Rev. ed. Reynolds, Susan. Kingdoms and Communities in Western Europe 900–​ Oxford: Oxford University Press 1997. ——​. “Tribes, Peoples, and States.” Medieval Worlds 2 (2015): 79–​88. Rykin, Pavel. “The Social Group and Its Designation in Middle Mongolian: The Concepts Irgen and Oboq,” translated by Sarah Turner. Antropologicheskii Forum /​Forum for Anthropology and Culture 1 (2004): 183–​21. Savant, Sarah Bowen. The New Muslims of Post-​ conquest Iran: Tradition, Memory, and Conversion. Cambridge: Cambridge University Press, 2013. Šmahel, František. “The Idea of the ‘Nation’ in Hussite Bohemia.” Historica 16 (1969): 143–​ 247 and Historica 17 (1969): 93–​197. Smith, Anthony D. Chosen Peoples: Sacred Sources of National Identity. Oxford: Oxford University Press, 2003. ——​. The Cultural Foundations of Nations: Hierarchy, Covenant, and Republic. London: Routledge, 2008. ——​. The Ethnic Origins of Nations. London: Blackwell, 1986. ——​. Ethno-​Symbolism and Nationalism: A Cultural Approach. London: Routledge, 2009. ——​. “History and National Destiny: Responses and Clarifications.” Nations and Nationalism, 10 (2004): 195–​209. ——​. Myths and Memories of the Nation. Oxford: Oxford University Press, 1999. ——​. National Identity. Reno: University of Nevada Press, 1991. ——​. “Review of Azar Gat with Alexander Yakobson, Nations: The Long History of Deep Roots of Political Ethnicity and Nationhood.” In Nations and Nationalism 19 (2013): 819–​34. ——​. “War and Ethnicity: The Role of Warfare in the Formation, Self-​Images, and Cohesion of Ethnic Communities.” Ethnic and Racial Studies 4 (1981): 375–​97. Sophoulis, Panos. Byzantium and Bulgaria, 775–​831. Leiden: Brill, 2012. Stouraitis, Yannis, ed. Identities and Ideologies in the Medieval East Roman World. Edinburgh: Edinburgh University Press, 2022. ——​. “Reinventing Roman Ethnicity in High and Late Medieval Byzantium,” Medieval Worlds 5 (2017): 70–​94. Tackett, Nicolas. The Origins of the Chinese Nation: Song China and the Forging of an East Asian World Order. Cambridge: Cambridge University Press, 2017.

229

Selected Bibliography

229

Ulrich, Brian. Arabs in the Early Islamic Empire. Edinburgh: Edinburgh University Press, 2019. Van Ess, Hans. “What is Chinese? Culturalism or Ethnicism in Song China? With an Appendix on Siku quanshu Censorship of Ethnic Prejudice.” Asia Major 34 (2021): 77–​80. Webb, Peter. “Ethnicity, Power and Umayyad Society: The Rise and Fall of the People of Maʿadd.” In The Umayyad World, edited by Andrew Marsham, 65–​102. London: Routledge, 2020. ——​. Imagining the Arabs: Arab Identity and the Rise of Islam. Edinburgh: Edinburgh University Press, 2016. Whalin, Douglas. Roman Identity from the Arab Conquests to the Triumph of Orthodoxy. Cham: Palgrave Macmillan, 2020. Wood, Ian. “Barbarians, Historians, and the Construction of National Identities.” Journal of Late Antiquity 1 (2008): 61–​81. Yang, Shao-​yun. The Way of the Barbarians: Redrawing Ethnic Boundaries in Tang and Song China. Seattle: University of Washington Press, 2019. Ziemann, Daniel. Vom Wandervolk zur Großmacht: Die Entstehung Bulgariens im frühen Mittelalter (7. bis 9. Jh.). Cologne: Böhlau, 2007.

230

231

INDEX

Abbasid Empire, 219 Abēl (Habēl), 65, 68, 72–​73, 75–​79, 81–​84 Abū Nuwās, 59–​61 Abū ʿUbayda, 37, 43, 55, 57–​60 Acerra, Adolph IV of, 165, 167 Acre, 156 Adamites, 149 Æthelflæd, 123–​24 Æthelred II, king of Wessex, 125–​26, 216 Æthelred, king of Mercia, 124 Æthelweard, 125 Æthelwold, 122–​24 Æthelwulf, 122 Agat‘angełos, 74 Aguilers, Raymond of, 154–​55, 157 ʿAjam (Persians, non-​Arabians), 39–​51 al-​Akhṭal, 54–​56 Alcuin, 121 Alexios II Komnenos, Byzantine emperor, 112–​13 Alfred the Great, 117, 119, 122–​23, 127, 216 Algeria, 153 allophylos, 111 allopistos, 111 Alp Arslan, 77 al-​Aʿshā (Aʿshā Qays), 44–​51 Andronikos I Komnenos, Byzantine emperor, 112 Andronikos II Palaiologos, Byzantine emperor, 105 Andronikos III Palaiologos, Byzantine emperor, 114 Angelcynn, 122, 124, 216 Angles, 124 Anglo-​Saxon Chronicle, 117, 119, 122–​26 Ani, 77, 79–​80, 82 cathedral of, 77–​78 Annals (Chunqiu), 179–​81, 186

Antioch, principality of, 163, 168 apostasia (apostasy/​defection), 102 Apulia, 155 Aquino, Henry of, 167 Arabness, 39–​41, 50, 56–​57, 61 Arabs, 33–​35, 39–​40, 48, 56–​58, 62–​63 Aragon, Peter II, king of, 167 Aramaic, 69 Arcruni family, 71, 77, 220 Armagnacs, 120, 138–​46, 150 Armenia, 65, 67, 69–​71, 74–​85 Armenian alphabet 76, 79 Armenian Church, 73, 76, 82–​85 Armenian language (hayerēn), 65, 68–​69, 72–​73, 75–​76, 78–​82, 84–​85 Armenianness, 65–​66, 74, 76–​77, 80, 84–​85, 220 “lost Armenians,” 65–​73, 77, 80–​83, 85 Arsacid Kingdom, 67 Arsen Sapareli, 74 Arsenios, Patriarch of Constantinople, 106 Arta, city of, 114 Artabasdos, Byzantine general, 108 Ascalon, 154 Ašot I Bagratuni, prince, then king, 71, 74n37, 220 Ašot II Bagratuni, 71 Ašot, son of Šapuh, 71 Asparuch, Bulgar khan, 88 Assises d’Antioche, 168 Athelstan, king of the Anglo-​Saxons, 123–​24 Auvergne, 155 Avars/​Avar qaghanate, 87, 89‒90 Awarayr, battle of, 69 Ayrivank‘, monastery of, 82 Azerbaijan, 71 azg, 68, 72, 75, 77, 78–​79, 81–​82

232

232

Index

Badr, battle of, 59, 61 Bagrationi, royal family of Georgia, 79 Bagratuni family, 65, 67, 71, 79–​81 Bagratuni kingdom, 76–​77 Bakr ibn Wāʾil, 36–​56, 61–​63 Baldwin I, emperor of Romania see Baldwin IV of Flanders Baldwin II, king of Jerusalem, 165 Baldwin IV of Hainaut see Flanders, Baldwin IX of Baldwin IX of Flanders, 163–​64, 166, 167 Balʿamī, 40, 61–​62 Bardas Phokas, Byzantine general, 111 Bardas Skleros, Byzantine general, 102 Barseł, bishop of Ani, 79 Basel, Council of, 150 Basil II, Byzantine emperor, 102, 114–​15 Bede, 121–​22 Benevento, battle of, 156 Bertrand, cardinal, 136–​37 Bethlehem, 140 boghol, 196–​99, 212 ötegü boghol, 212 Bohemia, 147–​51 Boris, Bulgarian tsar, 214 Bouillon, Godfrey of, 159, 165, 168 Bourgeois de Paris, 139–​40, 144–​46, 150–​51 Bourguignons, 120, 130, 139, 145 Bouvines, battle of, 138 Britons, 121, 123–​25 Brubaker, Rogers, 21 Brunanburh, 123–​24 Bulgar state, 11 Bulgars, 87–​89, 213–​14, 217 Burgundy, 155 Byzantine Empire, 8, 99, 208 Byzantine Empire, 65, 67, 70–​71, 73, 77; see also East Roman Empire Cabochiens, 141–​42, 144–​45 Cabochien uprising, 204 Caliphate, 67, 71 Canso de la Crozada, 130, 132, 135–​38 Capetians, 153–​54, 156

Časláv, Diet of, 150 Caspian Sea, 65 Catalan Company, 156, 159 Cathars, 131–​34, 203, 208 Central Lands, 211, 213 Chalcedon, council of, 74 Champagne, 162, 163 Charlemagne, 153–​54, 156, 167 Charles II, king of Sicily, 167 Charles IV of Luxemburg, emperor, 147 Charles of Anjou, king of Sicily, 156, 165 Charles VI, king of France, 139–​41 Charles VII, king of France, 139, 145–​46 Chartres, 154 Chateaubriand, François-​René de, 153 Chen Fuliang, 185 Chen Liang, 182–​84 Cheng Hao, 177–​78 Cheng Yi, 177 Chester, 124 Chinese Empire, 23 Chinese supremacism, 179–​82, 185, 187, 189 Chinggis Qan, 175, 194–​96, 198, 210 Christianity, Christians, 87, 88, 91, 93‒98, 155, 166, 167 eastern, 159, 160, 168 Latin, 153n1, 155 Chronique de Morée, 156, 164 Cilicia, 80–​81, 83 Cistercians, 131–​34, 165 civil war, 100–​101 Byzantine, 101–​2 clergy (criticized), 136–​37, 147–​48; see also Bertrand, Fulk Cnut the Great, 125, 127, 216 Collins, Randall, 22 commons, common people, 144, 146, 148 Confraternities (white and black, in Toulouse), 134–​35, 138, 151 Confucius, 179–​81, 185–​86, 188 Conrad II, king of Jerusalem, 155 Constance, Council of, 147–​48 Constantine IX, Byzantine emperor, 111

233



Constantine V, Byzantine emperor, 89, 92, 108 Constantinople, 159, 166, 167, 168; see also Romania Coser, Lewis, 16 Courson, Robert of, 166 Crusade, 153–​57, 162, 165, 167, 168 Albigensian, 130–​38, 156–​57, 160, 162, 166, 168, 203 First, 154, 155, 156, 157, 159 Fourth, 156, 159, 164, 166, 168 Second, 154 Sicilian, 157, 160, 165, 168 Crusades, 28 Czechs, 147–​51, 205; see also Bohemians Danelaw, 121 Darius I, shah, 69 Dehestān (Delhastan), 65, 68, 72, 83 Denmark, 120, 127 Dhū Qār, battle of, 33–​63, 217 Dhuhl (branch of the Bakr ibn Wāʾil), 42–​43, 46 dprut‘iwn, 68, 72, 75, 77, 79, 81–​82 Duin, 71 Duin, Cathedral of St. Grigor, 67, 71, 78, 81, 84n67 Duin, council of, 73 Dulo clan, 89, 214 Durkheim, Emile, 22 Eadred, king of Wessex, 216 East Anglia, 120, 123–​24 East Asia, 191–​92, 202 East Midlands, 120 East Roman Empire, 99, 211; see also Byzantine empire Eastern Jin dynasty, 182–​85, 187 Edbury, Peter, 158 Edington, 123 Edward I, king of England, 164 Edward the Elder, king of the Anglo-​ Saxons, 123–​24 Eldigüzids, 79 Election (religious), 149–​51

Index

233

emphyliospolemos, 102–​6, 205–​6 England, 117, 119–​22, 123, 125–​27, 153, 155–​56, 161 English, 117, 119–​27 epanastasis (rebellion), 102 Epirus, 114 ethnicity/​ethnic boundaries, 39–​42, 48–​51, 58, 60, 62 ethno-​symbolism, 2–​3 Ethnocentric moralism, 176–​77, 179–​80, 210 Eurasia, 191 Mongol​, 192 Eustache de Pavilly, Carmelite, 140–​41

Fang Cheng-​hua, 175, 188–​89 Feischmidt, Margit, 21 Filial revanchism, 171–​76, 184–​85, 189, 209 fitna, first, 66, 70 Flanders, 151, 156, 157 Former Qin dynasty see Fu Jian Fragmentary Annals of Ireland, 120 France, 126, 153–​68/​kingdom of, 204 Franks, 207 Frederick II, Holy Roman Emperor, 155, 158 French Wars of Religion, 138, 148, 152 French, Frenchness, 130, 133, 135–​36, 138, 139, 145–​46, 152, 203 Fu Jian, 182–​83, 185, 187 Fulcher of Chartres, 159–​60 Fulk, bishop of Toulouse, 131, 133–​34 Ganja, 77 Gaozong (Song emperor) see Zhao Gou gens Anglorum, 216 George Pachymeres, 105 George Tornikes, Byzantine general, 111 Germans, 147, 149–​50 Germany, 153, 154, 155, 159 Gesta Francorum, 154–​55 Giorgi III, king, 79 gir, 78–​79, 81, 85n68 Gongyang Commentary to the Annals, 172 Grand Empress Dowager Xie, 189

234

234

Index

Grandes Chroniques de France, 138 Greece, 156, 168; see also Greek, Romania, Morea Greek(s), 68–​70, 74, 83, 159, 160, 161 Grigor IV Pahlawuni, catholicos, 77 Grigor Lusaworič‘, St., 73–​76 Grigor, St., see of, 72–​73, 75, 77, 81, 84 Guilhèm de Tudela, 132 Guillaume (Guilhem) de Puylaurens, 131, 134–​35 Gumplowitz, Ludwig, 15–​16

Hainaut, 163–​64 Hāmarz, 36, 40, 46, 50–​51, 53, 61 Han Tuozhou, 174–​75 Han Yu, 186 Harič, monastery of, 78, 80 hayastan, 68, 72 hayastaneayk‘, 72, 75–​76, 78, 81, 85 hayk‘, 75, 81–​82, 85 Henri IV, king of France, 146 Hepthalites (kušank‘), 67 heretics, heresy, 131, 133–​34, 137–​38, 141, 143, 147, 149, 160, 161, 162 Holme, 124 homodoxoi see homopistoi homogeneis, 113 homophylos (homophyloi), 108–​10, 112, 114 homopistoi, 109 Hormizd IV, shah, 77 Hormuz Kharād, 36, 40, 46, 50, 60 Hu Hong, 176–​77, 180 Hu Yin, 180–​82, 184 Hugh, of Saint-​Cher, maître d’oeuvre of the Dominican Postilla, 141, 143 Huizong (Song emperor) see Zhao Ji Hülegü, Il-​khān, 81 Hundred Years’ War, 28 Hungary, Hungarians, 147–​49 Hussites, 204–​5 Hutchinson, John, 19–​22 hypêkooi, 103–​4, 108 Ibelin, 155 Iberia, 79 Ibn al-​Athīr, 59

Ibn ʿAbd Rabbihi, 47 identity, 129, 135, 138, 145–​46, 148, 150–​52 ideological power/​centrifugal ideologization, 23, 26–​28 Il-​Ghāzī, Artuqid, 82 ʿIjl (branch of the Bakr ibn Wāʾil), 36–​39, 42, 44, 46–​48, 50, 53, 56, 217 ʻImād al-​Dīn al-​Iṣfahānī, 140 inheritance see law, inheritance Innocent III, pope, 130–​32, 165 inscriptions, proto-​Bulgarian, 90‒93, 95‒97 Iranian nationalism, 34 Iraq, 33, 35, 37, 44, 52, 58–​59 Ireland, 125 Isaak II Angelos, Byzantine emperor, 104, 112 Isabeau of Bavaria, queen of France, 140 Islam, 59, 62–​63 Italy, 153, 156, 158, 160; see also Sicily Iwanē Ōrbēlean, 79–​80

Jacoby, David, 164, 168 James, William, 15–​16 Jan Žižka 149–​50 Jarīr, 54–​56 Jean de Troye, insurrection leader, 141, 145 Jean Gerson, 142, 144 Jean Juvénal des Ursins, 144–​45 Jean Petit, 144 Jean sans Peur (John the Fearless), duke of Burgundy, murdered in 1419, 139, 144 Jerusalem, 140, 154, 157, 158, 165, 168 High Court of, 160, 163 kingdom of, 155, 157–​59, 160, 162, 163n47, 168 Jews, 126 John II Komnenos, Byzantine emperor, 113 John II Tzimiskes, Byzantine emperor, 102, 111 John of Ibelin, 158–​59, 160 John Skylitzes, 102, 110–​11 John VI Kantakuzenos, Byzantine emperor, 114 John Wyclif, 147 Josheph Genesios, 108 J̌uanšir, 70–​71

235



Kaifeng, 170–​71, 173, 175, 176, 210 Kalamata, Benjamin of, 164 kana sybigi, 92, 97 Kars, 77 Kent, 124 Khazars, 87‒88, 92 Khosrow Parviz, 36, 50–​51, 54–​56, 58, 61 Khubilai (Qubilai) Khan, 189, 200, 213 Kid, Benjamin, 15–​16 Kitan (Khitan) people see Liao empire Kodrik‘, 68, 83 Koriwn, 76 Ḵosrow II, shah, 65, 67, 73, 77–​78, 81–​82 Krum, Bulgar khan, 90‒91, 95, 214

Language, 153, 155–​56, 160–​61, 163 Languedoc see Midi Later Liang dynasty, 185 Later Tang dynasty, 185 Lateran Council, Fourth, 165 law, 158, 159–​65 customary, 158–​59, 160, 162, 163–​65, 168 inheritance, 162–​63 marriage, 160, 161–​62 Roman, 158, 162 Lawrence of Brezowá (Vavřinec z Březové), 148–​50 Layman’s creed, 137 Leon Gabalas, Byzantine ruler of Rhodes, 106 Letres dou Sepulchre, 158–​59 Lewon I Ṙupenian, king, 83 Liao empire, 169–​70, 172 Lincolnshire, 120 Lindisfarne, 121 Lipany, battle of, 150–​51 Livre au roi, 160, 161, 162, 163, 167 Lombardy, 155, 158 Low Countries, 146–​51 Lu Jiuyuan, 181–​82 Lü Zhong, 175–​76, 188–​89 MacEvitt, Christopher, 160 Magna Carta, 161 Majapahit Empire, 213 Manfred, king of Sicily, 156

Index

235

Mann, Michael, 17–​19 marriage, 163, 166–​67; see also law, marriage martyrs, martyrdom, 133, 136, 140, 149 Marwanids 53–​57, 59, 63; see also Umayyads Mary (the Virgin), 135–​36 marzpan (marzbān), 65, 67, 69–​70, 72, 75–​78, 81–​82, 84n67 Maštoc‘, 76 Maurice, Byzantine emperor, 67, 73 Mauss, Marcel, 16–​17 Maʿadd, 41–​42, 48–​50, 52, 54–​57, 59, 63, 218 McDougall, William, 15–​16 Mediterranean Sea, 154, 158, 162, 164, 165, 168 Mencius, 186 Mercia, 120, 122–​24, 216 Mercian Register, 124 Mesembria, 93, 95 Michael I, Byzantine emperor, 111 Michael II, Byzantine emperor, 102, 108 Michael II, Byzantine ruler of Epirus, 105, 110 Michael Psellos, 111 Michael VIII Palaiologos, Byzantine emperor, 104, 106 Michaud, Joseph-​François, 153, 156 micro-​solidarity, 23, 28–​31 Middle Persian, 67n10, 69 Midi, 155n11, 156, 160, 161, 162, 163, 166, 167, 208 Mik‘ayēl Č‘amč‘ean, 65n4, 85n68 Minerve, 133 Miskawayh, 59 Mongol empire, 175, 188–​90, 194, 196, 198, 202, 211–​12 Mongolization, 12–​13, 193–​94, 212 Mongols, 193–​94, 212–​13 Montfort, 161, 162, 167 Moravia, Moravians, 147, 150–​51 Morea, 156, 159, 161, 163, 164, 168; see also Greek, Romania, Uxanze de Romania Movsēs Dasxuranc‘i, 70–​71, 85 Movsēs II, catholicos, 72–​73, 78–​79, 81, 220

236

236

Index

Muḥammad Balʿamī, Iranian historian, 219 Muhammad, Prophet, 34, 57–​59, 61–​63 Muntaner, Raymond, 156 Muret, battle of, 133, 135, 138, 157, 167 Muslim Conquests, 33–​34, 52, 55–​56, 59 Muslims, 154, 160, 167 Mxit‘ar Anec‘i, 78–​81, 83–​84, 221 Mxit‘ar Ayrivanec‘i, 82–​84, 221 Myinsaing Kingdom, 213 Nablus, council of, 160, 165–​66, 167n66 Narbonne, 154 Narratio de rebus Armeniae, 65n4, 74 nationalism, 4–​5, 34, 54 nationhood, 4 Neo-​Confucianism, 173–​82 Neogregorian reform, 165–​67 neōterismos (change by force), 102 Netherlands, 151 Neuilly, Fulk of, 166 Nikephoros I, Byzantine emperor, 89‒90, 95, 214 Niketas Choniates, 104, 112 Northern Wei dynasty, 183–​88 Northern Zhou dynasty, 183, 186 Northumbria, 119–​21, 123, 125 Norway, 120, 127 Novara, Philip of, 158, 159, 160, 161 Nuarsak, treaty of, 66

Oliver, paladin, 167 Omurtag, Bulgar khan, 91‒94, 96‒97 organisational power/​cumulative bureaucratization of coercion, 23–​25 Orleans, 126, 157, 204 Outremer, 156, 157–​59, 161, 162, 168; see also Antioch, principality of; Jerusalem, kingdom of Oxford, 126

Pallig, 126 Pamiers, Statutes of, 160, 161, 162, 163, 166, 168 Paris, 130, 138–​46, 150–​52, 153, 154, 156, 157, 162, 165, 166; see also France

peace, 134–​35, 142–​46 Pere II, king of Aragon, 132, 134 Persian Empire, 23 Persian identity, 61–​62; see also Iranian nationalism Persians, 33–​34, 39–​40, 50–​52, 57–​62, 73, 84n67, 85n68 Peter the Chanter, 141 Philip I of Piedmont, prince of Morea, 164 Philip II Augustus, king of France, 156 Philippe, duke of Burgundy, 145 Philostratus, Life of Apollonius of Tyana, 69–​70 Picards see Adamites Piedmont, Philip I of see Morea Pierre des Vaux-​de-​Cernay, 131–​33 Pliska, 90, 93 Poland, 27 Prague, 147–​51 Primogeniture see law, inheritance prophets (false), 142–​43 purge, 140–​41 Puy, Le, 155 Puylaurens, William of, 157 al-​Qādisiyya, battle of, 34, 52, 62 Qays ibn Masʿūd, 36–​38, 40, 51 Qays, branch of the Bakr ibn Wāʾil, 38, 42–​46, 48, 58 qi, 211 Qin Hui (Qin Gui), 170–​71, 175–​76 Qinzong (Song emperor) see Zhao Huan Raby, 120 rape, 139, 140 Rashīd al-​Dīn, 197 Rashidun Caliphate, 25 Ratzenhofer, Gustav, 15–​16 Raymond VI, count of Saint-​Gilles and Toulouse, 131, 133 Raymond VII, count of Toulouse, 132, 136–​38 Recoura, Georges, 164 Renier of Montferat, kaisar, 112–​13

237



Rheims, Robert of, 154 Rhōmaios (Rhōmaioi), 9, 100, 107, 114 Roland, 153–​54 Roman Empire, 23 Romania, 156, 159, 162, 166, 167; see also Greek, Morea Romanness (Byzantine), 99–​100, 107, 205 Rome, 154, 166 Ṙusudan Bagratuni, 79

Sagastan, 72, 75, 77–​78, 81–​82, 84, 85n68 Sahak, Armenian patriarch, 76 Saint-​Omer, Nicholas III of, 164 Sājid emirs, 71 Samūel Anec‘i, 77–​81, 83–​84, 85n68 Samuēl, 221 Samuel, Bulgarian tsar, 102 San Martino, parliament of, 163 Sancerre, Stephen I of, 162 Saracens, 139, 145 Saragossa, 154 Sasanian Empire, 24–​25, 219; see also Persian Empire Sasanians, 33–​42, 45, 49–​56, 60–​62, 65, 67, 70, 77 Saxons, 124 Scandinavia, 117–​22, 125–​27, 153, 155 Scotland, 155 Scots, 123–​24 Scyld, 122 Scythians, 82 Sebēos, 65–​73, 76–​79, 83–​84, 217 Secret History of the Mongols, 194, 196 Seljuks, 77 Sempad the Constable see Assises d’Antioche Sernin, saint, 136 Shaddādids, 77, 79 Shao Yong, 174 Shatuo Türks, 185 Shaybān (branch of the Bakr ibn Wāʾil), 36, 38–​39, 42–​51, 53–​55, 58, 217 Shenzong (Song emperor) see Zhao Xu Shizong (Jin emperor) see Wanyan Yong Sicily, 155, 163, 208 kingdom of, 156, 157, 159, 160, 161, 163, 165

Index

237

Sigismund, Emperor and King of Hungary, 147–​48, 150–​51 Simmel, Georg, 16, 22 Simon IV de Montfort, 131–​33, 136–​38 Simon of Montfort, 6th earl of Leicester, 155 Simon V of Montfort, 156–​57, 160, 162, 163, 166–​7 Smbat Bagratuni, 65–​73, 75–​84, 219–​20 Smith, Anthony D., 19–​22, 26 Social Darwinism, 15–​16 Song (dynasty), 192, 199, 202, 209–​10 networks, 200 Southern-​, 199 troops, 200 Song of Roland, 136, 148–​49 Southeast Asia, 191–​92, 199, 200–​201 Southern Qi dynasty, 186–​87 Spain, 154 St. Brice’s Day massacre, 117, 126, 216 St. Frideswide’s, 126 stasis (insurrection), 102 Step‘anos Tarōnec‘i, 71, 75–​76, 78–​81, 83–​84, 85n68, 221 Sui dynasty, 183–​84, 186 Sumner, William Graham, 16 Sword (biblical pericope from Matt. 10.34), 134–​35 Sykes-​Picot Agreement, 153 Syria, 65, 68, 154, 156, 159, 160, 162, 207 Syriac Orthodox Christians, 69, 85 Syrian Mandate, 153 Syrians, 83 al-​Ṭabarī, 58–​59 Tabor, Taborites, 149–​51 Tackett, Nicolas, 171 Taghlib, 36, 42, 54 Tagliacozzo, battle of, 160 Taktika of Emperor Leo VI, 103 Tang dynasty, 183, 185–​86 Taranto, Bohemond of, 155 Tares (biblical pericope from Matt. 13.24–​43), 141 Tettenhall, 123 Teutonic Knights/​Teutonic Order, 30–​31

238

238

Index

Theodore Mangaphas, Byzantine rebel, 104 Theophanes Continuatus, 109 Theophanes the Confessor, 108 theories of nationalism, 17–​22 Thingwall, 121 Thomas, Byzantine rebel, 102–​3, 108, 111 Tilly, Charles, 17–​19 Tong Guan, 169, 176 Toulouse, 130–​38, 151, 204 Trdat, king, 74 treason, traitors, 131–​32, 136, 140–​41, 145–​46, 162, 165 Tuoba Hong (Xiaowendi), 183–​84, 185–​88 Tuoba Xianbi see Northern Wei dynasty T‘urk‘astan, 68, 72, 75, 78, 81 Turkey, 156, 168; see also Romania tyrannis (usurpation), 102 Tyre, William II, archbishop of, 154n3, 157, 165

Umayyad Caliphate, 24–​25 Umayyads, 34, 42, 49, 51–​53, 57, 63; see also Marwanids Uxanze de Romania, 159, 161, 162, 168, 208

Valois, 204 Vardan Arewelc‘i, 74n37, 79, 81–​84, 221 Vaux-​de-​Cernay, Peter of, 157 Velay, 155 Venice, 159, 168 Versailles, 153 vikings, 12, 117–​21, 123, 127, 215–​17 Vramšapuh, king, 76 Vrkan, 65, 67–​68, 71–​73, 75, 77–​78, 81–​83, 219 Vrkanians, 82 Wang Anshi, 177 Wang Fu, 169 Wanyan Yong (Shizong), 178–​79 Warenne, John de, 164 Weber, Max, 16 Wenzel IV, king of Bohemia, 147 Wessex, 119, 121–​25, 216

William II, prince of Morea, 156 William the Breton, 156–​57 Wimmer, Andreas, 17 Wirral, 120

Xiaowendi see Tuoba Hong Xiaozong (Song emperor) see Zhao Shen Xor Virap, monastery of, 81

Yan-​Yun region, 169–​71 al-​Yaʿqūbī, 59 Yazdegird II, shah, 69 Ye Shi, 172–​73, 175, 182, 185, 187–​88 York, 121, 124 Yorkshire, 120 Yovhan, anti-​catholicos, 73 Yovhannēs V Drasxanakertc‘i, catholicos, 71–​75, 77, 81, 83–​84, 220 Yovhannēs-​Senek‘erim Arcruni, king, 77 Yuan (China), 191–​92, 194 dynasty, 192, 199 emperors, 200 empire, 192, 211 History of the Yuan Dynasty, 194 imperial documents, 192 imperial project, 202 period, 200 political and administrative elites, 19 state, 196 subjects, 199 Yuanshi, 194–​95, 198 Yue Fei, 170–​71, 175

Zak‘arē Mxargrjeli, 79–​80 Zhao Gou (Gaozong), 170–​73, 176–​77 Zhao Huan (Qinzong), 170, 172, 176 Zhao Ji (Huizong), 169–​70, 172, 176–​77 Zhao Ruyu, 174 Zhao Shen (Xiaozong), 171, 173, 182 Zhao Xian, 189 Zhao Xu (Shenzong), 173, 177 Zhu Fu, 185–​88 Zhu Xi, 173–​75, 177–​81