The Mongols' Middle East: Continuity and Transformation in Ilkhanid Iran

Home , AliOS, Baghatur, Baidar, Census in Armenia, Chobanids, Darugha

The Mongols’ Middle East Islamic History and Civilization !"#$%&' ()$ *&+"'

Editorial Board

Hinrich Biesterfeldt Sebastian Günther

Honorary Editor Wadad Kadi

,-./01 234

The titles published in this series are listed at brill.com/ihc The Mongols’ Middle East

Continuity and Transformation in Ilkhanid Iran

Edited by

Bruno De Nicola and Charles Melville

.15617 | 8-!*-7 Cover illustration: A Mongol prince studying the Qur’an, 14th century Illustration of Rashīd al-Dīn’s Jāmiʿ al-tawārīkh. Tabriz (?), 1st quarter of 14th century. Water colours on paper. Original size: 20.3 cm × 26.7 cm. Staatsbibliothek Berlin, Orientabteilung, Diez A fol. 70, p. 8 no. 1. This seems to be a tent mosque. The inscription above the arch on the left, which is either the entrance or the mihrab, reads “Allah is the ruler” (or similar). Dschingis Khan und seine Erben (exhibition catalogue), München 2005, p. 266. Library of Congress Cataloging-in-Publication Data Names: De Nicola, Bruno. | Melville, C. P. (Charles Peter), 1951– Title: The Mongols’ Middle East : continuity and transformation in Ilkhanid Iran / edited by Bruno De Nicola and Charles Melville. Description: Leiden ; Boston : Brill, 2016. | Series: Islamic history and civilization : studies and texts, 5!!7 0929-2403 ; volume 127 | Includes bibliographical references and index. IdentiGHers: .II7 2016004908 (print) | .II7 2016014826 (ebook) | 5!87 9789004311992 (hardback : acid-free paper) | 5!87 9789004314726 (e-book) | 5!87 9789004314726 (E-book) Subjects: .I!J: Mongols—Iran—History—To 1500. | Ilkhanid dynasty. | Iran—History—1256–1500. | Iran—Politics and government. | Iran—Social conditions. | Iran—Relations. | Social change—Iran—History—To 1500. ClassiGHcation: .II 6!289 .M66 2016 (print) | .II 6!289 (ebook) | 66I 955/.026—dc23 .I record available at http://lccn.loc.gov/2016004908

Want or need Open Access? Brill Open o9fers you the choice to make your research freely accessible online in exchange for a publication charge. Review your various options on brill.com/brill-open. Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface. %'') :;3;-3<:= %'>) ;4?-;:-:<-=22;;-3 (hardback) %'>) ;4?-;:-:<-=2<43-@ (e-book) Copyright 2016 by Koninklijke Brill 7,, Leiden, The Netherlands. Koninklijke Brill 7, incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijho9f, Brill Rodopi and Hotei Publishing. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill 7, provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, 0A 01923, /!A. Fees are subject to change. Brill has made all reasonable e9forts to trace all rights holders to any copyrighted material used in this work. In cases where these e9forts have not been successful the publisher welcomes communications from copyright holders, so that the appropriate acknowledgements can be made in future editions, and to settle other permission matters. This book is printed on acid-free paper and produced in a sustainable manner. Contents

Acknowledgements ,55 List of Illustrations K%%% Note on Transliteration %+ Contributors +

Introduction 1 Bruno De Nicola and Charles Melville

"#$% & The Mongol Conquest of the Middle East

1 Mongol Conquest Strategy in the Middle East 13 Timothy May

2 Continuity and Change in the Mongol Army of the Ilkhanate 38 Reuven Amitai

"#$% ' Internal Actors: Politics, Economy and Religion

3 Shams al-Dīn Juwaynī, Vizier and Patron: Mediation between Ruler and Ruled in the Ilkhanate 55 Esther Ravalde

4 The Economic Role of Mongol Women: Continuity and Transformation from Mongolia to Iran 79 Bruno De Nicola

5 Faith and the Law: Religious Beliefs and the Death Penalty in the Ilkhanate 106 Florence Hodous K% LM)"&)"'

"#$% ) Culture and the Arts

6 Music in the Mongol Conquest of Baghdad: Ṣafī al-Dīn Urmawī and the Ilkhanid Circle of Musicians 133 Michal Biran

7 Historical Epic as Mongol Propaganda? Juwaynī’s Motifs and Motives 155 Judith Kolbas

8 From the Mongols to the Timurids: Re+,nement and Attrition in Persian Painting 172 Karin Rührdanz

"#$% - Relationships with Neighbouring Actors

9 Champions of the Persian Language: The Mongols or the Turks? 195 Aptin Khanbaghi

10 Darughachi in Armenia 216 Bayarsaikhan Dashdondog

11 The Phoenix Mosque 237 George Lane

12 Mamluk and Mongol Peripheral Politics: Asserting Sovereignty in the Middle East’s ‘Kurdish Zone’ (1260–1330) 277 Boris James

Epilogue

13 The End of the Ilkhanate and After: Observations on the Collapse of the Mongol World Empire 309 Charles Melville

Index ==4 Acknowledgements

The editors would like to express their gratitude to all those colleagues who anonymously reviewed the papers published in this volume and o9fered their advice, comments and constructive criticism to improve the quality of the chapters. Apart from the collaboration of the contributors to this volume in reviewing the papers for publication, we have beneGHted from the knowledge of Mark Dickens (University of Alberta), Ladan Akbarnia (The British Museum) and Nur Sobers Khan (The Museum of Islamic Art, Doha). Also, special thanks go to Daniel Rolph for copy-editing the manuscript of this book. We would like to thank the Faculty of Asian and Middle Eastern Studies (University of Cambridge) for support during the period when we organized the symposium for N-I01! (World Congress for Middle Eastern Studies) and at the beginning of the preparation of this volume. We are grateful to our colleagues for their patience over the long process of assembling and preparing the book and for their faith in its GHnal appearance. Lastly, our appreciation goes to our families and friends who have helped throughout the process of making this book available. List of Illustrations

0.1 The Mongol Empire, c. 1294 9 0.2 The Ilkhanid territories 10 8.1 Prince being equipped with arms and armour by the king 179 8.2 A king on horseback encounters a slain warrior 180 8.3 On a terrace, courtiers take care of a man fainting in front of a ruler 182 8.4 A king on horseback is greeted by a group of people led by a man held up by a device installed between two buildings 185 8.5 Zangis in a castle attacked by warriors aiming to free captives 186 11.1 Annotated map of Qinsai based on locally produced Song dynasty maps from 1270s. In classical Chinese fashion, the map is south orientated with north at the bottom 242 11.2 An early 20th century depiction of the Phoenix Mosque complex before the implementation of the Town Planner’s modernisation scheme 251 11.3 The tombstone belonging to Khwāja Muḥammad bin Arslān Khānbāliqi, d.1317 271 12.1 Map of Kurdish dominated territories in the 13th century 279 Note on Transliteration

Place names have been given in their current anglicized forms where appli- cable, as have terms and names such as Mamluks, sultan, vizier, which have entered the English language. Due to the lack of academic agreement on the standardization of Mongolian names, we have followed the spelling of names used by John Andrew Boyle in both The Successors of Genghis Khan and History of the World Conqueror, with the exception of the name of Chinggis Khan. For Persian and Arabic script, we have followed the transliteration of names, works and speciGHc terminology proposed by the Library of Congress. Finally, Chinese characters have been transcribed using pinyin system while Russian has been maintained in cyrillic script. Contributors

Prof. Reuven Amitai (Hebrew University of Jerusalem) completed his 8A studies at the University of Pennsylvania and his 0A and PhD at the Hebrew University. He is Eliyahu Elath Professor of Islamic History and works on Middle Eastern and Central Asian history from 1000–1500 I1. His recent publications include Holy War and Rapprochement: Studies in the Relations between the Mamluk Sultanate and the Mongol Ilkhanate (1260–1335) (Brepols, 2013) and Nomads as Agents of Cultural Change: The Mongols and Their Eurasian Predecessors (co-edited with Michal Biran, University of Hawai’i Press, 2015). In 2014–16, Reuven Amitai is a fellow at the Annemarie Schimmel Kolleg for Mamluk Studies at the University of Bonn, Germany, and is engaged on a study of the southern Levant in the Ayyubid and Mamluk periods.

Prof. Michal Biran (Hebrew University of Jerusalem) is a historian of Inner Asia and a member of the Israeli Academy of Science and Humanities. She is the Max and Sophie Mydans Foundation Professor in the Humanities, and currently (2015) the director of the Louis Frieberg Center for East Asian Studies at the Hebrew University of Jerusalem, where she also leads the 1PI-funded project “Mobility, Empire and Cross-Cultural Contacts in Mongol Eurasia” (http://mongol.huji .ac.il/). Together with Hodong Kim she is now editing The Cambridge History of the Mongol Empire (2 volumes) for Cambridge University Press.

Dr. Bayarsaikhan Dashdondog (National University of Mongolia) holds a 8A in Philology from the State University of Erevan (Armenia), a MSt in Armenian Studies (University of Oxford, /Q) and a DPhil in Oriental Studies (University of Oxford, /Q). At present, she is a Lecturer at the National University of Mongolia, Department of History, teaching a History of Mongol Empire and Central Asia. She is engaged in the research projects of the Mongols in the 13–17th century (Monsudar Grant, 2015) and History of the Mongol Il-Khans (State Grant, 2014–2016). She is a part of the international scholarly project of the Cambridge History of Mongolia (eds. Michal Biran and Kim Hodong).

Dr. Bruno De Nicola (University of St. Andrews) holds a 8A in History from the University of Barcelona (Spain), an 0A in Middle Eastern Studies (!-A!, University of London) and a PhD in Middle Eastern History (University of Cambridge). IM)"R%>#"MR' +%

At present, he is Research Fellow in Middle Eastern Studies in the School of History, University of St Andrews in the United Kingdom. A specialist in Medieval History of the Middle East and Islamic manuscripts, he is now part of the project “The Islamisation of Anatolia (1100–1500)” (European Research Council, grant number 284076).

Dr. Florence Hodous (Hebrew University of Jerusalem) holds a 8A in History and Development Studies (!-A!, University of London), an 0A in Historical Research Methods (!-A!), and a PhD in History (!-A!). Currently she is researching judges and legal systems of the Mongol empire as a post-doctoral fellow in the research project “Mobility, Empire and Cross-Cultural Contacts in Mongol Eurasia” at the Hebrew University of Jerusalem (European Research Council, grant number 312397).

Dr. Boris James (Institut Français du Proche-Orient / Erbil) holds an 0A in Monde Arabe Moderne from the University Aix-Marseille 5 and a PhD in History (Université Paris-Ouest Nanterre-La Défense). He is Research Fellow at Institut Français du Proche-Orient (Ifpo) and head of its Erbil branch since 2014. He is a specialist of Medieval and Early modern History of the Middle East, especially Kurdish society and History. Nowadays he is implementing a research project on the Kurdish principalities and the role of Armenian actors in the vicinity of Lake Van during the 15th and 16th centuries.

Dr. Aptin Khanbaghi (5!0I—Aga Khan University in London) is a senior researcher at Aga Khan University. He received his doctorate from Cambridge University in Iranian studies. He is the series editor for the Muslim Civilisations Abstracts and the author of The Fire, the Star and the Cross: Minority Religions in Medieval and Early Modern Iran (58 Tauris).

Dr Judith Kolbas (Miami University, Ohio, /!A) received her 8A in International Studies from George Washington University in Washington, 6.I. in 1964 and then attended the American University of Beirut for an 0A in 1966. Returning to the United States, she attended the State University of New York at Bu9falo where she began her medieval Islamic and numismatic studies. Eventually, she achieved her PhD from New York University in 1991 after ten years of forming a catalogue of Mongol currency in Iran that required visits to museums and collectors around Europe and the Middle East. Since then, she has continued to +%% LM)"R%>#"MR' teach, research and publish on this topic as well as support related organizations. For example, she has been a vice president of the Royal Asiatic Society of Great Britain and Ireland and then re-founded the PA! branch in Shanghai in 2005. She is also currently a member of the Oriental Numismatic Society as well as the Royal Numismatic Society. She also founded and was Director of the Central Asian Numismatic Institute (Cambridge, /Q) until 2012. As soon as possible, she went to Mongolia in 1992, has lived and taught there and visited it most recently in 2014.

Dr George Lane (!-A!, University of London). After a varied career spent traveling throughout the Middle East and the Far East, George Lane joined !-A! both as a student in the History Dept and a teacher in the Academic English Dept. After receiving his PhD from the University of London in 2001 he began his career as a historian teaching, researching and publishing on all aspects of the Mongol Empire and the mediaeval Islamic world. His interest in the connections between Yuan China and Ilkhanid Iran led to his discovery of the tombstones in Hangzhou’s 1284 mosque and eventually to a generous grant from the British Academy’s Mid-Career Scholarship Award to further this research. A full record of the Hangzhou project with contributions from fellow scholars is currently being funded for publication by 85S!.

Dr. Timothy May (University of North Georgia) holds a 8A in History and Anthropology from the College of William & Mary, an 0A in Central Eurasian Studies from Indiana University, and a PhD from the University of Wisconsin (Madison). He is the author of The Mongol Art of War (2007) and The Mongol Conquest in World History (2012). When not writing or teaching about the Mongols, he practices the Dark Arts of Administration as the Associate Dean of the College of Arts and Letters.

Prof. Charles Melville (University of Cambridge) holds a 8A Hons. in Oriental Studies (University of Cambridge), 0A in Islamic History (!-A!) and PhD, in Oriental Studies (University of Cambridge). At present he is Professor of Persian History at the University of Cambridge and Fellow of Pembroke College. He has been a long- serving member of the Governing Council of the British Institute of Persian Studies and chairman of the Research Committee. Since 1999, he has been Director of the Shahnama Project, and since 2006 he has been President of The Islamic Manuscript Association (*50A). He has published extensively on the IM)"R%>#"MR' +%%% history and culture of Iran in the Mongol to Safavid periods and the illustration of Persian manuscripts as well as on the Shāhnama of Firdawsī, the illustration of the text and its reception.

Ms. Esther Ravalde (Independent Scholar) studied Persian at Wadham College, Oxford (0A Oriental Studies), and holds an 0A in Historical Research Methods from !-A!, University of London. She now teaches History at the Nelson Thomlinson School in Wigton, and trains teachers for the University of Buckingham.

Prof. Karin Rührdanz (Royal Ontario Museum-University of Toronto) obtained her PhD from Halle University in 1974. Currently she is Senior Curator of Islamic Art at the Royal Ontario Museum and Professor of Islamic Art at the Department of Near and Middle Eastern Studies, University of Toronto. A specialist in the arts of the book, she studies the text-image relationship in manuscripts from Iran, Central Asia and the Ottoman Empire. Recent research focuses on Central Asian miniature paintings of the 15th–17th centuries as well as the formation of genre- speciGHc pictorial programs for lyric poetry and other genres beyond the epic.

Introduction

Bruno De Nicola and Charles Melville

The Ilkhanate is the name by which historians refer to the political entity that governed present-day Iran, Azerbaijan and Eastern Anatolia between 1260 and 1335, with a short afterlife until 1353.! It is de"#ned by the establishment of a dynasty of Mongol rulers descended from Chinggis Khan (d. 1227) by his son Tolui (d. 1232) and especially by the descendants of Hülegü (d. 1265). The origin and meaning of the term ‘Ilkhan’ (or īl-Khān) has been subject of some academic discussion over time. The title Ilkhan had been used among the Great Seljuqs in the eleventh century and appeared in a Uyghur translation of a seventh-century Chinese biography of a Buddhist pilgrim.& Although the particle ‘Il’ originally meant ‘country,’ it seems that it became used as a term to refer to ‘subservient’ or ‘submissive’ in the Mongol period. This has raised discussions about the legitimacy of the establishment of the Ilkhanate as a separate ulus of the Mongol Empire, suggesting that the Mongols of Iran were seen as the guardians of the western lands of the Empire under the rule of the Great Khan living in China rather than a realm in its own right.' Despite this, the term ‘Ilkhan’ seems to have been in use among the Mongols of Iran and the people they conquered since at least 1259, when numismatic evidence con"#rms the use of the term, but also "#nds earlier references in other Middle Eastern sources.( The arrival of the Mongols in the Middle East has been interpreted from very di)ferent angles by historians of the region since the early twentieth century. The initial approach was to see it as an example of a confrontation between Civilization and Barbarism, where the former was represented by the high culture of the ‘Abbasid Caliphate and the latter by the in"#del and predatory

* Apart from the chapter in the Cambridge History of Iran by Boyle, “Dynastic and Political History of the Il-Khāns,” 303–421, there is remarkably little scholarly work dedicated solely to a political narrative of the Ilkhanid dynasty, the nearest modern contribution being by Bertold Spuler, Die Mongolen in Iran, of which the fourth printing was made in 1985. David Morgan’s chapter, “The Ilkhanate, 1260–1335,” in the forthcoming volume of The Cambridge History of the Mongol Empire, might be expected to update Boyle’s 50-year-old contribution; meanwhile, for a recent survey of the literature on the Mongol period (not exclusively concerned with Iran), see Morgan, The Mongols, 181–227. , Allsen, Culture and Conquest in Mongol Eurasia, 21. - Ibid., 21–22. . Amitai-Preiss, “Evidence for the Early Use of the Title īlkhān among the Mongols,” 353–61.

Mongols. This historiographical approach saw the Mongol-dominated Middle East merely as a dark period in Islamic history, which would gloriously revive with the rise of the Ottoman and Safavid Empires.: This reading was based on the accounts of some Islamic (mostly Arabic) sources that saw the Mongol invasions as a divine punishment for the sins of the Muslims in the Middle East and as a purely destructive interlude. Throughout the second half of the twentieth century, this idea lost ground and the period of Mongol domination of Iran and the Levant was re-examined in a new light that identi"#ed it as a remarkable period that o)fered unique opportunities for historical enquiry.; It is in this context that new studies on the political history of the Mongols of Iran were followed by research on their economic impact, their religious transformation and gender relationships between conqueror and conquered people. Studies in the second half of the twentieth century contributed to breaking the paradigm that viewed the Mongols as mere barbarians and prepared the "#eld for a new understanding of the Mongol Empire as a period when East and West were connected and in which the Mongols were not only the military elite of this vast territory, but acted as cultural brokers, facilitating the exchange of goods, people and ideas from the Mediterranean to China.< This approach drew attention onto the variety of political, economic, social and cultural processes that were unleashed in the Middle East by the speci"#c circumstances created by the conquest and settlement of the Mongols. For example, the Iranian region was under a non-Muslim sovereign for the "#rst time since the Arab conquest, o)fering a new religious scenario in the Middle East. Further, Iran’s integration into the main overland trade networks from the Mediterranean to China, freshly stimulated by the Mongols’ interest in trade, o)fered a new economic paradigm for the region. And "#nally, the interaction between nomadic rulers and sedentary subjects generated speci"#c sociopolitical dynamics between conquerors and conquered that shaped institutions and traditions in the area. These special characteristics and many others have contributed to the expansion of the historical enquiry on the Mongols in the Middle East in the last few decades.

= This has been mainly the approach of some nationalistic history produced in Iran, China and Turkey where the period of Mongol dominion is generally regarded as a dark period or transition from one glorious moment in the past to another. > See Morgan, “The Mongol Empire: A Review Article,” 120–25. ? The main promoter of this idea was Thomas Allsen with works such as “Biography of a Cultural Broker. Bolad Ch’eng-Hsiang in China and Iran,” Commodity and exchange in the Mongol Empire, and Culture and Conquest in Mongol Eurasia. 3ABCDEFGBHDA @

It is in this context that this book aims to contribute to the history of the period. It addresses various aspects of continuity and transformation in Mongol and local society in Iran in particular and the Middle East in general. The di)ferent contributions explore long- and short-term continuities and discontinuities in di)ferent layers of Iranian society. A previous attempt to explore these issues was made by A. K. S. Lambton more than a quarter of a century ago.I This pioneering work had a more comparative approach as it included sociopolitical developments in Persia during the pre-Mongol period. Lambton saw the Seljuq rulers as continuators of Islamic traditions and institutions while the Mongols represented in her view “a break with the past: the nomads [Mongols] were the state and political rule was in the hands of their leaders, who formed a kind of military aristocracy. They were hostile to settled life and exploited the townspeople and the peasants.”J Over 25 years since this observation, the research produced on the Mongol domination of the Middle East has shown that the Mongol invasions did not create such a clear-cut break with the past. Recent work by David Durand-Guédy, in particular, emphasizes ways in which Seljuq practice anticipated the Mongols.!K In fact, as much of the research presented here shows, continuity prevailed in some areas while in others transformations appear clearly. Further, the present works also look at elements of continuity and transformation not only among the Persian population or Islamic institutions but also among the Mongols who came from the Far East and settled in the Middle East. We hope that this dual approach will produce a more complex and coherent view of the Ilkhanid period. This book is the result of a joint endeavour between a mentor and his disciple. It represents, to some extent, both a beginning and an end of an academic journey into the Mongol Empire for both editors of this volume. For one it represents the closure of a fascinating period of academic activity as a gradu- ate student at the University of Cambridge and for the other, it is something of a valedictory engagement with the Mongols while he looks ahead to exploring more fully other periods of Persian culture, history and the arts. The idea of this book began back in 2009 with the possibility of proposing a symposium on the Mongols’ Middle East for the World Congress for Middle East Studies (L5481M) in Barcelona in 2010. It took a good number of years to put together the original papers presented in the Conference and complement them with other contributions from colleagues and friends. But after an arduous

N Lambton, Continuity and Change. O Lambton, Continuity and Change, 26. *P Durand-Guédy, “Ruling from the Outside” and “The Tents of the Saljuqs.” Q 01 234567 720 81693661 editorial journey, we are very grateful that the research contained in this volume is "#nally becoming available in print. The symposium on which the present publication is based aimed to challenge some pre-established concepts about the role and the legacy of the Mongols in the Middle East, where aspects of politics, religion, culture, society and gender were taken into consideration in searching for a more comprehensive understanding of the role of this ‘external’ ruler in the region. The symposium brought new ideas and concepts relating to the Mongol Empire to the forefront. Papers explored the inRSuence of Mongol dominion on the cultural, political and social life of the region in three panels, each devoted to analysing di)ferent patterns of continuity and transformation in a variety of disciplines. Not all the participants of the symposium were able to contribute to this volume but fortunately, colleagues like Esther Ravalde, Florence Horus, Aptin Khanbaghi and Boris James have generously contributed papers to enhance the quality of the book and enrich the debate. The present collection is divided into four parts, each dedicated to a speci"#c area of Mongol activity in the Middle East. The arrangement aims to thematically cover aspects of the Mongol conquest and settlement in Iran. The initial section contains chapters dealing with the conquest of the Middle East by the Mongols and the composition of the Mongol army that came to Iran in the 1250s. Timothy May explores the tactics of Mongol conquest that he describes as a “Tsunami Strategy.” In his view, this strategy can be explained simply as a pattern in which the Mongol army “invaded a neighbouring area, devastated a large region, but then receded back into the empire while retaining only a relatively small portion of the invaded region.” However, as he also points out, this tactic was more complex and required elaborate planning and coordina- tion. For that reason, the chapter also explores the tamma units of the Mongols and the fundamental role that they played in defending and expanding the Empire as border armies. In addition, May uses the conquest of Anatolia as a case study in which the Tsunami strategy seems to have been partially abandoned, but he highlights the continuity of the Mongols’ military institutions in the region up to the arrival of Hülegü. In the second chapter, Reuven Amitai addresses the Mongol army in the Ilkhanate by exploring aspects of the seden- tarization of the Mongols and the possibility that this process would (or would not) have a)fected changes in the composition of the Mongols’ military. His chapter draws comparisons with other nomadic societies that settled in the Middle East, like the Mamluks and the Great Seljuqs, and engages in dialogue with the secondary literature on the topic. Amitai concludes that the Mongol army did not undergo substantial change in its composition, technology and 3ABCDEFGBHDA T strategy, maintaining the nomadic elements that characterize the Mongol campaigns of the thirteenth century into the fourteenth century. The second part moves the centre of the analysis to the establishment of a Mongol-controlled Middle East with chapters dealing with politics, economy and religious issues in the Ilkhanate. In chapter 3, Esther Ravalde investigates the role of the famous Iranian vizier Shams al-Dīn Juwaynī (d. 1284) in the Mongol court. She explores the way in which the Ilkhanate was governed by looking at the role of patronage that the vizier used to establish his authority across the Mongol territories. Through patron-client relationships, Ravalde claims that Juwaynī managed to “collect taxes, strengthen Ilkhan rule in Anatolia, mediate between the ruler and the ruled, and develop alliances with the religious classes.” She suggests that although there is a continuity in the institution of the vizierate under the Mongols, the case of Juwaynī shows that this position had limitations if compared with previous dynasties that ruled the area. In this view, the Ilkhanid vizier was not as powerful as before and had to negotiate with Mongol elites and the Ilkhanid government to administer the realm. The transformation of a political institution such as the vizierate is followed by Bruno De Nicola’s chapter on patterns of continuity and transformation in the economic status of Mongol women. He observes both continuity and transformation in the way women acquired and accumulated wealth. By exploring the di)ferent ways in which women received and used their resources, he traces the evolution of these practices as the Mongols moved into the Middle East. He suggests that the economic status of women mutated and adapted when they reached Iran to the point of expanding the areas of these women’s economic activity as the resources at their disposal increased. However, he points out that faced with this economic success of Mongol women, change was triggered by Ghazan Khan, who identi"#ed the resources in female hands and appropriated them for his own political bene"#t. Finally in this section, Florence Horus deals with how religious beliefs a)fected the implementation of the death penalty in Mongol territory in the Middle East. By exploring di)ferent conceptions of capital punishment among the di)ferent religions that inRSuenced the Mongols, she suggests that punishment and amnesties were connected to the di)ferent religions that coexisted in the Ilkhanate. This multi-faith environment, she argues, meant that the inRSuence of one or another religion on the applicability of the death penalty prevailed, depending on the political context in which the Mongols of Iran found themselves at a given time. The third section is dedicated to the continuity and transformation of artistic production under the Mongols in the Middle East. In the initial chapter of this section, Michal Biran explores the scarcely researched "#eld of music U 01 234567 720 81693661 in Ilkhanid Baghdad. The chapter uncovers the life of Ṣafī al-Dīn Urmawī (613–93/1216–94), who was the musician of the ‘Abbasid Caliph at the time of the Mongol invasion of Iraq. Providing "#rst a translation of the musician’s own account of the fall of Baghdad and then exploring the cultural life of the city after the Mongol conquest, Biran suggests that at least in the "#eld of music, the Mongols did not provoke a rupture in the artistic production of the ‘Abbasid dynasty nor did they precipitate a decline in Islamic culture in Baghdad. In the following chapter, Judith Kolbas proposes to look at the Tār"̄kh-i Jahāngushā, the historical work of ʿAṭā Malik Juwaynī (d. 1283), from a literary point of view rather than the usual historiographical approach. She advocates seeing the work as an epic prose text with clear motifs that would have helped Juwaynī to explain to a Persian audience the new political order brought by the Mongols. The "#nal chapter in this section is dedicated to a di)ferent method of artistic representation that developed during the Mongol period. Karin Rührdanz explores the development in the illustration of manuscripts that occurred from the Mongol into the Timurid period. She notices that in the "#fteenth century there was an increasing preference for illustrating works that expressed mythi- cal ideas in the form of poetry, but also a continuity with the Mongol period in the illustration of classical Persian works such as the Shāhnāma and Kalīla wa Dimna. However, the more popular genres that she discusses, depicting subjects still not entirely identi"#ed, seem to have fallen out of the repertoire after the Jalayirid period. The fourth section contains chapters dedicated to the interaction of the Mongols’ Middle East with its neighbours. The "#rst chapter by Aptin Khanbaghi focuses mostly on Anatolia to explore the role played by the Mongols in ‘changing the fate’ of the Persian language. By looking back at the status of Persian in the pre-Mongol period and comparing it with the impact of the Mongol conquest, Khanbaghi suggests that the Mongols provided the umbrella under which a Persianized Turkish population that was already present in Iran could expand Persian language and literature to territories further a"#eld such as Anatolia, southern Khurasan and India. Continuing with territories subject to the Mongols but with a certain degree of autonomy, Dashdondog Bayarsaikhan’s contribution looks at the Mongol implementation of the military and administrative institution of the darughachi in Armenia. The institution served as a means to collect taxes and control the region newly conquered by the Mongols. Yet, in doing so, it not only fundamentally changed the status of Armenia in the Mongol Empire by directly incorporating the region into the Ilkhanate after its formation in the late 1250s but also helped the local rulers (in this case Armenians) to adapt to Mongol command. The 3ABCDEFGBHDA Y establishment of the Ilkhanate had a clear impact on life and institutions of regions bordering Iran such as Armenia and Anatolia. However, less intuitive is the inRSuence that Ilkhanid subjects had in areas far beyond the Ilkhanid borders such as Eastern China. The case discussed in the chapter by George Lane included in this section focuses our attention on the city of Hangzhou in the heart of Mongol China. Through the evidence provided mostly by inscriptions in steles and Chinese sources, the chapter explores the cultural, economic and religious activity of the Phoenix Mosque of Hangzhou, which played an important role as the place that brought together Iranian and Central Asian Muslims coming from the western parts of the Empire to China. Closing this section, the contribution of Boris James covers the relations between the Ilkhanate and the Mamluks by looking at the role that di)ferent Kurdish tribes played in the long- standing conRSict pursued by the Mongols of Iran and the Mamluks of Egypt. James argues that these tribes acted as a bu)fer in the conRSict, shifting sides and controlling bordering areas between the two major empires of the Middle East at the time. Finally, the book ends with an epilogue containing the contribution of Charles Melville. This chapter o)fers a discussion of the end of the Ilkhanate as a political entity and explores the reasons for its collapse. Melville confronts modern views and contemporary accounts of the end of Ilkhanid rule and looks at the di)ferent reasons that have been suggested to explain the fall of the Ilkhans, such as acculturation, Islamization and the ‘natural’ limitations of nomadic rule. With all these elements, he suggests that the end of the Ilkhanate might have been due to a ‘loss of identity’ and an erosion of Chinggisid legitimacy in a realm that remained an obstinately Mongol regime caught in the “turbulence of acculturation,” which provoked its collapse from within. In sum, the contributions to this volume serve as a good example of the fertile ground that Ilkhanid rule in the Middle East provides for historians of the period. They also help us to understand di)ferent aspects of the evolution that occurred in institutions, economic conditions, military organization and civil society both among the Mongol conquerors and the native populations. The conclusion that seems to emerge from the research presented here is that patterns of both continuity and transformation can be observed depending on where we focus our attention. The interaction between the Mongols and the people of the Middle East was a dynamic and complex one which cannot be covered fully in this one volume. The editors hope, nonetheless, that these contributions will contribute to maintaining interest in this fascinating period in the history of the Middle East and encourage further research into the history of the Ilkhanid dynasty of Iran. Z 01 234567 720 81693661

Bibliography

Allsen, Thomas T. “Biography of a Cultural Broker. Bolad Ch’eng-Hsiang in China and Iran.” In The Court of the Ilkhans 1290–1340, edited by Julian Raby and Teresa Fitzherbert, 7–22. Oxford: Oxford University Press, 1996. ———. Commodity and Exchange in the Mongol Empire: A Cultural History of Islamic Textiles. Cambridge: Cambridge University Press, 1997. ———. Culture and Conquest in Mongol Eurasia. Cambridge: Cambridge University Press, 2004. Amitai-Preiss, Reuven. “Evidence for the Early Use of the Title īlkhān among the Mongols.” Journal of the Royal Asiatic Society 1:3 (1991): 353–61. Boyle, John Andrew. “Dynastic and Political History of the Il-Khāns.” In The Cambridge History of Iran, vol. 5, edited by John Andrew Boyle, 303–421. Cambridge: Cambridge University Press, 1968. Durand-Guédy, David. “Ruling from the Outside: A New Perspective on Early Turkish Kingship in Iran.” In Every Inch a King. Comparative Studies on Kings and Kingship in the Ancient and Medieval Worlds, edited by Lynette Mitchell and Charles Melville, 325–42. Leiden: Brill, 2013. ———. “The Tents of the Saljuqs.” In Turko-Mongol Rulers, Cities and City Life, edited by D. Durand-Guédy, 149–89. Leiden: Brill, 2013. Lambton, Ann K. S. Continuity and Change in Medieval Persia: Aspects of Administrative, Economic, and Social History, 11th–14th Century. Albany, 2.[.: Bibliotheca Persica, 1988. Morgan, David. “The Mongol Empire: A Review Article.” Bulletin of the School of Oriental and African Studies 44:1 (1981): 120–25. ———. The Mongols. 2nd ed., Oxford: Blackwell, 2007. ———. “The Ilkhanate, 1260–1335.” In The Cambridge History of the Mongol Empire, edited by Michal Biran and Kim Hodong. Cambridge: Cambridge University Press, forthcoming. Spuler, Bertold. Die Mongolen in Iran. Politik, Verwaltung und Kultur der Ilchanzeit 1220–1350. 4th ed. Leiden: Brill, 1985. 3ABCDEFGBHDA \ a a Se East China u outh China Se S Hangzho ) T KHAN u

Daid (Beijing

THE GREA e

oru

i b

-Q

o G Bengal Qara EMPIRE OF Bay of

s n i a t n u o M

E y a t l AT A I KHAN a ATA GH Balkh bian Se a a Samarqand CHA Ar v Herat Bukhar E Mer Nishapur AT a al

f Se a Ar l

Shiraz u G n i ILKHAN ia Isfahan s er a P Sultaniyy -i Sulaiman) GOLDEN HORDE

Old-Sara Caspian Se

g i

briz l akht

o V Baghdad Ta Medina Sara Caucasus w- w Maragha Mecca Ne a a ny Mosco Sughurlakh (T Cairo Ko v Black Se Damascus Kie a The Mongol Empire, c. 1294. Empire, The Mongol

nean Se ra Mediter ]3^FC_ ` . a a` 01 234567 720 81693661

bHcFC_ `./ The Ilkhanid territories. ]CDd %&'(&)*, +,-.*, /'0 1234-*, e_Ff_A 7dHBgH gAE 8HGhgi jHCgA (_Ek.), /``T. !"#$ % The Mongol Conquest of the Middle East

∵

!"#$%&' ( Mongol Conquest Strategy in the Middle East

Timothy May

Since the Mongols created the largest contiguous empire in history, it is important to explore the process by which the Mongols incorporated territory into this empire. Through this it becomes clear that the Mongols possessed a standard strategy initiated during the reign of Ögetei, although antecedents clearly exist from the period of Chinggis Khan, whereby they invaded a region and caused considerable devastation but only intended to keep a small portion of the area they conquered. Meanwhile, the Mongols established a military force known as a tamma in this new borderland, which then used the region to control the Mongol frontier as well as to launch raids or intimidate neighbor- ing powers.) I term this strategy as the “Tsunami Strategy.”* An examination of the Mongol invasion of the Seljuq sultanate of Rūm is particularly useful in understanding this process as, at ,-rst glance, it appears to have been a devia- tion from the Tsunami Strategy. Furthermore, the resulting political relationship between the Mongols and the Seljuqs demonstrates the complexity of the repercussions of the use of the Tsunami Strategy as well as the role of the tamma within it. Finally, to fully appreciate the scale on which the Mongols applied this strategy, their actions within Rūm will be placed in the overall context of Mongol operations in the Middle East, demonstrating that the Mongols viewed it as a theatre of operations rather than haphazard conquests.

In its simplest explanation, the Tsunami Strategy meant that the Mongols invaded a neighbouring area, devastated a large region, but then receded back into the empire while retaining only a relatively small portion of the invaded

. For more on the tamma, including the term’s origins, see Aubin, “L’Ethnogenèse des Qaraunas,” 65–95; Buell, “Kalmyk Tanggaci People,” 41–59; Qu Dafeng, “On the Qusiqul Army,” 266–72; Ostrowski, “The tamma,” 262–77, and see below. The ,-rst detachment that appears to function in the manner of a tamma appears in 1209, when Chinggis Khan assigns Toquchar with troops to guard the western frontier of the Mongol state. 0 May, “The Mechanics of Conquest and Governance,” 131–36. My thanks to Michal Biran for her suggestion to change it from the more cumbersome “Tsunami Method of Conquest,” which I used in my dissertation, to the “Tsunami Strategy.”

© !"#$#!%$&!' ()$%% #*, %'$+'#, ,-./ | +"$ .-.../1/23425561.63,/_551 (1 234 region. Of course, the strategy was more complex than this and consisted of a set number of steps. The ,-rst step involved the gathering of complete intelligence on the enemy before the invasion was made or even before war was declared. This intelligence originated from earlier raids and diplomatic missions, as well as reports from merchants and other travellers. The next step was planning the campaign based on the gathered intelligence. This planning was essential as the Mongol military commanders ultimately operated on a very rigid time schedule with all operations being enacted within it.5 Although the independent commanders acted according to their own decisions, they still operated within the con- ,-nes of the master schedule. Thus columns hundreds of miles apart could still rendezvous against strongholds. These rendezvous points were decided at a quriltai or congress at the same time as the armies were called up. The next event was the invasion, which usually began with the Mongols striking at one point. This was the crashing of the wave. After this, the armies divided and created havoc on a wide front and then eventually converged upon a pre-designated region. Often, the convergence appeared as if the Mongols were retreating, but in reality this was the ebb of the tide, before it 67owed back to the frontier. By striking in many columns, the Mongols prevented their opponents from unifying, as each noble needed his own forces to defend his own locality. Thus, the local power could not risk assisting anyone else, thereby allowing the main Mongol force to concentrate on the prime target. Meanwhile, the surrounding regions were devastated. Again, the local forces remained on the defensive. It should be noted that this di8fered from a “scorched earth” policy. In the latter, the intent was to destroy and often literally scorch the earth so that the region could not be occupied (by invader or defender). The Mongols’ intent in the Tsunami Strategy was to destroy the defensive and o8fensive capabilities—not to make it uninhabitable. This method of invasion also confused the defenders. Often, while the defenders thought they were being attacked, the Mongols suddenly retreated. The Mongols did not conquer all of the areas their forces invaded; however, they withdrew to a now stabilized and conquered area. Thus the Mongols did not make sweeping conquests in which they gained vast amounts of territory, but rather, they slowly and steadily incorporated territory into the empire. Once a small amount of territory was ,-rmly in their control, they moved into the next land, which was already weakened. Therefore, the Mongols appeared to have conquered rapidly, but it was because they had already undermined the defence of the next area.

9 Sinor, “On Mongol Strategy,” 239; May, The Mongol Art of War, 82–83. ;<=>

One could argue that as the Mongol forces were divided, they could be destroyed piecemeal. This is not so, as this tactic only worked if the opponent was more mobile than the Mongols. Rarely, if ever, did the Mongols meet a foe who could equal their mobility, and the individual columns were strong enough to defend themselves. Thus, unless a force was ambushed, the Mongols could always retreat faster than their opponents could advance. Furthermore, because of their sophisticated screen of scouts and signalling techniques, the Mongols were rarely surprised.I If the Mongols had remained in a single column for the entire campaign, they ran the risk of being tied down by a united force. Thus, if the army was defeated, they could not avenge it as at ʿAyn Jālūt in 1260. The success of the Mongols was not that they were invincible, but that they did not allow many opportunities for their armies to be eliminated. One force could be defeated, but it was almost guaranteed the victors would have to face another force very quickly, one which often arrived only days after its predecessor was defeated. Thus the Mongols could rely on reinforcements while the enemy, no matter how resilient, risked being worn down by the constant dispatch of new armies. Normally, the Mongols took the most heavily forti,-ed or most important strongholds last, thereby exhausting their opponents psychologically and materially. Refugees, those who were not enlisted or slaughtered, 67ed to these points, creating a strain on food supplies and most importantly, water. Furthermore, most refugees were not soldiers or even capable of assisting in the city’s defence due to their weakened condition.L The key to the success of the ‘Tsunami’ was the transition from military rule to civilian rule. Based on Mongol activities in Iran and the Jin Empire, a pattern of transition from military to civil government occurred. Paul Buell, in his study of the Mongol ruling establishment in China, noted that the chiliarchy or the minqan (pl. minqad) was the ,-rst ruling establishment which the Mongols utilized.M It was the basic unit of organization in the early Mongol government, both for taxes and for military levies. Over time, this changed as the empire expanded and became increasingly complex. Buell also noted that initially in China, the most important early institution through which the Mongols

N Polo, The Travels, 261. The Mamluks of Egypt are among the few who did surprise the Mongols, as they did at ‘Ayn Jalūt. While the Mongols were aware the Mamluks were in the vicinity, they did not appear to know their exact location until the Mamluks seized the better position. O Carpini, “History of the Mongols,” 36, 38; Paris, English History, P:312; Hung, Men-da bei-lu, 67. Q Buell, “Kalmyk Tanggaci People,” 47. (R 234 ruled was the tamma.S These forces were military in nature and situated on the frontier of the Empire, usually newly conquered lands. Indeed, the tamma was the main instrument through which territory was incorporated and held for the Mongol Empire, and the tamma often remained at its post for several years.T If possible, the tamma also was used to expand the Empire into bordering territories. The ,-rst appearance of this term connected to a unit used in military operations appears in The Secret History of the Mongols, §273, which discusses Ögetei’s establishment of tammachin in the former Jin Empire. Igor de Rachewiltz views these as mounted garrison troops. The Chinese term is tamma or scout horse.U Although the term ,-rst appears in §273 of The Secret History of the Mongols, which is situated in 1234, the ,-rst detachment appears earlier. Buell indicates that it appeared with Muqali in 1217 or 1218 when that Mongol general received “,-ve ‘tribal’ groupings comprised of Onggirat, Ikeres, Mangqut, Uru’ut, and Jalayir, the latter belonging to his own family, to secure conquered Chinese domains in the qan’s absence. To this tamma nucleus were associated various local groupings as an ‘outer’ circle around the Mongol ‘royal’ element.”)V Evidence of units that functioned in the manner of a tamma appears even earlier. In the winter of 1211, Chinggis Khan prepared for war against the Jin. In addition to gathering his army, he still maintained vigilance against Güchülüg of the Naiman in the west and the possibility of internal rebellion in his absence. To this end, he dispatched Toquchar to the west.)) Meanwhile, Temüge Otchigin remained in Mongolia to maintain order. Thus with an army of roughly

W Buell, “Kalmyk Tanggaci,” 45; Aubin, “L’Ethogenèse des Qaraunas,” 74–75; Ostrowski, “The tamma,” 264. X Chormaqan’s tamma remained in Iran from 1230 to 1237 before moving to the Mughan plain in Azerbaijan in 1234–1235. It remained there until Baiju moved it to Rūm with Hülegü’s campaign, as will be discussed in this paper. Also see Aubin, “L’Ethnogenèse des Qaraunas,” 65–95, on how a tamma in Afghanistan became a distinct group over time. Y Rachewiltz, ed., The Secret History, §273 (hereafter !"#). Also see Igor de Rachewiltz’s brief commentary on the term in Idem, !"#, 2: 1002–1003. .Z Buell, “Kalmyk Tanggaci People,” 46. Buell bases his dates on the Shengwu Qin Zheng Lu. See He Qiutao, Shengwu Qin Zheng Lu, 50 (Hereafter !$%&'). Muqali received a tümen from the Onggud, four minqad of Uru’ut, one mingqan of Arulat, one Mangghut minqad, The Onggirat provided three minqad, while the Ikeres and Jayilar provided two minqad each. This force was also supplemented by Khitans who joined the Mongols in 1218. .. !$%&', 49–50; Rashīd al-Dīn, Jāmi‘ al-tawārīkh, ed. B. Karīmī, 320 [hereafter Rashīd/ Karīmī]. Idem Jami‘u’t-tawarikh [sic] Compendium of Chronicles, translated by Wheeler M. Thackston, 1: 213 [hereafter Rashīd/Thackston]. ;<=>

100,000 men, Chinggis Khan marched south in April, 1211. Toquchar’s position placed him in position to not only protect Mongolia against the Naiman prince Güghülüg, who took refuge in Qara Khitai after being defeated at the Battle of the Irtysh River in 1209, but also to render aid to the Uighurs and to Arslan Khan of the Qarluq, who also submmited in January or February 1211.)* Toquchar served as the ,-rst tammachi on the western border of Mongolia as a deterrent to Güchülüg, a monumental role. Although Toquchar is not speci,-- cally referred to as a tammachi in the sources, he may have been the ,-rst such person in that position. Members of the tamma were known as tammachi (pl. tammachin), although the sources usually refer speci,-cally to the commander of the tamma when using this term. Attempting to de,-ne the tamma has, nevertheless, generated a series of scholarly debates.)5 Donald Ostrowski questions whether the tamma/ tammachi refers to both the individual and the group, or whether one term refers to an individual and the other to a group, or if there is even another way of understanding the term.)I Grammatically, however, the meanings of tamma and tammachi are very distinct. The –chi ending in Middle Mongolian, as well as Classical Mongolian, is a nomen actoris su8,-x, thus making the person it refers to one who performs or is a part of the tamma.)L One would never refer to the institution of tamma as a tammachi, as that would only signify a member of it. The sources also clearly indicate that individuals were tammachi/ tammachin and that they belonged to a tamma. The tamma consisted of a main force and an advance force of scouts known as the alginchi (pl. alginchin).)M The alginchin were stationed closer to the cities, while the tamma remained in better pasture lands. The tamma consisted of troops from various tribes and regions, or minqad, from across the empire,

.0 !$%&', 41; Rashīd/Karīmī, 320; Rashīd/Thackston, 1: 213. .9 For a discussion of the linguistic arguments on the origin of the word as well as its meaning, please refer to Buell, “Kalmyk Tanggaci People,” 41–59. Also see Ostrowksi, “The tamma,” 262–77. Ostrowski focuses on the non-Mongol sources beginning on page 268. In my opinion, the Mongol sources are the most relevant as they are the sources that indicate how they used the word. The Persian sources and others simply translate it or attempt to use it in conjunction with another word (e.g. lashkar tamma) so that they can better understand the meaning of the new loan word. .N Ostrowski, “The tamma,” 268. .O Refer to Poppe, Grammar of Written Mongolian, 40–41. Also see Grǿnbech and Krueger, An Introduction to Classical (Literary) Mongolian, 46. .Q Buell, “Tribe, Qan, and Ulus,” 70. (^ 234 while the commander was not necessarily of Mongolian origin.)S Despite its importance, it seems clear that the tamma was a separate unit or army from the regular forces. The Yuan Shi notes there were two distinct forces, the Meng-ku chün or Mongol army, and the Tanmachi jun or tammachi army. The distinction lay in the composition of these forces. The Mongol army was solely comprised of Mongols whereas members of various tribes or nations formed the tammachi army.)T While the antecedents of the tamma appeared in 1211 and 1218, a regular tamma system was implemented in North China in 1236. With the alginchi of the tamma established near the cities of Yidu, Jinan, Pingyang, Daiyuan, Zhenting, Da-ming, and Dongping, the tamma aided the expansion of the Mongol Empire. However, these units eventually moved forward. As Ch’i- ch’ing Hsiao noted, “Nevertheless, since the Mongols were at this stage looking for further conquest, troops were moved forward as campaigns proceeded, and no permanent garrison system was possible.”)U Another unit known as the cherig was also associated with the tamma. The term is Mongolian. In the Secret History of the Mongols, it is used to refer to a soldier.*V It can also be used to refer to many warriors in the sense of ‘troops.’*) Cherig remained in use as a word for ‘soldier,’ ‘warrior,’ and even ‘troops’ or ‘army.’ But at the same time, a distinction was recognized between cherig and tammachin. The members of the tamma are never referred to as cherig. Also, the cherig tended to be recruited from the local forces in the occupied lands. Structurally, the cherig tended to mirror the tamma; however, its primary purpose was to garrison locations.** The terminology of cherig and tamma may have simply been used to distinguish between the nomadic tammachi and

.W Yuan Shih (Ch’i-Ch’ing Hsiao), 72–91 (chapter 98) (hereafter _`); Grigor of Akanc, “The History of the Nation of the Archers,” 333; Ostrowski, “The tamma,” 264. .X _`, 73 (chapter 98). Mongol in this sense should also include Turkic groups such as the Kereit, Naiman, etc., who were conquered as part of Chinggis Khan’s uni,-cation of the Mongolian steppes. .Y Ch’i-ch’ing Hsiao, The Military Establishment, 53. 0Z See !"#, §100. Here, the term is ,-rst used in reference to the Merkit warriors who attacked Temüjin’s camp and carried o8f the abandoned Börte. 0. !"#, §107. In this passage, Temüjin is bringing his troops to join Toghril and Jamuqa in order to rescue his wife Börte from the Mongols. The Mongolian passage reads “Temüjin tendece cerik e’üsgejü” or “Temüjin started from there with [his] troops.” The translation is mine. For the Mongolian transliteration see Rachewiltz, Index to the Secret History, line 2639. This usage appears again in !"# §133; Rachewiltz, Index to the Secret History, line 3613, 3619 and on multiple other occasions throughout `"2. 00 Buell, “Tribe, Qan, and Ulus,” 73–74. ;<=>Kucha. Stationed in Syria in 1260, this tamma took part in the battle of ʿAyn Jālūt and lived up to their reputation.*L Although the tamma existed on the borders between the Mongol Empire and other realms, and thus was in a position to defend or expand the empire, the Mongols did not rebuild fortresses to defend their territory. Rather, they carried out a general policy of razing city and garrison walls.*M In general, the Mongols did not use the tamma as a means of garrisoning a region. Rather, they used auxiliaries who were more accustomed to such duties. The Mongols, however, did use tamma camps to control regions and repel hostile attacks. The camps were spread out for pasture purposes and protected by patrols while messengers maintained communications. H. Desmond Martin noted that, “As an additional precaution, in hostile or semi-hostile country, areas containing such camps were frequently surrounded by a belt of desolation.”*S In this manner, the tamma essentially served the same purposes as did a Western European castle, the various walls of China,*T or any fortress for that matter.

09 Allsen, Mongol Imperialism, 192–93. 0N Martin, The Rise of Chinggis Khan, 11. 0O Allsen, “The Yüan Dynasty and the Uighurs of Turfan,” 243–79; Rossabi, ed., China Among Equals, 265–66. 0Q ʿUmar Ibn Ibrāhīm al-Ansī al-Anṣārī, Tafrīj al-kurūb fī tadbīr al-ḥurūb, 119. 0W Martin, Chinggis Khan, 32–33. 0X Di Cosmo, Ancient China and its Enemies, 139, according to whom, “These walls e8fectively inaugurated a new type of defensive system, possibly, as a consequence of the creation of cd 234

Defensively, the castle might serve to guard borders, as did the tamma. Like the castles in the Crusades, however, the tamma was also an o8fensive force that gave the Mongols a base of operations and a means of intimidating neighbouring realms.*U Although the tamma functioned much as fortresses in other kingdoms and empires, the Mongols themselves (or most nomads for that matter) did not construct fortresses. The Mongols viewed ,-xed forti,-cations with disdain and razed them.5V The importance of the tamma cannot be overstated and examples of it are numerous: Muqali’s army in Northern China, Toquchar’s forces in Western Mongolia, Chormaqan and Baiju in Azerbaijan, the forces under Köketei and Sönitei in the Volga-Ural region prior to Batu’s invasion of the Rus’. Buell further demonstrates that the tamma was always established on the fringes of the empire, literally on the border between nomadic and sedentary cultures.5)

infantry armies whose maneuverability required more extensive use of, and control over, natural features. The ‘long’ walls appear also to have been strategic forti,-cations aimed at asserting a state’s political and military control over a given area. Those areas could be strategically important not just defensively but also o8fensively, in as much as control over a mountain pass or a river ford could either block an advancing army or secure passage to one’s own troops. Moreover, like roads, walls provide the logistic infrastructure to facilitate communication and transportation, vital elements, for armies employed in the occupation or invasion of a foreign territory. Thus military walls could be an integral part of an expansionist ‘o8fensive’ project, and we can assess their function properly only in their historical context.” Comparable examples would be the Norman castles built immediately after 1066 by William the Conqueror in England, and later in Scotland in the thirteenth century. A more modern comparison is with the forts built by Americans along their Western frontier from the eighteenth century through the nineteenth century. 0Y France, Western Warfare in the Age of the Crusades, 1000–1300, 105; Di Cosmo, Ancient China and its Enemies, 143. The use of the castle as an o8fensive device was particularly common in Western Europe and in the Latin East. In these cases, the castle served as a base from which to attack as well as a source of intimidation and domination for neighbouring lands. The walls of ancient China are commonly thought to have been built to repel nomadic attacks. However, Cosmo points out that the walls were built usually after the Chinese had conquered formerly nomadic territory: “after a substantial drive into foreign lands. The building of forti,-cations proceeded in tandem with the acquisition of new territory and the establishment of new administrative units. In other words, the state of Ch’in, Chao, and Yen needed to protect themselves only after they had taken large portions of territory from people and had chased the nomads away from their homelands. Surely at some point the forti,-cations did acquire a ‘defensive’ function, but the context suggests strongly that this defensive role was subordinate to a grander strategy, one that was militarily o8fensive and territorially expansionist, pursued by all three Chinese states.” 9Z Matthew Paris, English History, 312; al-Anṣārī, Tafrīj al-kurūb fī tadbīr al-ḥurūb, 119, fn. 7. 9. Buell, “Kalmyk Tanggaci,” 45. ;<=>

Typically pasture determined the exact location of the tamma. When Batu invaded Hungary, he left thirty thousand men with the herds to support them to control the Rus’ and Qipchaq lands. These had to be stationed at intervals between steppes and sedentary areas primarily to provide enough pasture. This left the Mongols with an e8fective force to maintain control in both zones. Over the course of time, the governance structures of the conquered lands changed. Thus, darughachi or governors would be sent in accompanied by bichikchi or secretaries.5* The darughachi or basqaq “controlled the activities of [the local elite], collected taxes and was endowed with military might.”55 Eventually, an administrator known as the yeke jarquchi (great or high lawgiver, one who decrees) gradually replaced the tammachin.5I While the Mongols did manipulate local administrative structures for their own use, in the newly conquered lands the Mongols had to apply their own system in order to maintain order after the armies advanced into unconquered areas. In doing so, the various local and native systems of governance were integrated into a Mongol administrative framework. This is what allows for the ‘Tsunami’ form of conquest. While an advance force created the political vacuum in the neighbouring regions, the tammachin ruled the core of the recently conquered area. Gradually, it was converted to civil and bureaucratic government along the lines that Buell suggests. The power vacuum and disturbances which the advance force created in the outlying regions allowed this transition to occur. To simplify, the advance force disrupts everything in the neighbouring regions, thus allowing the tammachin to rule and pacify without interference. Thus when the Mongols invaded the lands across their border, the territory behind them was being prepared for civil authority. Eventually, a civil government controlled by the darughachin replaced the tamma, thus allowing the tammachin to move forward into new territory. This is not to say that the military governors of the tamma enjoyed the reduction of their power in those regions. By the time the darughachin were

90 Vernadsky, The Mongols and Russia, 220. The darughachi were also known as basqaq in Turkic or shahna in Persian. See also István Vásáry, “The Golden Horde Term Daruġa and its Survival in Russia,” 187–97; idem, “The Origin of the Institution of basqaqs,” 201–206. This may also be found as chapters f and fPP in Vásáry, Turks, Tatars and Russians in the 13th–16th Centuries. On the darughachi see also Dashdondog’s article in this volume. 99 Petrushevsky, “Socio-Economic Conditions of Īrān,” 530. 9N For a detailed analysis see May, “The Conquest and Rule of Transcaucasia,” 142–49; see also the article by Dashdondog, “Darughachi in Armenia” in the present volume. Also see the following primary sources: Juwaynī/Qazwīnī, PP:221–23; Juwaynī/Boyle, 485–86. cc 234 in position, there was no need for a regular military unit as each darughachi possessed a force of troops in order to maintain control of his region.5L Because these forms of government moved in behind the armies, the Mongols could continue to advance as the main armies were not needed to put down revolts. Furthermore, the military and the civil government extracted material support from the populace, which the main armies needed in order to continue the war e8fort. Finally, although the Mongols appeared to be over- extended, they were not because of the Tsunami Strategy. They expanded at a rate they could maintain. To garrison the newly conquered areas, it required more resources than the Mongols possessed, whether in terms of manpower or livestock. By keeping their troops mobile in the steppe, they easily moved to trouble spots rather than trying to rule directly. The transition from military to civil control impacted the empire in two ways. Firstly, the conquests moved forward. Secondly, it prevented the generals from becoming too ,-rmly entrenched in one area. The Mongols implemented this strategy in all instances. During the invasion of Khwārazm, the Mongol forces conquered Māwarānnahr, then Tolui devastated Khurasan, and Chinggis Khan campaigned in Afghanistan and pushed forward into India. Meanwhile, Sübedei and Jebe entered Iran. However, at the end of the campaign the armies withdrew, leaving a holding force in Māwarānnahr. While this was happening, the former state of Qara Khitai was annexed into the empire. Māwarānnahr was then annexed at a later date. In Chormaqan’s subsequent invasion of Iran, that region quickly succumbed. However, the latter sent a 67ying column led by Taimaz into Azerbaijan, Arran, and Armenia in pursuit of Jalāl al-Dīn Khwarazmshah. Thus while Chormaqan laid the foundations for Mongol rule, the only e8fective resistance was by the Khwārazmians far to the west and then it too was destroyed, upon which Taimaz withdrew from Armenia. Only after a few years had passed did Chormaqan invade Armenia and Georgia in full force, simultaneously sending a few probes towards Baghdad.

If one compares the borders of the empire with territories actually occupied and garrisoned by Mongols, a clear pattern develops. First, by devastating Khurasan, Khwarazm, and Mazandaran, the new Mongol territory of Māwarānnahr was protected by a broad belt of destruction. The inhabitants of this belt were either dead or dispersed, the rulers were in 67ight, and there was simply no organized military to threaten Mongol control. Furthermore, the land that had been cultivated could be occupied by nomads, and thus serve

9O Vernadsky, The Mongols and Russia, 70. ;<=>Buqa and Baidar’s positions in Syria and Palestine.

War with the Seljuqs

Prior to 1240 the Mongols expressed little interest in the Seljuq Sultanate of Rūm. The tammachi in the region at the time, Chormaqan Noyan, spent most of the decade consolidating Mongol power in Iran and Transcaucasia after his

9Q Allsen, Mongol Imperialism, 7. c1 234 forces defeated Jalāl al-Dīn. After his death in 1240 the Mongols stationed in Transcaucasia, now commanded by Baiju Noyan, renewed their expansion. This is notable for several reasons with the ,-rst being that it is one of the few instances in which Mongol armies conquered territory during a period in which a Great Khan (Khaghan) did not sit on the throne. Secondly, the campaign was a departure from the steady use of the Tsunami method of conquest. This, however, raises the question of whether the latter was an aberration or an intentional departure. During the reign of Sultan ʿAlāʾ al-Dīn Kayqubād ibn al-sulṭān Kaykhusraw (1219–37), hostilities between the Seljuqs and the Mongols did not amount to anything more than a few raids.5S After Kayqubād’s death, Ghiyāth al-Dīn Kaykhusraw (1237–45) came to the throne. During the initial years of Sultan Ghiyāth al-Dīn’s reign, his rule coincided with that of Chormaqan Noyan. Upon the death of Chormaqan Noyan, a new tammachi and a lieutenant of Chormaqan’s, Baiju Noyan, altered the existing status quo between the Mongols and the Seljuqs.5T It appears that the Seljuqs believed there was little indication the Mongols might invade Rūm as they carried out their own military campaigns close to Mongol territory. In 638 H./1240–41 the Seljuqs laid seige to ʿAmid but withdrew after obtaining a peace treaty.5U Only a year later in 639 H./1242, a Mongol force accompanied by contingents of Georgian and Armenian troops invaded Rūm. Their forces raided up to the fortress of Zārīd, thus engaging the Mongols and Seljuqs in war.IV Grigor of Akner wrote that Baiju, along with Armenian and Georgian troops, attacked the town of Karin, besieging it for two months. Upon capturing the city, they sacked it and even ransacked churches; notably, the Christian Armenians and Georgians took part in the church looting as well.I)

9W al-Dhahab ī, Kitab Duwal al-Islam, 233; al-Ḥadīd, Sharḥ nahj al-balāghah, 81; Bar Hebraeus, The Chronography, 405; Aḥmad ibn ʿAbd al-Wahhāb al-Nuwayrī, Nihāyat al-arab fī funūn al-adab, 348. The incursion took place in 634 H./1236–37, although Bar Hebraeus lists a date of 1240. Nuwayrī and al-Ḥadīd agree on the date. 9X See Melville, “Anatolia under the Mongols,” 51–54. 9Y al-Dhahab ī, Kitab Duwal al-Islam, 242; Golden, An Introduction, 290. Golden believes that by the time of Köse Dāgh, the Saljūqs had been weakened by the 1240 revolt of Turkmen tribes under Baba Isḥāq. NZ Bar Hebraeus, The Chronography, 406. N. Grigor of Akanc, “The History of the Nation of the Archers,” 307; Brosset and Chubinov, Histoire de la Géorgie, 518. Grigor indicates this took place in 1239; however, as no other sources corroborate this, it seems likely that this took place in 1242 at the earliest. ;<=>

The Seljuq Sultan Ghiyāth al-Dīn assembled his army, including a sizeable force of mercenaries, in 1243.I* The mercenaries comprised a variety of soldiers including troops from Aleppo, as well as Greeks and Franks.I5 The Seljuq army met the Mongols at Köse Dāgh, located between Karin and Erzincan, in 641 H./1243. Many of the Seljuq forces 67ed after Aghbagha, a Georgian prince of Gag, defeated the right wing of the Seljuq army; the Sultan himself 67ed to Ankara. In addition to the defeat, Ghiyāth al-Dīn was also concerned that some of his amirs were considering submitting to the Mongols.II The complete withdrawal occurred during the night. Thus on the second day of the battle, the Mongols discovered a deserted Seljuq camp. Initially, the Mongols believed the Seljuqs’ 67ight to be a trap, but once they realized the Seljuqs truly had abandoned the ,-eld of battle they advanced and conquered the rest of the sultanate.IL During the course of their conquests the Mongols sacked Sivas and destroyed much of the city wall. Then they attacked Kayseri, using siege engines.IM In addition, Konya and Ankara fell to the Mongols.IS On their return to Armenia, they attacked Erzincan after it refused to pay tribute. Furthermore, they installed a darughachi to govern Rūm.IT With his realm conquered, Sultan Ghiyāth al-Dīn submitted and agreed to pay tribute in gold, horses, cattle,

N0 Grigor of Akanc, “The History of the Nation of the Archers,” 309. Grigor puts the Sultan’s force at a highly improbable 160,000 men. N9 Bar Hebraeus, The Chronography, 407–409; Jūzjānī, Ṭabaqāt-i Nāṣirī, 1:313; Idem, (Raverty), 162. Jūzjānī viewed Ghiyāth al-Dīn as a great monarch. He also wrote that rather than simply hiring Frankish mercenaries, Ghiyāth al-Dīn entered into an alliance with them against the Mongols. NN Ibn Bībī, Saljūq Nāmah, 205; Grigor of Akanc, “The History of the Nation of the Archers,” 309. NO Bar Hebraeus, The Chronography, 407–9; al-Dhahabī, Kitab Duwal al-Islam, 243; Jūzjānī, Ṭabaqāt-i Nāṣirī, 1: 313; Idem, (Raverty), 162–63; Grigor of Akanc, “The History of the Nation of the Archers,” 309. According to Grigor, Aghbagha’s victory over the Seljuqid right wing was the turning point of the battle, as prior to it the Mongols had been hard pressed. He also notes that the Georgians and Armenians formed the vanguard of Baiju’s army. NQ Bar Hebraeus, The Chronography, 407–9; al-Dhahabī, Kitab Duwal al-Islam, 246; Grigor of Akanc, “The History of the Nation of the Archers,” 309; al-Ḥadīd, Sharḥ nahj al-balāghah, 82. NW Ibn Bībī, Saljūq Nāmah, 205–6; Grigor of Akanc, “The History of the Nation of the Archers,” 309. NX Grigor of Akanc, “The History of the Nation of the Archers,” 309. cR 234 sheep, and slaves.IU Al-Dhahabī noted that the tribute amounted to 400,000 dinars in value.LV In addition to the Seljuq sultanate, the Mongols acquired other territory and otherwise impacted the various powers in Anatolia. William of Rubruck, who wrote in the 1250s, noted that Trebizond was subject to the Mongols.L) Mongols initially attacked Trebizond in 1240, setting ,-re to the citadel, but they did not conquer it.L* Trebizond probably submitted to the Mongols after 1243. It is not mentioned at Güyük’s coronation, at least not by that name, but this is not unusual. Manuel Comnenus I was the ruler. As Trebizond was small and isolated it is rarely mentioned in the Latin, Arab or Persian sources. Lippard, however, thinks that Juwaynm̄ made a mistake concerning the identi,-cation of the sultan of Takavor who attended the ceremony.L5 Juwaynī and Boyle thought that the name was Armenian. Boyle took Takavor to be the Armenian word T’aghavor or King, and so he thought it referred to Haithon or Smpad. However, Takavor could refer to Armenia, Nicea, or Trebizond. Furthermore, since Haithon never went to Güyük’s enthronement in 1246, nor was John Vatatzes PPP of Nicea present, Manuel P Comnenus of Trebizond seems the likely attendee.LI Nicea feared either Turkmen or Mongol invasion, or both. Indeed, John Vatatzes PPP met with Sultan Ghiyāth al-Dīn Kaykhusraw’s envoys and made a treaty. Neither wanted a war as the Greeks were busy in Europe and Ghiyāth al-Dīn was too occupied with the Mongols to enter a second front. John also

NY Ibn Bībī, Saljūq Nāmah, 205–6; Bar Hebraeus, The Chronography, 407–9; Cahen, “The Turks in Iran and Anatolia Before the Mongol Invasions,” 691–92. The treaty and terms came through the e8forts of the Salūq wazīr Muhadhdhib al-Dīn. OZ al-Dhahab ī, Kitab Duwal al-Islam, 246. O. William of Rubruck, “The Mission of William of Rubruck,” 91; Idem, The Mission of Friar William of Rubruck (Jackson), 65. O0 Lippard, “The Mongols and Byzantium, 1243–1341,” 179; Miller, Trebizond, the Last Greek Empire of the Byzantine Era, 1204–1461, 23–24. Trebizond was also friendly with the Sultan of Erzerum and Jalāl al-Dīn before, all enemies of the Mongols, thus giving the Mongols plenty of reason to attack it if the general incentive of plunder was not su8,-cient. Trebizond provided a haven for Jalāl al-Dīn’s army after it was defeated by the Seljuqs and Ayyubids near Erzinjan in 8/1230. Jalāl al-Dīn’s army at that battle may have also been originally from Trebizond. Afterwards Andronicus Gidon, the ruler of Trebizond, submitted to Kaykhusraw and sent troops. O9 Juwaynī/Qazwīnī, 1:207; Juwaynī/Boyle, 250. ON Lippard, “The Mongols and Byzantium, 1243–1341,” 179–80; Het’um, “The Journey of Het’um P,” 178–79. Smbat did attend Güyük’s coronation, but it is unlikely that he would have presented himself as the King of Cilicia. ;<=>

A Second Seljuq-Mongol War

Hostilities 67ared anew in Rūm during the reign of ʿIzz al-Dīn. In 1251, after Baiju went to Iraq, ʿIzz al-Dīn, accompanied by Turkmen, Arabs and Kurds, attacked Mongol detachments and the city of Tokat. The Seljuq prince accomplished little, however, as his brother, Rukn al-Dīn Qilij Arslān, counter-attacked him and liberated Tokat.LU Rūm’s status as a client state or appendage of the Mongol

OO Lippard, “The Mongols and Byzantium, 1243–1341,” 176. OQ Fine, The Late Medieval Balkans, 134. OW Lippard, “The Mongols and Byzantium, 1243–1341,” 177. OX Lippard, “The Mongols and Byzantium, 1243–1341,” 163; Cahen, “The Turks in Iran and Anatolia,” 691–92. OY Cahen, Pre-Ottoman Turkey, 277. c^ 234

Empire changed with the arrival of Hülegü in the Middle East. Accompanied by a large army and as the new overlord of the south-western portion of the Mongol Empire, Hülegü positioned himself in the traditional headquarters of the tammachi, the lush Mughan plain. Thus, when Hülegü came to the Mughan plain to winter in 1256, Baiju had little choice but to seek pasture land for his troops in Rūm.MV Contrary to the way the Mongol commanders have often been portrayed, Baiju in this instance did not simply move his forces into the region and take over the territory by the sword. As Rūm was now a part of the Mongol Empire, he requested pastures where his troops and horses could winter from Sultan ʿIzz al-Dīn in August 1256. ʿIzz al-Dīn, however, viewed the arrival of Hülegü as a diminishment of Baiju’s power and in67uence. Therefore, Sultan ʿIzz al-Dīn did not desire to assist Baiju, viewing the noyan’s status as below that of a tribute- paying client of the Great Khan.M) Baiju quickly corrected this impression by defeating the Seljuqs in battle at Aksaray. As he advanced, towns surrendered to the Mongols and some allies deserted the Seljuqs.M* ʿIzz al-Dīn did perhaps have some justi,-cation for denying pasture to Baiju. Baiju’s army hurt the local economy as the Seljuq sultan could not tax the Mongol army. Furthermore, he risked losing in67uence in the region, as Baiju would be viewed as the real power in Rūm.M5 There was some discussion among ʿIzz al-Dīn’s court about whether to attack Baiju. Some simply wished to comply with Baiju’s request, while others who had the most to lose in terms of status and privilege, argued for war.MI Of course, in hindsight, not only did ʿIzz al-Dīn lose considerable in67uence with Baiju’s victory, but he possibly laid himself open to the possibility of losing considerably more power. After the battle of Aksaray (14 October 1256/23 Ramadan 654 H.), Baiju remained in Rūm, pasturing the herds of his armies. He also spent four months at the Byzantine frontiers. He camped in the north, in Paphlagonia or near Ilgin, ,-fty kilometres west of Akşehir in central Anatolia. Besütei, Baiju’s uncle, advanced to Denizli in November 1256.ML

QZ Bar Hebraeus, The Chronography, 424–25; Juwaynī/Qazwīnī, P, 93; Juwaynī/Boyle, 608–609. Q. Bar Hebraeus, The Chronography, 424–25. Q0 Bar Hebraeus, The Chronography, 424–25; al-Nuwayrī, Nihāyat al-arab fī funūn al-adab, 349–50; Cahen, Pre-Ottoman Turkey, 276. Michael Palaeologus deserted and returned to Nicea. Q9 Cahen, Pre-Ottoman Turkey, 275–76. One may wonder whether, if he had simply granted Baiju’s request, the Mongols may have lessened his tribute or granted certain favours in appreciation for his cooperation. QN Cahen, Pre-Ottoman Turkey, 275–76. QO Lippard, “The Mongols and Byzantium, 1243–1341,” 23. ;<=>

Perhaps as a rami,-cation of ʿIzz al-Dīn’s actions and disrespect for the Mongol military, Hülegü summoned ‛Izz al-Dīn and Rukn al-Dīn to him. During his meeting with them, the Mongol prince divided Rūm between the two rivals. Rukn al-Dīn received the territory from Caesarea to Armenia and ruled from his capital at Sivas. Meanwhile, ʿIzz al-Dīn received Aksaray to the sea coast and ruled from Konya to the frontier of the Greeks.MM Although his power diminished with the division of Rūm, ʿIzz al-Dīn still challenged the Mongols. According to Bar Hebraeus, out of fear of Baiju, ‛Izz al-Dīn attempted to assemble an army as well as form an alliance with the polity of Melitene or Malatya against Baiju. Again, Baiju’s response was swift and decisive. Baiju’s forces spread out in April 1257/655 into Galatia and Cappadocia, wreaking destruction upon ʿIzz al-Dīn’s domains. As a further punishment, Baiju gave forts that his troops captured to Rukn al-Dīn. Melitene did not escape his notice either. For its part in the rebellion, Baiju forced Melitene to swear fealty to Rukn al-Dīn, e8fectively giving Rukn al-Dīn control of Seljuqid Rūm. Perhaps fortunately for ʿIzz al-Dīn, Baiju then left Rūm to join Hülegü.MS Somewhat oddly, ʿIzz al-Dīn sent his younger brother Rukn al-Dīn to Möngke to ward o8f further Mongol attacks. Rukn al-Dīn, instead of assisting his brother, attempted to gain Mongol backing. In support of Rukn al-Dīn, Möngke gave the Seljuq prince the daughter of Aljaktu and sent Aljaktu to assist him.MT Rukn al-Dīn, with help from Aljaktu, forced ʿIzz al-Dīn to 67ee so that Rukn al-Dīn Qilij Arslān became the ruler of Rūm. ʿIzz al-Dīn 67ed to Aor Khan who then assisted him in defeating the Mongols. They also captured Rukn al-Dīn and imprisoned him, thus restoring some territory to ʿIzz al-Dīn.MU The Byzantine prince, Michael Palaeologus, lived with the Seljuqs and fought the Mongols alongside the Seljuqs at Aksaray. He realized the importance of the Mongols and made an alliance with them once he usurped the throne of Nicea. Being a small military power, the focus of his rule was on securing Greek Orthodoxy and he sought sympathy from Hülegü and Doquz Khatun. This may have included submitting to the Mongols. In 1260 Hülegü had Bohemund, Prince of Antioch and Mongol vassal, install Euthymius, an Orthodox bishop, as the Patriarch of Antioch. Bohemund did so reluctantly, but nonetheless was

QQ Bar Hebraeus, The Chronography, 434–35; Melville, “Anatolia under the Mongols,” 55. QW Bar Hebraeus, The Chronography, 426; al-Dhahabī, Kitab Duwal al-Islam, 264. Dhahabī indicated that Baiju carried this out on Hülegü’s orders in 654 H. QX J ūzjāni, Ṭabaqāt-i Nāṣirī, 1: 314; Idem, tr. Raverty, 164. This may have been Eljigedei, the former tammaci, or another Eljigedei. QY J ūzjāni, Ṭabaqāt-i Nāṣirī, P, 314; Idem, tr. Raverty, 164. hd 234 excommunicated for this by the Bishop of Bethlehem. In 1263 Bohemund and Hethum kidnapped Euthymius and sent him to Armenia. Hülegü chastised them for this, but does not appear to have taken further action.SV In 1259, Mehmed Beg al-Ūji founded the ,-rst Turkmen emirate around Denizli, and included Khūnās and Dalaman. In 1261, Mehmed Beg, his brother Ilyās Beg, son-in-law ʿAlī Beg, and relative Sevinch rebelled against ʿIzz al-Dīn and did not recognize Rukn al-Dīn as a legitimate ruler as well. Mehmed Beg wrote to Hülegü and wanted to submit. Hülegü ordered him to come in person, but Mehmed Beg declined. Hülegü then sent an army of Rūm to the Uj. In 1262, Mehmed Beg was defeated by the Seljuq–Mongol army on the plain of Tālamānī or Dalaman. Mehmed Beg’s son-in-law ʿAlī Beg switched sides and was named ruler over that area, thus dividing Turkic Anatolia into three Mongol controlled spheres. Two were Seljuqid and the other was Turkmen.S) The remaining independent state was Nicea, which may or may not have been a client of the Mongols. With their dominance in Anatolia now complete, Mongol activities in the region declined. They did distribute ʿiqṭas in Rūm, and this may have assisted in later diminishing their control over Rūm, but the silence in the chronicles on other activities indicates a level stability.S* Their declining interest in Anatolia a8fected border states as well. After the Greeks took Constantinople in 1261, the Byzantines’ eastern frontier defences were neglected and Turkmen raids, and often settlement, increased in these areas. Although not as successful as the Cilicians in the 1250s, the Byzantines could occasionally get the Mongols to rein in the Turkmen attacks.S5

The Tsunami Method of Conquest

The 1240s was a curious decade for the expansion of the Mongol Empire. The reign of Ögetei served as a driving force for expansion as he introduced the concept of divine right to rule the world, which remained a primary rationale for the Mongols from this point on. Indeed, in a letter to Pope Innocent Pf, the Great Khan Güyük sent the following message after the Pope chastised him for killing Christians:

WZ Lippard, “The Mongols and Byzantium, 1243–1341,” 159. W. Lippard, “The Mongols and Byzantium, 1243–1341,” 24. W0 Lippard, “The Mongols and Byzantium, 1243–1341,” 25. W9 Lippard, “The Mongols and Byzantium, 1243–1341,” 34–35. ;<=>

God ordered us to destroy them and gave them up to our hands. For otherwise if God had not done this, what could man do to man? But you men of the West believe that you alone are Christians and despise others. But how can you know to whom God deigns to confer His grace? But we worshipping God have destroyed the whole earth from the East to the West in the power of God. And if this were not the power of God, what could men have done? . . . But if you should not believe our letters and the command of God nor hearken to our counsel then we shall know for certain that you wish to have war. After that we do not know what will happen, God alone knows.SI

Thus, while the divine right of conquest remained from the Ögeteid era, the sustained drives of conquest did not occur. The most notable actions took place in Anatolia under the command of Baiju. It is not clear whether he attempted to follow the process of Tsunami conquests, as the Seljuqs collapsed after the battle of Köse Dāgh. Quite simply, the Mongols acquired the entire Sultanate after one battle. Despite this, they still adhered to the principle of the Tsunami Strategy. With the collapse of the Seljuq Sultanate of Rūm, the rest of Anatolia lay open. Trebizond fell quickly and Cilicia saw the wisdom in submitting. With only Nicea and perhaps a few remaining unruly Turkmen tribes, it seemed likely that the Mongols would conquer all of Anatolia. Yet they did not, primarily because it did not ,-t into the Tsunami Strategy at that time. While Baiju’s army defeated the Seljuqs in 1243, another force led by Yassur conquered the Diyar Bakr region and raided Aleppo and Antioch. Although both cities had formidable forti,-cations, the Mongols convinced the cities in the Jazīra to submit through a combination of force and diplomacy. While Aleppo, and perhaps Antioch, paid tribute, the Mongols did not leave a darughachi in either city as they did in the other cities. Furthermore, they did not extend their raids further south. Meanwhile, Mongol forces marched against a number of towns that sought protection from the Caliph in Baghdad. The Mongol forces fared well in most encounters in this region; however, they do not appear to have sought to establish themselves there. They only raided and plundered. When a larger enemy force appeared, the Mongols left, whereas in Rūm the Mongols fought at Köse Dāgh.

WN John de Plano Carpini, “History of the Mongols,” 83–84. hc 234

Thus, when viewed at the microcosmic level, the Mongols of Baiju’s tamma did not carry out the Tsunami Strategy. Yet at the macrocosmic level they did. In Rūm, Jazīra, and even in Iraq, the Mongols used the Tsunami Strategy. They conquered Rūm, the cities of the Jazīra, Trebizond, added Cilicia through submission, and forced Aleppo and Antioch to pay tribute. Nicea, the rest of Syria, and Iraq were not conquered, although they were placed on guard or actually su8fered invasions, but in Rūm and Jazīra, the Mongols conquered some territory and made e8forts to incorporate it by placing darughachin to administer the new territory. Meanwhile, the tammachin returned to the pastures of Azerbaijan. The Mongols did not overextend their armies. It is apparent that Baiju did not have su8,-cient forces to continue a campaign of conquest. He did, however, ensure that the states that bordered his territory were aware the Mongols could strike them. Some, such as Aleppo and Antioch, merely paid them to stay away. Baghdad remained de,-ant while Nicea prepared its defences. Meanwhile in the east, the Mongols continued their protracted war with the Song. Their armies there accomplished little other than demonstrating that the Song armies were no match for the Mongols in the ,-eld. Only through an intricate system of forti,-cations could the Song slow the Mongol advance. The ascension of Güyük should have allowed the Mongols to devote more troops to both theatres. Although armies were dispatched, Eljigedei appears to have accomplished virtually nothing. In the eastern theatre the Mongols continued to raid and probe the forti,-cations of the Song and Koryŏ, but without any substantial gain. The campaigns in the Middle East during the reign of Möngke marked a substantial divergence in the operational strategies of the Mongols. Whereas they had previously operated on a broad and far-ranging front, the armies under Hülegü’s command adopted a di8ferent strategy: overwhelming force. Against their main targets, the Assassins, the Caliphate, and Ayyubid Syria, the Mongols attacked their enemies one at a time. The conquest of the Isma’ilis and ʿAbbasid Caliphate were essentially campaigns of sieges. During the course of these, the Mongols isolated the main target by taking the outlying cities, fortresses, and towns. Then with the full weight of their military they captured the ,-nal target. Finally, mop-up operations then occurred to destroy any remaining resistance. In Rūm and the Jazīra, the Mongols encountered some rebellions. These were dealt with promptly. Meanwhile in Ghur, the Mongols secured their hold over the territory and annexed Kashmir. Forays into India produced little other than plunder. The Delhi Sultanate remained vigilant and could prevent the tamma forces in ;<=>

Ghur from acquiring any of its own territory.SL Nonetheless, it could do little to prevent the Mongols from raiding and even sacking towns such as Lahore. Furthermore, the presence of the tamma in Ghur served as a beacon of resistance to those unsatis,-ed with the Sultan of Delhi. Of course these were other princes who vied for the Delhi throne, but they did establish client principalities on the border with the Delhi Sultanate, thus providing the Mongols with a potential entrepôt into Indian politics and conquest at a future date should the possibility arise.SM During the conquest of Syria, the main force destroyed a politically and militarily important target, Aleppo, decimating it before any attempt at relief could be made. Most of the remaining cities decided that resistance was futile, and submission was preferable to destruction. The only thing which the Syrian campaign lacked was a ,-eld battle. Traditionally, the Mongols ,-rst defeated any armies in the ,-eld, leaving forces behind to lay siege to cities. The ,-eld army of the enemy was attacked until it was destroyed. Once it was defeated, the Mongols could then reduce the main fortresses at their leisure without fear of relief armies or their lines of communication being cut. As this never happened in Syria, the conquest was made easier. The death of Möngke and the subsequent civil wars altered the course of the history of the Mongol Empire. It is uncertain what the Mongols’ intentions towards the Latin States or Egypt were; however, considering their other operations in which those who had not submitted were eliminated one at a time, it is likely that Hülegü would have invaded the ‘rebellious’ Mamluk Sultanate had circumstances been di8ferent. The method of governance did not change substantially. The Mongols still used the tamma system to control their borders as demonstrated in Ghur. The possibility of the Mongols invading the Kingdom of Jerusalem as well as the Mamluk Sultanate was enhanced by the establishment of Baidar and Ket-Buqa’s tammas in Gaza and the Biqāʿ Valley respectively. The Mongols meanwhile created a rudimentary administration to control their newly acquired territories. Ket-Buqa stationed himself in the Biqāʿ Valley, which provided the pasture his troops needed. Maqrīzī believed him to be the gover- nor of Aleppo, while Baidar, who was stationed in Gaza, was the governor of

WO Aḥmad, “Mongol Pressure in an Alien Land,” 182; Digby, War-Horse and Elephant in the Delhi Sultanate, 21. Digby wrote that Sultan Ghiyāth al-Dīn Balban apparently limited his military operations against the rest of India because of the threat of Mongol raids and invasions. This kept his army close to Delhi and the north. WQ Jackson, The Delhi Sultanate, 49. h1 234

Damascus. The prince of Homs, Malik al-Ashraf Mūsā, served as the viceroy.SS Considering where their armies were stationed, neither Baidar nor Ket-Buqa should be considered governors. As both were stationed on the western frontiers of the Mongol Syrian holdings, they were probably tammachin. Ket-Buqa, in the northern part of Syria and stationed in the Biqāʿ Valley, could defend against the Kingdom of Jerusalem, which stretched to Beirut, and impose his will over the northern territories. Baidar, meanwhile, could do the same to the southern borders of Jerusalem and also guard against the Mamluk Sultanate in Egypt. Additionally, one must ask if Baiju’s tamma that was once stationed in the Mughan also marked the establishment of a tamma in Rūm. With the defeat of the Isma’ilis, Baghdad, and the complete conquest of the Jazīra, there was no rationale to keep a tamma in an area now surrounded by conquered areas. It was time to move to the new borderlands in Rūm. While Rūm was conquered in 1243, other areas surrounding the Mughan were not, thus there was cause to maintain it there. Although Baiju moved there in 1256, before Hülegü attacked the Isma’ilis, the Mongols, particularly during Möngke’s reign, did not operate under the assumption that they might not be victorious. It was simply a matter of how quickly they would win. Thus, it was proper for Baiju to re-establish the tamma in Rūm.

Bibliography

Aḥmad, Aziz. “Mongol Pressure in an Alien Land.” Central Asiatic Journal 6 (1961): 182–193. Allsen, Thomas T. Mongol Imperialism the Policies of the Grand Qan Möngke in China, Russia, and the Islamic Lands, 1251–1259. Berkeley: University of California Press, 1987. ———. “The Yüan Dynasty and the Uighurs of Turfan in the 13th Century.” In China Among Equals, Edited by Morris Rossabi, 243–279. Berkeley: University of California Press, 1983. al-Anṣārī, ʿUmar Ibn Ibrāhīm al-Ansī. Tafrīj al-kurūb fī tadbīr al-ḥurūb. Cairo: American University at Cairo Press, 1961. ———. A Muslim Manual of War. Edited and translated by George Scanlon. Cairo: American University in Cairo Press, 1961. Aubin, Jean. “L’Ethnogenèse des Qaraunas.” Turcica 1 (1969): 65–95. Bar Hebraeus. The Chronography of Gregory Abū’l-Faraj. 2 vols. Translated by Ernest A. Wallis Budge. Amsterdam: #$#-Philo Press, 1932.

WW al-Maqrizi, Histoire des Sultans Mamlouks de L’Egypte, 99–100. ;<=>

Ibn Bībī, Nāṣir al-Dīn Ḥusayn ibn Muḥammad. Saljūq Nāmah. Edited by Muḥammad Zakariyā Mā’il. Lahore: Markaz-Urdū Bord, 1975. Brosset, Marie Félicité and David Iessevich Chubinov, ed. and trans. Histoire de la Géorgie depuis L’antiquité jusqu’au 0102 siècle. St. Pétersbourg: L’imprimerie de l’Academie Impériale de Sciences, 1849. Buell, Paul D. “Kalmyk Tanggaci People: Thoughts on the Mechanics and Impact of the Mongol Expansion.” Mongolian Studies 6 (1980): 41–59. Cahen, Claude. Pre-Ottoman Turkey. Translated by J. Jones-Williams. New York: Taplinger Publishing Company, 1968. ———. “The Turks in Iran and Anatolia before the Mongol Invasions.” In A History of the Crusades, vol. 2: The Later Crusades, 1189–1311, edited by R. L. Wol8f and H. W. Hazard, 660–92. Madison: The University of Wisconsin Press, 1969. Dafeng, Qu. “On the Qusiqul Army and the Tamaci Army.” Central Asiatic Journal 45, no. 2 (2001): 266–72. al-Dhahabī, Muḥammad ibn Aḥmad. Kitāb Duwal al-Islām (Les Dynasties de L’Islam). Translated by Arlette Negre. Damascus: Institut Français de Damas, 1979. Di Cosmo, Nicola. Ancient China and its Enemies: The Rise of Nomadic Power in East Asian History. Cambridge: Cambridge University Press, 2002. Digby, Simon. War-Horse and Elephant in the Delhi Sultanate: A Study of Military Supplies. Oxford: Oxford University Press, 1971. Fine, John V. A. The Late Medieval Balkans. A Critical Survey from the Late Twelfth Century to the Ottoman Conquest. Ann Arbor: University Of Michigan Press, 1987. France, John. Western Warfare in the Age of the Crusades, 1000–1300. Ithaca: Cornell University Press, 1999. 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 (Turcologica). Wiesbaden: O. Harrassowitz, 1992. Grigor of Akanc. “The History of the Nation of the Archers by Grigor of Akanc.” Translated and edited by R. P. Blake and R. N. Frye. Harvard Journal of Asiatic Studies 12 (1949): 269–399. Grønbech, Kaare and John R. Krueger. An Introduction to Classical (Literary) Mongolian. Wiesbaden: O. Harrassowitz, 1976. al-Ḥadīd, ʿAbd al-Ṣamīd ibn Hibat Allāh Ibn Abī. Sharḥ nahj al-balāghah. Beirut: Dār Maktabah al-Hayāt, 1963. He Qiutao. Shengwu Qin Zheng Lu (Bogda bagatur bey-e-ber tayilagsan temdeglel). Translated by Arasaltu. Qayilar: Obor Monggol-un Soyul-un Keblel-un Qoriy-a, 1985. Het’um. “The Journey of Het’um P, King of Little Armenia to the Court of the Great Khan Möngke.” Translated by John A. Boyle. Central Asiatic Journal 9 (1964): 175–89. Hsiao, Ch’i-ch’ing. The Military Establishment of the Yuan Dynasty. Cambridge, 2#: Harvard University Press, 1978. hR 234

Jackson, Peter. The Delhi Sultanate, A Political and Military History. Cambridge: Cambridge University Press, 1999. Juwaynī, ʿAṭā Malik. Ta’rīkh-i Jahāngushā. 3 vols. Edited by Mīrzā Muḥammad Qazwīnī. Leiden: Brill, 1912, 1916, 1937. ———. The History of the World Conqueror. Translated by John Andrew Boyle. Seattle: University of Washington Press, 1997. Jūzjānī, Minhāj Sirāj. Ṭabaqāt-i Nāṣirī. 2 vols. Edited by ʿAbd al-Ḥayy Ḥabībī. Kabul: Anjuman-i Tārīkh-i Afghānistān, 1964–65. ———. Ṭabaqāt-i Nāṣirī. 2 vols. Translated by Major H. G. Raverty. New Delhi: Oriental Book Reprint Corp, 1970. Lippard, Bruce G. “The Mongols and Byzantium, 1243–1341”. PhD diss., Uralic and Altaic Studies, Indiana University, 1983. al-Maqrīzī, Aḥmad Ibn ʿAlī. Histoire des Sultans Mamlouks de L’Egypte. 2 vols. Translated by Etienne Quatremère. Paris: Oriental Translation Fund of Great Britain and Ireland, 1837, 1845. Martin, H. Desmond. The Rise of Chinggis Khan and his Conquest of North China. Baltimore: The Johns Hopkins Press, 1950. May, Timothy. “The Mechanics of Conquest and Governance: The Rise and Expansion of the Mongol Empire, 1185–1265.” PhD diss., Department of History, The University of Wisconsin-Madison, 2004. ———. The Mongol Art of War. Barnsley, vw: Pen & Sword, 2007. ———. “The Conquest and Rule of Transcaucasia: The Era of Chormaqan.” In Caucasus during the Mongol Period—Der Kaukasus in der Mongolenzeit, edited by Jürgen Tubach, Sophia G. Vashalomidze, and Manfred Zimmer, 129–52. Wiesbaden: Reichert Verlag, 2012. Melville, Charles. “Anatolia under the Mongols.” In The Cambridge History of Turkey. Vol. 1, edited by Kate Fleet, 51–101. Cambridge: Cambridge University Press, 2009. Miller, William. Trebizond, the Last Greek Empire of the Byzantine Era, 1204–1461. Chicago: Argonaut Inc, 1969. al-Nuwayrī, Aḥmad ibn ʿAbd al-Wahhāb. Nihāyat al-arab fī funūn al-adab. Edited by Saīd ‘Āshūr. Cairo: Al-hay’ā al-Misriyya al-‘āmma lil-kitāb, 1975. Ostrowski, Donald. “The tamma and the dual-administrative structure of the Mongol Empire.” Bulletin of the School of Oriental and African Studies 61 (1998): 262–77. Paris, Matthew. English History. Translated by J. A. Giles. New York: #2` Press, 1968. Petrushevsky, I. P. “Socio-Economic Conditions of Iran under the Il-Khans.” In Cambridge History of Iran: The Saljuq and Mongol Periods, edited by J. A. Boyle, 483– 537. Cambridge: Cambridge University Press, 1968. Plano Carpini, John of. “History of the Mongols.” In The Mongol Mission: Narratives and Letters of the Franciscan Missionaries in Mongolia and China in the Thirteenth and Fourteenth Centuries, edited by Christopher Dawson, translated by A Nun of Stanbrook Abbey, 3–72. London: Sheed and Ward, 1955. ;<=>

Polo, Marco. The Travels of Marco Polo. Translated by Henry Yule. New York: Dover Publications, 1993. Poppe, Nicholas. Grammar of Written Mongolian. Wiesbaden: O. Harrassowitz, 1954. Rachewiltz, Igor de, ed. and trans., The Secret History of the Mongols. Brill’s Inner Asian Library, Vol. 7. Leiden: Brill, 2004. Rashīd al-Dīn. Jāmiʿ al-tawārīkh. Edited by B. Karīmī. Tehran: Iqbāl, 1983. ———. Jami‘u’t-tawarikh [sic] Compendium of Chronicles. Translated by Wheeler M. Thackston. Cambridge, 2#: Department of Near Eastern Languages and Civilizations, Harvard University, 1998–99. Rubruck, William of. “The Mission of William of Rubruck.” In The Mongol Mission: Narratives and Letters of the Franciscan Missionaries in Mongolia and China in the Thirteenth and Fourteenth Centuries, edited by Christopher Dawson, translated by A Nun of Stanbrook Abbey, 89–237. London: Sheed and Ward, 1955. ———. The Mission of Friar William of Rubruck: His Journey to the Court of the Great Khan Mongke, 1253–1255. Edited by Peter Jackson and David Morgan, translated by Peter Jackson. London: the Hakluyt Society, 1990. Sinor, Denis. “On Mongol Strategy.” In Proceedings of the Fourth East Asian Altaistic Conference, edited by Ch’en Chieh-hsien, 238–49. Taipei: Tainan, 1977. Vásáry, István. “The Golden Horde Term daruġa and its Survival in Russia,” Acta Orientalia Hungarica 30 (1976): 187–97. ———. “The Origin of the Institution of basqaqs,” Acta Orientalia Hungarica 32 (1978): 201–6. ———. Turks, Tatars and Russians in the 13th–16th Centuries (London: Ashgate- Variorium, 2007). Vernadsky, George. The Mongols and Russia. New Haven: Yale University Press, 1953. Yuan Shih. Chapters 98 and 99 in The Military Establishment of the Yuan Dynasty. (72–91, 92–124). Translated by Ch’i-ch’ing Hsiao. Cambridge, 2#: Harvard University Press, 1978. Zhao Hung. Men-da bei-lu: Polnoe opisanie Mongolo-Tatar. Translated by N. Ts. Munkeyev. Moscow: Akademiia nauk, 1975. !"#$%&' ( Continuity and Change in the Mongol Army of the Ilkhanate

Reuven Amitai

One of the important trends of late medieval societies in much of the Islamic world is that the military elite became increasingly identi)*ed with the political ruling class. This tendency possibly reached its height in the Mamluk Sultanate of Egypt and Syria. One cannot go quite so far when describing the nature of Mongol rule in Iran and the surrounding countries, but it would be fair to state that the army was the most important institution in the Ilkhanid state (and Mongol states as a whole). It is thus not surprising that the study of military aspects of the Ilkhanate (and the Mongol armies that preceded it) have received some serious attention in modern scholarship. Mention can be made of the important studies on the Mongol armies in the Middle East (some of them parts of larger works) by Bertold Spuler,+ David Morgan,, C. E. Bosworth,- John M. Smith, Jr.. and Arsenio Martinez./ To this can be added very useful studies on the Mongol army as a whole by H. Desmond Martin,0 S. R. Turnbull,1 Robert Reid,2 Witold Świętosławski,6 Timothy May,+7 and others, which help us understand the background of much of the Mongol military activity in the countries today known as the Middle East. My own contributions to the study of the Mongol military machine have generally been in connection with the ongoing Ilkhanid war with the Mamluks, and I have tried to put this war in

8 Spuler, Die Mongolen in Iran, 330–48. : Morgan, “The Mongol Armies in Iran,” 81–96. ; Bosworth, “Army <<,” 2, 499–503. = Smith, Jr., “ ‘Ayn Jālūt,” 307–44. A Martinez, “Some Notes on the Īl-Xānid Army,” 129–242. C Martin, The Rise of Chingis Khan, 11–47. D Turnbull, The Mongols, 105. E Reid, “Mongolian Weaponry,” 85–96. F Ś więtosławski, Arms and Armour of the Nomads of the Great Steppe, 103–10. I am thankful to Kate Raphael for bringing this work to my attention. 8G May, The Mongol Art of War.

© !"#$#!%$&!' ()$%% #*, %'$+'#, ,-./ | +"$ .-.../1/23425561.63,/_556 !JKLMKNMLO #KP !QRKST gunpowder in siege warfare; the nature and extent of auxiliary units, both of East Asian and Middle Eastern provenance; the tactical and related changes that the Mongol army may have undergone over the many decades of its residence in Iran and the surrounding countries; the importance of logistics—i.e. pasturage and water—in determining Mongol strategy and the outcome of campaigns; and the relative quality of the Mongol troops when compared to their enemies, not least the Mamluks. In my previous work I have dealt with the last two questions, perhaps not to everyone’s satisfaction, but I will not delve into them again here. The question of gunpowder is an important one which refuses to go away, and I hope to attempt an investigation of this matter—at least as far as impinges on the frontier with the Mamluks— in the future; in the meantime, we have a very important pioneering study on the matter by Dr. Kate Raphael, which has appeared in the Journal of the Royal Asiatic Society.+, The role and identity of auxiliaries will also be left to another occasion. By process of elimination, one can surmise that this present discussion will focus on the possible changes that the Mongol military underwent in the realm of tactics during the almost one hundred and twenty years that began with the )*rst Mongol conquests in the Islamic world in 1219 and lasted until the breakup of the Ilkhanate in the years after Abū Saʿīd’s death in 1335. This question ties in with matters of armament, equipment and horses, which will also be touched upon below.

88 Amitai-Preiss, Mongols and Mamluks, esp. chapter 10; Amitai, The Mongols in the Islamic Lands, esp. part 3. 8: Raphael, “Mongol Siege Warfare,” 355–70. Cf. Khan, “The Coming of Gunpowder,” 27–45. The forthcoming work of Stephen J. Haw may shed further light on this matter. ^_ #XMLRM

Much of the evidence for this present paper is derived from the Arabic sources emanating from the Mamluk Sultanate, which have been read in tandem with the sources (written in Persian, Syriac, Armenian and Arabic) deriving from the Ilkhanate. There are several reasons why the Mamluk sources are so important in both the present context and the whole study of the Ilkhanate. Firstly, there was a phenomenal `aowering of historical writing under the Mamluks, the cause for which has not yet been satisfactorily explained.+- Secondly, as be)*tting a state ruled by a military elite, military abfairs—not only local ones—play a notable role in this corpus. Thirdly, as the main enemies of the Sultanate in the )*rst generations of its existence, the Mongols received plenty of attention, not only in the military sphere. Fourthly, much notice was given to battles with this enemy, where usually the Mamluks held the upper hand. The latter aspect is of particular signi)*cance in trying to understand the nature of the Mongol military. The British military historian John Keegan correctly stated that military history must ultimately deal with battles, since that is where armies )*nally prove their mettle, and where we can learn much about their preparedness, equipment, tactics, command structure, morale and other important matters.+. The descriptions of battles between the Mongols and the Mamluks—where most details are provided from sources composed in the Sultanate—can also be a source of information about the changes—if any—in the nature of the Ilkhanid army. My position is that on the whole the Ilkhanid armies did not undergo a massive transformation in their basic nature, equipment and tactics, certainly into the early fourteenth century; for the )*rst decades of the Ilkhanate we have relatively good evidence, although subsequently we are on less )*rm ground. This does not mean that in this early period there were no changes among the Mongol society elites and common tribesmen: Islamization is clearly one of them,+/ and another would be the growing connections with the Iranian cultural and administrative elite among the Mongol upper classes.+0 This paper,

8; See the concise but comprehensive survey by Little, “Historiography of the Ayyūbid and Mamlūk Epochs,” 412–44. For the importance of Arabic sources—mostly written in the Mamluk Sultanate—for the study of Mongol history, see Amitai and Biran, “Arabic Sources.” 8= Keegen, The Face of Battle, 15–36. 8A Melville, “Pādshāh-i Islām,” 159–77; Pfeibfer, “Conversion Versions,” 35–67; Amitai, “Su)*s and Shamans,” 27–46; Amitai, “The Conversion of Tegü der Ilkhan to Islam,” 15–43; Amitai- Preiss, “Ghazan, Islam and Mongol Tradition,” 1–10. (The last three studies are republished in Amitai, The Mongols in the Islamic Lands, part 2). 8C See Aubin, Émirs mongols, as well as Lane, “Persian Notables and the Families,” 182–213. This latter study argues that there was a virtual merging of the Mongolian and Iranian !JKLMKNMLO #KP !QRKST

In order to trace the development of the Mongol army in Iran and the surrounding countries, one should )*rst ascertain its original state. Actually I was perhaps too quick above when stating that there is a consensus among scholars regarding the fact that the early Mongol army was composed overwhelmingly of light cavalry, i.e. lightly armed and armoured mounted archers

elite in the Ilkhanate from quite early on, which to my mind is a somewhat exaggerated position. 8D For a brief comparison between the two states, see the discussion in Amitai, Holy War and Rapprochement, 87, 108. ^( #XMLRM

)*ghting on steppe ponies.+2 Some popular books suggest that “heavy” cavalry, better armoured and carrying a lance of some type, was interspersed with the mounted archers, and their main task was to assault the enemy directly.+6 This seems to go back to Martin’s early study, whose reasoning was that if the Jurchens (i.e., the Jin dynasty, 1115–1234 !&) in China had such a system of integrated light and heavier cavalry, the Mongols must have had one too.,7 Why a Chinese dynasty of Manchurian provenance should provide an automatic precedent for the Mongols on the steppe is not clear to me. In any event, as far as I am aware, there is no evidence that the Mongol armies who entered the Islamic world from Inner Asia were on the whole anything else but light archers, as described by the papal envoy John of Plano Carpini who visited the Mongol capital of Qaraqorum in the mid-1240s.,+ Carpini, however, notes that these archers could at times also carry other weapons too: lances, swords, axes, etc., and some of them also wore rudimentary armour.,, The only possible exception might be the royal guard unit, known as the keshig, which was recently investigated in the Ilkhanate by Charles Melville.,- However, in this important article, there is little discussion of the military aspect of this formation, whose members were known as baghatut/ba’atut (singular baghatur), literally ‘heroes.’ The Arabic sources mention at times bahāduriyya,. and bahāduriyyat al-mughul,,/ and it is reasonable that these can be identi- )*ed as members of the local royal guard unit that probably entered Iran with Hülegü in the 1250s. Such a unit is also mentioned by Rashīd al-Dīn at the end

8E For these ponies, see Smith Jr., “ ‘Ayn Jālūt,” 336–40. 8F Turnbull, The Mongols, 23; James Chambers, The Devil’s Horsemen, 69–71, 78; Marshall, Storm from the East, 40. :G Martin, The Rise of Chingis Khan, 33. See also Martin, “The Mongol Army,” 69. :8 John of Plano Carpini, “History of the Mongols,” 36–37; the Latin original is found in Anastaas van den Wyngaert, ed., Sinica Franciscana, vol. 1: Itinera et Relationes Fratrum Minorum saeculi #$$ et #$% (Quaracchi-Firenze, 1929), 81–82. See the comments of Świętosławski, Arms and Armour of the Nomads, 104, which express a similar opinion, as well as Amitai-Preiss, Mongols and Mamluks, 216; Biran, “The Battle of Herat,” 204–5. :: John of Plano Carpini, (Dawson), 33–35 (Latin original in Wyngaret, Sinica Franciscana, 77–80). :; Melville, “The Keshig in Iran,” 135–64. := Ibn ‘Abd al-Ẓāhir, al-Rawḍ al-zāhir fī sīrat al-malik al-ẓāhir, 177; al-Maqrizi, Kitāb al-sulūk li-ma‘rifat al-duwal wa’l-mulūk, <, 500. A translation of the latter passage is found in Quatremère, Histoire des sultans mamlouks, 1; 221. :A Al-Y ūnūnī, Dhayl mir’āt al-zamān, 2: 115; cf. the parallel passage in Ibn Wāṣil, Mufarrij al-kurūb fī akhbār banī ayyūb, (Rahim), 224: dhurriyyat al-mughul (“descendants of the Mongols”). !JKLMKNMLO #KP !QRKST

:C Rash īd al-Dīn, Geschichte Ġāzān Ḫān’s, 125–26 (describing the events before the battle of Wādī al-Khanzadār in 1299); see also 127. :D For these in the Mongol Empire as a whole, particularly in the east, see: Hsiao Ch’i-ch’ing, The Military Establishment of the Yuan Dynasty, 36; Allsen, Mongol Imperialism, 21–22 and fn. 14; Allsen, “Guard and Government,” 495–521; Martin, “The Rise of Chingis Khan,” 23–24, 36. :E Wa ṣṣāf, Ta’rīkh i Waṣṣāf, 75; see the discussion in Biran, “The Battle of Herat,” 204–5; cf. Martinez, “Īl-Xānid Army,” 152–56. :F “Did the Mongols in the Middle East Remain Nomadic Pastoralists?”, found at https:// www.academia.edu/21768134/Did_the_Mongols_in_the_Middle_East_Remain_Pastoral_ Nomads. ^^ #XMLRM

Every tribe has land to reside in and the descendant inherits it from the forefathers since Hülegü conquered the country. Their abodes (manāziluhum) are in it. They have in it crops for their substance, but they do not live by tilling and sowing.-7

Here we learn that the Mongols have not adopted sedentary forms of economy. In other words, they have remained pastoral nomads into the 1340s, when al-‘Umarī composed his work, receiving information from many informants who had left the Ilkhanate in previous years. Perhaps, the objection can be raised: what about the famous passage in Rashīd al-Dīn’s Jāmiʿ al-tawārīkh, in which the author describes Ghazan Ilkhan’s promulgation of the iqṭā‘ system? This passage, discussed in the previously mentioned paper and at greater length in an article published some )*fteen years ago,-+ goes as follows:

At this time [ca. 1300 or later], most of the soldiers had the desire for estates (amlāk) and agriculture. Upon acquiring iqṭā‘ land (milkī iqṭā‘ī), they will have reached their goal.-,

The basic arguments presented in the article just mentioned were that iqṭā‘ here is referring to a dibferent conception than usually understood by this term in both Iran and countries bordering the eastern Mediterranean; the farmān (“order”) commanding this change was issued late in Ghazan’s reign and therefore was probably not widely executed, if at all; and, there is little actual evidence of such iqṭā‘āt. Finally, it should be added that a desire for landed property did not entail the adoption of agriculture by the Mongols per se, but probably more likely only the desire to enjoy its fruits, i.e. the sur- plus extracted by taxes. By the same reasoning, we can see that the contemporary Mamluks of Egypt and Syria did not practise agriculture (or have a desire to do so) just because they possessed iqṭā‘āt, which was the basis of their personal wealth, and by which they also maintained their military units. In short, there is no reason to assume that the adoption of some type of iqṭā‘ system by

;G al-‘Umar ī, Masālik al-abṣār, in Lech, ed. and trans., Das Mongolische Weltreich, 95. ;8 Amitai,“Turco-Mongolian Nomads,” 152–71. ;: Rash īd al-Dīn, Geschichte Ġāzān Ḫān’s, 302. Cf. Morgan, “The Mongol Armies in Persia,” 92–93, who suggested that before the word “agriculture” we should add “the practice of,” implying therefore that indeed the Mongols were intent on settling down and changing their lifestyle. !JKLMKNMLO #KP !QRKST

;; Bosworth, “Army,” 503. There is a short discussion in Amitai, “Armies and their Economic Basis,” 558–60. ;= Amitai, “Whither the Ilkhanid Army?” 221–64. ;A Wa ṣṣāf, Ta’rīkh i Waṣṣāf, 373, lines 20–23 (discussed by Smith, “ ‘Ayn Jālūt,” 329). î #XMLRM practice reàecting the use of relatively small steppe horses, which needed to be rotated on campaign and during the battle.-0 While originally the Mongols in the Centre had met the initial Mamluk attack dismounted, so as inter alia to improve the accuracy of their archery, they eventually remounted, and “they attacked [the Mamluks] in the Turkish fashion” (bar ishān turk tāz kardand).-1 This I take as an allusion to the traditional Steppe tactic of launching a uni- )*ed assault of mounted archers in a concerted manner.-2 At an apparent later stage, when fresh Mongol troops came up, the Mongols—according to the contemporary Mamluk historian and senior ob)*cer Baybars al-Manṣūrī, who was not at the battle—used archery ebfectively against the Mamluk horses,-6 indicating the continued importance of this traditional nomadic mainstay. There is thus some clear evidence that leads to the conclusion that two generations after Hülegü’s arrival in the region, the Mongol troops on the whole still maintained the military techniques that had been employed by their forefathers. This further implies that generally the Mongol army of circa 1300 was one of light cavalry. If this was indeed the case, we would expect that a couple of years later, when Ghazan launched his third campaign into Syria, there would not be a marked change. (As an aside, the his second campaign, of 1300, was cancelled due to particularly harsh weather, and the Mongol army withdrew back over the frontier from north Syria without confronting the Mamluks). I have only partially reviewed the sources for the campaign of 1302–3, in preparation for a full-àedged study of the battle of Marj al-Ṣubfar, which was a humiliating defeat for the Mongols. So far, I have found nothing which hints at any major paradig- matic shift in the Mongol military. This, of course, is not very satisfactory, and it would be desirable to have more explicit evidence. There is information about how the Mongol mounts subfered terribly from thirst during the )*ghting,.7

;C Smith, “ ‘Ayn Jālūt,” 331, fn. 75; Amitai-Preiss, Mongols and Mamluks, 215; Amitai, “Whither the Ilkhanid Army?” 253, fn. 129. ;D Rashīd al-Dīn, Geschichte Ġāzān Ḫān’s, 128. ;E See May, The Mongol Art of War, 71–74; Smail, Crusading Warfare, 75–83; as well as the discussion in Amitai, “Whither the Ilkhanid Army?” 247, fn. 108. ;F al-Man ṣūrī, Zubdat al-./kra, 331. =G Rash īd al-Dīn, Geschichte Ġāzān Ḫān’s, 148; Het‘um, “La Flor des estories de la Terre d’Orient,” <<, 202; Smith Jr., “ ‘Ayn Jālūt,” 343. Incidentally, but not without relevance, I have found evidence that contradicts Smith’s suggestion that the Mongols had only three tümens (units theoretically of 10,000 men) at this battle (ibid., 341). See al-Manṣūrī, Zubdat, 372, who states categorically that the Mongols had twelve tümens; on 276, 12 senior ob)*cers are mentioned, all of them evidently tümen commanders. All told, this army had 100,000 men, composed of Mongols (al-mughul), Georgians, Armenians, and !JKLMKNMLO #KP !QRKST

There remains at this point the question of the comparison between the Ilkhanids and the Seljuqs, or rather, the armies of these two dynasties. The researches of Claude Cahen, A. K. S. Lambton and David Morgan, and more recently those of Andrew Peacock and David Durand-Guédy have shown clearly that fairly early on, the Seljuq sultans began to rely less and less on the Turkoman nomads. In their place a relatively large corps of slave soldiers was created, and this remained the backbone of the Seljuq army.., The ghilmān/ mamlūks certainly help to save the day at Manzikert in 1071..- The Mongols, on the other hand, never adopted the institution of military slavery. (We can assume that Ibn Baṭṭūṭa’s use of the word ‘Mamluk’ to describe soldiers in the Ilkhan’s entourage is based on ignorance compounded by memories of Egypt and Syria)... This may be due to the fact that unlike the Seljuqs, the Ilkhans came into the Middle East not as Muslims and without any commitment to Muslim administrative culture, while the former adopted this relatively quickly, including those aspects dealing with the army and its payment

others. The majority of these would have been Mongols, and if I am right, still )*ghting as light cavalry. =8 This is the position suggested by Raphael, in private conversations and in her article, “Mongol Siege Warfare,” 367–70. =: Cahen, “The Turkish Invasions,” 158; Lambton, Continuity and Change, 240–43; Morgan, Medieval Persia, 28–29; Peacock, Early Seljūq History; Durand-Guédy, “Goodbye to the Türkmen?”. The last two mentioned provide a nuanced and detailed analysis of this process, showing that this was a somewhat drawn out matter, and emphasizing that the Turkomans were never fully eliminated from Seljuq service. In any case, the growing and decisive role of the military slaves is clearly presented by all authors. =; al-Jawz ī, Mir’āt al-zamān, 147–52, cited in Ayalon, “From Ayyūbids to Mamlūks,” 44 (reprinted in Ayalon, Islam and the Abode of War). == Ibn Baṭṭūṭa, Voyages d’Ibn Batoutah, <<, 119, 127; English translation in H. A. R. Gibb, The Travels of Ibn Baṭṭūṭa, 338, 343. An indication of the in`auence of the nomenclature from the Mamluk Sultanate on Ibn Baṭṭūṭa’s description of the Ilkhanate is seen in his use of khāṣṣikiyya to describe the Ilkhan’s bodyguard (Arabic text, 2: 119; tr. Gibb, 2: 338). ^m #XMLRM

(I am referring, of course, to the iqṭā‘). Secondly, it may be that the royal family and elite of the Ilkhanate had more of a personal commitment to pastoral nomadism, therefore creating a cultural and political atmosphere conducive to the continuation of this lifestyle among the Mongol tribesmen. The work of Charles Melville has shown that the Ilkhan and his court remained nomads,./ although obviously their nomadism was not that of their forefathers in the pre- imperial period, and certainly was conducted now with a great deal of pomp and luxury..0 The impression gained is that the Ilkhanid elite did a better job of keeping their nomads under control than their Seljuq predecessors (although the former surely had to deal with much larger numbers spread over a wider area). Nostalgia for the nomadic past may have played a part in maintaining an army essentially based on pastoral nomads. Finally, it may be that the Ilkhans and their senior commanders were sincerely convinced that their army was superior to those of their enemies, including the hated Mamluks, and thus continuity was a wise strategic move. In short, I am suggesting here a great deal of continuity of the military system of the Ilkhanid army, implying and based on continuity of the lifestyle of the majority of its troopers. This is in spite of the great changes that took place in the nomadic society of the Ilkhanate: Turki)*cation, Islamization, the nature of the state, relations with the Iranian elite and surely in other areas. I think that in general we should be looking for continuity in Iran and the surrounding areas in the sphere of military abfairs from the beginning of the Mongol period into the early Safavid era, although further research will be necessary to con)*rm this statement.

Bibliography

Allsen, Thomas T. Mongol Imperialism: The Policies of the Grand Qan Möngke in China, Russia and the Islamic Lands. Berkeley: University of California Press, 1987. ———. “Guard and Government in the Reign of the Grand Qan Möngke, 1251–59.” Harvard Journal of Asiatic Studies 46 (1986): 495–521. Amitai-Preiss, R. Mongols and Mamluks: The Mamluk-Īlkhānid War, 1260–1281. Cambridge: Cambridge University Press, 1995. ———. “Ghazan, Islam and Mongol Tradition: A View from the Mamluk Sultanate.” Bulletin of the School of Oriental and African Studies 59 (1996): 1–10.

=A Melville, “The Itineraries of Sultan Öljeitü,” 55–70. =C George Lane has brought this point home to me in several conversations. !JKLMKNMLO #KP !QRKST

Amitai, R. “Su)*s and Shamans: Some Remarks on the Islamization of the Mongols in the Ilkhanate.” Journal of the Economic and Social History of the Orient 42/1 (1999): 27–46. ———. “The Conversion of Tegüder Ilkhan to Islam.” Jerusalem Studies in Arabic and Islam 25 (2001): 15–43. ———. “Turco-Mongolian Nomads and the iqṭā‘ System in the Islamic Middle East (1000–1400 #n).” In Nomads in the Sedentary World, edited by A. Wink and A. M. Khazanov, 152–71. Richmond, Surrey: Curzon, 2001. ———. “Whither the Ilkhanid Army? Ghazan’s First Campaign into Syria (1299–1300).” In Warfare in Inner Asian History, edited by N. Di Cosmo, 221–64. Leiden: Brill, 2002. ———. The Mongols in the Islamic Lands: Studies in the History of the Ilkhanate. Aldershot, op and Burlington, q%, or#: Ashgate, 2007. ———. “Armies and their Economic Basis in Iran and the Surrounding Lands, ca. 1000–1500 !&” In The New Cambridge History of Islam, 3: The Eastern Islamic World Eleventh to Eighteenth Centuries, edited by D. O. Morgan and A. Reid, 539–60. Cambridge: Cambridge University Press, 2010. ———. Holy War and Rapprochement: Studies in the Relations between the Mamluk Sultanate and the Mongol Ilkhanate (1260–1335). Turnhout: Brepols, 2013. ———. “Did the Mongols in the Middle East Remain Nomadic Pastoralists?” https:// www.academia.edu/21768134/Did_the_Mongols_in_the_Middle_East_Remain_ Pastoral_Nomads. Amitai, R. and M. Biran, “Arabic Sources for the History of the Mongol Empire.” In The Cambridge History of the Mongol Empire, edited by M. Biran and Hodong Kim. Cambridge: Cambridge University Press, forthcoming. Aubin, J. Émirs mongols et vizirs persans dans les remous de l’acculturation. Paris: Association pour l’avancement des études iraniennes, 1995. Ayalon, D. “From Ayyūbids to Mamlūks.” Revue des études islamiques 49 (1981): 44. Reprinted in Ayalon, D. Islam and the Abode of War. Art. <<<. Aldershot, U. K.: Ashgate, 1994. Baybars al-Manṣūrī. Zubdat al-./kra fī ta’rīkh al-hijra. Edited by D. S. Richards. Beirut and Berlin: Das Arabische Buch, 1998. Biran, M. “The Battle of Herat (1270): A Case of Inter-Mongol Warfare.” In Warfare in Inner Asia, edited by N. Di Cosmo, 175–219. Leiden: Brill, 2002. Bosworth, C. E. “Army <<. Islamic to the Mongol Era.” In Encyclopaedia Iranica, <<: 499–503. Cahen, C. “The Turkish Invasions: The Selchükids.” In A History of the Crusades, vol. 1: The First Hundred Years, (general editor K. M. Setton), edited by M. W. Baldwin, 135–76. Philadelphia: University of Pennsylvania Press, 1955. g_ #XMLRM

Chambers, James. The Devil’s Horsemen: The Mongol Invasion of Europe (2nd edn.). London: Cassel, 1988. Ch’i-ch’ing, Hsiao. The Military Establishment of the Yuan Dynasty. Cambridge, U#: Harvard University Press, 1978. Durand-Guédy, D. “Goodbye to the Türkmen? The Military Role of Nomads in Iran after the Saljūq Conquest.” In Nomad Military Power in Iran and Adjacent Areas in the Islamic Period, edited by K. Franz and W. Holzwarth, 107–36. Wiesbaden: Dr. Ludwig Reichert Verlag, 2015. Gibb, H. A. R. The Travels of Ibn Baṭṭūṭa 12 1325–1354. 3 vols. London: Hakluyt Society, 1958–71. Het‘um [Hayton/Hathoum]. “La Flor des estories de la Terre d’Orient.” In Recueil des historiens des croisades, documents arméniens, 2:111–253. Paris: Académie des Inscriptions et Belles-lettres, 1869–1906. Ibn ‘Abd al-Ẓāhir (Muḥyī al-Dīn). al-Rawḍ al-zāhir fī sīrat al-malik al-ẓāhir. Edited by Abd al-Azīz al-Khuwayṭir. Riyad: n.p. 1396/1976. Ibn Baṭṭūṭa (Shams al-Dīn Muḥammad b. ‘Abdallāh). Riḥla. Published as Voyages d’Ibn Batoutah. 4 vols. Edited and translated by C. Defrémery and B. R. Sanguinetti. Paris: Imprimerie Nationale, 1874–1879. Ibn Faḍlallāh al-‘Umarī (Shihāb al-Dīn Aḥmad b. Yaḥyā). Masālik al-abṣār fī ’l-mamālik al-amṣār. Partial edition and translation in Das Mongolische Weltreich: al-‘Umarīs Darstellung der mongolischen Reiche in seinem Werk Masālik al-abṣār fī ’l-mamālik al-amṣār, edited and translated by Klaus Lech. Wiesbaden: Harrossowitz, 1968. Ibn Wāṣil (Jamāl al-Dīn Muḥammad b. Sālim). Mufarrij al-kurūb fī akhbār banī ayyūb. In Die Chronik des Ibn Wāṣil, edited and translated by M. Rahim. Wiesbaden: Harrassowitz, 2010. Keegen, J. The Face of Battle. New York: Viking Press, 1976. Khan, Iqtadar Alam. “The Coming of Gunpowder to the Islamic World and India: Spotlight on the Role of the Mongols.” Journal of Asian History 30 (1996): 27–45. Lambton, A. K. S. Continuity and Change in Medieval Persia: Aspects of Administrative, Economic and Social History, 11th–14th Centuries. London: I. B. Taurus 1988. Lane, G. “Persian Notables and the Families which Underpinned the Ilkhanate.” In Eurasian Nomads as Agents of Cultural Change: The Mongols and Their Eurasian Predecessors, edited by M. Biran and R. Amitai, 182–213. Honolulu: Hawaii University Press: 2015. Little, D. P. “Historiography of the Ayyūbid and Mamlūk Epochs.” In The Cambridge History of Egypt, vol. 1: Islamic Egypt, 640–1517, edited by C. P. Petry, 412–44. Cambridge: Cambridge University Press, 1998. al-Maqrīzī (Taqī al-Dīn Aḥmad b. ‘Alī). Kitāb al-sulūk li-ma‘rifat al-duwal wa’l-mulūk. Edited by M. M. Ziyāda et al. Cairo: Dār al-Kutub, 1934–73. !JKLMKNMLO #KP !QRKST

Marshall, R. Storm from the East: From Genghis Khan to Khubilai Khan. London: ss!, 1993. Martin, H. D. “The Mongol Army.” Journal of the Royal Asiatic Society 1 (1943): 46–85. ———. The Rise of Chingis Khan and his Conquest of North China. Reprint, Baltimore: John Hopkins University Press, 1950. Martinez, A. P. “Some Notes on the Īl-Xānid Army.” Archivum Eurasiae Medii Aeivi 6 (1986): 129–242. May, T. M. The Mongol Art of War: Chinggis Khan and the Mongol Military System. Yardley, $#: Westholme, 2007. Melville, C. “Pādshāh-i Islām: The Conversion of Sultan Maḥmūd Ghazan Khān.” Pembroke Papers 1 (1990): 159–77. ———. “The Itineraries of Sultan Öljeitü, 1304–1316.” Iran 28 (1990): 55–70. ———. “The Keshig in Iran: The Survival of the Royal Mongol Household.” In Beyond the Legacy of Genghis Khan, edited by L. Komarobf, 135–64. Leiden: Brill, 2006. Morgan, D. O. “The Mongol Armies in Iran.” Der Islam 56 (1979): 81–96. ———. Medieval Persia 1040–1797. London: Longman1988. Peacock, A. Early Seljūq History: A New Interpretation. London and New York: Routledge, 2010. Pfeibfer, J. “Conversion Versions: Sultan Öljeitü’s Conversion to Shi‘ism (709/1309) in Muslim Narrative Sources.” Mongolian Studies 22 (1999): 35–67. Plano Carpini, John of. “History of the Mongols.” In The Mongol Mission: Narratives and Letters of the Franciscan Missionaries in Mongolia and China in the Thirteenth and Fourteenth Centuries, edited by Christopher Dawson, translated by A Nun of Stanbrook Abbey. London and New York: Sheed and Ward, 1955. Quatremère, M. Histoire des sultans mamlouks de l’Égypte. Paris and London: Oriental Translation Fund, 1837. Rashīd al-Dīn. Geschichte Ġāzān Ḫān’s aus dem Ta’rīḫ-i mubārak-i-ġāzānī. Edited by Karl Jahn London: Luzac, 1940. Raphael, K. “Mongol Siege Warfare on the Banks of the Euphrates and the Question of Gunpowder (1260–1312).” Journal of the Royal Asiatic Society, 3rd Ser., 19/3 (2009): 355–70. Reid, R. W. “Mongolian Weaponry in The Secret History of the Mongols.” Mongolian Studies 15 (1992): 85–96. Sibṭ ibn al-Jawzī. Mir’āt al-zamān. Ankara: Türk Tarih Kurumu, 1968. Smail, R. C. Crusading Warfare, 1097–1193. Cambridge: Cambridge University Press, 1995. Smith , Jr., J. M. “ ‘Ayn Jālūt: Mamlūk Success or Mongol Failure.” Harvard Journal of Asiatic Studies 44 (1984): 307–44. Spuler, B. Die Mongolen in Iran: Politik, Verwaltung und Kultur der Ilchanzeit 1220–1350. (4th edn.) Leiden: Brill, 1985. g( #XMLRM

Świętosławski, W. Arms and Armour of the Nomads of the Great Steppe in the Times of the Mongol Expansion. Lodz: O)*cyna Naukowa Ur, 1999. 103–10. Turnbull, S. R. The Mongols. London: Osprey, 1980. Waṣṣāf (‘Abd Allāh b. Faḍl Allāh). Ta’rīkh-i Waṣṣāf [=Tajziyat al-amṣār wa tazjiyat al-a‘ṣār]. Reprint, Teheran: Ibn Sīnā, A.Sh. 1338/1959–60, of Bombay: Mīrzā Muḥammad Ḥasan Kāshānī, A. H. 1269/1852–53. van den Wyngaert A., ed. Sinica Franciscana, vol. 1: Itinera et Relationes Fratrum Minorum saeculi u<< et u

∵

!"#$%&' ( Shams al-Dīn Juwaynī, Vizier and Patron: Mediation between Ruler and Ruled in the Ilkhanate

Esther Ravalde*

This paper aims to establish the political position and role of the vizier Shams al-Dīn Juwayn*̄ (d. 1284) by looking at his relationships as patron and client, and at how he used these relationships to govern. The upheaval of the Mongol invasions of Iran initially left a power vacuum because many of the former members of the administrative class had been dispersed and the continuity of the tradition of Persian political administration by a vizier was by no means inevitable., In the Seljuq period, the vizier was head of the bureaucracy and of the supreme dīwān (dīwān-i aʿlā).. His duties included supervising a/fairs on behalf of the Sultan: in particular, increasing agricultural production and populousness so as to maximize revenue, and managing state 01nances. He also had general supervision of the religious institutions and conducted relations with other rulers on behalf of the Sultan.2 Lambton argues that the role of the vizier temporarily contracted during the Mongol period,3 while Lane suggests that, with the gradual return to stability under Hülegü and his successors, Juwayn*̄ brought about “a recognisable continuity.4 This paper will argue that there is insu/01cient evidence to claim that the temporary contraction had eased by the end of Juwayn*̄’s life, and that what continuity there was by 1284 was limited. While his skilful negotiation of patronage networks did allow him to achieve much, this does not mean he restored the role of vizier. Lane says that Juwayn*̄ served the o/01ce and not the

* This article is based upon an 5# dissertation which was made possible by the generous support of the Soudavar Memorial Foundation. 6 Aigle, “Iran under Mongol Domination,” 72. See also May’s article in this volume, which argues that the Mongols created a power vacuum. 8 Lambton, “Wazīr,” 192. While there were also other viziers who were the heads of other estab- lishments, this essay is concerned with the head of the supreme dīwān, sometimes known as the grand vizier. 9 Lambton, Continuity and Change, 29. : Lambton, Continuity and Change, 50. ; Lane, Early Mongol Rule, 211.

© !"#$#!%$&!' ()$%% #*, %'$+'#, ,-./ | +"$ .-.../1/23425561.63,/_557 <= '>?>@AB person of the Ilkhan.C However, the o/01ce was of DEuctuating value, and the power of a vizier was determined at the discretion of the Ilkhan.F Aigle goes further than Lambton and Lane, arguing that there was a fundamental change to the role of vizier. He was no longer ‘prime minister,’ responsible for supervising all administration. At times he was not even in charge of 01nances: occasionally this role was given to an assistant vizier. The vizier always operated in collaboration with a Mongol o/01cial and required their patronage.G This paper examines Juwayn*̄’s use of patronage ties to collect taxes, strengthen Ilkhan rule in Anatolia, mediate between the ruler and the ruled, and develop alliances with the religious classes. To a certain extent, patronage enabled Juwayn*̄ to reconcile the seemingly incommensurable demands placed upon the vizier by the ruler and the ruled. It allowed a degree of negotiation to take place between Mongol and local elites, and also enabled him to project a successful image (which may explain why he is often seen as having restored the traditional role of the vizier). However, his position in the network of patronage suggests that Juwayn*̄ was dependent upon, and shared responsibility for administration with, Mongol elites. Furthermore, the personal and DEuctuating nature of Juwayn*̄’s patronage means we cannot claim that in his lifetime the traditional roles of the vizier were regained. Rather, the evidence suggests—as Aigle argues—that the vizier’s norms and functions were circumscribed.H

I Lane, Early Mongol Rule, 178. J Aigle, “Iran under Mongol Domination,” 73. K Aigle, “Iran under Mongol Domination,” 72. L This study is based on eight letters and documents in the Safīna-yi Tabrīz, including a hitherto unknown testament (vaṣīyatnāma), which is a rare act of confession in the medieval world, written by Shams al-Dīn Juwayn*̄ near the end of his life. Other letters and documents by Shams al-Dīn Juwayn*̄ can be found in later narrative chronicles, including in the Jāmiʿ al-tawārīkh, and Mujmal-i Faṣīḥī (Rashīd al-Dīn, Jāmi‘ al-tawārīkh, ed. M. Raushan and M. Mūsavī, 2: 158–59 [hereafter Rashīd/Raushan & Mūsavī]) 158–59/Rashīd al-Dīn, Jami‘u’t- tawarikh [sic] Compendium of Chronicles, translated by Wheeler M. Thackston, 3: 564–65 [hereafter Rashīd/Thackston]; Faṣīḥ, Mujmal, 2: 354–56). Yet more can be found in two scribal anthologies in St. Petersburg and Istanbul (St Petersburg, Asiatic Museum, C 816 (P$) and Istanbul University, Farsça Yazmalar, 552). While, unfortunately, I have not been able to see these two manuscripts, versions of some of these texts are preserved in the Safīna and in narrative chronicles (for instance a letter to 01ve shaykhs of Tabriz in Safīna, 76, 205a is also found in Rashīd/Raushan & Mūsavī, 2: 1158–59; Rashīd/Thackston, 3: 564–65; and in the Safīna, 414). Others from these two manuscript collections have been published over the years, including three letters in Humām al-Dīn, Dīwān, Introduction, 47–49; and two letters published anonymously (neither of which give a precise reference to their sources) in PR>ST #@-AUV WXY>ZVU, [\]\B^ #VA $>_^`V

The precise role of the vizier in 01nancial administration in the Ilkhanid period is unclear. In part this is due to a lack of evidence,,a but this problem also reveals the complex relationship between local and Mongol practices and responsibilities. However, Juwayn*̄ did play a part in taxation. He used the proceeds both to rebuild the economy and to develop ties of loyalty in a way that would have consolidated both his own position and that of the Ilkhanid rulers. Although Lambton claims that the vizier’s “fundamental and most important duty was to oversee the 01nances of the state,”,, in the Ilkhanid period he nevertheless shared this duty with several other o/01cials whose roles remain to be fully examined. It is not known, for instance, how the vizier related to the ulugh bitikchi, a secretarial position established by the Mongols,,. nor is it entirely clear how duties were divided between the mustawfī (revenue accoun- tant) and the mushrif-i mamālik (head of the dīwān-i ishrāf, the o/01ce charged with audit).,2 Indeed, since power was highly personal in nature, positions shifted according to force of personality.,3 After having waged a bitter campaign against Juwayn*̄, Majd al-Mulk, the mustawfī, was, for a time, made joint vizier with Juwayn*̄—again, how responsibilities were divided is not made clear.,4 At times there was an assistant vizier in charge of 01nance, meaning that the vizier sometimes lost his traditional main function.,C

Armagh ān, v, 1923, 284–87 and xiii, 1932, 379–81. Jürgen Paul has also helpfully written an introductory article to the Juwayn*̄ letters, in which he provides a brief synopsis of the content of many of the ‘business’ letters found in the manuscript (Paul, “The Juvayni Letters,” 277–85). As I hope to demonstrate, the published letters and documents are of immense value for the study of Juwayn*̄’s career and his patronage, yet up until now they have been rarely used. At the time of writing only Aubin, Émirs mongols, and Koprülü, “Cüveynî,” had used the letters. 6b Petrushevsky, “The Socio-Economic Condition of Iran,” 529. 66 Lambton, “Wazīr,” 193. 68 Lambton, “Wazīr,” 193; see also Dashdondog Bayarsaikhan’s article in the present volume. 69 Lambton, Continuity and Change, 60. 6: Aigle, “Iran under Mongol Domination,” 73. 6; Rash īd al-Dīn, Jāmiʿ, 2:1114/tr. Rashiduddin, Compendium, 3: 543. Rashīd al-Dīn does, however, say that when Abaqa decreed that the two were to govern jointly the area from the Oxus to the Gates of Egypt, the Mongol princesses and amirs were struck by how much power a Mongol ruler had given to Persians. 6I Aigle, “Iran under Mongol Domination,” 73. ?>@AB

The stereotypical view is that the Mongol rulers were somewhat distant, ruling with the sword and leaving Persian elites to run the administration.,F This view suggests that there was continuity in administration. However, Aigle argues this is not the case. Under the Mongols, there was a break with the Iranian sedentary tradition and unlike the Seljuqs, the Mongols made structural changes.,G For example, the Chinese practice of dual administration was introduced into Iran.,H On the whole this consisted of local civilian governors and Mongol military governors. The two roles overlapped and coexisted, and each had their own sta/f. The level of paci01cation of the land determined which of the two had precedence..a Mongols did not always leave administration to the locals., and sometimes there was a Mongol military 01gure in charge of 01nance... Martinez argues that capable Ilkhans and Mongol commanders pursued successful economic policies such as agricultural colonization, the restoration of debased currency and the consolidation of tax farms..2 Juwayn*̄’s contributions should be seen in this context. The role of vizier was shifting rather than clear-cut, and his relationships with both local and Mongol elites, while not fully clear, were close. He played a part in a new style of government which was Mongol-led, and which incorporated elements of Mongol and Persian practice. Ruined land combined with high, inexact tax rates and arbitrary methods of tax collection meant that taxation under the Ilkhans was more onerous than before..3 Lambton writes that Mongol taxation no longer remained within the bounds of the “traditional” or the “tolerable” and that “the restraints on the arbitrary and innovative use of power were minimal.”.4 However, our only sources are from conquered peoples,.C and Rashīd al-Dīn’s evidence of extor- tions, mismanagement and corruption were intended to highlight his own achievements and therefore cannot be counted on as reliable evidence..F The practical advice Rashīd al-Dīn puts into the mouth of Ghazan may, however,

6J Morgan, “Mongol Empire,” 135. 6K Aigle, “Iran under Mongol Domination,” 72. 6L Ostrowski, Muscovy and the Mongols, 36 and 41. 8b Ostrowski, Muscovy and the Mongols, 40–42. 86 Aigle, “Iran under Mongol Domination,” 73. 88 Aigle, “Iran under Mongol Domination,” 72. 89 Martinez, “Institutional Development,” 90. 8: Petrushevsky, “The Socio-Economic Condition of Iran,” 529–30. 8; Lambton, Continuity and Change, 199. 8I Lambton, Continuity and Change, 200. 8J Rashīd/Raushan & Mūsavī, 2: 1349; Rashīd/Thackston, 3: 672. English translations are by Thackston. PR>ST #@-AUV WXY>ZVU, [\]\B^ #VA $>_^`V Tajiks. If it were in my interests to pillage them all, no one would be more able to do it than I, and we would pillage together; but if henceforth you expect to requisition supplies and meals, I will reprimand you severely. Just think—if you use violence on the peasants and take their cattle and grain or let your horses graze in their 01elds, what will you do next?’ ”.G Although new taxes were introduced, such as the tamgha, qubchur, and qalan, their application and meaning varied. The qubchur and the qalan were sometimes imposed alongside the traditional kharāj and sometimes replaced it. The application of the qubchur to nomads was di/ferent from its application to the sedentary population..H Despite the gaps in our knowledge, it is clear that there was an increased burden which fell most heavily on the raʿiyyat in towns and the countryside.2a An agricultural recession meant that by the end of the period, revenue had fallen steeply since Seljuq times.2, As a consequence of these pressures, the peasants sometimes DEed the land and left it uncultivated, compounding the problem.2. Some have even argued that the consequent deserti01cation is still visible in Iran today.22 These economic problems made the vizier’s job di/01cult. Should he fail, Ilkhans had a tendency to 01ne and execute the vizier, and con01scate his property—a sure indication that what they wanted most of all was immediate cash wealth rather than long- term stability. Despite the lack of clarity about his role and the di/01culties involved in managing Ilkhan 01nances, Nakhchivānī writes that Juwayn*̄ played an important role in balancing revenue and expenditure through taxation.23 There is evidence to suggest that, through patronage, Juwayn*̄ succeeded both in organizing taxation e/01ciently and in ensuring a degree of consensus from local elites. Much of the collection and administration of taxes, in Anatolia in particular, took place under the auspices of the vizier. Juwayn*̄ imposed the tamgha in Anatolia, a levy on commercial goods, which provided a substantial part of the revenue of the Ilkhanate,24 and which had not been customary in Anatolia before.2C Some versions of the Jāmiʿ al-tawārīkh also tell us that Juwayn*̄ sent Majd al-Mulk

8K Ghazan Khan to his Mongol amirs. Rashīd al-Dīn, Jāmiʿ, I2: 1443/tr. Rashiduddin, Compendium, 3: 714. 8L Petrushevsky, “The Socio-Economic Condition of Iran,” 530–32. 9b Petrushevsky, “The Socio-Economic Condition of Iran,” 530. 96 Lambton, Continuity and Change, 219–20. 98 Rashīd/Raushan & Mūsavī, 2: 1567; Rashīd/Thackston, 3: 756. 99 Morgan, The Mongols, 81–82. 9: Nakhchiv ānī, Dastūr, 2: 52. 9; Lambton, Continuity and Change, 199. 9I Rashīd/Raushan & Mūsavī, 2: 1104; Rashīd/Thackston, 3: 538. =f '>?>@AB

Yazdī to conduct a census and to collect taxes in Mosul and Diyarbakır.2F That Juwayn*̄ had the power to spend the proceeds is also clear. From the revenue of Anatolia, he provided a stipend of an annual one thousand dinars to Humām al-Dīn’s khānaqāh.2G By doing so, Juwayn*̄ was attempting to gain the loyalty of Humām al-Dīn, and by extension his followers, by giving them some of the proceeds of taxation. More than simply a just and legitimate use of revenue, this cemented the community’s loyalty to Juwayn*̄ and established their dependence upon the Ilkhan taxation system. Shams al-Dīn Juwayn*̄’s personal patronage in Anatolia was, as Melville notes, probably due to his “private interest alert to the need to develop this important nodal point in the regional trade network.”2H It also coincided with, and may have smoothed, Ilkhanid annexation of the area. Prior to imposing the tamgha on Anatolia in 1277 and arranging the unprecedented submission of the tribes of Caucasus and Lagzistān, which may have been accompanied by a tribute, Juwayn*̄ attempted to “placate the peasantry” and he “restored the cities to habitation.”3a He had attempted intercession (shifāʿat) on behalf of Anatolia following a defection to the Mamluks and had ransomed several towns (chand pāra shahr-rā bāz kharīd).3, He built a hospital in Tokat, three madrasas in Sivas,3. and two in Nakhchivān.32 Juwayn*̄’s restoration, building works, and intercession may have been intended as a means of procuring the loyalty of the populace, for whom obedience and the payment of taxation may well have been a just exchange for protection. The fact that Juwayn*̄’s patronage and taxes coincided suggests this was the case. The practical e/fects of his patronage further reinforce this view. Some of Shams al-Dīn Juwayn*̄’s patronage was designed to raise productivity, and thus, tax revenues. Ḥamd Allāh Mustawfī writes that the irrigation of Āveh and Sāveh was dependent upon the dam Juwayn*̄ built on the Gāvmāsā river.33 In one letter, Juwayn*̄ calls on a certain Muḥyī al-Dīn34 to justify his name by reclaiming wastelands and repairing irrigation works.3C These

9J Rashīd/Thackston, 3: 542, fn. 1. 9K Humām al-Dīn, Dīwān, 48. 9L Melville, “Anatolia under the Mongols,” 72. :b Rashīd/Raushan & Mūsavī, 2: 1104–5; Rashīd/Thackston, 3: 38–39. :6 Rashīd/Raushan & Mūsavī, 2: 1102; Rashīd/Thackston, 3: 537. :8 Togan, “Economic and Cultural Life,” 102. :9 Abū al-Majd Muḥammad, Safīna, 412. :: Ḥamd Allāh Mustawfī, Nuzhat, 221. :; This may be the same Muḥyī al-Dīn who was chief qāḍī (judge) of Tabriz in 1295. Rashīd/ Thackston, 3: 628 (not in Raushan and Mūsavī’s version of the Persian text). :I Paul, “Inshāʾ,” 281. PR>ST #@-AUV WXY>ZVU, [\]\B^ #VA $>_^`V =j projects were not merely for the bene01t of the peasants. The tradition of “no kharāj on land in a bad condition” meant that uncultivated land was often untaxed,3F whereas land irrigated by running water was liable to twice the taxation of land irrigated by carried water or a water wheel.3G This looks like an attempt to create a symbiotic relationship between Ilkhanid rule and local prosperity. Shams al-Dīn Juwayn*̄’s building projects made Anatolia more accessible, and hence, more open to trade and taxation. He and his son Hārūn oversaw the construction of two branches of a highway between Sivas and Istanbul, part of the ‘Shāhrāh,’ a trade route which extended from Transoxiana to Byzantium and the Mediterranean.3H He also built bridges, including one over the Hasht river in Mīānīj.4a Greater communications, and the likely increase in accessible trade along the route, not only provided a justi01cation for taxation but the possibility of the stable and long-term collection of revenue. While one may not wish to go as far as Zeki Velidi Togan, who argues that this was part of an ‘Islamist’ colonization of Anatolia,4, the coincidence of increased building activity, in particular road building, which extended beyond Ilkhanid territory and the introduction of a new tax, are signs of a successful, and relatively peaceful, Ilkhan annexation of Anatolia. Although complaints about Ilkhanid taxes are not reliable, criticisms of Juwayn*̄’s methods of taxation are notably limited. Indeed, as Lane points out, the stark contrast between how Juwayn*̄ and his son Bahāʾ al-Dīn4. are remembered suggests that Shams al-Dīn Juwayn*̄ was widely regarded as a fair administrator. He raised taxes in Anatolia and also invested in the infrastructure, agriculture, trade and accessibility of the area. He also took care to use patronage to develop ties of loyalty. In doing so, Juwayn*̄ was acting in the manner of a traditional Persian vizier, and others after him were to continue this. While some accredit Rashīd al-Dīn with focusing on agricultural prosperity,42 it is worth noting that Juwayn*̄ anticipated this, and this reminds us of the danger of taking Rashīd al-Dīn’s self-propaganda too seriously. However, the evidence does not allow us to argue that Juwayn*̄ consolidated or regained the functions of the Persian vizier. Although he played a part, he was operating within a new

:J Lambton, Continuity and Change, 185. :K Lambton, Continuity and Change, 187. :L Togan, “Economic and Cultural Life,” 102. ;b Ḥamd Allāh Mustawfī, Nuzhat, 224. ;6 Togan, “Economic and Cultural Life,” 96. ;8 Lane, Early Mongol Rule, 211. ;9 Lambton, Continuity and Change, 65. =k '>?>@AB context. As May argues, the local system of government was integrated into a Mongol framework, and not the other way around.43 Furthermore, Juwayn*̄’s actions in Anatolia were part of the Ilkhanid colonization of the area. He had to maintain the patronage of Mongol elites in order to survive, and his success rested less on his occupation of the o/01ce than on his personal relationships. Juwayn*̄’s actions in Anatolia reveal that his patronage was not purely personal, but functioned to gain the loyalty of the community, both to himself and to Ilkhan rule. Juwayn*̄’s patronage in Anatolia operated alongside the use of force and was part of the colonization of the region. Juwayn*̄’s patronage and its links to Ilkhanid consolidation also reveal the complex boundaries between public and private actions for the vizier. His relationship to the Ilkhans suggests that they, at least, saw him and his possessions as ultimately belonging to them and that they made little distinction between public and private. The hegemony of the Ilkhans was yet to be established in Anatolia, and their rule was often challenged, particularly by the Mamluks. The Mamluks repeatedly attacked and they also had allies within the region.44 In response, the Ilkhans used force to maintain control: Abaqa went in person with the army to subdue a rebellion and defection to the Mamluks.4C Abaqa also sent Juwayn*̄ there to pacify the tribes and to collect taxes.4F Juwayn*̄ was therefore closely associated with the actions of the military in Anatolia. Furthermore, he was also known to use force himself, burning a notorious criminal in a forest.4G Juwayn*̄’s patronage was part of the process of state building.4H After the use of force came attempts to gain the loyalty of elites within the community. In this context, Juwayn*̄’s patronage appears to support and consolidate Ilkhanid policy rather than being personally motivated. We see this particularly in his use of funds from the Ilkhan to support Humām al-Dīn’s khānaqāh. The nature of Juwayn*̄’s patronage in this area raises a question about the distinction between public and private. That there was a distinction between private and o/01cial, albeit a disputed one, may be inferred from the prevalence

;: May, “Mongol Conquest Strategy in the Middle East,” p. 21 of this volume. ;; Rashīd/Raushan & Mūsavī, 2: 1115; Rashīd/Thackston, 3: 544. ;I Rashīd/Raushan & Mūsavī, 2: 1102; Rashīd/Thackston, 3: 537. ;J Rashīd/Raushan & Mūsavī, 2: 1104; Rashīd/Thackston, 3: 538. ;K Rashīd/Raushan & Mūsavī, 2: 1104; Rashīd/Thackston, 3: 538. ;L By state I mean daulat, the kingdom or realm, and not the modern idea of the nation state. It is worth noting, however, that Bert Fragner argues that the conception of the geographical and political entity of Iran, Irānzamīn, stretching from the Nile to the Oxus, came into being in the Ilkhan period. Fragner, “Ilkhanid Rule,” 80. PR>ST #@-AUV WXY>ZVU, [\]\B^ #VA $>_^`V =( of corruption cases. Majd al-Mulk Yazdī alleged to Prince Arghun (r. 1284–91) that Shams al-Dīn Juwayn*̄ had paid out large bribes to the amirs and courtiers to prevent them from criticizing him. His brother, the historian ʿAlāʾ al-Dīn ʿAṭā Malik Juwayn*̄, was also repeatedly accused of embezzlement.Ca This suggests that there was a distinction between private and public funds and corresponding expectations about how such money was used. Lambton goes further by saying that the Ilkhans went to great lengths to obtain money from their ministers, part of which was their private wealth.C, Equally, this could suggest that the Ilkhans di/fered from the Persian commentators and considered their ministers and their property to be legitimately theirs. The vizier’s inDEuence in the Ilkhanid period was invested more in his person than in the o/01ce itself, and this challenges the idea of there being a 01rm division between private and o/01cial. An illustration of this is Juwayn*̄’s patronage in Yazd. In his commission of a hospital there, we see the interweaving of private initiative and public funds. Juwayn*̄ neither saw the hospital during construction, nor paid for all of it, but sent his deputy, Shams al-Dīn Tāzīgūī, to draw up plans and to oversee the construction in his stead. Part of the cost was borne by the dīwān. Of course he took an active, if distant role, taking the initiative, overseeing plans, and converting private wealth into waqf. Despite the involvement of others, however, the hospital was most clearly recognized as having been Juwayn*̄’s project. The historians of Yazd make a distinction between this project and projects Shams al-Dīn Tāzīgūī made under his own initiative, and the poet Imāmī of Herat gives Juwayn*̄ full credit for it.C. What this episode reinforces is the fact that the distinction between private and o/01- cial is an anachronistic one to make. As vizier, Juwayn*̄ could legitimately use a combination of state and private resources for projects in the public good. Rather than an example of corruption, this was an expected part of his job, and in his job he used patronage. Thus, many accusations of corruption may have been cynical and opportunistic attacks on practices which were not only the norm, but which tended to support Ilkhanid consolidation of power. As such, 01nes and con01scations demonstrate not so much a distinction between private and o/01cial, but that the Ilkhans considered the vizier’s life and property to belong to them, and that what distinction there was between private and public was fragile at best.

Ib E.g. Rashīd/Raushan & Mūsavī, 2: 1115; Rashīd/Thackston, 3: 544. I6 Lambton, Continuity and Change, 65. I8 Aḥmad b. Ḥusain, Tārīkh, 131–33; Ja‘far b. Muḥammad, Tārīkh, 89–91. See also Lane, Early Mongol Rule, 197, for an anecdote about Tāzīgūī’s oversight of building projects. =n '>?>@AB

Although Juwayn*̄’s independence and inDEuence were limited by his dependence upon the Mongol military amirs, his position as a Persian administrator within Ilkhanid government gave him a role as arbitrator and mediator. Through his patronage relationships, he could administer justice, and, to a degree, protect the local population. The dynamics of his relationships with the rulers and the ruled also demonstrate his position within the system. Thus Aubin’s dim view of many Persian o/01cials, “destined by nature to work with any regime,”C2 who sided with the Mongols against their compatriots and co- religionists, can be re01ned. Rather than the agent of rapacious tyrants, Juwayn*̄ was able to ful01l an important role in society: that of arbitrator. Mongol rule was alien and had little legitimacy in the eyes of many Muslim subjects. Unlike the Seljuq Turks, who took power as Muslims and who ruled under the theoretical sovereignty of the ʿAbbasid Caliph, the Mongols in the Islamic lands were largely non-Muslim from their arrival in 1220 to Ghazan’s conversion in 1295 (r. 1295–1304). Hülegü, who established the Ilkhanate, a Mongol Khanate subservient to the Great Khan in China, killed the Caliph in 1258, thus extinguishing a convenient source of legitimation for Turkish military commanders in the Persian and Muslim world. The Ilkhans’ claims to legitimacy rested on their belonging to the Toluid branch of the Chinggisid family, and from having been sent to the Islamic lands by the Great Khan Möngke.C3 While the Chinggisid line was to become a central tenet of the legitimation narrative of future rulers in the Islamic lands, and of the Timurids in particular, initially, this claim made little sense to the Ilkhan’s subjects. In addition to this, their predatory attitude to their subjects was felt, although it may have been exaggerated. As we have already seen, even the most ‘cultured’C4 and Persianized of the Ilkhans, the Muslim Ghazan Khan, shared little sympathy with his subjects and instead saw them as a source of revenue.CC However, Mottahedeh argues that it was precisely in that predatory lack of identi01cation with their subjects that the potential for impartial arbitration lay. Like the Shīʿī Buyids, who ruled over predominantly Sunni Muslims, the

I9 Aubin, Émirs mongols, 81. The translation is the author’s own. I: However, whether Möngke intended Hülegü to set up a Khanate is not clear. Some have also argued that the establishment of the Ilkhanate by the Toluids was contested by another branch of the family, the Jochids. Morgan, The Mongols, 148–49. I; According to Rashīd al-Dīn, who describes the many branches of learning with which Ghazan Khan is familiar. Rashīd/Raushan & Mūsavī, 2: 1331–41; Rashīd/Thackston, 3: 663–69. II Rashīd/Raushan & Mūsavī, 2: 1443; Rashīd/Thackston, 3: 714. PR>ST #@-AUV WXY>ZVU, [\]\B^ #VA $>_^`V =<

Mongol ruler may have had the “role of arbiter, distant from the society for which it arbitrated, known to live largely for its own interest and not for any particular interest in society [. . .]. The king who ful01lled this role and saw that each interest got its due, but no more than its due, was ‘just.’ ”CF As representative of the Ilkhans, and object of personal loyalty to many groups in society, Juwayn*̄ occupied a pivotal position between the ruler and the ruled. Both the poet Humām al-Dīn and Juwayn*̄ himself saw it as the vizier’s role to administer justice, to protect from oppression, and to provide stability and strength to the realm and the faith. Humām al-Dīn writes that Juwayn*̄ was the “lion of justice” (shīr-i ʿadl) who protected the people from the “jackal of oppression” (shiqāl-i ẓulm), that he gave the world security, and that he made “the foundation of the kingdom and the faith prosperous and 01rm” (asās-i mulk va dīn maʿmūr va muḥkam). After his death, the whole world fears injustice.CG Of course Humām al-Dīn idealizes Juwayn*̄ and the concept of justice, as used by both Humām al-Dīn and Juwayn*̄ himself, function in part as literary topoi. However, the symbolic value of justice would have meant little if it had not been, at least to a certain extent, acted upon. At the end of his life, Juwayn*̄ recognised this. He apologizes, should he have oppressed anyone. He writes that if he has, his sons are to o/fer compensation.CH Throughout Juwayn*̄’s career we see him attempting to apply justice—often in relatively small but practical ways through his patronage. While direct access to the Ilkhan was limited, even to the vizier, the vizier’s court was to be open to all petitioners. Dawlatshāh tells us that Juwayn*̄ held court, seated on a throne where he received both high and low (dar ṣadr-i jāh qabūl-i ʿavāmm va khāṣṣ bar masnad-i khwājagī mutamakkin būd).Fa His correspondence supports this: Dawlatshāh preserves a petition and reply written in verse.F, Juwayn*̄’s letters sometimes appear to have been written in response to a particular request, such as his grant to Humām al-Dīn’s khānaqāh. Many of Juwayn*̄’s arbitrations may have been small token gestures, such as the granting of three hundred sheep from his special DEock,F. but the fact it was remembered suggests the symbolic act of justice was appreciated.

IJ Mottahedeh, Loyalty, 175. Ghazan Khan’s comment about the Persians is a good illustration of this (see fn. 29). IK Humām al-Dīn, Dīwān, 168–70. IL Abū al-Majd Muḥammad, Safīna, 414. Jb Daulatsh āh, Tadhkirat, 105. J6 Daulatsh āh, Tadhkirat, 105–6. J8 Daulatsh āh, Tadhkirat, 106. == '>?>@AB

Juwayn*̄’s position and use of patronage and intercession enabled him to mediate between the ruler and the ruled. On numerous occasions he mediated on behalf of provinces and villages. He wrote to Abaqa on behalf of Tabriz, saying that the city su/fered in his absence.F2 When Abaqa wanted to send an army to Khurasan to seize the ruler of Herat, Juwayn*̄ argued that Khurasan was already in ruins and could not bear the arrival of the army.F3 When parts of Anatolia defected to the Mamluks, under the governorship of Muʿīn al-Dīn Parvāna, Juwayn*̄ attempted to stop Abaqa wreaking revenge on the province, arguing that it was unjust to punish a community for the crime of an individual. While this intercession did not entirely succeed, as we have seen, Juwayn*̄ did manage to ransom several villages, thereby protecting them.F4 As an intermediary, communicating the wishes of each party to the other, Juwayn*̄ may have played a part in the gradual synthesis and accommodation of locals to Mongol rule and of Ilkhans to their subjects. However, it is important to remember the limits to what Juwayn*̄ could achieve. He himself was the client of Mongol amirs and princesses and he needed their intercession to gain access to the Ilkhan. The dual nature of administration meant that Mongol elites may have played at least as important a part in mediation as Persian elites. In order to survive, Juwayn*̄’s loyalty ultimately rested with his masters. In the Safīna-yi Tabrīz, Juwayn*̄ writes that he never attacked the life or property of any Muslim, except under the orders of his masters, and that he carried out such orders with a heavy heart: “tā ghāyat qaṣd-i khūn va māl-i hīch musulmān nakarda bāsham va agar vaqtī az ḥukmī khārijī az īn nawʿ ṣādir shuda bi dil-i rāḍī nabūda va bi-dān nimīpayvasta.”FC While he may have regretted these actions in later life, it is clear that in the course of his career, his loyalty to his masters took precedence over his loyalties to the people. Nevertheless, while Mongol government may have been seen as tyrannical, we must not overlook the fact that many poets and contemporary historians viewed Shams al-Dīn Juwayn*̄ in a positive light. To a certain extent, this was the result of successful propaganda; after all, he fostered many poets, who acted as publicists. Yet it also points to the fact that Juwayn*̄, while constrained and con- ditioned by his loyalties to his masters, did provide a level of both arbitration among the people and mediation between the ruler and the ruled. Juwayn*̄’s ambiguous position, which meant people showed loyalty to his person and

J9 Anon., “Makātib,” 285–86. J: Rashīd/Raushan & Mūsavī, 2:1105; Rashīd/Thackston, 3:539. J; Rashīd/Raushan & Mūsavī, 2:1102; Rashīd/Thackston, 3:537. JI Abū al-Majd Muḥammad, Safīna, 414. PR>ST #@-AUV WXY>ZVU, [\]\B^ #VA $>_^`V =Q that he was closely, but conditionally, connected to the Ilkhans, enabled him to transcend the distance between distant rulers and the local population. Through his patronage, which often produced mutual bene01ts, such as revenue for the ruler and security and prosperity for the ruled, Juwayn*̄ was able, to a degree, to reconcile the demands of the ruler and the ruled. In the Seljuq period, one of the functions of the vizier was the supervision of the religious institutions.FF Because of the growing inDEuence of Su01 groups, particularly among the Mongol elites, Juwayn*̄ was not in a position to ‘super- vise’ them. Juwayn*̄’s patronage of religious institutions is interesting because although he and his family tend to be remembered as orthodox Sunnis, he sometimes supported Shīʿī shrines. This tells us little about Juwayn*̄’s beliefs and personal piety—beyond his tolerance and willingness to adapt—and more about the shifting realities of power. Su01 groups were growing in inDEuence, particularly among Mongol and Turkic groups.FG Denominational boundaries were increasingly blurred and Su01 groups often incorporated Shīʿī elements.FH Juwayn*̄’s patronage in this area suggests that he needed the support of these institutions. As well as projecting his good name, support of these institutions may have made it easier for Juwayn*̄ to justify taxes, given him powerful supporters, and also indirect access to Mongol elites. Much evidence does suggest that Shams al-Dīn Juwayn*̄ was a staunch Sunni and “strictly Muslim,” and that he pursued Islamic policies, such as alliance with the Mamluks, to the extent that he was able.Ga His brother ʿAlāʾ al-Dīn ʿAṭā Malik Juwayn*̄ ended his history with the glorious Mongol conquest of the Shīʿī Isma’ilis, a triumph for orthodox Islam, but omitted to mention (according to most manuscripts) the death of the ʿAbbasid Caliph.G, In his history, he also goes to great lengths to reconcile Mongol custom with Shariʿa law. One could take this as an indication of the brothers’ traditional orthodox Sunni stance, which Spuler suggests contributed to Shams al-Dīn Juwayn*̄ falling out of favour with Abaqa later in his reign.G. Shams al-Dīn Juwayn*̄ founded schools, gave support to khānaqāhs, shaykhs, and the ʿulamā (orthodox Sunni clerics),

JJ Lambton, Continuity and Change, 29. JK Melville, “Pādshāh,” 161, argues that Ghazan converted to Islam to secure the support of the Mongol military. Bausani, “Religion under the Mongols,” 543, argues that Ghazan, like other Ilkhans, leant towards Shi‘ism. JL Bausani, “Religion under the Mongols,” 543 and 545–46. Kb Spuler, “Djuwaynī,” 607. K6 G. M. Wickens argues that the additional text was written by Naṣīr al-Dīn Ṭūsī. Wickens, “Nasir ad-Din Tusi,” 23–35. K8 Spuler, “Djuwaynī,” 607. =c '>?>@AB and converted much wealth into awqāf (charitable endowments). He also followed Quranic injunctions as to the use of wealth. When asked to pay a 01ne of 20,000,000 dinars, in accordance with the Quran, which condemns hoard- ing but encourages the use of wealth,G2 Juwayn*̄ replied “I have not stashed gold away under the earth like some fools (jāhilān).”G3 The Quran frequently exhorts Muslims to care for the poor and needy, for orphans and for slaves.G4 Juwayn*̄ ful01lled these requirements, freeing a slave,GC permitting his family’s slaves to leave should they wish and stressing the importance of treating them humanely.GF He also ensured beggars and travellers would be provided for at Humām al-Dīn’s khānaqāh.GG However, while his patronage was pious, it did not always tend to represent the traditional, orthodox Sunni, religious elite. Although Shams al-Dīn Juwayn*̄ built madrasas, which would usually have included mosques, there is a notable absence of mosque building in the sources to which I have access, which reDEects a general trend that Rashīd al-Dīn remarks upon during his time as vizier.GH The role of the ʿulamā in Juwayn*̄’s projects appears to have been restricted to administering awqāf, appointments of qāḍīs (judges) and to teaching positions.Ha In contrast, Juwayn*̄ appeared to have particularly close relations with several shaykhs, including the poet Saʿdī, Humām al-Dīn and Fakhr al-Dīn, and his support of khānaqāhs outweighs his interest in schools and mosques. In addition to this, and perhaps surprisingly, were we to accept Spuler’s analysis, Juwayn*̄ also supported the “mashāhid-i ṭūs va jurjān,” the tomb of Imām Riḍā, in Mashhad, and a Shīʿī shrine in Jurjān. Part of the explanation for Juwayn*̄’s support of Shīʿī shrines is that a recent blurring and softening of religious boundaries made such support acceptable.H,

K9 Sura 104. K: Rashīd/Raushan & Mūsavī, 2: 1158; Rashīd/Thackston, 3: 564. Jāhilān refers to those who lived in ‘ignorance’ before the coming of Islam. K; For example Sura 2:43, 83 and 177 and Suras 90 and 107. KI Paul, “Inshāʾ,” 280. KJ Abū al-Majd Muḥammad, Safīna, 414. KK Humām al-Dīn, Dīwān, Introduction, 47–48. KL Rashīd al-Dīn complains that upon the accession of Ghazan Khan, many towns and villages did not have their own mosque. Rashīd/Raushan & Mūsavī, 2: 1502; Rashīd/ Thackston, 3: 743. Lb For example Paul, “Inshāʾ,” 281 on the appointment of a qāḍī, Abū al-Majd Muḥammad, Safīna, 412–13 on the appointment of a teacher, and 414, the Qāḍī Muḥyī al-Dīn is made overseer of Juwayn*̄’s awqāf in Tabriz. L6 Bausani, “Religion under the Mongols,” 543–49. PR>ST #@-AUV WXY>ZVU, [\]\B^ #VA $>_^`V =d

The Shīʿī Fatimids had been replaced by the Sunnī Ayyubids and then the Mamluks, while in Persia, the Isma’ilis had been destroyed. Political reasons for the suppression of Shiʿism had therefore become less pressing. In addition to this, following the extinction of the caliphate and the works of two inDEuen- tial Shīʿīs, Naṣīr al-Dīn Ṭūsī (d. 1274) and ʿAlāma Ḥillī (d. 1326), Shiʿism gained new con01dence. At the same time, the relative status of the ʿulamā declined, while that of Su01 shaykhs increased. Under the auspices of more heterodox (but still Sunni) Su01 movements, pro-Shīʿī beliefs became more widespread. In this climate, the distinction between Shiʿism and Sunnism could become blurred.H. Indeed, many of the inhabitants of Tus, who no doubt venerated the tomb there, were “orthodox of belief.”H2 In this intellectual climate, Shīʿī and other non-orthodox groups could be supported without incurring disapproval. In addition to the blurring of denominational boundaries, the leaders of khānaqāhs and other heterodox, particularly Su01, groups were becoming increasingly powerful, and Shams al-Dīn Juwayn*̄ needed their support in order to maintain control and to collect tax. Shams al-Dīn Juwayn*̄ writes in a letter that Humām al-Dīn’s khānaqāh was thronging with people, high and low.H3 Nuzhat al-qulūb tells us the shrine at Tus was highly venerated and that a small town had grown up around it.H4 In the Timurid period, Su01 leaders were to become powerful protectors of the local population and mediators between the ruler and the people. Khwāja Aḥrār, a Naqshbandī shaykh in Timurid Central Asia, was able to negotiate tax bene01ts for his followers.HC The Safavids were powerful enough to wrest land from the Juwayn*̄ family in the 1350s.HF If Juwayn*̄ was to successfully negotiate settlements and collect taxes from the local populace, it was necessary to establish good relationships with the leaders of those communities, in particular relationships of patronage, which established the dependence of the community upon the good will of the patron. Su01 groups were particularly popular among Turko-Mongol converts, and especially, among members of the military. Indeed, Ḥamd Allāh Mustawfī writes that “the Mongols are utterly devoted to him [Shaykh Ṣafī al-Dīn Ardabīlī]; many of that tribe have ceased to oppress people and this is a great

L8 Bausani, “Religion under the Mongols,” 545 and 549. L9 Ḥamd Allāh Mustawfī, Nuzhat, 151. L: Humām al-Dīn, Dīwān, Introduction, 47–48. L; Ḥamd Allāh Mustawfī, Nuzhat, 151. LI Paul, “Forming a Faction,” 540. LJ Gronke, Derwische, 306–308 and 323. Qf '>?>@AB deed.”HG Rather than a conversion of the heart, as Rashīd al-Dīn tells us it was, Melville has argued that Ghazan’s conversion in 1295 was engineered to appeal to Muslim converts in the military.HH ReDEecting the inclinations of key supporters, although Ghazan converted to Sunnism, his conversion included a Su01 element,,aa and he took care to pay his respects to the shrine at Tus,,a, and to ʿAlids and Sayyids.,a. Bearing this in mind, much of Juwayn*̄’s patronage of the religious classes may have had the similar aim of gaining the support of the military, and an aura of popular legitimacy among Turco-Mongol elites. It does seem to have been the case that Shams al-Dīn Juwayn*̄ cultivated friendships with the religious classes in order to gain their blessing and the support of parts of the military, and Mongol elites, including women. During the civil war between Aḥmad, who was Muslim, and Arghun, who was not, Aḥmad and his entourage visited the Ṣafavī Shaykhs at Ardabil. The author of Ṣafwat al-ṣafā tells us that Juwayn*̄, who was present, greeted one of the Su01s with an embrace. Following this, Aḥmad went to meet Arghun in battle.,a2 Just as Ghazan later attempted to appeal to the Muslim element in the army by converting (in contrast to his non-Muslim rival Baidu), so, it seems, did Aḥmad. Mongol women also supported Su01 shaykhs. Öljei built a khānaqāh over the place where her father, Hülegü’s, grave stood, and Būlūghān, the wife of Abaqa and Arghun, supported shaykhs and their communities.,a3 Juwayn*̄’s patronage mirrored the preferences of his own patrons, showing the relative lack of inDEuence of the vizier in the period and his need to maintain the support of local elites who did have the ear of Mongol elites. We should not, however, entirely overlook personal piety as a motive for religious patronage. Prior to his death, Shams al-Dīn Juwayn*̄ wrote a letter to 01ve great shaykhs of Tabriz, including Humām al-Dīn and Fakhr al-Dīn, as well as Muḥyī al-Dīn, who may have been the chief qāḍī of Tabriz, reminding them of the good news (bashārat) of Islam and asking them to pray for him when he is gone.,a4 At a time of crisis, while his friends may not have been able to help materially (he had been unable to raise the 01ne of 20,000,000 dinars), he

LK Ḥamd Allāh Mustawfī, Guzīda, 675. LL Melville, “Pādshāh,” 159–77. 6bb Ḥamd Allāh Mustawfī, Guzīda, 675. 6b6 Rashīd/Raushan & Mūsavī, 2: 1376; Rashīd/Thackston, 3: 685. 6b8 Rashīd/Raushan & Mūsavī, 2: 1358; Rashīd/Thackston, 3: 676–77. 6b9 Ibn Bazzāz, Ṣafwat, 218. 6b: De Nicola, “Unveiling the Khatuns,” 249. 6b; Rashīd/Raushan & Mūsavī, 2: 1157; Rashīd/Thackston, 3: 565. PR>ST #@-AUV WXY>ZVU, [\]\B^ #VA $>_^`V Qj did look to his friends for spiritual support, perhaps in return for his support of their khānaqāhs. “Rašīddadīn 01rmly believed, that whosoever inaugurates a perpetual charity or pious institution will rise in the other world’s hierarchy step by step as long as this institution exists.”,aC A similar belief may explain why Shams al-Dīn Juwayn*̄ went to such great pains to provide for the perpetuation of the institutions he founded. Just before his death, he constituted many schools, hospitals and khānaqāhs he had built into waqf.,aF Waqf was in theory exempt from taxation and could not be con01scated. While in practice the inalien- able status of waqf land was often not respected in the Ilkhanid period,,aG the increase in waqfs over the Mongol period suggests that this arrangement was considered more secure than other options. Nevertheless, while Shams al-Dīn Juwayn*̄ was a believing Muslim, for most of his life he was certainly not a strict Muslim, as Spuler would have us believe. Ful01lling pious duties was secondary to his career as vizier. In addition to molesting fellow Muslims, while he writes that he committed no major sins (makrūhī-yi kullī), he does confess to having drunk wine, committed sins of the DEesh and committed usury (shurb-i khamr va maḥrumāt-i shahvānī va ribāʾ). He also asks his sons to employ several people to say the prayers he has missed. The confessional tone of this testament, written at the end of his life, should not be considered representative of his life. It suggests rather that he now regrets some of the aims he pursued throughout much of his life. He for- bids his sons to work in government service and instructs them to be content with limited wealth, all of which has a rather regretful air.,aH While Shams al-Dīn Juwayn*̄’s patronage of religious institutions may have been motivated by piety to an extent, by far the greater reason was that shrines and khānaqāhs were becoming popular, and that their leaders were becoming powerful 01gures, especially among the Turko-Mongol population and the military. This, rather than personal conviction, may well explain why Juwayn*̄ gave greater attention to these groups, including to the Shīʿī shrines, rather than to traditional forms of orthodox Sunni Islam. Good relationships with the shaykhs may have given Juwayn*̄ greater access to taxation, the loyalty of their followers, and therefore a level of popularly granted legitimacy. It also gave him access to some parts of the military and to some Mongol women, and

6bI Ho/fmann, “Perfect Organiser,” 288. 6bJ Abū al-Majd Muḥammad, Safīna, 414. 6bK Nakhchivānī, Dastūr, t/1: 175. 6bL Abū al-Majd Muḥammad, Safīna, 414. Qk '>?>@AB through them he could sometimes gain access to the Ilkhans. Thus, in this area, rather than consolidating the role of the vizier, Juwayn*̄ was adapting to a new trend and following the inclinations of the powerful. Juwayn*̄ was from a long line of administrators and one of his aims was 01rstly to use patronage to secure his good name for posterity and secondly to secure an inheritance for his sons. The degree to which contemporaries praised him suggests he succeeded in his 01rst aim.,,a This issue is worth considering because his successful self-propaganda means we have to be cautious about using contemporary sources, which may suggest that Juwayn*̄ had more inDEu- ence than he really did. Juwayn*̄ is remembered as being more than the servant of the Ilkhans, even though much of his patronage was intended to further his political role. He was less successful in his second aim. The Safīna-yi Tabrīz suggests that Juwayn*̄ was intent on providing an inheritance for his sons. When he converted many of his schools, hospitals and khānaqāhs into waqf, he made his son Hārūn (mutavallī-yi dīgar awqāf farzand hārūn), and after him, his sons who were worthy and learned (ki bi ʿilm va dīyānat makhṣūṣ bāshand), the administrators of much of his waqf land.,,, While waqf property was a charitable endowment and was therefore not owned by the founder or his descendants, the founder had the power to appoint the administrators. In turn the administrator could take a large payment from the proceeds of the endowment. In the case of Rashīd al-Dīn’s endowment in Tabriz, for example, the administrator, who was to be a son, was paid half the endowment’s income.,,. The constitution of awqāf, then, was a relatively secure means of securing inheritance for nominated members of his family. At the time of his death, Juwayn*̄ probably believed his e/forts to accumulate vast family wealth had failed. He thus converted much land to waqf at the last minute and warned his sons to accept any con01scation of land.,,2 To an extent, his concern was justi01ed: much of his land was con01scated by Amīr Buqa.,,3 However, despite the Ilkhan’s purge of the Juwayn*̄ family and con01scation of land, as mentioned earlier, one of Shams al-Dīn Juwayn*̄’s grandsons, Shams al-Dīn, son of Nawruz, lived to contest, and lose, claims for land with the Safavids in the 1350s. Despite falling from political favour, it is worthy of note that the Juwayn*̄ family managed to hold on to land for a

66b Lane, Early Mongol Rule, 211. 666 Abū al-Majd Muḥammad, Safīna, 414. 668 Sheila Blair, “Endowment Deed,” 79; Lambton, “Mongol Fiscal Administration,” 79–99. 669 Abū al-Majd Muḥammad, Safīna, 414. 66: Rashīd/Raushan & Mūsavī, 2: 1160; Rashīd/Thackston, 3: 565. PR>ST #@-AUV WXY>ZVU, [\]\B^ #VA $>_^`V Q( further sixty years: in the face of adversity, Shams al-Dīn Juwayn*̄ did succeed in passing down some wealth and inheritance to two generations. Charity was central to providing an inheritance: while also providing institutions for public bene01t, such as schools and hospitals, waqf also gave the founder’s family a greater chance of retaining some control of family land and wealth, albeit not ownership. This type of patronage not only bridged the gap between the ruler and the ruled, as we have seen, but it also gave great bene01t to its founder and his heirs. Shams al-Dīn Juwayn*̄ believed that through good works, his name could be immortalized. At the start of the document appointing a teacher to the schools he founded is written: the Ṣāḥib Dīwān “deserves great praise and he had stored up for himself great rewards,” and “indeed our works point to us, so when we are gone, look to our contribution.”,,4 He wanted his contribution to be remembered and so places he founded were named after him, such as the “new Ṣāḥibābād garden of Charandāb.”,,C One of his 01nal charitable acts was to commemorate his own life: in the testament, he gives detailed stipulations for the provision of all kinds of sweets, tallow and oil for his “shab-i ghurbat,” the night of his death.,,F Commemoration was not a legacy that could be achieved by force. More important was to establish the good will of poets, who operated as publicists and propagandists. This he did, establishing close friendships with Humām al-Dīn and Saʿdī,,,G and by paying poets generously to commemorate his patronage.,,H Despite being dependent on the patronage of Mongol amirs and princesses and needing the support of religious institutions, Juwayn*̄ nevertheless succeeded in projecting an image as a patron in his own right. When assessing his real power and inDEuence, therefore, we have to be aware of this bias in the sources. Despite the image Juwayn*̄ projected, his real power and inDEuence were limited. Although he developed networks of patronage which included Persian and Turko-Mongol elites, ultimately Juwayn*̄ remained at the mercy of higher networks of patronage. His ability to govern by using patronage was therefore limited. He often did not have direct access to the Ilkhan. On at least two occasions when Abaqa was displeased with him, he had to write to plead his

66; Abū al-Majd Muḥammad, Safīna, 412. 66I Abū al-Majd Muḥammad, Safīna, 414. Ṣāḥib referring to his title, Ṣāḥib dīwān. 66J Abū al-Majd Muḥammad, Safīna, 414. 66K Daulatshāh, Tadhkirat, 106. 66L Aḥmad b. Ḥusain, Tārīkh, 131–33; Jaʿfar b. Muḥammad, Tārīkh, 89–91. Imāmī was paid 1,000 dinars to commemorate the new hospital. Qn '>?>@AB loyalty and innocence and to request an audience. He was only able to go to Arghun’s court under the protection of Amīr Buqa. More successful than his own protestations were the intercessions of the Mongol princesses. When Majd al-Mulk accused Juwayn*̄ of corruption and bribery, he sought the aid of Öljei who “wrote a petition for him” which “rescued him from the brink of disaster.”,.a Later, when the same accusations were raised again, Juwayn*̄ sought protection with “Armini Khatun, under whose patronage he DEourished as before” (bi Armini Khatun iltijāʾ namūd va bi tarbiyat-i ū bar qarār-i sābiq muʿtabar-i tamām shud).,., When Arghun, whose accession Juwayn*̄ had not supported, gained the throne, Juwayn*̄ hoped that he might have intercession through Amīr Buqa, “who is an old friend” (dūst-i dīrīna). This failed and so did his attempts to raise the required cash 01ne from friends and family.,.. These events reveal the limitations of Shams al-Dīn Juwayn*̄’s position. The networks of alliances he had created through ‘downward’ patronage, especially with the inDEuential religious classes, may have ensured the smooth running of government, but they were not enough to release him from his debt, nor to defend him against accusations of corruption and treachery. Even his downward relationships revealed his need for acceptance among the growing Su01 groups. Whatever power and inDEuence Juwayn*̄ may have wielded among his allies was irrelevant in the face of the fact that he was the Ilkhan’s man—in this case, he had allied himself to the losing side, and as Aḥmad was defeated, so must he be. The only factions that could have saved him, as they had before, were the Mongol elite—the princesses and the amirs and they did not save him this time. The land, connections and wealth he had accumulated through patronage proved to be a double-edged sword. Vulnerable to accusations of embezzlement and bribery, Juwayn*̄’s wealth was also coveted by others. None other than Amīr Buqa, in whom Juwayn*̄ had placed the last vestiges of hope, (and who also became vizier) took possession of a large part of Juwayn*̄’s property.,.2 Lambton suggests that by the end of the Ilkhanid period the vizierate had mostly regained the functions it had had in the Seljuq period.,.3 By the time of

68b Rashīd/Raushan & Mūsavī, 2: 1112; Rashīd/Thackston, 3: 543. Mongol princesses often acted as protectors of inDEuential characters. For example, see De Nicola, “Unveiling the Khatuns,” 126. 686 Rashīd/Raushan & Mūsavī, 2: 1127; Rashīd/Thackston, 3: 549 688 Rashīd/Raushan & Mūsavī, 2: 1157; Rashīd/Thackston, 3: 564. 689 Rashīd/Raushan & Mūsavī, 2: 1160; Rashīd/Thackston, 3: 565. 68: Lambton, Continuity and Change, 67–68. PR>ST #@-AUV WXY>ZVU, [\]\B^ #VA $>_^`V Q<

Juwayn*̄’s death in 1284 few signs of this could be seen. It was true that he had some responsibility for collecting tax and for state expenditure. However, this role was shared with Mongol amirs and occasionally with other Persian 01gures. The ambiguity surrounding the role of the vizier in the dual system of administration also means that we cannot say Juwayn*̄ or the o/01ce of the vizirate had 01rmly regained its position. That was liable to change, at the discretion of the Ilkhan. Despite this, Shams al-Dīn Juwayn*̄ did use patronage to achieve a great deal. Through patronage he not only organized and collected taxes but developed agriculture and trade, supported the religious classes and acted as nego- tiator, moderating some of the harshness of Ilkhan rule. To a degree, patronage made possible the reconciliation of the demands of the ruler and the ruled: to the ruler Juwayn*̄ could present tax receipts and prosperous, loyal, subjects. To the ruled Juwayn*̄ could o/fer negotiation, prosperity, and justice. Because power was invested in his person by the Ilkhan and because simultaneously, thorough patronage, he had developed ties of loyalty among certain sectors of society, Juwayn*̄ could o/fer a degree of real dialogue, mediation, and synthesis between the ruler and the ruled. However, rather than Ilkhans adapting to Persian practice, Juwayn*̄ adapted to the Ilkhans. He deferred to Mongol amirs and princesses and he followed the trend in supporting Su01 groups, Shīʿī causes and the consolidation of power in Anatolia. His position in the network of patronage reveals that he was not particularly powerful. Although he spent his money in a way which would project a pious, cultured and inDEuential image, this was illusory. He did succeed in creating a legacy for his sons but his life and power remained precarious. While Juwayn*̄ succeeded in gaining some inDEuence, he did not revive the institution of the vizier. The role of the vizier remained unresolved and much less inDEuential than it had been under the Seljuqs.

Bibliography

Abū al-Majd Muḥammad. Safīna-yi Tabrīz. Facsimile of manuscript in the Islamic Consultative Assembly. Tehrān, 2003. Aḥmad b. Ḥusain b. ʿAlī al-Kātib. Tārīkh-i jadīd-i Yazd. Edited by Īraj Afshār. Tehrān: Intishārāt-i Farhang-i Irān-i Zamīn, 1345/1966–67. Aigle, Denise. “Iran under Mongol domination: The E/fectiveness and Failings of a Dual Administrative System.” Bulletin d’études orientales 57 (March 2008): 65–78. Anon. “Makātib-i tārīkhī.” Armaghān 5 (1923): 284–87. Anon. “Du nāma-yi muhimm-i tārīkhī.” Armaghān 13 (1932): 379–82. Q= '>?>@AB

Aubin, Jean. Émirs mongols et vizirs persans dans le remous de l’acculturation. Paris: Studia Iranica, Cahier 15, 1995. Blair, Sheila. “Īl-Khānid Architecture and Society: An Analysis of the Endowment Deed of the Rabʿ-i Rashīdī.” Iran 22 (1984): 67–90. Bausani, A. “Religion under the Mongols.” In The Cambridge History of Iran, vol. 5, edited by J. A. Boyle, 283–302. Cambridge: Cambridge University Press, 1968–1991. Dawlatshāh Samarqandī. Tadhkirat al-shuʿarāʾ. Edited by E. G. Browne, London: Luzac & Co., 1901. De Nicola, Bruno. “Unveiling the Khatuns: Some Aspects of the Role of Women in the Mongol Empire.” Unpublished PhD diss., University of Cambridge, Pembroke College, 2011. Faṣīḥ. Mujmal-i Faṣīḥī, vol. tt. Edited by Maḥmūd Farrukh. Mashhad: no publisher named, 1339–41/1958–61. Fragner, Bert, “Īl-Khānid Rule and its Contribution to Iranian Political Culture.” In Beyond the Legacy of Genghis Khan, edited by Linda Komaro/f, 68–80. Leiden and Boston: Brill, 2006. Gellner, Ernest. “Patrons and Clients.” In Patrons and Clients in Mediterranean Societies, edited by E. Gellner and J. Waterbury, 1–6. London: Duckworth, [for] the Center for Mediterranean Studies of the American Universities Field Sta/f, 1977. Gronke, Monika. Derwische im Vorhof der Macht: Sozial- und Wirtschaftsgeschichte Nordwestirans im 13. und 14. Jahrhundert. Stuttgart: Steiner, 1993. Ḥamd Allāh Mustawfī Qazvīnī. Nuzhat al-qulūb, 2 vols. Edited and translated by G. Le Strange. London: Luzac & Co; Leiden: E.J. Brill: 1915 and 1919. ———. Tārīkh-i guzīda. Edited by ʿAbd al-Ḥusain Navāʾī. Tehrān: Amir Kabir, 1362/1983. Ho/fmann, Birgitt. “Rašīdaddīn Faḍlallāh as the Perfect Organiser: The Case of the Endowment Slaves and Gardens of the Rabʿ-i Rašīdī.” In Proceedings of the Second European Conference of Iranian Studies, 287–96. Rome: Istituto Italiano per il Medio ed Estremo Oriente, 1995. Humām al-Dīn Tabrīzī. Dīwān. Edited with an introduction by Rashīd Aivażī. Tabriz: Dānishkada-yi adabiyyāt va ʿulūm-i insānī-yi Tabriz, 1351/1972. Ibn Bazzāz Ardabīlī. Ṣafwat al-ṣafā. Edited by Ghulām Riḍā Ṭabāṭabāʾī Majd. Tabriz: Ghulām Riḍā Ṭabāṭabāʼī Majd, 1373/1994. Jaʿfar b. Muḥammad b. Masan Jaʿfarī. Tārīkh-i Yazd. Edited by Īraj Afshār. Tehrān: Bungāh-i Tarjuma va Nashr-i Kitāb, 1338/1960. Juwaynī, ʿ Aṭā Malik. Ta’rīkh-i Jahāngushā. Edited by Mīrzā Muḥammad Qazwīnī. London: Luzac, 1912. ———. The History of the World-Conqueror. Vol. 1. Translated by John Andrew Boyle. Manchester: Manchester University Press, 1958. Köprülü, M. F. “Cüveynî.” In Islâm Ansiklopedisi, ttt, 255–59. Istanbul: Maarif Matbaası, 1945. PR>ST #@-AUV WXY>ZVU, [\]\B^ #VA $>_^`V QQ

Lambton, Ann K. S. “Mongol Fiscal Administration in Persia.” Studia Islamica 64. (1986): 79–99. ———. Continuity and Change in Medieval Persia: Aspects of Administrative, Economic and Social History, 11th–14th Century. London: Biblioteca Persica, 1988. ———. “Wazīr, ii, in Persia.” In Encyclopaedia of Islam, Second Edition xttt, 192–94. Lane, George. Early Mongol Rule in Thirteenth-Century Iran: A Persian Renaissance. London: RoutledgeCurzon, 2003. Martinez, Arsenios Peter. “Institutional Development, Revenues and Trade.” In The Cambridge History of Inner Asia: The Chinggisid Age, edited by Nicola Di Cosmo, Allen J. Frank and Peter B. Golden, 89–108. Cambridge: Cambridge University Press, 2009. May, Timothy, “Mongol Conquest Strategy in the Middle East.” In The Mongols’ Middle East: Continuity and Transformation in Ilkhanid Iran, edited by Bruno De Nicola and Charles Melville, 13–37. Leiden: Brill, 2016. Melville, Charles. “Pādshāh-i Islām: The Conversion of Sulṭān Maḥmūd Ghazan Khān.” Pembroke Papers 1 (1990): 159–77. ———. “Anatolia under the Mongols.” In The Cambridge History of Turkey, vol. t, edited by K. Fleet: 51–101. Cambridge: Cambridge University Press, 2009. Morgan, David. “Who Ran the Mongol Empire?” ,-./ 1 (1982): 124–36. ———. The Mongols. Oxford: Blackwell, 2001. Mottahedeh, Roy. Loyalty and Leadership in an Early Islamic Society. London: I. B. Tauris, 2001. Muḥammad b. Hindūshāh Nakhchivānī. Dastūr al-kātib fī taʿyīn al-marātib, vols. t and tt, edited by A. A. ‘Alīzāda. Moscow: Nauka, 1976. Ostrowski, Donald G. Muscovy and the Mongols: Cross-cultural In0luences on the Steppe Frontier, 1304–1589. Cambridge: Cambridge University Press, 1998. Paul, Jürgen. “Forming a Faction: The Himayat System of Khwaja Ahrar.” 1,23/ 23:4 (1991): 533–48. ———. “Some Mongol Inshāʾ-Collections: The Juvayni Letters.” In Proceedings of the Third European Conference of Iranian Studies, part 2, edited by C. Melville, 277–85. Wiesbaden: Dr Ludwig Reichert Verlag, 1999. Petrushevsky, I. P. “The Socio-Economic Condition of Iran under the Īl-Khāns.” In The Cambridge History of Iran, vol. 5, edited by J. A. Boyle, 483–537. Cambridge: Cambridge University Press, 1968. Rashīd al-Dīn. Jāmiʿ al-tawārīkh. Edited by M. Raushan and M. Mūsavī. 4vols. Tehran: Nashr-i Alburz, 1373/1994. ———. Jami‘u’t-tawarikh [sic] Compendium of Chronicles. 3 vols. Translated by Wheeler M. Thackston. Cambridge, 5#: Department of Near Eastern Languages and Civilizations, Harvard University, 1998–99. Spuler, B. “Djuwaynī, Shams al Dīn Muḥammad b. Muḥammad.” In Encyclopaedia of Islam, Second Edition tt:607. Qc '>?>@AB

Togan, Zeki Velidi. “References to Economic and Cultural Life in Anatolia in the Letters of Rashīd al-Dīn.” Translated by Gary Leister. In History and Historiography of Post- Mongol Central Asia and the Middle East: Studies in Honor of John E. Woods, edited by Judith Pfei/fer and Sholeh A. Quinn in collaboration with Ernest Tucker, 84–111. Wiesbaden: Harrassowitz, 2006. Wickens, G. M. “Nasir ad-Din Tusi on the Fall of Baghdad.” Journal of Semitic Studies 7 (1962): 23–35. !"#$%&' ( The Economic Role of Mongol Women: Continuity and Transformation from Mongolia to Iran

Bruno De Nicola

Introduction

The economic history of the Mongol Empire has puzzled historians for decades. The information we have on the subject for the di)ferent parts of the vast territory conquered by the Mongols varies radically. This becomes apparent if we compare, for example, the fair amount of data available for the territories of the Mongol Yuan Dynasty of China (1279–1368) with the almost non-existent information we have on how these nomadic rulers administered the economy of, for example, Central Asia.* For the territories of the Ilkhanate, that is Iran, Khurasan and West Asia, it is important to highlight the relevance of the studies carried out by Lambton and Petrushevsky some time ago.+ More recently, a more comprehensive framework for understanding the development of the Mongols’ role in the economy of the territories they conquered was provided by Allsen, who suggested a more nuanced view of the Mongols as rulers and administrators of a multicultural empire., The Mongol Empire underwent a bi-directional cultural and economic transformation whereby di)ferent territories had to accommodate their economies to the needs not only of the new rulers, but also of the conquered populations. Consequently, the di)ferent Mongol rulers had to adapt, with the help of the local elites, to di)ferent circumstances and economic opportunities presented by the newly conquered territories. Because the Mongol conquest was not only a military enterprise but also a migration of people, women were not excluded from the economic activity of the empire. Further, the relatively high status enjoyed by Mongol Khatuns

- Schurmann, Economic Structure of the Yuan Dynasty; Endicott-west, “Merchants Associations in Yüan China,” 127–54. / Lambton, Landlord and Peasant in Persia; Petrushevsky, “The Socio-Economic Condition of Iran,” 483–537. 0 Allsen, “Mongolian Princes and their Merchant Partners,” 83–126; Allsen, Culture and Conquest, 17–56.

© !"#$#!%$&!' ()$%% #*, %'$+'#, ,-./ | +"$ .-.../1/23425561.63,/_55/ 12 34 56789: across the empire made them, at times, important actors in the empire’s economy. This article will focus on how the role of Mongol women in the economy of the empire was transformed when these ladies settled in Iran after the conquest of Hülegü during the 1250s. I will use Allsen’s ideas on the economic activity of the Mongols in the early empire and contrast it with the information we have on female economic activity in Ilkhanid Iran.; The ultimate aim is to show that women played a signi<=cant role in the economic life of the empire and that this role was transformed and adapted depending on the dif- ferent circumstances faced during the territorial expansion of the empire.

Female Economic Activity from the Steppe to World Empire

Although recent archaeological research has documented the existence of agricultural practices in the steppe at the time of Chinggis Khan, in pre- imperial Mongolia wealth was based on two basic resources: cattle and people.> As Chinggis Khan was subduing his rivals in the steppe, a process of systematized plunder, characteristic of this early stage, was carried out by his followers and relatives.? In the Secret History, this process is especially vigor- ous when Temüjin (the original name of Chinggis Khan) conquers a rival faction. For example, immediately after he “crushed and despoiled” the Kereits, Chinggis Khan started “distributing them on all sides,” giving to some of his allies a full sub-group of the conquered people.@ He also took the nieces of his defeated rival from him, marrying one of them and giving the other to his son Tolui.A He was merciful to the father of the two women, allowing him to keep his subordinates but placing under his control all the resources (mostly cattle) belonging to him and his daughters. The Secret History goes on to detail how the Kereit people were distributed among Chinggis’ allies according to their merit in battle, attesting not only to a system of plunder but also to one of distribution of human and material resources.B

C I would also like to thank Timothy May for sharing with me an early version of his forthcoming paper. See May, “Commercial Queens.” D Allsen, “The Princes of the Left Hand,” 27. E Allsen, Commodity and Exchange, 27. F It is mentioned that he gave to the Suldus Taqai the Jirgin branch of the Kereit people. Anonymous, The Secret History of the Mongols, §186 [hereafter G"H]. I These women are Ibaqa Beki and Sorghaghtani Beki respectively. J G"H, §187. Similar examples can be observed during the defeat of other steppe factions such as the Merkits; see G"H, §198; on the Tatars’ extermination and the inKLuence of women in saving some of their relatives, see Rashīd al-Dīn, Jāmi‘ al-tawārīkh, ed. M. Raushan and %O4 &78P8Q67 '894 RS H8PT89 U8Q4P 1N

In this strategy of plunder and distribution, women were part of the booty and had a share in it.*V During this early period, one woman who received a considerable amount of wealth in this manner was Chinggis Khan’s mother Hö’elün. She was always a bene<=ciary in the allocation of people conquered by Chinggis Khan. Sources disagree about her share of the booty, a subject that has generated a certain amount of discussion among scholars.** However, despite di)ferences of opinion regarding the number of people received, she certainly was a bene<=ciary of this allocation practice. Her position as mother of the ruler was the determining factor in her always receiving more people than the rest of the family. Further, having all these people under their command lent a military capability to members of the royal family, and an economic one too, because herds and KLocks came with the conquered people and were incorporated into the Chinggisid family’s ordos (Mongol camps including properties and people).*+ Other women in Chinggis Khan’s family also received people who had been seized in war. His wife Börte had an ordo and at least occasionally was the recipient of such people.*, She is not mentioned in either the Chinese, Persian or Mongolian sources that refer to the Khan’s distribution of people, but Rashīd al-Dīn comments on the fate of a Tangut boy who was brought to her, presumably after a raid against the Xi-Xia Kingdom.*; Generals from di)ferent backgrounds from across the steppe are mentioned as being part of Börte’s personal appanage, con<=rming that she also received a share of the

M. Mūsavī, -:I0 [hereafter Rashīd/Raushan & Mūsavī]; Rashīd al-Dīn, Jami‘u’t-tawarikh [sic] Compendium of Chronicles, translated by Wheeler M. Thackston, -: CE–CF [hereafter Rashīd/Thackston]. -X Another appropriation of women by the victorious side after a battle occurred when the Mongols defeated the Tumat people; see G"H, §241. -- The Secret History says that she received 10,000 people while Rashīd al-Dīn reduces this to 5,000 with another 3,000 going to the youngest son Otegin; these, however, remain within the mother’s control, taking her tally to 8,000 people. See G"H, §242; Rashīd/Raushan & Mūsavī, 1:611; Rashīd/Thackston, 2:281. -/ On the inKLuence of women in the Mongol army see De Nicola, “Women’s Role and Participation in Warfare,” 95–112. For an introduction to female ordos in the Mongol empire see De Nicola, “Ruling from Tents,” 126–36. -0 The incorporation of people into Börte’s ordo is not presented in quantitive but in ‘quali- tative’ terms, meaning that is the background of the people she received (for example Buda Noyan) that matters for Rashīd al-Dīn rather than the number of people given to her. See Rashīd/Raushan & Mūsavī 1:137; Rashīd/Thackston, 1:74. -C Dunnel, “The Hsi Hsia,” 206–14. 1Y 34 56789: steppe people who submitted to her husband.*> Other wives (Qulān Khatun) and some of Chinggis Khan’s daughters (Tümälün Khatun and Chächäyigän) are also reported to have received people.*? This suggests that, though it may only be Hö’elün who is mentioned in the major distribution of people carried out by the Khan among his male relatives, other women in the royal family participated in the system of conquest and distribution during the formative period of the empire. A story concerning another of Chinggis’ wives gives us not only further evidence of the existence of ladies’ ordos and the share Khatuns had in the revenues produced by the consolidation of the empire, but also some details about what the wealth in their camps consisted of. There is a famous anecdote in the Secret History of the Mongols and the Jāmiʿ al-Tawārīkh about Chinggis Khan giving away one of his chief wives, the Kereit Ibaqa Beki, to one of his generals.*@ The story di)fers signi<=cantly between the two versions but there are certainly correspondences in the two renditions with regard to the wealth of this lady.*A The story is useful to us for a number of reasons but here it is important to obtain a clearer idea of the property which these ladies had at their disposal.*B The economic structure of the empire changed as it expanded and so too did the nature of women’s involvement in economic activity. In Iran, at least until the fourteenth century, most of their economic activity was concentrated in and around their ordo. Military campaigns to capture booty a)fected women in di)ferent ways. On the one hand, it increased patrimony among the Mongol Khatuns, as they shared in the fruits of conquest. On the other, the expansion of the empire deeply a)fected the lives of women in the conquered countries, since those who were not killed were generally considered part of the loot and distributed as slaves among members of the royal family (both men

-D Rashīd/Raushan & Mūsavī, 1:593; Rashīd/Thackston, 2:272–73. -E There is a famous reference in Juwaynī to a daughter of Chinggis Khan (Tümälün) entering the city of Nishapur after the Mongol conquest; after massacring the majority of the prisoners she took four hundred of them for their craftsmanship skills and brought them to Turkestan. See Juwaynī/Qazwīnī, 1:139–40; Juwaynī/Boyle, 1:177. On the veracity and the confusion generated in the sources about this event see the commentary in Boyle’s translation (174–75, fn. 1); De Nicola, “Women’s Role and Participation in Warfare,” 101–2. -F Her case is also mentioned by Sneath, The Headless State, 175. -I Ibaqa had people under her command (at least two hundred), given to her by her father; she also possessed horses, slaves and cattle, which would have generated revenue. See Rashīd/Raushan & Mūsavī, 1:197; Rashīd/Thackston, 1:104. Jagambo was the father of Ibaqa Beki and the brother of the Ong Khan of the Kereits. He was pardoned when Chinggis Khan conquered his people, but he later rebelled and was destroyed. See G"H, §208. -J For a more detailed analysis of the story see De Nicola, “Unveiling the Khatuns,” 156–57. %O4 &78P8Q67 '894 RS H8PT89 U8Q4P 1[ and women). If the captive women were of noble stock they were married to princes and generals.+V The search for booty was the driving force which kept the Mongol military machine en marche, and the early Mongol thirst for booty continued into the fourteenth century. During the time of Ghazan Khan in Iran, pillage was still important as a source of income among princes and generals.+* However, this direct exaction of resources by military conquest was not the only way in which the Mongols expanded their wealth. Trade is stressed by modern historians as one of the most valuable economic activities favoured by nomadic empires in general and by the Mongols in particular.++ Yet views vary about how this trade was articulated and what its implications were for the Mongols.+, Some scholars have observed that the development of Mongol involvement in trade varied, depending on the region they occupied, but overall the nomadic promotion of trade has generally been regarded as ‘bene<=cial’ for those sedentary societies that fell under the control of the Mongols.+; Trade expanded not only beyond their frontiers but also beyond their political control. Commercial centres were established that connected China, Iran, India, and Europe allowing, by the thirteenth and fourteenth centuries, many foreign traders (such as Italians) to operate across Asia.+> Those in charge of carrying out commercial transactions were the ortaqs or merchants who had close relationships with the royal family. However, the relationship between these commercial agents and the Mongols went through di)ferent stages as the empire expanded and divided and as its economic policies and resources came under more

/X Lambon, Continuity and Change, 289. /- It is mentioned that during their <=rst campaign into Syria in 1299, the Mongols captured the treasury of the Sultan of Egypt and distributed it among the o)<=cials. See Rashīd al-Dīn, Jāmi‘ al-tawārīkh, ed. B. Karīmī, 2:940 [hereafter Rashīd/Karīmī]; Rashīd/ Thackston, 3:646. // Khazanov distinguishes between two di)ferent kinds of trade among pastoral societies: one in which there is direct exchange of goods between nomadic and sedentary people and another in which the nomads act as mediators in exchanges between sedentary societies. See Khazanov, Nomads and the Outside World, 202. /0 Di Cosmo, “State Formation and Periodization,” 11. /C Halperin, “Russia in the Mongol Empire,” 243; see, for example, the study on the promotion of the fur trade in the Golden Horde by Martin, “The Land of Darkness and the Golden Horde,” 401–21; see also Noonan, “Russia’s Eastern Trade, 1150–1350,” 201–64. /D See, for example, the commercial importance of the Persian Gulf in the Mongol period for international trade in Aubin, “Les Princes d’Ormuz du \]]] au \^ siècle,” 77–137. See also Allsen, “Ever Closer Encounters,” 22; Kauz, “Bengali Textiles as Tribute items to Ming China,” 39–55; Aigle, Le Fars sous la domination mongole, 71–77. 1( 34 56789: centralized control.+? Just as women had shared in the riches gained through plunder, so they played a role in trade. The expanding wealth of the new Mongol nobility in the early Empire meant that women occupied a new role in the economy. As Allsen has noted,

The Mongols, heretofore a society with limited purchasing power, now suddenly found themselves with vast and unaccustomed wealth, and the ruling strata, the main bene<=ciaries of the booty and tribute, were prone to an extravagance typical of the nouveau riche.+@

Hence, Mongol women sought expensive goods which stimulated trade. Objects of luxury had been present in the steppe before the empire,+A but con- sumption rapidly increased as more and more resources found their way into the hands of the Khatuns, promoting trade and the movement of goods across Eurasia.+B Demand for such goods seems to have been especially strong in the early period of the empire when massive amounts of money were at the disposal of the royal family. The accumulation of such highly elaborate artefacts seems to have increased in the Khatuns’ ordos as the empire expanded, eventually reaching the levels of opulence described by Ibn Baṭṭūṭa among the ladies of the ulus of Berke in the fourteenth century.,V Despite this, Khatuns were not only interested in luxury artefacts. Chinese sources describe the purchase of KLour by Mongol women. It was necessary to transport it over a distance of up to 1,150 kilometres, but the Khatuns had no problems in paying the bill.,* In order to satisfy increasing demand among Mongol princes and princesses, some kind of structure was needed which would enable trade to KLourish and merchants to bring their products safely to their customers. Accounts of the relationship between the Mongols and the merchants in this early period are not very clear. It seems that both parties actively collaborated, the new empire stimulated and protected trade, and Mongolia became a new market for Inner Asian traders.,+ The presence of

/E On this phenomenon see especially Allsen, “Mongolian Princes and their Merchant Partners,” 83–126. /F Ibid., 93. /I See for example the present given by Chotan (the mother of Börte) to Chinggis Khan’s mother when the marriage between their sons was arranged in G"H, §96. /J Especially textiles and gold brocades were demanded by Mongol women; see Allsen, Commodity and Exchange, 16. 0X Ibn Baṭṭūṭa, The Travels of Ibn Baṭṭūṭa, 485–86. 0- See Allsen, “Mongolian Princes and their Merchant Partners,” 93. 0/ Ibid., 94; Rossabi, “The Muslims in the Early Yüan Dynasty,” 260–61. %O4 &78P8Q67 '894 RS H8PT89 U8Q4P 1` commercial agents from Central Asia before the rise of Chinggis Khan,,, and according to Juwaynī, the Mongol lifestyle and the scarcity of well-established merchants meant that those traders who did reach the camps could expect high pro<=ts. Citing the pro<=ts made by a particular entrepreneur called Aḥmad of Khojend, Juwaynī highlights the fortunes that it was possible to make by bringing “gold embroidered fabrics, cottons, zandanichi and whatever else they thought suitable” to be sold to the Khan and his family.,; Whilst the khans dealt with merchants, references to women doing the same thing in this period are not abundant. However, women’s capital was represented in the commercial expedition sent by Chinggis Khan to the neighbouring Khwarazmshah kingdom in 1218 (this expedition became the catalyst for the Mongol invasion of Central Asia). The Khan gathered the merchants together and “ordered his sons, daughters, wives, and military commanders to select Muslims from their respective retinues and to supply each with gold and silver ingots (balish) so that they might trade in the land of the Khwarazmshah.”,> This suggests that even at the time of Chinggis Khan, women were investing in long-distance trade in accordance with the Mongol policy of engaging in commerce. However, it seems that when Chinggis Khan died in 1227, trade was still in its infancy. Guards were appointed at the borders of Mongol domains to secure the unhindered entry of merchants,,? but this only happened when Chinggis Khan had consolidated his power in the steppe.

Towards a Mercantile Economy: Mongol Women in the United Empire

The accumulation of wealth through the exaction of resources continued after the death of Chinggis Khan, with the Khatuns garnering riches when the

00 G"H, §§182 and 657. For the presence of merchants from Bukhara in the court of Chinggis Khan mentioned in the Yuanshi, see Pelliot, “Une ville musulmane,” 264–68. 0C Juwaynī/Qazwīnī, 1:59; Juwaynī/Boyle, 77–78. Rashīd al-Dīn also mentions Chinggis Khan dealing with Muslim merchants. See Rashīd/Raushan & Mūsavī, 1:472–73; Rashīd/ Thackston, 1:233–34. 0D Allsen, “Mongolian Princes and their Merchant Partners,” 88. He refers to Juwaynī/ Qazwīnī, 1:60; Juwaynī/Boyle, 78, but in this account, only the sons (pesarān), the noyans (nūyanān) and the amirs (āmirān) are mentioned and no speci<=c reference to women is made. The Khatun is mentioned in Rashīd al-Dīn’s version, see Rashīd/Raushan & Mūsavī, 1:473; Rashīd/Thackston, 1:234. 0E Juwaynī/Qazwīnī, 1:59; Juwaynī/Boyle, 77–78. 1b 34 56789:

Mongols conquered Northern China and Russia.,@ Wealth exaction quickly adopted a more ‘imperial form’ whereby not only cattle and soldiers but also skilled artisans were captured and brought to Mongolia to begin the construction of the imperial city of Qaraqorum in the heart of the steppe.,A Juwaynī comments that the building of the new Mongol capital was an enterprise in which:

artisans of every kind were brought from Khitai, and likewise craftsmen from the lands of Islam; and they began to till the ground. And because of Qa’an’s great bounty and muni<=cence people turned their faces hither- ward from every side, and in a short space of time it became a city.,B

Ögetei abandoned his <=ef in eastern Turkestan and moved towards the new city, where resources could be accumulated easily and economic activity centralized.;V In the two decades following the death of the founder of the empire, new ways of extracting and maximising resources developed as the empire grew. For example, Ögetei Khan expanded the network of relay stations created by Chinggis Khan. It was actually the second emperor of the Mongol Empire who made the <=nancial e)fort to transform the famous yam system from a military function into a commercial enterprise.;* The consolidation of this system of provision posts greatly stimulated merchants, who also bene<=ted from Mongol investment in <=xing the roads that connected di)ferent parts of the empire.;+ The period of Ögetei’s reign (r. 1229–41) seems to have been a golden period for trade in the Mongol empire. Infrastructures, bene<=ts and an easy-spending nobility secured pro<=ts for those who ventured into the north-eastern parts of

0F See for example how in 1236 Ögetei granted Sorghaghtani Beki those areas in Northern China conquered by Tolui. See Abramowski, “Die Chinesischen Annalen von Ögödei und Güyük,” 131–32; Abramowski, “Die Chinesischen Annalen des Möngke,” 16. 0I Although the artisans were already being taken to the Mongol court at the time of Chinggis Khan, this practice appears to be more extensive under his successor. For a description of the city of Qaraqorum and the diversity of origin of its inhabitants see Dawson, The Mongol Mission, 177–80. See also Allsen, Commodity and Exchange, 34–35; Allsen, “Biography of a Cultural Broker,” 130. 0J Juwaynī/Qazwīnī, 1:192; Juwaynī/Boyle, 236. CX Barthold, Four Studies on the History of Central Asia, 1: 45. C- On the yam system see Morgan, The Mongols, 91–94; Silverstein, Postal Systems in the Pre- Modern Islamic World, 141–64; Allsen, “Imperial Posts, West, East and North,” 237–76. C/ Allsen, “Mongolian Princes and their Merchant Partners,” 96. %O4 &78P8Q67 '894 RS H8PT89 U8Q4P 1c

Asia.;, However, towards the end of Ögetei’s reign some new measures were implemented in order to further bene<=t from all this commercial activity. Prominent o)<=cials such as Yelü Chucai and Maḥmūd Yalavach made the <=rst attempts to establish a taxation system in order to regulate and tax trade.;; Nonetheless, after the death of Ögetei and during the regency of his wife Töregene Khatun (r. 1241–46), Yalavach was replaced by a certain ʿAbd al-Raḥmān and the stimulus for commercial activities became a priority once again, diminishing the political interest in centralizing measures and control of land taxation carried out before the Khans’ death. As Allsen puts it, “the regent was well disposed toward the merchants, who quickly resumed their former position at court. ʿAbd al-Raḥmān, a personal favourite of Töregene, received an imperial seal that gave him administrative as well as <=scal control over north China.”;> This transfer of attribution from the empress to her subordinates (including a woman like Fāṭima Khatun) and her commitment to a system less keen on central taxation and more open to ‘free trade’ could be interpreted as weakness in Töregene’s government.;? But it seems that this strategy was in accordance with a certain idea of what the Mongol Empire should be, which was opposed to that of the Persian sources of the period. In her defence though, we can add that the system considerably bene<=ted the Khatun’s treasury.;@ But the organisation of commerce and the distribution of pro<=t were not as smooth as the sources suggest. As the empire grew, so too did competition between members of the royal family. At stake was the assignment of merchants, who became increasingly sought after as more and more members of the royal family acquired wealth and became involved in trade. Powerful and wealthy women were able to claim particular merchants (ortaqs) even if they belonged to the Khan.;A Khatuns with less inKLuence at the court had to force “people of the provinces” to give their sons up not only for domes- tic service but also to become ortoy [merchants] in the service of Mongol

C0 For the taxation organized to maintain the yam system see Smith, “Mongol and Nomadic Taxation,” 52. CC Rachewiltz, “Yeh-Lü Ch’u-Ts’ai (1189–1243),” 189–216; Allsen, “Mahmud Yalavach,” 122–27. CD Allsen, “Mongolian Princes and their Merchant Partners,” 103. CE This is the interpretation given by the Persian sources, which are generally more sympathetic to the Toluid branch of the Chinggisid family. On Fāṭima Khatun see De Nicola, “Unveiling the Khatuns,” 73–75. CF Rossabi, “The Muslims in the Early Yüan Dynasty,” 267–68. CI Sorghaghtani speci<=cally asked Ögetei to give her a merchant, but the Khan was opposed to the request. She used the name of her husband to change his mind. See Rashīd/ Raushan & Mūsavī, 1:789; Rashīd al-Dīn, The Successors of Genghis Khan, 168. 11 34 56789: princesses for a small remuneration.;B The involvement of women in trade was not restricted to granting authorizations to merchants, however. Although references to women are not abundant, some of them sent members of their personal ordos on expeditions to seek trade beyond the frontiers of their <=efs. Sorghaghtani Beki (d. 1252), wife of Tolui,>V being wealthy and favoured by the Persian sources, received much attention in matters of the economy. Rashīd al-Dīn recalls a particular occasion when she sent a thousand men in a ship to sail northward along the Angara River into deep Siberia.>* Three commanders of the lady’s ordo were in charge of the expedition which had the objective of reaching “a province near which is a sea of silver.” The group accomplished their mission and obtained so much silver that they were not able to put it all in the ship.>+ There are no other descriptions of women ordering expeditions for raw materials in the sources. However, this is not to say that other Khatuns who had important appanages in this period did not also <=nance exploration, considering the men and resources they had at their disposal. It is interesting to see how the same sources dealing with similar topics can change their perception depending on their political outlook.>, Sorghaghtani was praised for the expedition that found so much silver, but when referring to the regency of Oghul Ghaimish, Rashīd al-Dīn says the latter’s relationship with merchants was the reason she neglected the empire’s governance. The reference indicates the direct involvement of Khatuns in the commercial life of the empire. The regent is described as spending most of her time with shamans (qams), which provoked the remark that in her reign “Little was done, however, except for dealings with merchants,” referring to her role as a consumer but probably also to her role as a promoter of trade.>; Thus it appears that during the reign of Güyük and his wife Oghul Ghaimish, trade KLourished with the Silk Roads con- necting the Far East and Europe via the Russian steppes, Central Asia, Iran and Anatolia. Nevertheless, the hugely pro<=table unregulated trade system

CJ See Rashīd/Karīmī, 2:1045. Allsen also mentions this fact; see Allsen, “Mongolian Princes and their Merchant Partners,” 103. On the term “ortoy” or “ortaq” see Doerfer, Turkische und Mongolische Elemente im Neupersischen, 2:25–27. DX On Sorghaghtani Beki see Rossabi, “Khubilai Khan and the Women in His Family,” 158–62; Lane, Daily Life in the Mongol Empire, 239–43. D- Today, the Angara River is located in Russia, and is the only river KLowing out of Lake Baikal to join the Yensei River and thence into the Kara Sea, which is part of the Arctic Ocean. D/ Rashīd/Raushan & Mūsavī, 1:76–77; Rashīd/Thackston, 1:43. D0 Morgan, “Persian Historians and the Mongols,” 109–24. DC Rashīd/Raushan & Mūsavī, 2:810; Rashīd al-Dīn, The Successors of Genghis Khan, 186. %O4 &78P8Q67 '894 RS H8PT89 U8Q4P 1e supported by Töregene and afterwards by her son and daughter-in-law seems to have clashed with the necessity, by the end of 1240, to have a more centralized economic administration. The relationship between the royal family and the merchant community functioned in this period at a personal level, with presumably lax central control over expenditure and the issuing of drafts based on anticipated revenues.>> The over-issuing of these drafts and the inadequacy of the resources to pay for them may have provoked a cycle of speculation and inKLation leading to what was seen as a chaotic economic period.>? At this point in the historical narrative of the Mongol Empire, Persian sources present the reign of Möngke Khan as a return to economic stability. While prior to the death of Chinggis Khan, economy was marked by a system based on the exaction and distribution of wealth via conquest, the rule of the Ögeteids and the regencies of their women can arguably be seen as a golden age in the expansion of trade across the empire.>@ The consolidation of the yam system led to rapid and lucrative commerce that not only enriched the multinational merchant community of Eurasia but allowed Mongol women to invest in commercial expeditions, establish commercial enterprises with traders and satisfy their personal desire for luxury products, which in itself helped to stimulate the economy. However, this model depended upon the constant exaction of wealth through military means and complete freedom of action for the ortaqs.>A By 1250, the model which had been proposed ten years earlier by Yelü Chucai and discarded by Töregene was recovered. Möngke Khan (r. 1251–59) assumed control with a new set of advisors who were more in tune with the policies developed by his mother Sorghaghtani in her sedentary dominions in Northern China.>B The new Khan implemented a series of measures to better control the imperial treasury, to establish a more settled economy based on taxation, to promote farming and control the merchants.?V This did not mean that trade was out of favour or that business with merchants had to stop. On the contrary, Rashīd al-Dīn says that debts owing to traders were paid by

DD There is evidence in Chinese sources of <=nancial ruin among the peasantry in Northern China due to the speculation of these courtly merchants. Farmers had to sell their property and even their wives and children to pay their debts to the ortaqs. See Endicott-West, “Merchant Associations in Yüan China,” 149. DE Allsen, “Mongolian Princes and their Merchant Partners,” 108–9. DF Ibid., 104–5; Rossabi, “The Muslims in the Early Yüan Dynasty,” 269–70. DI For trading conditions prior to the reign of Möngke Khan, see Allsen, “Mongolian Princes and their Merchant Partners,” 104–5. DJ The most complete account of Möngke’s reign is Allsen, Mongol Imperialism. EX Juwaynī/Qazwīnī, 3;79; Juwaynī/Boyle, 2:600. e2 34 56789:

Möngke and the KLow of entrepreneurs continued in Eurasia.?* The Franciscan missionary Rubruck, who travelled to the court of Möngke in the early 1250s, constantly relied on the advice of merchants of various origins on his trip to the court of the Khan.?+ Yet a new model was implemented and it aimed to tidy up the imperial accounts and produce a more e)<=cient revenue system which would favour the rulers. In 1251, Möngke Khan and his administrators introduced a ‘Mongol taxation system’ for the conquered population, and it was brought by Hülegü into Iran when he settled in the area after the fall of Baghdad in 1258. While women had their share of the booty in the early period and were active in the period of ‘free trade’ during the uni<=ed Mongol empire, the Mongol conquest of Iran would prove a new land of opportunity for the rapidly transforming economic status of Mongol Khatuns.

The Khatuns’ Economic Participation in Ilkhanid Iran: Continuity and Transformation

The economic development of the Ilkhanate has been generally described as evolving in two stages, divided by the reign of Ghazan Khan.?, The economic reforms he and his vizier Rashīd al-Dīn introduced have been regarded as a turning point in the government of Mongol Iran. New economic policies were introduced which focused on land productivity and new modes of taxation, abandoning the Mongol exaction model of the early empire. Despite this, doubts have been raised as to whether these measures were actually implemented and the degree to which the sedentary Persian population really bene<=ted from them.?; Those reforms begun under Möngke Khan and were continued by Hülegü when he conquered Iran, as the <=nancial notes left by

E- Rashīd/Raushan & Mūsavī, 2:861; Rashīd al-Dīn, The Successors of Genghis Khan, 236; Allsen, Mongol Imperialism, 89. However, under Möngke some restrictions were introduced in the use of the yam system by merchants, ibid., 80. E/ These merchants were mostly Christian; the clerics were more trusting of coreligion- ists—Byzantines or Ruthenians, for example. See Allsen, “Mongolian Princes and their Merchant Partners,” 108–9. E0 For a description of Ghazan’s reforms see Lambton, Landlord and Peasant, 83–92; Petrushevsky, “Rashid al-Din’s Conception of the State,” 148–62. EC While Rashīd al-Dīn tends to exaggerate the bene<=ts of Ghazan’s reforms, Waṣṣaf and Mustawfī for example are more cautious in their description of the bene<=ts of the new economic policies. See Petrushevsky, “The Socio-Economic Condition of Iran,” 494–500; Lambton, Landlord and Peasant, 92. %O4 &78P8Q67 '894 RS H8PT89 U8Q4P eN his advisor Naṣīr al-Dīn Ṭūsī suggest.?> However, the conquest of Iran and the further campaigns carried out by the Mongols in the Middle East under the Ilkhans brought with them the opportunity for more booty, and women were once again among the bene<=ciaries. Hülegü’s conquest produced enormous booty. Great amounts of gold, silver, horses and slaves were taken, for example, during the fall of Baghdad in 1258, and further tributes were received from the Caliph when he capitulated.?? Furthermore, the continuation of the military campaigns in the Mamluk territories in Syria and Palestine in 1260 yielded substantial riches, some of which were distributed among Hülegü’s family members and some sent back to the court of Möngke in Mongolia.?@ There is evidence that at least part of this booty accumulated during Hülegü’s conquests was set aside for the women’s personal treasuries in their ordos. An example of this phenomenon happened when the <=rst Ilkhan died in 1265. His successor Abaqa (d. 1282) informed his widows of his passing. Among them was one of his principal wives (Qutui Khatun), who was on her way from Mongolia to Iran to meet her husband. The news surprised the Khatun in the area of Badakhshan in modern day southern Tajikistan and north-eastern Afghanistan. When she and her party arrived they were received by the new Ilkhan. The interesting part of this story is that Rashīd al-Dīn not only says that Abaqa “enriched with money and goods” the wife of his father, but adds that a concubine called Arighan who had attended Hülegü also belonged to Qutui Khatun now.?A Her actual role in the court of the Ilkhan is made clear when Rashīd al-Dīn mentions:

Qutui Khatun’s share of booty and plunder had been turned over to her [a woman called Arighan]. She had accumulated vast amounts of valuable items and property, so when Qutui Khatun arrived in the ordu she found it well stocked with all sorts of things.?B

First, this story shows that women still had a share in the revenues produced by conquest during the invasion of Iran and after the establishment of the

ED For a transcription and translation of this document see Minovi and Minorsky, “Naṣīr al-Dīn Ṭūsī on Finance,” 755–89. EE Anonymous, Histoire de la Géorgie, 519–20; on the fall of Baghdad and the appropriation of the Caliph’s treasury by the Mongols see Ibn al-Fuwaṭī, al-Ḥawādith al-jāmiʻa, 325–28. EF Amitai, “Mongol Raids into Palestine,” 236–55. EI Rashīd/Raushan & Mūsavī, 2:967–68, 1051–52; Rashīd/Thackston, 2:474; 3:514. EJ Rashīd/Raushan & Mūsavī, 2:1064–65; Rashīd/Thackston, 3:520. eY 34 56789:

Ilkhanate. Secondly, it underlines once again the di)ference between the status of wife and concubine among the Khatuns, since the properties belonged to the chief consort of the khan while the concubine played a role in accumulat- ing and administrating the Khatun’s revenues if she was absent. Gifts were important in maintaining alliances between di)ferent members of the royal family and this can be documented from the very beginning of the Mongol Empire. Empresses in the 1240s distributed gifts among their political allies, whether they were members of the royal family or the local nobility.@V Other women were among the bene<=ciaries of this wealth distribution. Whenever a new Ilkhan ascended the throne the ladies of the court would receive gifts. Hülegü gave gifts to his sisters, sons and generals immediately after appointing a man to oversee the treasury in Iran.@* His two immediate successors Abaqa and Tegüder also made gifts to the Khatuns when they took the throne.@+ Similarly Geikhatu and especially Ghazan Khan, whose generosity was remarked upon by Rashīd al-Dīn, gave money to the women on several occasions.@, It is not easy to be precise about the quantity of riches transferred from the treasury to the ladies’ ordos. Such gifts generally included money and luxury goods such as goblets, jewels and especially expensive textiles.@; In their turn the women bestowed gifts of money on local nobles and religious leaders, contributing in this way to a further distribution of wealth in ways similar to the male members of the royal family.@> These personal gifts and presents need to be distinguished from the other sources of income which noble women had in Iran. In the Ilkhanate there was a larger sedentary population than among uluses such as the Golden Horde or the Chaghatai Khanate. The taxation of Mongol Iran is confusing and has

FX See for example Juwaynī/Qazwīnī, 3:7; Juwaynī/Boyle, 551; Dunlop, “The Kereits of Eastern Asia,” 285. F- Ward, “Ẓafarnāmah of Mustawfī,” 17. F/ Rashīd/Raushan & Mūsavī, 2:1060, 1126; Rashīd/Thackston, 3:517, 549. F0 Rashīd/Raushan & Mūsavī, 2:1195, 1331; Rashīd/Thackston, 3:582, 660. The local administrators and the Ṣāḥib Dīwān also o)fered gifts to the ladies; see Rashīd/Raushan & Mūsavī, 2:1323–24; Rashīd/Thackston, 3:661. FC Allsen, Commodity and Exchange, 56. See also the allocation of money given to Ābesh Khatun by Hülegü in Waṣṣāf, Taḥr)̄r-i tār)̄kh-i Waṣṣāf, 113. FD See for example the money given by Konchak Khatun, wife of Amir Irinjīn, to the Nestorian church in the early fourteenth century. See Budge, The Monks of Kublai Khan, 304–5. Terken Qutlugh Khatun provided gifts to inKLuential personalities such as Arghun, Anonymous, Tārīkh-i Shāhī-i Qarākhitāi᾿yān, 183. %O4 &78P8Q67 '894 RS H8PT89 U8Q4P e[ generated a considerable amount of literature.@? Generally speaking, the Ilkhanate functioned with a dual administration system which maintained the existing Islamic-Persian system and incorporated mainly three new <=scal measures of Mongol origin. The di)ference in the Mongol taxation system appears to rest on the fact that it was irregular in its timetable and was based on the census to determine the amounts to be paid. In contrast, the Islamic- Persian system was based on land productivity. This di)ference between the two systems allowed for their coexistence but doubled the <=nancial pressure on the conquered population.@@ Three new taxes were introduced by the Mongols with the clear purpose of generating income for the royal family.@A The taxation system helped the Ilkhanate to accumulate resources that were distributed among the inKLuen- tial Mongol women of Iran. However, noble women participated more directly in the Ilkhanate economy through “estate taxes,” which were divided between those lands under the supervision of the dīwān and those which were the direct property of the royal family.@B These included land con<=scated by the Mongols from the conquered Persian nobility, with its revenues being used in the maintenance of the khan, the khatuns, the o)fspring of the royal family and members of the ordos.AV Consequently, when references are made in the sources to Mongol women in Iran being allocated land to hold in usufruct we should not understand that they governed such land, but that they enjoyed its productivity.A* The practice of allotting land to women took place from the very beginning. Abaqa distributed the resources of sedentary populations among the Khatuns. He gave:

FE On the taxation system of the Mongols see among many others Lambton, Continuity and Change, 79–99; Smith, “Mongol and Nomadic Taxation,” 46–85; Schurmann, Economic Structure of the Yuan Dynasty, 304–89; Petrushevsky, “The Socio-Economic Condition of Iran,” 483–537; Morgan, The Mongols, 87–90; Ostrowski, “The “Tamma,” 262–77. FF Lambton, Continuity and Change, 84; Aigle, “Iran under Mongol Domination,” 65–78. FI These taxes are generally known as qubchur, qalān and tamghā. See Lambton, Landlord and Peasant, 77–99; Lambton, Continuity and Change, 199–220; Petrushevsky, “The Socio- Economic Condition of Iran,” 533; Morgan, The Mongols, 88; Allsen, “The Yüan Dynasty and the Uighurs of Turfan,” 263; Smith, “Mongol and Nomadic Taxation,” 46–85. FJ Mostly those known as ḵāṣṣa and injü, see Paul, “Fiscal System ]]].” IX The revenues produced from the implementation of these taxes varied and are di)<=cult to assess; for an interpretation of the numbers see Petrushevsky, “The Socio-Economic Condition of Iran,” 499. I- For the allocation of lands and shares across the Mongol Empire see Allsen, “Sharing out the Empire,” 172–90. e( 34 56789:

a portion of Mayyafariqin [in Syria] to Qutui Khatun, part of Diyarbekir and the province of Jazīra [Iraq] to Oljäi Khatun, Salmas [North-western Iran] to Jumghur’s wife Tolun Khatun and his sons Jüshkab and Kingshü. He also gave some territories to his sons and concubines.A+

Nonetheless, from the end of Abaqa’s reign, the allocation of estate taxes for Khatuns’ ordos was replaced by a system whereby regions of the Ilkhanate had to pay a tax to contribute to the ordo’s wealth, and the women’s servants were charged with collecting these revenues from the assigned territories.A, According to Rashīd al-Dīn, funds were squandered and corruption among those servants and the governors of the provinces soared, leading to an ever- increasing loss of revenue that culminated in the <=nancial chaos of Geikhatu’s reign (1291–95).A; This situation prepared the way for the Persian vizier to justify the reforms of his patron Ghazan Khan when he assumed control of the realm. It is said that trials were held to punish corrupt servants and provincial deputies, whilst the administration of the ladies’ ordos was reformed and their autonomy restricted.A> Despite the corruption and the possible impov- erishment of the Khatuns’ ordos during the period of <=nancial chaos, women retained control over property—attested to by the fact that Ghazan Khan gained the support of many women’s injü to <=nance his claim to the throne in 1294.A? Further, women having land revenues under their command seems to have persisted up to the reign of Abū Saʿīd (1316–35).A@ Apart from the novelty of land-taxes allocation, in the Ilkhanate trade continued to play a pivotal role in the economy as it did in the period of the united empire.AA Opening trade routes had been a clear policy from the time

I/ Rashīd/Raushan & Mūsavī, 2:1110; Rashīd/Thackston, 3:541. On Qutui Khatun see also Rashīd/Raushan & Mūsavī, 2:1064–65; Rashīd/Thackston, 3:520. I0 Rashīd/Raushan & Mūsavī, 2:1507; Rashīd/Thackston, 3:745. IC There is as yet no in-depth analysis of Geikhatu’s reign available, and so far, Boyle’s description of the political history of his reign remains the best account of his years in o)<=ce. See Boyle, “Dynastic and Political History of the Il-Khans,” 372–79. ID On this see De Nicola, “Unveiling the Khatuns,” 173–85; speci<=c reference to the appropriation of ordos can be seen in Rashīd/Raushan & Mūsavī, 2:1509; Rashīd/Thackston, 3:745. IE Aubin, Émirs mongols et vizirs persans, 58. According to Mustawfī, Ghazan’s reforms maintained the economic wellbeing of the Khatuns by rearranging the injü system. See Ward, “Ẓafarnāmah of Mustawfī,” 429. IF Ibn Baṭṭūṭa, The Travels of Ibn Baṭṭūṭa, 340. II See the reference to the transit of merchants from Iran to China and vice versa in Bar Hebraeus, The Chronography, 456; Rashīd/Raushan & Mūsavī, 2:1339–40; Rashīd/ %O4 &78P8Q67 '894 RS H8PT89 U8Q4P e` of Möngke, and this remained so under Hülegü.AB The immense booty gained by the conquest of Alamūt and the sacking of Baghdad might have acted like a magnet, drawing merchants to the Ilkhanate.BV This interest in keeping commerce KLowing in Iran is seen from the beginning of the Mongol conquest of Iran. When Hülegü occupied Baghdad after defeating the last Abbasid Caliph, he sent two of his commanders to begin the reconstruction of the city. The Ilkhan commanded that once the dead had been buried the markets of the city were to be restored as quickly as possible.B* There are other examples of this determination to restore trade after military conKLict. During Ghazan Khan’s campaign to repel the Golden Horde incursion into Azerbaijan, the area was “obliterated” according to Rashīd al-Dīn, but merchants returned to the region of Tabriz to continue trading.B+ In the <=rst period of the Ilkhanate, Mongol women in Iran appear to have managed to maintain their lifestyle with the revenues produced in their ordos. With booty, land and gifts pouring in from the khan and the amirs, the Khatuns seem to have been well provided for. In this period of plenty, Rashīd al-Dīn notes, merchants were a common feature in female camps, where they moved goods to bene<=t the Khatuns and were depositaries of the ladies’ revenues.B, Marco Polo observed that Mongol women were constantly involved in trading activities, selling and buying all that they and their dependants needed.B; These women, who belonged to the highest strata of the Ilkhanid nobility, shared their

Thackston, 3:668. On long-distance trade in this period see Lopez, “Les méthodes commerciales des marchands occidentaux,” 343–48; Delumeau and Richard, “Sociétés et compagnies de commerce en Orient,” 823–43; for the fourteenth century see Jackson, The Delhi Sultanate, 252–53; Di Cosmo, “State Formation and Periodization,” 1–40; Allsen, Culture and Conquest, 41–50; Yasuhiro, “Horses in the East-West Trade,” 87–98; in a later period, textiles also circulated between China and India; see Kauz, “Bengali Textiles,” 39–55. IJ Mustawfī suggests that one of the reasons why Möngke sent Hülegü against the Isma’ilis was because they were cutting the trading routes between Persia and Mongolia. See Ward, “Ẓafarnāmah of Mustawfī,” 12–13. JX Rashīd/Raushan & Mūsavī, 2:1349; Rashīd/Thackston, 3:672. J- Rashīd/Raushan & Mūsavī, 2:1069; Rashīd/Thackston, 2:499. J/ Rashīd/Raushan & Mūsavī, 2:1303; Rashīd/Thackston, 3:651. J0 From Rashīd al-Dīn we can interpret that gifts and cattle were included among the resources that women invested in trade through the mediation of merchants; Rashīd/ Raushan & Mūsavī, 2:1507; Rashīd/Thackston, 3:745. JC Marco Polo, The Description of the World, 1:, 169. eb 34 56789:

connexion with the great commercial companies and with big wholesale and transit trade. They invested a part of their income in the companies of the great wholesale merchants, called usually urtaq (ortaq) [. . .] who returned the feudal lords [or ladies] their share of the pro<=t in goods, mostly textiles.B>

However, the growing dependence of the nobility upon the merchant community led to corruption and <=nancial speculation, and the manipulation of the currency to the bene<=t of the merchants themselves, which led inevitably to the draining of treasuries and economic instability.B? One of the areas where trade became especially pro<=table and women very active in the Ilkhanate was the Persian Gulf and the provinces of Southern Iran that connected the Arabian Peninsula, the Levant and India.B@ Ruled up to the early fourteenth century by di)ferent women of the Salghurid and Qutlughkhanid dynasties, these areas were autonomous entities under Mongol rule with special control over trade revenues in the Gulf.BA Ābesh Khatun (d. 1286), the ruler of Fars, or later Bībī Maryam (d. early fourteenth century) are some examples of women being mentioned in the sources as being implicated in the lucrative trade of the region.BB Thus, when Ghazan Khan ascended to the throne of the Ilkhanate, female economic activity in the Ilkhanate had already passed through several phases.

JD Petrushevsky, “The Socio-Economic Condition of Iran,” 509–10. Among these nobles was also Rashīd al-Dīn himself, who owned an impressive fortune and used to invest in trade as did his Persian and Mongol counterparts. See Rashīd al-Dīn, Kitāb-i Mukātibāt-i Rash)̄d)̄, 220–40 (letter 36) or 183–207 (letter 34). Although there has been debate about the authenticity of these letters, it seems to me that either Rashīd al-Dīn traded with the goods mentioned in the book, or those who produced the letters in the Timurid period considered the participation of these personalities in trading activities common practice. On the debate see Morton, “The Letters of Rashīd al-Dīn,” 155–99; Soudavar, “In Defence of Rashid,” 77–122. JE See Martinez, “Regional Mint Outputs,” 147–73; Allsen, “Ever Closer Encounters,” 21. JF On the trade in the region of Hormuz and the Persian Gulf in this period see Aubin, “Les Princes d’Ormuz du \666 au \n siècle,” 77–137; Yasuhiro, “Horses in the East-West Trade,” 87–98; for the <=fteenth and sixteenth century see Kauz, “Trade and Commerce on the Silk Road after the End of Mongol Rule in China, Seen from Chinese Texts,” 54–59. JI Wa ṣṣāf, Taḥr)̄r-i tār)̄kh-i Vaṣṣāf, 100. JJ On Bībī Maryam see Spuler’s entry “Ābeš Ḵātūn” in Encyclopaedia Iranica; Rashīd/ Raushan & Mūsavī, 2:936–37; Rashīd al-Dīn, The Successors of Genghis Khan, 307; Shabānkāra’ī, Majma‛ al-ansāb, 215–16. It is not clear if this is the same Bībī Maryam who had a mausoleum built in the Omani city of Qalhat, see Ibn Baṭṭūṭa, The Travels of Ibn Baṭṭūṭa, 396. %O4 &78P8Q67 '894 RS H8PT89 U8Q4P ec

Initially, women continued to bene<=t from the distribution of wealth carried out by the rulers as gifts and presents coming from the conquest of Iran and west Asia. The sedentary nature of Iran and the implementation of Möngke’s new approach to the land gradually made way for women to be allocated a share in the dual taxation system. Finally, the constant interest of the Mongol nobility in trade and commercial exchange allowed women to invest at least part of their income with the merchants of their camps, in order to <=nance the increasing demands of their lifestyle. Speculation and corruption appear to have been widespread due to the uncontrolled KLood of money into these activities. The economic situation of the Ilkhanate in the early 1290s propitiated the attempt for an economic transformation under Ghazan that would have a direct e)fect on women’s autonomy, marking a clear distinction in the attitude of the rulers towards queenly ordos before and after his reign.*VV In the period that began with the struggle for power between Baidu (d. 1295) and Ghazan Khan (r. 1295–1304) and the subsequent conversion of the latter to Islam, noblewomen in the Mongol court maintained certain privileges and their prominent role in society. For example, their involvement in religion, particularly with Islam, became more notorious.*V* Similarly, as mentioned above, some women continued their economic participation in trade and to enjoy revenues coming from taxation into the fourteenth century. However, the rise of Ghazan Khan to power initiated a centralizing process that a)fected the autonomy and economic independence of women by changing the way the ordos, which represented the core of the ladies’wealth, were controlled.*V+ This essay is not the place to delve deeper into the particularities of this appropriation process. However, it is worth mentioning an example of how this process was articulated. Rashīd al-Dīn claims in his Jāmiʿ al-tawārīkh a supposed inheritance right held by Ghazan over the ordo of the powerful Būlūghān Khatun Buzurg.*V, To what extent the claim was legitimate or whether it was all part of Ghazan’s strategy, supported by the historical narrative of his vizier Rashīd al-Dīn, is di)<=cult to tell.*V; Yet, it seems clear that Ghazan bene<=ted greatly from the ladies’ revenues when he had to confront his cousin Baidu for the throne of the Ilkhanate in 1295.

-XX On this process of appropriation see De Nicola, “Unveiling the Khatuns,” 172–84. -X- Se De Nicola, “Patrons or murids?” -X/ For a detailed description of this phenomenon see De Nicola, “Unveiling the Khatuns,” 172–84. -X0 Rashīd/Raushan & Mūsavī, 2:1213; Rashīd/Thackston, 3:593. -XC De Nicola, “Unveiling the Khatuns,” 137–8. e1 34 56789:

The use of female ordos’ resources to <=nance and support his claims to the throne might have helped Ghazan to realize, if he did not know it already, just how extensive were the resources accumulated by Khatuns in previous decades. He and his administrators saw quickly how useful these resources could be in overcoming the economic crisis the new khan found in the realm and in undertaking some of the centralizing measures in the Ilkhanate administration that were already underway. Accompanied by a narrative of chaos and corruption, Rashīd al-Dīn mentions that the newly appointed Ilkhan issued a decree whereby the revenues and maintenance of queenly ordos would be from that moment onwards under the jurisdiction of the Dīwān.*V> Without questioning the veracity of the accusations of corruption among the ladies’ servants and administrators which certainly existed, this order also removed the autonomy of the ladies over their resources by placing these resources under the control of a royal administrator. This meant that the revenues of the female ordos obtained from the mechanisms explained above were to be dedicated, under the supervision of the dīwān, only to the maintenance of the supply of horses and camels, cloth and food, and if there was extra money available from the revenues, it was to be locked in the Khatun’s treasury to be used only in case of emergency.*V? In addition to this control of the revenues, the system of inheritance of female ordos were also modi<=ed. A tradition of maintaining female ordos among women was changed by Ghazan, who decreed that a lady must endow her male o)fspring with the properties of her ordo after her death instead of her female descendants.*V@ There have always been justi<=ed doubts about the real implementation of Rashid al-Dīn and Ghazan’s reforms to the economy of the Ilkhanate.*VA Yet it appears that although some ladies kept their ordos and autonomy into the fourteenth century, the intervention of the state in the properties of the majority of female members of the royal family increased and was present even into the reign of Abū Saʿīd (r. 1317–35).*VB Women continued to be politically inKLuential in fourteenth-century Iran as the cases of Baghdad Khatun and others demonstrate. However, their economic autonomy and wealth had grown so much by the end of the thirteenth century that Ghazan and his advisers targeted their rich encampments as sources for resources that

-XD Rashīd/Raushan & Mūsavī, 2:1508; Rashīd/Thackston, 3:745. -XE Rashīd/Raushan & Mūsavī, 2:1508; Rashīd/Thackston, 3:745. -XF Rashīd/Raushan & Mūsavī, 2:1509; Rashīd/Thackston, 3:745. -XI Amitai, “Turko-Mongolian Nomads,” 153; Morgan, Medieval Persia, 75–76; Lambton, Continuity and Change, 115–29; Miller, “Local History,” 78. -XJ De Nicola, “Unveiling the Khatuns,” 189; Mustawfī, The Ta’r)́kh-i-Guz)́da, 148–49. %O4 &78P8Q67 '894 RS H8PT89 U8Q4P ee could help the depleted Ilkhanid treasury. The exact degree to which resources were drained from the ladies into the royal treasury is not possible to estimate, but the direct intervention on the women’s properties and the limitations imposed on the expansion of queenly ordos mark a clear transformation from the late thirteenth century onwards in the economic activity and status of the Khatuns of Iran.

Conclusion

From late twelfth-century Mongolia all the way to fourteenth-century Iran, noble Mongol women received a share of the resources that the hordes of Chinggis Khan and his successor were taking from conquest across Asia. Initially, the nomadic tradition of taking into account the female members of the royal family in the distribution of booty exacted by military conquest was maintained as the empire began to expand. The resultant accumulation of wealth in the camps stimulated the KLow of merchants into the Mongol territories who were attracted by the high prices which these nomadic “nouveau riche” were prepared to pay for goods. In this process, women participated both as consumers of luxury products and as investors in the commercial enterprises of Eurasian and European traders. However, this system carried within itself speculation and instability when the KLow of resources from conquest diminished or stopped. The reign of Möngke Khan seems to have marked the implementation (or the intention) of reform in the economic model of the Mongols with the newly conquered territories of Iran being an interesting laboratory for these new measures. Evidence suggests that the process of distribution of wealth among Mongol women continued in the early years of the Ilkhanate but added into their attribution other sources of income apart from the direct exaction of booty and the bene<=ts from trade. Three speci<=c areas of revenue can be identi<=ed in which women had a share in Ilkhanid Iran. Firstly, as in the united empire, there were the customary presents received by the Khatun from the royal camp every time a new ruler took control of the realm. This was a two- way distribution system: Khatuns received gifts, but they also gave them to members of the royal family and important members of the Persian aristocracy. Secondly, settlement in Iran gave the Mongols control over the resources of sedentary populations, with urban districts and cultivated lands being a source of income for both royal female and male camps. Finally, when a new system of taxation was implemented in Iran, women continued to bene<=t as recipients of certain taxes imposed on lands allocated to them. N22 34 56789:

The economic status of Mongol Khatuns can be seen as a journey of transformation in which noblewomen increased their wealth by incorporating into their original economic attribution of “booty-share” di)ferent sources of revenue as the empire expanded and settled. The Khatuns’ involvement in the economic activity of the empire (be it booty, trade or taxation) was comple- mented by their capacity—grounded in the ordos which they commanded— to assume a degree of <=nancial autonomy. In the territories of the Ilkhanate, prior to the ascension of Ghazan Khan, Mongol women appear to have maintained their income from booty and trade but added to their resources the revenues obtained from the productivity of the lands allocated to them for taxation. The political, economic and social circumstances of Iran from the reign of Ghazan Khan onwards triggered a set of reforms by the Mongol Ilkhan under the supervision of Persian dignitaries that particularly a)fected female property. This new process is beyond the scope of this article, but serves as a new reminder of the constant transformation that Mongol women experienced in economic terms in their passage from their original homeland in the Mongolian steppe into the Middle East.

Bibliography

Abramowski, W. “Die Chinesischen Annalen von Ögödei und Güyük: Übersetzung Des 2. Kapitels des Yüan-Shih.” Zentralasiatische Studien 10 (1976): 117–67. ———. “Die Chinesischen Annalen des Möngke: Übersetzung des 3. Kapitels des Yüan-Shih.” Zentralasiatische Studien 13 (1979): 7–71. Aigle, D. Le Fars sous la domination mongole: politique et 01scalité, 23334–2354 s. Paris: Studia Iranica Cahier 31. Paris : Association pour l’avancement des études iraniennes, 2005. ———. “The Mongol Invasions of Bilād al-Shām by Ghazan Khān and Ibn Taymīyah’s Three “Anti-Mongol” Fatwas.” Mamluk Studies Review 11:2 (2007): 89–120. ———. “Iran under Mongol Domination: The E)fectiveness and Failings of a Dual Administrative System.” Bulletin d’Ėtudes Orientales 57:2 (2008): 65–78. Allsen, Thomas. T. “The Yüan Dynasty and the Uighurs of Turfan in the 13th Century.” In China among Equals: The Middle Kingdom and Its Neighbors, 10th–14th Centuries, edited by M. Rossabi, 243–80. Berkeley, !#.: University of California Press, 1983. ———. “The Princes of the Left Hand: An Introduction to the History of the Ulus of Orda in the Thirteenth and Early Fourteenth Centuries.” Archivum Eurasiae Medii Aevi 5 (1985–87): 5–40. ———. Mongol Imperialism: The Policies of the Grand Qan Möngke in China, Russia, and the Islamic Lands. 1251–1259. Berkeley, !#.: University of California Press, 1987. %O4 &78P8Q67 '894 RS H8PT89 U8Q4P N2N

———. “Mongolian Princes and their Merchant Partners, 1200–1260.” Asia Major 2:2 (1989): 83–126. ———. “Mahmud Yalavach.” In In the Service of the Khan: Eminent Personalities of the Early Mongol-Yüan Period (1200–1300), edited by Igor de Rachewiltz, et al., 122–27. Wiesbaden: Harrassowitz, 1993. ———. “Biography of a Cultural Broker: Bolad Ch’eng-Hsiang in China and Iran.” In The Court of the Ilkhans 1290–1340, edited by J. Raby and T. Fitzherbert, 7–22. Oxford: Oxford University Press, 1996. ———. Commodity and Exchange in the Mongol Empire. A Cultural History of Islamic Textiles. Cambridge: Cambridge University Press, 1997. ———. “Ever Closer Encounters: The Appropriation of Culture and the Apportionment of Peoples in the Mongol Empire.” Journal of Early Modern History 1 (1997): 2–23. ———. Culture and Conquest in Mongol Eurasia. Cambridge: Cambridge University Press, 2001. ———. “Sharing out the Empire: Apportioned Lands under the Mongols.” In Nomads in the Sedentary World, edited by A. M. Khazanov and A. Wink, 172–90. Richmond: Curzon, 2001. ———. “Imperial Posts, West, East and North: A Review Article: Adam J. Silverstein, Postal Systems in the Pre-Modern Islamic World.” Archivum Eurasiae Medii Aevi 17:1 (2011): 237–76. Amitai, R. “Mongol Raids into Palestine (#3 1260 and 1300).” The Journal of the Royal Asiatic Society of Great Britain and Ireland 2 (1987): 236–55. ———. “Turko-Mongolian Nomads and the Iqṭā System in the Islamic Middle East (Ca. 1000–1400 !&).” In Nomads in the Sedentary World, edited by A. M. Khazanov and A. Wink, 152–71. Richmond: Curzon, 2001. Anonymous. Histoire de la Géorgie, depuis l’Antiquité jusqu’au 2324 siècle. Translated by M. Brosset. St. Petersburg: Imprimerie de l’Académie Impériale des Sciences, 1849–58. Anonymous. Tārīkh-i Shāhī-yi Qarākhitāi᾿yān. Edited by M. I. Bāstānī Pārizī. Tehran: Intishārāt-i Bunyād-i Farhang-i Īrān, 1976–77. Anonymous. The Secret History of the Mongols. A Mongolian Epic Chronicle of the Thirteenth Century. Translated by I. de Rachewiltz, 2 vols. Leiden and Boston: Brill, 2004. Aubin, J. “Les Princes d’Ormuz du \]]] au \^ siècle.” Journal Asiatique 241 (1953): 77–137. ———. Émirs mongols et vizirs persans dans les remous de l’acculturation. Studia Iranica Cahier 15. Paris: , 1995. Bar Hebraeus. The Chronography of Gregory Abû’l-Faraj 1225–1286, the Son of Aaron, the Hebrew Physician Commonly Known as Bar Hebraeus. Translated by E. A. Wallis Budge, 2 vols. Amsterdam: #$#, 1976. Reprint, New Jersey, 2003. N2Y 34 56789:

Barthold, V. Four Studies on the History of Central Asia. Translated by V. Minorsky, 3 vols. Leiden: Brill, 1956–62. Boyle, J. A. “Dynastic and Political History of the Il-Khans.” In The Cambridge History of Iran, vol. 5, edited by J. A. Boyle, 303–421. Cambridge: Cambridge University Press, 1968. Budge, W., trans. The Monks of Kublai Khan, Emperor of China. London: The Religious Tract Society, 1928. Dawson, C. The Mongol Mission. Narratives and Letters of the Franciscan Missionaries in Mongolia and China in the Thirteenth and Fourteenth Centuries. London and New York: Sheed and Ward, 1955. Delumeau, J. and J. Richard. “Sociétés et compagnies de commerce en Orient et dans l’Océan Indien.” Annales 23:4 (1968): 823–43. De Nicola, B. “Women’s Role and Participation in Warfare in the Mongol Empire.” In Soldatinnen. Gewalt und Geschlecht im Krieg vom Mittelalter bis Heute, edited by Klaus Latzel, Silke Satjukow and Franka Maubach, 95–112. Paderborn: Schöningh, 2010. ———. “Unveiling the Khatuns: Some Aspects of the Role of Women in the Mongol Empire.” PhD diss., University of Cambridge, 2011. ———. “Ruling from Tents: the Existence and Structure of Women’s ordos in Ilkhanid Iran.” In Ferdowsi, The Mongols and Iranian History: Art, Literature and Culture from Early Islam to Qajar Persia. Studies in Honour of Charles Melville, edited by Robert Hillenbrand, Andrew C. S. Peacock and Firuza Abdullaeva, 126–36. London: I. B. Tauris, 2013. . ———. “Patrons or murids? Mongol Women and Shaykhs in Ilkhanid Iran and Anatolia.” Iran: Journal of British Institute of Persian Studies 52 (2014): 143–56. Di Cosmo, N. “State Formation and Periodization in Inner Asian History.” Journal of World History 10:1 (1999): 1–40. Doerfer, G. Turkische und Mongolische Elemente im Neupersischen: Unter Besonderer Berücksichtigung älterer Neupersischer Geschichtsquellen, vor Allem der Mongolen- und Timuridenzeit. 3 vols. Wiesbaden: Harrassowitz, 1963–67. Dunlop, D. M. “The Keraits of Eastern Asia.” Bulletin of the School of Oriental and African Studies 11:2 (1944): 276–89. Dunnel, R. “The Hsi Hsia.” In The Cambridge History of China. Alien Regimes and Border States, 907–1368, edited by D. Twitchett and Herbert Franke, 154–214. Cambridge: Cambridge University Press, 1994. Endicott-West, Elizabeth. “Merchant Associations in Yüan China: The ‘Ortoy’.” Asia Major 2:2 (1989): 127–54. Halperin, C. J. “Russia in the Mongol Empire in Comparative Perspective.” Harvard Journal of Asiatic Studies 43 (1983): 239–61. %O4 &78P8Q67 '894 RS H8PT89 U8Q4P N2[

Ibn Baṭṭūṭa. The Travels of Ibn Baṭṭūṭa, 78 1325–1354. Edited and translated by C. Defrémery, B. R. Sanguinetti and H. A. R. Gibb. 4 vols. Cambridge: Cambridge University Press, 1962. Ibn al-Fuwaṭī, ʻAbd al-Razzāq ibn Aḥmad. al-Ḥawādith al-jāmiʻa. Edited by M. Jawād. Baghdad: al-Maktabah al-ʻArabīyah, 1932. Jackson, P. The Delhi Sultanate: a Political and Military History. Cambridge, Cambridge University Press, 1999. Juwaynī, ʿ Aṭā Malik. Ta’rīkh-i Jahāngushā. Edited by Mīrzā Muḥammad Qazwīnī. 3 vols. Leiden and London: Brill and Luzac, 1912, 1916, 1937. ———. The History of the World-Conqueror. Translated by John Andrew Boyle. 2 vols. Manchester: Manchester University Press, 1958. Reprinted with a foreword by D. O. Morgan. 1997. Kauz, R. “Bengali Textiles as Tribute items to Ming China.” In Textiles from India, edited by R. Crill, 39–55. Calcutta: Seagull Books, 2006. ———. “Trade and Commerce on the Silk Road after the End of Mongol Rule in China, Seen from Chinese Texts.” The Silk Road 4: 2 (2006–7): 54–59. Khazanov, A. M. Nomads and the Outside World. Translated by J. Crookenden. Madison, U]: University of Wisconsin Press. 1994. Lambton, Ann K. S. Landlord and Peasant in Persia; a Study of Land Tenure and Land Revenue Administration. London: Oxford University Press, 1953. ———. Continuity and Change in Medieval Persia. Aspects of Administrative, Economic and Social History. Albany, 5s: Bibliotheca Persica, 1988. Lane, G. Daily Life in the Mongol Empire. Westport, !%.: Greenwood Press, 2006. Lopez, R. S. “Les méthodes commerciales des marchands occidentaux en Asie du \]t au \]^t siècle.” In Colloque International d’Histoire Maritime, edited by M. Mollat, 343–51. Paris: G.&.^.$.&.5., 1970. Marco Polo. The Description of the World. Translated by A. C. Moule and P. Pelliot, 2 vols. London: Routledge, 1938. Martin, J. “The Land of Darkness and the Golden Horde: The Fur Trade under the Mongols \]]]–\]^th Centuries.” Cahiers du monde russe et soviétique 19:4 (Oct.– Dec., 1978): 401–21. Martinez, A. P. “Regional Mint Outputs and the Dynamics of Bullion Flows through the Ilkhanate.” Journal of Turkish Studies 8 (1984): 147–73. May, T. “Commerical Queens: Mongolian Khatuns and the Silk Road.” Forthcoming. Miller, I. “Local History in Ninth/Fifteenth Century Yazd: The “Tārīkh-i Jadīd-i Yazd.” Iran 27 (1989): 75–79. Minovi, M., and V. Minorsky. “Naṣīr al-Dīn Ṭūsī on Finance.” Bulletin of the School of Oriental and African Studies 10:3 (1940): 755–89. Morgan, D. O. “Persian Historians and the Mongols.” In Medieval Historical Writing in the Christian and Islamic Worlds, edited by D. O. Morgan, 109–24. London: School of Oriental and African Studies, 1982. N2( 34 56789:

———. Medieval Persia, 1040–1797: A History of the Near East. New York: Longman, 1988. ———. The Mongols. 2nd edition, Oxford: Blackwell, 2007. Morton, A. H. “The Letters of Rashīd al-Dīn: Ilkhanid Fact or Timurid Fiction?” In The Mongol Empire and Its Legacy, edited by R. Amitai and D. O. Morgan, 155–99. Leiden: Brill, 1999. Mustawfī Qazvīnī, Ḥamd Allāh. The Ta’ríkh-i-Guzída: or “Select History” of Hamduʾlláh Mustawfí-i-Qazwíní. Edited and Translated by E. G. Browne, 2 vols. Leiden: Brill, 1910–13. Noonan, T. S. “Russia’s Eastern Trade, 1150–1350: The Archaeological Evidence.” Archivum Eurasiae Medii Aevi 3 (1983): 201–64. Ostrowski, D. “The “Tamma” and the Dual-Administrative Structure of the Mongol Empire.” Bulletin of School of Oriental and African Studies 61:2 (1998): 262–77. Paul, J. “Fiscal System ]]]. Islamic Period.” In Encyclopaedia Iranica. (online version at http://www.iranicaonline.org/articles/<=scal-system-iii-islamic-period) Pelliot, P. “Une ville musulmane dans la Chine de nord sous les Mongols.” Journal Asiatique 211 (1927): 261–79. Petrushevsky, I. P. “The Socio-Economic Condition of Iran under the Il-Khans.” In The Cambridge History of Iran, vol. 5, edited by John A. Boyle, 483–537. Cambridge. Cambridge University Press, 1968. ———. “Rashid al-Din’s Conception of the State.” Central Asiatic Journal 14 (1970): 148–62. Rachewiltz, Igor de. “Yeh-Lü Ch’u-Ts’ai (1189–1243): Buddhist Idealist and Confucian Statesman.” In Confucian Personalities, edited by A. F. Wright and D. Twitchett, 189–216. Stanford, !#.: Stanford University Press, 1962. Rashīd al-Dīn. Kitāb-i Mukātibāt-i Rashīdī. Edited by M. Abarqūhī and M. Shafī. Lahore: University of Punjab, 1947. ———. The Successors of Genghis Khan. Translated by J. A. Boyle. New York: Columbia University Press, 1971. ———. Jāmiʿ al-tawārīkh. Edited by B. Karīmī. 2 vols. Tehran: Iqbāl, 1367/1988–89. ———. Jāmiʿ al-tawārīkh. Edited by M. Raushan and M. Mūsavī. 4 vols. Tehran: Nashr-i Alburz, 1373/1994. ———. Jami‘u’t-tawarikh [sic] Compendium of Chronicles. Translated by Wheeler M. Thackston. 3 Vols. Cambridge, H#: Department of Near Eastern Languages and Civilizations, Harvard University, 1998–99. Rossabi, M. “Khubilai Khan and the Women in His Family.” In Studia Sino-Mongolica, edited by Wolfgang Bauer, 153–80. Wiesbaden: Harrassowitz, 1979. ———. “The Muslims in the Early Yüan Dynasty.” In China under Mongol Rule, edited by J. D. Langlois, 257–95. Princeton, 5v: Princeton University Press, 1981. %O4 &78P8Q67 '894 RS H8PT89 U8Q4P N2`

Schurmann, Herbert Franz. Economic Structure of the Yuan Dynasty. Cambridge, H#: Harvard University Press, 1956. Shabānkāra’ī. Majma‛ al-ansāb. Edited by H. Muḥaddith. Tehran: Amīr Kabīr, 1363/1984. Silverstein, A. J. Postal Systems in the Pre-Modern Islamic World. Cambridge: Cambridge University Press, 2007. Smith, J. M. “Mongol and Nomadic Taxation.” Harvard Journal of Asiatic Studies 30 (1970): 46–85. Sneath, D. The Headless State: Aristocratic Orders, Kinship Society and Misrepresentations of Nomadic Inner Asia. New York: Columbia University Press, 2007. Soudavar, A. “In defence of Rashid-od-Din and his Letters.” Studia Iranica 32 (2003): 77–122. Spuler, B. “Ābeš Ḵātūn.” In Encyclopaedia Iranica. 1:2, 210. Waṣṣāf, ʻAbd Allāh ibn Fazḍl Allāh. Taḥrīr-i Tārīkh-i Waṣṣāf. Edited by A. M. Āyatī. Tehran: Pizhūhishgāh-i ʻUlūm-i Insānī va Muṭālaʿāt-i Farhangī, 2004. Yokkaichi, Yasuhiro. “Horses in the East-West Trade between China and Iran under Mongol Rule.” In Pferde in Asien: Geschichte, Handel und Kultur, edited by G. Bert Fragner, Ralph Kauz, Roderich Ptak and Angela Schottenhammer, 87–98. Vienna: Verlag, 2009. Ward, L. J. “Ẓafarnāmah of Mustawfī.” PhD diss., University of Manchester, 1983. !"#$%&' ( Faith and the Law: Religious Beliefs and the Death Penalty in the Ilkhanate

Florence Hodous*

Religious developments during the period of Mongol rule in Persia are particularly complex due to the multitude of in)*uences a+fecting Persia at this time. While the Mongols were originally shamanist, they were tolerant of various faiths, showing what has been described as either openmindedness, indi+ference, or the subsuming of other religious in)*uences into their own worldview., While the in)*uence of Islam was strong within the Ilkhanate, as a faith it no longer enjoyed undisputed primacy. Rather, faiths such as Christianity, Judaism and Buddhism gained increased prominence, and occasionally, political in)*uence. This paper will consider some aspects of how religious beliefs a+fected the death penalty in the Ilkhanate. In such a prosaic matter as the death penalty and the manner of its implementation, practical considerations of course played a role, as did political calculations. Nevertheless, the death penalty also sometimes re)*ected religious beliefs. The method of execution could indicate allegiance to a particular religion; it was also sometimes held to in)*uence the afterlife of the executed person, and was adjusted accordingly. For example, elite -.gures could be executed di+ferently from ordinary people. In addition, religious beliefs could inspire, or provide justi-.cation for, amnesties which meant that the death penalty was annulled. The beliefs that in)*uenced execution methods and amnesties, sometimes directly and sometimes indirectly, were manifold. It should be noted that it will not be possible here to make a categorical distinction between ‘religious beliefs,’ such as the doctrines of world religions, and ideas that could be de-.ned as ‘superstitions.’ Rather, any in)*uence which derives from some kind of a belief in the supernatural will be included in the analysis.

* This paper has grown out of my PhD research under Dr. George Lane, and has been supported by funding from the European Research Council under the European Union’s Seventh Framework Program (/$/2007–2013) / !"# Grant Agreement 312397. 0 Atwood, “Validation by Holiness or Sovereignty,” 237–56.

Re)*ecting the role of the Mongols as cultural mediators between West and East Asia, this paper will identify three di+ferent sources of religious in)*uence on executions in the Ilkhanate: Shamanism, Islam and -.nally Confucianism and Buddhism. The diversity of these in)*uences sheds light on the lively cultural and religious exchange which took place in the Ilkhanate. Not only did Mongols interact with Persian culture, but in)*uences from the Yuan dynasty with which the Ilkhans were closely linked also played a part.

The In$%uence of Shamanism

The belief system which the Mongols brought with them when they invaded Persia was shamanism.> This involved seeing a numinous quality in many natural elements, including in the sky (Tengri, which can also be translated as ‘heaven’ and which was thought to in)*uence the destinies of man), mountains, rivers, and other natural features.? In addition, the supernatural realm included spirits representing animals or ancestors, which frequently interacted with humans and the natural world.@ The wellbeing of the community depended on their relationship with the spirits. The interactions were often mediated by a religious specialist, the shaman,A in public ceremonies. In the Mongol case, the belief that the soul can leave the body, thus enabling the shaman’s soul to reach distant places to interact with the spirits, was crucial.B The role of the shaman is to enable and, wherever possible, in)*uence exchanges between the natural world and the human world, to the good of his community.C

D As shamanism is the better-known term for the Mongols’ belief system, it will be used here, although ‘animism’ may be more accurate, as the shaman is merely the ritual specialist dealing with the spirits which the Mongols believed ‘animated’ nature. Thanks to Prof. Veronika Veit for pointing out this distinction. Eliade, Shamanism, 4–8. See also DeWeese, Islamization and Native Religion, 32–39. E Tucci and Heissig, Die Religionen Tibets und der Mongolei, 350–51; Be+fa, “Le concept de täng- gäri,” 218; Even, “L’Au-delà dans les représentations religieuses des Mongols,” 166. F Roux, La religion des Turcs et des Mongoles, 62. G The precise de-.nition of ‘shaman,’ and the method of di+ferentiating shamans from other specialists employing altered states of consciousness, is disputed. A characteristic of the shaman may be that he is able to control the spirits, his personality not being displaced while he is interacting with spirits. Sidky, “Ethnographic perspectives,” 231. H Ruotsala, Europeans and Mongols, 26; Roux, La religion, 157–58; this aspect may not be true of all forms of shamanism, see Humphrey, “Shamanic Practices and the State in Northern Asia,” 197–200. I Ruotsala, Europeans and Mongols, 26. 23J "K:KLM

Roberte Hamayon, who has researched Siberian (including Mongol) shamanism in depth, describes how in forest shamanism, the (super)natural world and the human world were perceived to be in a constant exchange of “life- force” (force de vie): humans took life-force from nature when hunting, and gave life-force back when people fell ill or died.N This “life-force” was believed to be primarily found in several important elements: in particular the blood,O the respiratory apparatus,,P and some of the bones.,, Because of this, great care was taken not to break or damage certain bones of the animals, and forest Mongol tribes would leave the skull, and sometimes the respiratory organs, in the forest in order to symbolically permit a new animal to appear in nature.,> Due to the sacred nature and “magical” properties,? of blood, the blood of animals should also not be “lost” by shedding it on the ground. For this reason, animals were put to death in a way which avoided the shedding of blood, often through slitting open the belly and squeezing the heart.,@ These beliefs applied not only to the blood, respiratory organs and bones of animals, but in the case of humans the same beliefs, at least in regards to blood and bones, can be observed. This is because humans were considered to be part of the same exchange with the supernatural world as animals. Their blood should not be “lost” by being shed on the ground but should rather return to nature. If humans were killed, their afterlife depended on the manner in which they had been killed. Souls of people who had died were held to be able to protect and help, or they could harm the living,,A and the souls of those who had died badly were considered to be particularly dangerous. Therefore, these beliefs were of primary concern in carrying out executions. It is for this reason that Mongol execution methods can be divided into those where blood is shed, and those which avoid the shedding of blood. A ‘regular’

Q Hamayon, La chasse à l’âme, 397–402; see also Roux, La religion, 157. R Anonymous, Secret History of the Mongols (hereafter !"#), 754. The blood was thus considered a seat of the soul. 0S The ‘respiratory system’ consisted of “the head, the trachea, the heart and the lungs, bearing the ‘vital breath,’ ” Aigle, “Le ‘grand jasaq’ de Gengis-Khan,” 66. 00 Roux, La religion, 163; see also Hamayon, La chasse à l’âme, 548. 0D Hamayon, La chasse à l’âme, 397–400. 0E Thanks to Prof. Veronika Veit for pointing this out; see Uray-Köhalmi, “Sibirische Parallelen,” 260; Roux, La mort, 80. 0F Juwaynī/Qazwīnī, 162; Juwaynī/Boyle, 206; al-ʿUmarī, Das mongolische Weltreich, 8–9 (Arabic), 96 (German translation); Polo, The book of Ser Marco Polo, W, 420; Aigle, “Le grand jasaq de Gengis-Khan,” 66. 0G Ibid., 575–82. /5678 #9: %8; <5= 23X death (where the blood was not spilt and bones were not burnt) allowed the person in question to still become an ancestor, their lineage to be continued, and clan identity to be transmitted.,B Other execution methods, including all those in which blood was shed, and also boiling to death,,C did not allow the person to become an ancestor. The cremation of bodies was occasionally practised as a particular insult.,N The ‘privilege’ of being allowed to ‘survive’ and become an ancestor, even after having committed a transgression, was usually accorded to nobles, not commoners.,O The privilege was ostensibly extended to Jamuqa, Chinggis Khan’s anda (blood-brother), and the belief is expressed that the soul of a powerful, dead enemy could exert a powerful favourable in)*uence on the one who had conquered him.>P In Jamuqa’s speech in the Secret History, he promises to be a blessing if his blood is not spilt:

If you want to show favour to me, let me die swiftly and your heart will be at rest. And if you condescend to have me put to death, let them kill me without shedding my blood. When I lie dead, my bones buried in a high place, for ever and ever I shall protect you and be a blessing to the o+f- spring of your o+fspring.>,

However, in reality, Jamuqa was probably not so honoured. Rashīd al-Dīn says that Jamuqa was in fact hacked to death,>> although publicly he had been assured that he would be put to death without shedding his blood. Therefore, publicly the principle of not shedding the blood of nobles was upheld and its importance thereby rea+-.rmed.

0H Aigle, “Le grand jasaq de Gengis-Khan,” 63. 0I Y"Z, 54 (§129). 0Q The rulers Mahmud of Ghazna and the Khwarazmshah ‘Alā’ al-Dīn Muḥammad were cremated. Al-Nasaw]̄, Histoire du sultan Djelal ed-Din, 193 (Arabic), 319–21 (French translation). The burning of executed people’s bodies had been on occasion practised in pre- Mongol Persia: Lange, Justice, Punishment and the Medieval Islamic Imagination, 69. 0R The exact ranks of those considered ‘noble’ in steppe society were of course )*uid, but included the leaders of ‘tribes’ or political groupings and their relatives. For a criticism of the concept of ‘tribe’, see Atwood, “How the Mongols Got a Word for Tribe,” 63–89. DS Y"Z, 754. D0 Y"Z, 132 (§201). DD Rash īd al-Dīn, Jāmi‘ al-tawārīkh, ed. M. Raushan and M. Mūsavī, 1:204 [hereafter Rashīd/ Raushan & Mūsavī]; Rashīd al-Dīn, Jami‘u’t-tawarikh [sic] Compendium of Chronicles, translated by Wheeler M. Thackston, 1:108 [hereafter Rashīd/Thackston]. 223 "K:KLM

It should be noted that these beliefs were not unique to the Mongols but were shared by Turks as well. Thus, long before the Mongols came on the scene, the principle of not shedding the blood of nobles already had an impact. An early example is Caliph al-Mutawakkil (r. 847–61) putting the Turkish army leader Itakh to death by weighing him down with iron chains and possibly making him die of thirst. Then he summoned the judges of Baghdad and the head of the postal service to examine the corpse and witness that Itakh had not been beaten or marked in any way.>? In addition, bloodless execution methods were regularly used in the Seljuq Empire for members of the ruling family.>@ Because of the centuries-long rule of Turks and Mongols in the Persian realm, the in)*uence of the shamanistic belief system on execution methods in Persia both predated Mongol rule and outlived it. The Mongols used various execution methods speci-.cally to avoid shedding blood, but the principle of not shedding the blood of nobles was more enduring than particular execution methods. Several execution methods used by the Mongols in pre-imperial days stopped being used, while others continued in use. Looking at the methods of execution that were commonly used to avoid the spilling of blood, one that seems to have disappeared quickly is execution by ordeal through wrestling. Though in theory this meant establishing right and wrong through appealing to a heavenly power, both cases recorded in the Secret History seem to have been ‘rigged’ in advance; they were in fact thinly disguised executions.>A For the losing party, the -.ction of a wrestling match may have been a more honourable way to die than being executed in another manner.>B In practice, executions in the context of wrestling matches ended in the loser’s back being broken.>C Breaking the back continued to be used as an execution method in the Ilkhanate throughout the thirteenth century, at least until the reign of the Muslim Ilkhan Ghazan, though not necessarily in the context of a wrestling match. The Ilkhan Aḥmad Tegüder was put to death by breaking his back in 683/1284.>N The last instance of breaking the back was during Ghazan’s reign, when Süge, a rebel general, had his back broken by the leaders of the army.

DE Kennedy, The Court of the Caliphs, 238. DF Köprülü, “La proibizione di versare il sangue,” 18–19; Lange, Justice, 27. DG Y"Z, 61–62 (§140), 172–73 (§245). DH Roux, La religion, 249–50. DI Y"Z, 61–62 (§140), 172–73 (§245). DQ Wa ṣṣāf, Kitāb-i Waṣṣāf al- Ḥaz̤rat, 13; Aayat]̄, Taḥrīr, 80. /5678 #9: %8; <5= 222

[Ghazan Khan] said to Horqudaq: “Go from here and remove Süge from the face of the kingdom of the world, lest he suddenly escape.” [. . .] There was a famous amir, Sati by name, a brave leader of a thousand, who had always hated Süge. Horqudaq said to him, “Go and take his soul from his body in retribution, but do not let his blood )*ow on the ground, for Chingiz Khan said, ‘when blood of a king )*ows on the ground, that is a crime.’ Sati left his presence, to put Süge to death, but when Süge saw the face of his enemy, he pulled a dagger from his boot and struck Sati so that the blow pierced his heart. [. . .] There and then, the leaders of the army broke [Süge]’s back, and he died as a result.>O

The reason for the rapid disappearance of breaking the back as an execution method may be its connection with wrestling, which was integral to Mongol culture but not necessarily so for the peoples conquered by the Mongols. It is signi-.cant that the last instance of breaking the back took place within the army among Mongol amirs, who may have maintained the tradition of wrestling. Another bloodless execution method that disappeared quickly was trampling to death; it was a ‘bloodless’ execution method because the victim was wrapped in a felt carpet -.rst. A very famous example is Hülegü Khan granting the Caliph al-Mustaʿṣim to be sewn in a piece of cloth and trampled to death rather than su+fering beheading or another ignominious penalty. Following the conquest of Baghdad in 1258, Hülegü had been warned that killing the caliph could lead to natural disasters,?P although others assured him that there would be no supernatural consequences.?, In any case, Hülegü decided to give him the treatment of a noble by not letting his blood be shed: “they put him on a piece of sackcloth and then sewed it up round about him, and with kicks of their feet they killed him.”?> This execution method was not used after this by the Ilkhans, although in 1287, Prince Nayan was given a similar treatment by Qubilai who had him “wrapt in a carpet, and tossed to and fro so mercilessly that he died.”??

DR Mustawf ī, Ẓafarnāmah, 1354; Ward, “The Zafar-Namah of Hamdallah Mustau-.,” 399–400. ES Bar Hebraeus, The Chronography of Gregory Abū’l Faraj, 431. E0 Rashīd/Raushan & Mūsavī, 2:1007; Rashīd/Thackston, 2: 492–93. ED Bar Hebreaus, The Chronography of Gregory Abū’l Faraj, 431. Other sources, such as Grigor of Akanc‘, History of the Nation of the Archers, 333–34, say that he was starved to death; but not a single source implies that his blood was shed. EE Polo, The Book of Ser Marco Polo, W, 343. 22c "K:KLM

Despite the bloody nature of this execution method, in the Mongol belief system it still counted as bloodless, since no blood was shed on the ground. The use of a carpet or piece of cloth was therefore crucial. It appears that the key was to avoid blood being shed on the ground—where the “life-force” would be lost?@—or which could cause the executed person’s spirit to be particularly dangerous for the living.?A A further bloodless execution method was drowning, though this was used mainly for women. This was the fate of Fāṭima, the regent Töregene’s con- -.dant, who was executed by Güyük Khan after Töregene’s own death. Rashīd al-Dīn says that: “Her bodily ori-.ces sewn closed, she was wrapped in felt and thrown into the river.”?B The purpose of sewing her bodily ori-.ces closed may have been to prevent her soul from exiting her body and seeking vengeance;?C or it may simply be due to a desire not to pollute the ‘sacred’ water.?N Shera, a Central Asian and one of Fāṭima’s accusers, later su+fered nearly the same fate: “he too was detained until he confessed to a crime he had not committed and was thrown into the river, and his women and children were put to the sword.”?O During the Ilkhanate, drowning was used as an execution method in the cases of Gontsa, the wife of the Georgian king David, who was executed at Hülegü’s command;@P for Toghachaq Khatun and other women who were put to death for allegedly causing Arghun’s illness in 1291;@, and, later, for Baghdad Khatun.@> However, wrapping in carpet is not mentioned in the later references, nor is sewing closed the bodily ori-.ces. The disappearance of breaking the back and trampling to death as execution methods did not mean that the principle of not shedding the blood of nobles disappeared. The principle was in)*uential throughout the Ilkhanate, although the precise ranks of those who were considered eligible for treatment as a ‘noble’ in case they were executed did shift. As for the Ilkhans themselves, there are no accounts of their blood being shed. In 1284, the Ilkhan Tegüder

EF Roux, La religion, 162, 261–62. EG Shirokogoro+f, “Versuch einer Erforschung der Grundlagen,” 53–54. EH Rashīd/Raushan & Mūsavī, 2:803; Rashīd/Thackston, 2:391. Felt was an everyday material used by the Mongols for yurts and countless utensils, but felt carpets were also associated with funerary practices as well as enthronements, although their precise signi-.cance remains di+-.cult to determine. Olschki, The Myth of Felt, 13–20, 23–34. EI Hamayon, La chasse à l’âme, 421. EQ Thanks to Charles Melville for this suggestion. ER Rashīd/Raushan & Mūsavī, 2:803; Rashīd/Thackston, 2:391. FS Orbélian, Histoire de la Siounie, 1 :, 234. F0 Rashīd/Raushan & Mūsavī, 2:1180; Rashīd/Thackston, 2:575. FD Shab ānkāra´ī, Majmaʿ al-ansāb, 299. /5678 #9: %8; <5= 22e

Aḥmad had his back broken.@? Even such a late source as Ibn Bazzāz, who gives an erroneous account of his death (he says that Aḥmad Tegüder was wrapped in felt and kicked to death),@@ still re)*ects the Mongol concern for not shedding royals’ blood. In 1295, the Ilkhan Geikhatu was strangled with a bowstring, according to Waṣṣāf.@A This not only re)*ected concern for not shedding any blood, but also indicated considerations of power and status through the use of a very symbolic item—the bowstring—to perform the operation. It must have been considered a very ‘honourable’ dispatch. Baidu was also strangled.@B For the remaining Ilkhans, either the execution method is not mentioned or they died natural deaths. Nevertheless, it is noteworthy that none of the sources even suggests that the blood of any Ilkhan was shed. Other members of the royal family were not necessarily accorded the same privilege. While Qonqurtai, a son of Hülegü, was accorded a bloodless execution method, breaking the back, in 682/1283–4,@C in the year 703/1303–4 prince Horqudaq was beheaded.@N What this shows is that the boundaries of privilege and status were )*exible and could be rede-.ned. Some subordinate rulers in the Ilkhanate also knew the principle of not shedding the blood of nobles and used it. Ironically, while some speci-.cally Mongol bloodless execution methods were disappearing from use, the principle itself only gained in in)*uence. This can be seen in further bloodless executions reported from the Mongol period. One of these is the case of Sayf al-Dīn Gharjistānī, who refused to recognize Shams al-Dīn, the ruler of Herat and founder of the Kart dynasty. When in 647/1249–50 Shams al-Dīn wanted to force him to surrender, Sayf al-Dīn took refuge with the Mongol governor Arghun Aqa; however, “Arghun Aqa without hesitation handed the tied up Malik Sayf al-Dīn over to the Malik Shams al-Dīn’s messenger so that he be brought to Herat. Malik Shams al-Dīn ordered that he be trampled at the gate of Khosh.”@O Although this source does not give much detail, it is possible that there was Mongol in)*uence here because trampling to death was only rarely used in the Persian world.AP

FE Wa ṣṣāf, Kitāb-i Waṣṣāf al- Ḥaz̤rat, 136; Aayat]̄, Taḥrīr, 80. FF Ibn Bazzāz, Ṣafwat al-ṣafā, 219. FG Wa ṣṣāf, Kitāb-i Waṣṣāf al- Ḥaz̤rat, 279; Aayat]̄, Taḥrīr, 169–70. FH Brosset, Histoire de la Géorgie, 615. FI Wa ṣṣāf, Kitāb-i Waṣṣāf al- Ḥaz̤rat, 125; Aayat]̄, Taḥrīr, 74. FQ Wa ṣṣāf, Kitāb-i Waṣṣāf al- Ḥaz̤rat, 465; Aayat]̄, Taḥrīr, 273. FR Zamch ī Is-.zārī, Rawz̤āt al-jannāt, 412. The second source recording this event, the Tār.̄khnāmah-yi Harāt, says that he was beaten to death; al-Harav]̄, Tār.̄khnāmah-yi Harāt, 188–89. GS Lange, Justice, 66, 70–71. 22f "K:KLM

In addition, strangulation, a bloodless execution method already in use for members of the royal family under the Seljuqs,A, became a well-established execution method for people of status during the Ilkhanate, perhaps because it was already somewhat familiar to the subjects of the Mongols. The ruler of Kirman Terken Khatun had Soyurghatmish, the son of her husband who had been challenging her position, put to death by strangulation.A> In 1327, Ghiyath al-Dīn, the ruler of Herat, had Amir Chūpān put to death with a bowstring.A? Later, Arpa Khan had Malik Sharaf al-Dīn Maḥmūd Shāh Inju, Shahzade and other descendents of Hülegü put to death by strangulation.A@ This shows how the Turko-Mongol concern with bloodless execution methods was an important factor in Persia during the Ilkhanate and beyond. However, the wider use of bloodless execution methods did not necessarily entail an understanding of the shamanistic belief system underlying them. As Jean-Paul Roux has pointed out, the use of bloodless execution methods was based on the belief that all blood (including that of non-nobles and of animals)AA is sacred, or a seat of the soul or “life-force,” though there is little evidence that Persian commentators understood this. The Persian sources give various justi-.cations for the use of bloodless execution methods, but they usually focus only on the idea that the blood of nobles should not be shed, rather than on the belief that all blood is sacred. For example, Mustawfī says merely that shedding the blood of a noble on the ground is “a crime,” and attributes the principle to Chinggis Khan, as if it were a new law rather than an ancient principle. In describing the execution of Süge, Mustawfī implies that Chinggis Khan himself forbade the shedding of royal blood on the ground.AB Bar Hebraeus links a bloody execution with lack of rain and “coals of -.re” bursting from the earth.AC Jūzjānī also links the spilling of blood on the ground with natural disasters, reporting the astrono- mer Husām al-Dīn’s words to Hülegü as: “if [the Khalifah] is slain, with the sword, when his blood falls upon the ground, an earthquake will take place, and people will be destroyed.”AN While Bar Hebraeus and Jūzjānī correctly capture the idea that blood has magical properties and will bring about negative

G0 Ibid., 27. GD Shab ānkāra’ī, Majmaʿ al-ansāb, 201. GE Ibid., 284. GF Ḥā-.ẓ-i Abrū, Ẕayl jāmiʿ al-tawārīkh Rashīdī, 192. GG Roux, La mort, 77–79. GH Mustawf ī, Ẓafarnāmah, 1354; Ward, “The Zafar-Namah of Hamdallah Mustau-.,” 399–400. GI Bar Hebraeus, The Chronography of Gregory Abū’l Faraj, 431. GQ J ūzjānī, Ṭabaqāt-i Nāṣirī, 2:1252. /5678 #9: %8; <5= 22( consequences if it is shed on the ground, neither they nor Mustawfī show evidence of any deeper understanding of blood as one of the seats of the soul. Therefore, the Mongol and Turkish conception of the magical quality of blood had an important in)*uence on execution methods during the Ilkhanate, as the Ilkhans and others used bloodless execution methods for people of high status. Despite some typically Mongol execution methods such as trampling to death and breaking the back falling into disuse, the idea that people of high status should be executed without shedding their blood remained in)*uential.

The In$%uence of Islam

The executions carried out in the Ilkhanate were also in)*uenced by Islamic execution methods and their underpinning beliefs. Such execution methods were used not only by Muslims in the Ilkhanate but also occasionally by Mongols. Here I will discuss relevant execution methods that can be de-.ned as Islamic and how they were used in the Ilkhanate. Executions in Persia in the pre-Mongol period included penalties speci-.cally advocated in the religious literature for various o+fenses, as well as a variety of execution methods which had developed over time and become customary. A large number of executions both before and during the Ilkhanate were carried out by the sword—usually through beheading or cutting in two.AO For example, among the viziers, Shams al-Dīn Juwaynī was beheaded after torture and Saʿd al-Dawla Yahūdī was also beheaded, whereas Ṣadr al-Dīn Aḥmad Zanjān]̄, Rashīd al-Dīn, and Malik Nuṣrat al-Dīn ʿĀdīl were cut in two.BP Many amirs were beheaded, and such executions were also very frequent at the regional level.B, These execution methods, because of their frequency and because they were shared by di+ferent cultures, are not a very useful object of analysis. On the other hand, less frequently imposed execution methods which can nonetheless be identi-.ed as ‘Islamic’ will be more useful for the present analysis.

GR On the Seljuq period see Lange, Justice, 61–22; many of the accounts of executions ordered by the Ilkhans are collected in Ma‘dan’kan, Bih yāsā rasīdagān dar ‘aṣr-i Īlkhānī. HS Ma‘dan’kan, Bih yāsā rasīdagān, 25–35. H0 Thus many executions reported by the major sources such as Rashīd are beheadings, and local sources such as Is-.zārī’s history of Herat or pseudo-Ibn al-Fuwaṭ]̄’s al-Ḥawādith al-jāmiʿa concentrating on Baghdad also mention many executions carried out by the sword. 22k "K:KLM

Although diverse, the Shariʿa includes the following categories of o+fenses: qiṣāṣ, taʿzīr and ḥudūd.B> Qiṣāṣ (or talio) concerns o+fenses of murder and bodily harm, which are meant to be punished like for like.B? Taʿzīr are discre- tionary penalties for lesser crimes, which are discussed more in manuals of statecraft than in Islamic theological works.B@ Ḥudūd are o+fenses the punishment for which is stated in the Qurʾan.BA While punishments in the qiṣāṣ and taʿzīr categories are of necessity diverse, punishments in the ḥudūd category are relatively -.xed, though by no means universally practised. Among the ḥudūd punishments of interest here are stoning, cruci-.xion,BB and the cutting o+f of hands and feet which sometimes preceded the actual execution.BC These punishment methods were not used at all, traditionally, by Mongols; the Secret History of the Mongols mentions neither stoning, cruci-.xion nor the cutting o+f of hands and feet. In addition, with regards to cruci-.xion, the penalty would seem unsuitable for nomads: since it takes time (usually several days), and requires a -.xed place, it seems a more attractive option for use in crowded urban public spaces rather than among mobile nomads.BN Although the Mongols in the Ilkhanate did adopt some of these execution methods that can be de-.ned as ‘Islamic,’ their adoption of such methods was sporadic rather than systematic. ‘Islamic’ execution methods were used a few times by the Ilkhans, and once, according to [pseudo]-Ibn al-Fuwaṭ]̄, by a Mongol who had settled in Baghdad. Even in Ghazan’s time and afterwards, such execution methods were employed rarely by Mongols, and there is no evidence they were used because they were considered ‘Islamic,’ rather they may have been seen as simply another alternative to execution methods already in use. The -.rst report of a Mongol imposing cruci-.xion regards the activities of Küchlüg, leader of the Naiman Mongols who had )*ed from Chinggis Khan and usurped power in Hotan. Juwaynī reports that after getting into a theological

HD Tellenbach, “Die Strafe im islamischen Strafrecht,” 10. HE Ibid., 10. HF Ibid., 11. HG Except for zinā (fornication), the punishment for which is mentioned in the ḥadīth but not the Qur’an; ibid., 10. HH ‘Cruci-.xion’ took a variety of forms in Persia, and it is not always possible to determine whether nailing to a cross or gibbeting is meant. Lange, Justice, 62–63. HI Seidensticker, “Responses to Cruci-.xion in the Islamic World,” 206; Kennedy, The Court of the Caliphs, 238–39. HQ Ibn ‘Arabshāh’s allegation that among the Mongols “Thieves may be cruci-.ed [. . .] on the evidence of a single witness . . . ” mentioned by Irwin in his article: “What the Partridge Told the Eagle,” 8–9, is unfounded according to the other sources. /5678 #9: %8; <5= 224 dispute with an imām, he had the imām cruci-.ed. However, Michal Biran has rightly drawn attention to the questionable nature of Küchlüg’s anti-Muslim persecution, which is reported only by the Muslim historians Juwaynī and Rashīd al-Dīn, and not supported by the Chinese sources.BO It was only in 689/1290–91 that the Ilkhan Arghun (or possibly his vizier Saʿd al-Dawla) ordered a cruci-.xion. As a scapegoat for the alleged misdeeds of some important Muslims, the order was given that “Jamāl al-Dīn ibn Halavī, who was in charge of the tamghas of Baghdad, should be cruci-.ed. They cruci- -.ed him on the al-Nubi gate. He was still wearing his garment when they gave his corpse to his family.”CP Although Arghun himself was Buddhist, by the time of his rule Mongol amirs were increasingly becoming Muslims.C, Another cruci-.xion took place in 711/1311 when Zayn al-Dīn ʿAlī Kīmīāchī was among those accused of earlier plotting against the vizier Ṣadr al-Dīn. He was cruci-.ed and died on the second day.C> Stoning was used infrequently by Muslims in the Ilkhanate, although it was used occasionally as a penalty for adultery. ʿAlāʾ al-Dīn, the governor of Iraq and brother of the vizier and historian Shams al-Dīn, imposed this penalty in 1276 in Baghdad.

In the month of Ramadan, a woman and a man were found in the hammam who had committed the crime of zinā (adultery). ʿAlāʾ al-Dīn gave the order that they should be stoned. They were stoned outside the city of Baghdad and up to this time, there is nothing recorded in the histories with regards to anyone having been stoned in Baghdad.C?

This incident happened during the governorship in Baghdad of ʿAlẚ̄ al-Dīn Juwaynī, acting on behalf of the Ilkhans. It shows the freedom which Mongol governors had in legal matters, even when they may have consciously attempted to revive ‘Islamic’ punishments. Another case of stoning for adultery occurred in Baghdad in 1289.C@ Nevertheless, the use of stoning by an Ilkhan only occurred much later, perhaps because it was so rarely used among Muslims themselves. The only record of an Ilkhan ordering a stoning is from

HR Biran, The Empire of the Qara Khitai, 195. IS Ibn al-Fuwaṭ]̄, al-Ḥawādith al-jāmiʿa, 461–62 [year 689]. I0 Pfei+fer, “Re)*ections on a ‘Double Rapprochement,’ ” 374–75. ID Q āshānī, Tārīkh-i Uljāytū, 132; Waṣṣāf, Kitāb-i Waṣṣāf al- Ḥaz̤rat, 538; Aayat]̄, Taḥrīr, 278. IE Ibn al-Fuwaṭ]̄, al-Ḥawādith al- jāmiʿa, 386. IF Ibn al-Fuwaṭ]̄, al-Ḥawādith al- jāmiʿa, 486–87; Gilli-Elewy, Bagdad nach dem Sturz des Kalifats, 179. 22J "K:KLM near the end of the dynasty, when Abū Saʿīd ordered Konjak, the daughter of Aḥmad Tegüder and wife of Amir Irinjīn, to be stoned to death in the year 719/1319.CA A further example regards the cutting o+f of hands and feet before the execution. An example of a Mongol who had settled in Baghdad is reported by Ibn al-Fuwaṭ]̄. When the Mongol was assassinated his son took revenge by ordering the assassin’s hands and feet to be cut o+f before he was executed:

In this year [i.e. 692/1292–93], a bāţinī attacked Neqājū, the leader of the converts in al-ʿIrāq, at the beginning of the ʿAz̤udi bridge in Baghdad and struck [him] with a stone repeatedly and killed [him]. The killer )*ed. An Isfahani man caught him on the road. The killer fell to the ground and was captured. He kept saying, ‘Fidā’ī al-Malik al-Ashraf, ‘Fidā’ī al-Malik al-Ashraf.’CB They handed him over to the son of the Mongol Neqājū. They made an example of him and cut o+f his hands and feet. He was still alive. Then they [. . .] back and he did not utter a sigh. Then the bāṭinī said, ‘Aye, you old e+feminates, whatever you have done is less than what I have in my heart so do whatever you are able.’ Then they killed him and buried him in the place where Neqājū had been murdered.CC

Although it is possible that Ibn al-Fuwaṭ]̄ exaggerated the report, the mention of this particular punishment in the context of an execution ordered by a Mongol is nonetheless remarkable. It seems to be a rare glimpse into the assimilation of a Mongol who had settled locally with his family. The cutting o+f of hands and/or feet was used occasionally in the Islamic world and is referred to occasionally in medieval Arabic poems; it was not a Mongol practice.CN However, the Ilkhans’ use of these execution methods seems to -.t more into a general pattern of )*exibility rather than a progressive adaptation to Islamic execution methods. Some of this )*exibility can be explained by the concept of talio, where the execution method was related to the o+fense committed by the person being executed. Talio is seen for example in the execution of Körgüz, an

IG Wa ṣṣāf, Kitāb-i Waṣṣāf al- Ḥaz̤rat, 645; Aayat]̄, Taḥrīr, 370. Other sources say she was trampled to death (presumably without being -.rst wrapped in carpet), Melville, “Abu Sa⁠’īd and the Revolt of the Amirs in 1319,” 109 n. 118. IH See Atwood, Encyclopedia of Mongolia, 255. The murderer would seem to be an Isma’ili ‘self-sacri-.cer’. II Ibn al-Fuwaṭ]̄, al-Ḥawādith al-jāmiʿa, 475. Thanks to George Lane and Konrad Hirschler for helping with the translation. IQ Seidensticker, “Responses to Cruci-.xion,” 6. /5678 #9: %8; <5= 22X administrator in Persia who was put to death in Chaghatai’s court. His execution was linked to the words he had spoken: “The order was given that since Körgüz had been arrested on account of words spoken in the ordu of Ulugh-Ef he should be taken back thither and tried upon the spot. As usual he had spoken hard words without considering the outcome. Qara Oghul ordered his men to -.ll his mouth with stones and so put him to death.”CO In other cases too, the Ilkhans used varied execution methods, some apparently suggested by the circumstances and others in)*uenced by execution methods practised in Persia, including ones that were not Islamic as such. Ghazan Khan ordered Pīr Yaʿqūb Bāghbān]̄, who was held responsible for leading prince Alafrang to rebel, to be “thrown o+f the top of the mountain they were on.”NP This example resembles execution methods from the Seljuq period such as throwing people from minarets or citadels or putting them into pits.N, The Ilkhan Abū Saʿīd also employed a form of gibbeting, usually more frequently used among Persians:N> “They hung them [Aras, Tuqumaq and Isbuqa, amirs who had fought with prince Irinjīn against Abū Saʿīd] up on hooks, and from this, their blood )*owed painfully. Then they put -.re beneath them; such was the order of their execution.”N? Therefore, the Ilkhans’ occasional use of ‘Islamic’ execution methods can be explained by their )*exibility, and does not show that they had any deep understanding of Islamic execution methods or the religious principles underlying them. On the local level, ‘Islamic’ as well as other execution methods continued to be used and even in)*uence Mongols, as can be seen from the example of Neqājū.

The In$%uence of Confucianism and Buddhism

A third source of religious in)*uence on executions during the Ilkhanate was through Confucianism and Buddhism. A notable feature of the early Ilkhanate were some amnesties of criminals, which the Ilkhans were encouraged by several o+-.cials and advisors to issue. While the numbers of those amnes- tied in this way is unknown and likely very small, the issuing of amnesties is

IR Juwaynī/Qazwīnī, 242; Juwaynī/Boyle, 505. QS Rashīd/Raushan & Mūsavī, 2:958–9; Rashīd/Thackston, 3:659. Q0 Lange, Justice, 69–70. QD Wa ṣṣāf, Kitāb-i Waṣṣāf al- Ḥaz̤rat, 62; Aayat]̄, Taḥrīr, 42–43. QE Mustawf ī, Ẓafarnāmah, 1459; Ward, “The Zafar-Namah of Hamdallah Mustau-.,” 641; see Melville, “Abu Sa’īd and the Revolt of the Amirs in 1319,” 109. 2c3 "K:KLM nonetheless signi-.cant as it points to the ties of the Ilkhans with China, where amnesties were a long-standing tradition, promoted by both Confucians and Buddhists. The importance of the entourage of the khans, namely, the people who surrounded them every day and provided practical and spiritual counsel, really comes to the fore in the case of the amnesties. Unlike bloodless execution methods which may have been adopted simply to distinguish ‘nobles’ from common people, and Islamic execution methods which were used by the Mongols apparently without a religious purpose, the ‘religious’ signi-.cance of amnesties was important. The in)*uence of advisors over the Ilkhans may have been partly channelled through the Mongol religious worldview and more speci-.cally through the shamans, but the impetus for the issuance of amnesties came from Confucianism and Buddhism, and to a lesser extent also from other religions.N@ Amnesties were not originally a Mongol practice, not only because there were never large numbers of criminals imprisoned at any one time on the Mongol steppe, but also because the Mongol legal system was based on the concept of vengeance, which sits uneasily with the idea of commuting penalties.NA Some of the personnel serving the early Mongol Empire and its administration in Central Asia and Persia came from the former Qara Khitai Empire or from China, and thus from Buddhist or Confucian-in)*uenced backgrounds. In addition, the Ilkhans speci-.cally summoned Buddhists, according to Rashīd al-Dīn, from “India, Kashmir, Cathay, and Uighur lands,”NB and Buddhists from all four of these regions were involved in shaping Buddhism in the Ilkhanate.NC Abaqa Khan also appointed bakhshis (Buddhist monks) to educate Ghazan when he was a child.NN Buddhist monks were present at the court of Arghun

QF Talking about omens was often a good way to approach the Ilkhans: al-Ṣafad]̄ and al- Kutub]̄ report that Hülegü Khan’s Shi‘a advisor Naṣīr al-D]̄n Ṭūsī used this tactic to get ʻAlāʼ al-Dīn Juwaynī released from prison. Sayılı, The Observatory in Islam, 202–3. The Muslim Ilkhan Aḥmad Tegüder also mentioned forgiving those who had done evil deeds in his -.rst letter to Qalawun. Bar Hebraeus, Tārīkh mukhtaṣar al-duwal, 508; Waṣṣāf, Kitāb-i Waṣṣāf al-Ḥaz̤rat, 113–18. QG But good actions could be used to remedy illness. Roux, La religion, 82. QH Rashīd/Raushan & Mūsavī, 2:1331–32; Rashīd/Thackston, 3:664; see also Rashīd/Raushan & Mūsavī, 2:1356, Rashīd/Thackston, 3:676, where Rashīd indicates that the native countries of Buddhist bakhshis included India, Kashmir and Tibet; Bulliet, “Naw Bahar and the Survival of Iranian Buddhism,” 144; Melikian-Chirvani, “The Buddhist Ritual in the Literature of Early Islamic Iran,” 272–79. QI Elverskog, Buddhism and Islam on the Silk Road, 157–62. QQ Rashīd/Raushan & Mūsavī, 2:1331–32; Rashīd/Thackston, 3:591, 664. /5678 #9: %8; <5= 2c2

(who was a devout Buddhist), and debated with Muslims at the Ilkhan’s command.NO Rashīd al-Din received information about Buddhism from the Kashmiri monk Kamalashri,OP as well as from two other Chinese collaborators, Litaji and Kamsun.O, While the origins of these Buddhists and Confucians were diverse, they or people like them may have promoted amnesties in the Ilkhanate. In the Yuan dynasty too Confucians and Buddhists were promoting amnesties. Amnesties had been a feature of Chinese political life since the Qin and Han dynasties. In this context, they were connected with (though perhaps not caused by)O> Confucian ideas, in particular the belief that the human and natural worlds are interlinked and a+fect each other.O? Amnesties, as acts of mercy, were held to have positive e+fects in the natural world and beyond, and were thus often deployed at times of unpropitious events or natural disasters such as omens, earthquakes, and eclipses,O@ and even on occasions of rebellion or unrest.OA Many were issued in the spring or summer,OB since it was believed that the emperor’s acts should be appropriate to the season, and unseasonal acts could “disturb the natural rhythm of the world and bring on natural calamities.”OC These beliefs are re)*ected in the Book of Changes, in hexagram 40, (‘loosen- ing’), where natural phenomena are brought into relation with the pardoning of crimes: “Thunder and rain perform their roles: this is the image of Release. In the same way, the noble man forgives misdeeds and pardons wrongdoing.”ON Throughout imperial Chinese history, amnesties continued to be framed in Confucian terms.OO However, during the Yuan dynasty, some of the staunchest promoters of this practice were Buddhists. This is connected with the great in)*uence of Buddhists at the Yuan court. Rashīd al-Dīn relates how Tanba the Tibetan bakhshi was successful in in)*uencing the emperor Temür Khan () to give an amnesty for 100 prisoners, after previous attempts to con- vince him to free the prisoners had failed.,PP Buddhists promoted amnesties so

QR Lane, “Persian Notables and the Families,” 38, 43. RS Schopen, “Hinayana Texts in a 14th Century Persian Chronicle,” 226–35. R0 Elverskog, Buddhism and Islam on the Silk Road, 149; Allsen, Culture and Conquest, 92. RD McKnight, The Quality of Mercy, 18. RE Ibid., 21. RF Ibid., 16. RG Ibid., 16. RH Ibid., 22. RI Ibid., 21. RQ Ibid., 21. RR Ibid., 45 and 140 n. 14. 0SS Rashīd/Raushan & Mūsavī, 2:958–59; Rashīd/Thackston, 2:469. 2cc "K:KLM zealously that people such as Chen Tianxiang (1237–1316) complained about them and said that “amnesties are good fortune for vile people, and a misfortune for noble people,” arguing that if amnesties are given too frequently, people would not be afraid of the law.,P, The eminent o+-.cial and advisor Yelü Chucai, of syncretist Confucian and Buddhist leanings,,P> was instrumental in getting Ögetei to issue amnesties; the -.rst was in 1229 when he obtained “a decree to the e+fect that crimes committed before the -.rst month of 1229 should not be punished.”,P? This was the -.rst time that an amnesty had been granted by the Mongols.,P@ Later, in 1241, Yelü Chucai recommended another amnesty which was accepted, as Ögetei’s health was not that good.,PA Yelü Chucai had earlier used the Buddhist precepts of non-killing, compassion and forbearance as justi-.cations for preventing Buddhist monks from being drafted in the army;,PB whether or not he recommended the amnesties from genuine religious feeling, these precepts were part of his thinking and could be drawn on to justify acts of mercy. An ‘almsgiving’ amnesty was issued by Sorqaqtani Beki while she was ill: “But Beki being ill and her illness having grown worse, as an almsgiving for her long life those who had that day been condemned to death all received their pardon.”,PC Since Sorqaqtani was a Christian, perhaps she was in)*uenced by Christian practices of charity, although religious specialists of many faiths were all active at the court. What is certain is that some illnesses were considered troubling in shamanistic culture, a sign of spiritual attack,,PN so that it is not surprising that Ögetei and Sorqaqtani adopted this new measure in the time of their need. Arghun, who was the most devout Buddhist among the Ilkhans, is connected with two amnesties. He declared his -.rst amnesty when he ascended the throne: “Since God the eternal has shown me favour and awarded me my late father’s crown and throne, I hereby pardon all criminals’ crimes. If the Ṣaḥib Dīwān [Shams al-Dīn Juwaynī] comes to court, I will receive him with

0S0 Yuan Shi , juan 168, 3947; Langlois, “Law, Statecraft and the Spring and Autumn,” 105. 0SD Rachewiltz, “Yeh-lü Ch’u-ts’ai,” 157, 166, 169; Jan Yün-hua, “Chinese Buddhism in Ta-tu,” 380–84. 0SE Rachewiltz, “Yeh-lü Ch’u-ts’ai,” 149. 0SF Ibid., 149. 0SG Ibid., 161. 0SH Jan Yün-hua, “Chinese Buddhism in Ta-tu,” 382. For the in)*uence of Buddhism on law in later Mongol history see Aubin, “Some Characteristics of Penal Legislation,” 140. 0SI Juwaynī/Qazwīnī, 38; Juwaynī/Boyle, 52. 0SQ Hamayon, La chasse à l’âme, 581. /5678 #9: %8; <5= 2ce honour.”,PO It should be noted that amnesties in China also often accompanied enthronements.,,P Although Rashīd al-Dīn mentions this amnesty in particular because it a+fected the vizier Shams al-Dīn Juwaynī, there is nothing to show that it was not intended to be a general amnesty; though to what extent it was implemented is another question. Another amnesty was issued by Arghun shortly before his death in 1292, the reason for which is described in some detail by Riccoldo di Monte Croce:

When [. . .] Argon had shed much blood and had killed many innocent women and children, he became gravely ill. A terrible, awe-inspiring lady appeared to him in a dream. Growling, she took him by the chest and said: ‘Come, answer to the Lord about the blood you have spilled.’ And he said: ‘What Lord? Am I not the Lord of the world?’ (For he was generally called so by everyone, even the Christians, namely the lord of the world.) She said, ‘There is another Lord.’ Terri-.ed, he remained awake, and quickly summoned a baxitas and his high priests and asked them who is this lord who had called him with such authority, and how could he be free of him? They told him that there is someone in particular who has a blood vengeance and wishes to investigate him about the blood of the many innocents he had shed, and that it is not possible to evade him except through many alms. And then he wrote to all the eastern cities that he would be freeing all the captives, and sent great treasures to them and gave the greatest alms. Soon after, he died.,,,

Baxita are described elsewhere in Riccoldo’s text as coming from India and working various kinds of magic. Thus, it is most likely that Buddhist monks are meant.,,> On the other hand, the word ponti56ces is less clearly de-.ned,,? and could possibly refer to shamans. In any case, it seems that both Buddhists and other religious specialists were involved in recommending amnesties. Buddhists were still in)*uential at the enthronement of the following Ilkhan Geikhatu. At the enthronement ceremony, in which Buddhist bakhshis took a visible part and which was held in “relative proximity” to Labnasagut, where

0SR Rashīd/Raushan & Mūsavī, 2:1158; Rashīd/Thackston, 3 :564. 00S McKnight, The Quality of Mercy, 14. 000 Rita George-Tvrtković, A Christian Pilgrim, p. 198. 00D Ricoldo de Monte Croce, Pérégrination en Terre Sainte et au Proche Orient, 92–93 and fn. 141, Ricoldo de Monte Croce, translated by Rita George-Tvrtković, A Christian Pilgrim, 192. 00E Rita George-Tvrtković, A Christian Pilgrim, 192. 2cf "K:KLM

Hülegü had founded a Buddhist sanctuary,,,@ Geikhatu was given a Buddhist name, Irinjīn Dorji, from the Tibetan Rin-ch’en Do-rje (“Jewel Diamond”).,,A It is not surprising therefore that one of the acts of the newly enthroned Geikhatu was to issue an amnesty: he granted alms and support to ʿulamāʾ and sayyids and also ordered prisoners freed.,,B The particular resonance of the concept of an amnesty seems to have been due to the fact that amnesties were held to solve a very practical problem, namely illness. In shamanistic culture, some illnesses were considered a sign of spiritual attack by other people,,,C and thus susceptible to being in)*uenced by human action, in particular the ritual actions of the shaman in shamanistic culture. While shamans became less in)*uential with the expansion of the empire and the growth of the imperial ideology of the Mandate of Heaven, the belief that illnesses could be caused by other humans does not seem to have faded quickly, if at all.,,N Therefore, this was an area in which shamanist and Buddhist belief coincided, and the healing of illnesses was an area of perceived need which Buddhist monks and other advisors could seek to ful-.l by o+fering the issuance of amnesties as an e+fective remedy. This is why many of the occasions for issuing an amnesty involved illness. Amnesties were employed because they were perceived as powerful means to in)*uence the spiritual world and to obtain healing from illness. In addition, it is interesting to note the importance given to blood in Riccoldo di Monte Croce’s account of Arghun’s amnesty. As in Mongol belief the blood is one of the seats of the soul, it does not seem coincidental that it is not just any kind of killing or oppression, but speci-.cally the blood that has been shed which cries out for vengeance. The amnesty is seen here as an e+fective means of forestalling such vengeance. Thus, Confucian and Buddhist in)*uence connected with Mongol shamanist ideas about the importance of blood. In this cultural exchange the Mongols were not merely passive trans- mitters of concepts and beliefs but rather active agents in furthering ideas and practices which speci-.cally appealed to them.,,O

00F Grupper, “The Buddhist Sanctuary,” 51, see also 58–61. 00G Rashīd/Thackston, 3:579 n. 3. 00H Waṣṣāf, Kitāb-i Waṣṣāf al- Ḥaz̤rat, 264; Aayat]̄, Taḥrīr, 159. 00I Hamayon, La chasse à l’âme, 581. 00Q Accusations of witchcraft, whether real or politically motivated, were frequent in the Ilkhanate. 00R Allsen, Culture and Conquest, 193–97, 207. /5678 #9: %8; <5= 2c(

Conclusions

The death penalty in the Ilkhanate was in)*uenced by diverse strands of belief, re)*ecting the cultural exchange brought about by the Mongol conquest. This does not mean that every execution method was decided purely based on religious principles—on the contrary, practical and governance matters also played a role in the decision-making. But religious principles were able to have concrete in)*uence, even without those religious principles themselves being fully understood by those who were in)*uenced by them. This was the case with the Mongol principle of not shedding the blood of nobles, based on the sacred- ness of blood as one of the seats of the soul. It also seems to have been the case with the ‘Islamic’ execution methods which were used by Ilkhans, perhaps without a clear perception of these execution methods as ‘Islamic.’ With amnesties on the other hand, it is likely that Ilkhans declared them precisely because they understood their religious or supernatural justi-.cations. Thus the law was in)*uenced not directly through legal arguments or policies but through religious belief, whether this was genuine or simply used as a justi-.cation for some of the amnesties. The possibility for this cultural transfer was given by the presence of Confucians and Buddhists who were able to have personal contact with Mongols and even with the Ilkhans themselves. These in)*uences re)*ect the changes and transformation which Mongol rule brought to Persia. In fact, the diversity of the in)*uences re)*ects the political position of the Ilkhanate, as well as the composition of the elite. The early Ilkhanate’s strong ties with Yuan China are re)*ected in the presence and in)*u- ence of Buddhists and Confucians at the Ilkhanid court. And while their in)*u- ence waned with the Islamization of the Ilkhanate, the Mongol (and Turkish) in)*uence from shamanism on the other hand was much more enduring.

Bibliography

Aigle, Denise. “Le grand jasaq de Gengis-Khan, l’empire, la culture mongole et la sharî‘a.” Journal of the Economic and Social History of the Orient 47:1 (2004): 31–79. Allsen, Thomas. Culture and Conquest in Mongol Eurasia. Cambridge: Cambridge University Press, 2001. Anonymous. Secret History of the Mongols: A Mongolian Epic Chronicle of the Thirteenth Century. Translated by Igor de Rachewiltz. Leiden: Brill, 2004. Atwood, Cristopher P. “Validation by Holiness or Sovereignty: Religious Toleration as Political Theology in the Mongol World Empire of the Thirteenth Century.” International History Review 26:2 (June 2004): 237–56. 2ck "K:KLM

———. Encyclopedia of Mongolia and the Mongol Empire. New York: Facts on File, 2004. ———. “How the Mongols Got a Word for Tribe—and What It Means.” Menggu shi yanjiu /Studia Historica Mongolica 10 (2010): 63–89. Aubin, Françoise. “Some Characteristics of Penal Legislation among the Mongols (13th–21st Centuries).” In Central Asian Law: An Historical Overview. A Festschrift for the Ninetieth Birthday of Herbert Franke, edited by Wallace Johnson and Irina F. Popova, 119–151. Lawrence, tY: Society for Asian Legal History, 2004. Āyatī, ʿAbd al-Muḥammad. Taḥrīr-i Tārīkh-i Waṣṣāf. Tehran: Muʼassasa-yi Muṭālaʻāt va Farhangī, 1993. Bar Hebraeus, Gregory John. The Chronography of Gregory Abū’l Faraj . . . commonly known as Bar Hebraeus, 2 vols. Translated by E. A. Wallis Budge. Repr. Amsterdam: Philo Press, 1976. Be+fa, Marie-Lise. “Le concept de tänggäri, ‘ciel’ dans l’Histoire Secrète des Mongols.” Etudes Mongoles et Sibériennes 24 (1993): 215–36. Biran, Michal. The Empire of the Qara Khitai in Eurasian History: Between China and the Islamic World. Cambridge: Cambridge University Press, 2005. Brosset, Marie-Félicité, trans. Histoire de la Géorgie, depuis l’antiquité jusq’au 898: siècle. Vol. 1. St. Petersburg: Imprimerie de l’Académie Impériale des Sciences, 1849. Bulliet, Richard W. “Naw Bahar and the Survival of Iranian Buddhism.” Iran 14 (1976): 140–45. DeWeese, Devin. Islamization and Native Religion in the Golden Horde: Baba Tükles and Conversion to Islam in Historical and Epic Tradition, Pennsylvania: Pennsylvania State University Press, 1994. Eliade, Mircea. Shamanism: Archaic Techniques of Ecstasy. Princeton, uv: Princeton University Press, 2004. Elverskog, John, Buddhism and Islam on the Silk Road. Philadelphia: University of Pennsylvania Press, 2010. Even, Marie Dominique. “L’Au-delà dans les représentations religieuses des Mongols.” In La mort et l’au-delà: une rencontre de l’Orient et de l’Occident, edited by Paul Servais, 149–96. Louvain-la-Neuve: Bruylant-Academia, 1999. George-Tvrtković, Rita. A Christian Pilgrim in Medieval Iraq: Riccoldo da Montecroce’s Encounter with Islam. Turnhout: Brepols, 2012. Gilli-Elewy, Hend. Bagdad nach dem Sturz des Kalifats. Berlin: Klaus Schwarz, 2000. Grigor of Akanc‘. History of the Nation of the Archers. Translated by Robert P. Blake and Richard N. Frye. Cambridge, Z#: Harvard University Press, 1954. Grupper, Samuel M. “The Buddhist Sanctuary of Labnasagut and the Il-Qan Hülegü: An Overview of Il-Qanid Buddhism and Related Matters.” Archivum Eurasiae Medii Aevi 13 (2004): 5–78. Ḥā-.ẓ-i Abrū. Ẕayl jāmiʿ al-tawārīkh Rashīdī. Edited by Khānbābā Bayānī. Tehran: Anjuman-i Āsār-i Millī, 1972. /5678 #9: %8; <5= 2c4

Hamayon, Roberte. La chasse à l’âme: esquisse d’une théorie du chamanisme sibérien. Nanterre: Société d’ethnologie, 1990. Humphrey, Caroline. “Shamanic Practices and the State in Northern Asia: Views from the Centre and Periphery.” In Shamanism, History and the State, edited by Nicholas Thomas and Caroline Humphrey, 191–228. Ann Arbor: University of Michigan Press, 1994. Ibn Bazzāz, Tawakkul ibn Ismāʿīl ibn Ḥājī Ardabīlī. Ṣafwat al-ṣafā. Edited by Ghulām Riz̤ā Ṭabāṭabāʾī Majd. Tehran: Intishārāt-i Zaryāb, 1997. Ibn al-Fuwaṭī [pseudo], ʿAbd al-Razzāq ibn Aḥmad. Al-Ḥawādith al-jāmiʿa. Edited by Muṣṭafā Jawād. Baghdad: Al-Maktaba al-ʿArabīya, 1932. Irwin, Robert. “What the Partridge Told the Eagle: A Neglected Arabic Source on Chinggis Khan and the Early History of the Mongols.” In The Mongol Empire and its Legacy, edited by Reuvan Amitai-Preiss and David O. Morgan, 5–11. Leiden; Boston: Brill, 1999. Jan, Yün-hua. “Chinese Buddhism in Ta-tu: The New Situation and New Problems.” In Yüan Thought: Chinese Thought and Religion under the Mongols, edited by Hok-lam Chan and William Theodore de Bary, 375–417. New York: Columbia University Press, 1982. Juwaynī, ʿ Aṭā Malik. Ta’rīkh-i Jahāngushā. Edited by Mīrzā Muḥammad Qazwīnī. London: Luzac, 1912. ———. The History of the World-Conqueror. Translated by John Andrew Boyle. Reprint, Manchester: Manchester University Press, 1997. Jūzjānī, ‘Uthmān ibn Sirāj al-Dīn. Tabakāt-i-Nāṣirī. 2 vols. Translated by Henry George Raverty. London: Gilbert & Rivington, 1881. Kennedy, Hugh. The Court of the Caliphs: The Rise and Fall of Islam’s Greatest Dynasty. London: Weidenfeld & Nicolson, 2004. Köprülü, Mehmed Fuad. “La proibizione di versare il sangue nell’esecuzione d’un membro della dinastia presso i Turchi e i Mongoli.” Annali dell’Istituto universitario orientale di Napoli, nuova serie 1 (1940): 15–23. Lane, George. “Persian Notables and the Families which Underpinned the Ilkhanate.” In Eurasian Nomads as Agents of Cultural Change: The Mongols and Their Eurasian Predecessors, edited by M. Biran and R. Amitai, 182–213. Honolulu: Hawaii University Press, 2015. Lange, Christian. Justice, Punishment and the Medieval Islamic Imagination. Cambridge: Cambridge University Press, 2008. Langlois, John D. “Law, Statecraft and the Spring and Autumn Annals in Yüan Political Thought.” In Yüan Thought: Chinese Thought and Religion under the Mongols, edited by Hok-lam Chan and William Theodore de Bary, 89–152. New York: Columbia University Press, 1982. Ma‘dan’kan, Ma‘ṣūmah. Bih yāsā rasīdagān dar ‘aṣr-i Īlkhānī. Tehran: Markaz-i Nashr-i Dānishgāhī, 1996. 2cJ "K:KLM

McKnight, Brian E. The Quality of Mercy: Amnesties and Traditional Chinese Justice. Honolulu: University of Hawaii Press, 1981. Melikian-Chirvani, Assadullah Souren. “The Buddhist Ritual in the Literature of Early Islamic Iran.” In South Asian Archeology: Proceedings of the Sixth International Conference of South Asian Archeologists in Western Europe held in Cambridge University, 5–10 July 1981, edited by Bridget Allchin, 272–79. Cambridge: Cambridge University Press, 1984. Melville, Charles. “Abu Sa’īd and the Revolt of the Amirs in 1319.” In L’Iran face à la domination mongole, edited by Denise Aigle, 89–120. Tehran: Institut français de recherche en Iran, 1997. Mustawfī Qazwīnī, Ḥamd Allāh. Ẓafarnāmah-i Ḥamd Allāh Mustawfī. 2 vols. Tehran: Markaz-i Nashr-i Dānishgāhī, 1999. al-Nasawī, Muḥammad ibn Aḥmad. Histoire du sultan Djelal ed-Din Mankobirti prince du Kharezm. 2 vols. Edited and translated by O. Houdas. Paris: Ernest Leroux, 1891–1895. Olschki, Leonardo. The Myth of Felt. Berkeley: University of California Press, 1949. Orbélian, Stéphannos. Histoire de la Siounie. 2 vols. Translated by Marie-Félicité Brosset. St. Petersburg, Académie Impériale des Sciences, 1864–1866. Pfei+fer, Judith. “Re)*ections on a ‘Double Rapprochement’: Conversion to Islam among the Mongol Elite during the Early Ilkhanate.” In Beyond the Legacy of Genghis Khan, edited by Linda Komaro+f, 369–389. Leiden: Brill, 2006. Polo, Marco. The Book of Ser Marco Polo, the Venetian, concerning the Kingdoms and Marvels of the East. 2 vols. Translated by Sir Henry Yule and revised by Henri Cordier. London: John Murray, 1903. Qāshānī, ʿAbd Allāh ibn ʿAlī. Tārīkh-i Uljāytū. Edited by M. Hambly. Tehran: Shirkat-i Intishārāt-i ‘Ilmī va Farhangī, 2005. Rachewiltz, Igor de. “Yeh-lü Chu-ts’ai (1189–1243), Yeh-lü Chu (1221–1285), Yeh-lü Hsi- Liang.” In In the Service of the Khan: Eminent Personalities of the Early Mongol-Yüan Period (1200–1300), edited by Igor de Rachewiltz et al., 136–75. Wiesbaden: Harrassowitz, 1993. Rashīd al-Dīn. Jāmiʿ al-tawārīkh. Edited by M. Raushan and M. Mūsavī. 4 vols. Tehran: Nashr-i Alburz, 1373/1994. ———. Jami‘u’t-tawarikh [sic] Compendium of Chronicles. Translated by Wheeler M. Thackston. 3 vols. Cambridge, Z#: Department of Near Eastern Languages and Civilizations, Harvard University, 1998–99. Ricoldo de Monte Croce. Pérégrination en Terre Sainte et au Proche Orient; Lettres sur la chute de Saint-Jean d’Acre. Edited and translated by René Kappler. Paris: Champion, 1997. Roux, Jean-Paul. La mort chez les peuples altaïques anciens et médiévaux. Paris: Maisonneuve, 1963. /5678 #9: %8; <5= 2cX

———. La religion des Turcs et des Mongols. Paris: Payot, 1984. Ruotsala, Antti. Europeans and Mongols in the Middle of the Thirteenth Century: Encountering the Other. Helsinki: Finnish Academy of Science and letters, 2001. Sayf ibn Muḥammad ibn Ya‘qūb al-Haravī. Tārīkhnāma-yi Harāt. Edited by Muḥammad Zubayr al-Ṣiddīqī. Calcutta: Khan Bahadur K. M. Asadullah, 1944. Sayılı, Aydın, The Observatory in Islam and its Place in the General History of the Observatory, Ankara: Turk Tarih Kurumu Basimevi, 1988. Schopen, Gregory. “Hinayana Texts in a 14th Century Persian Chronicle.” Central Asiatic Journal 26 (1982): 226–35. Seidensticker, Tilman. “Responses to Cruci-.xion in the Islamic World (1st–7th/7th– 13th Centuries).” In Public Violence in Islamic Societies, edited by Christian Lange and Maribel Fierro, 203–16. Edinburgh: Edinburgh University Press, 2009. Shabānkāra’ī, Muḥammad ibn ʿAlī ibn Muḥammad. Majma‘ al-ansāb. Edited by Mīr Hāshim Muḥaddith. Tehran: Amīr Kabīr, 1376/1997. Shirokogoro+f, Sergei M. “Versuch einer Erforschung der Grundlagen des Schamanismus bei den Tungusen.” Baessler-Archiv 18:2 (1935): 41–96. Sidky, H. “Ethnographic Perspectives on Di+ferentiating Shamans from Other Ritual Intercessors.” Asian Ethnology 69:2 (2010): 213–40. Tellenbach, Silvia. “Die Strafe im islamischen Strafrecht.” In La Peine/ Punishment, 4th part, 9–40. Bruxelles: De Boeck-Wesmael, 1991. Tucci, Giuseppe and Walther Heissig. Die Religionen Tibets und der Mongolei. Stuttgart: Kohlhammer, 1970. al-ʿUmarī. Das mongolische Weltreich: Al-‘Umarī’s Darstellung der mongolischen Reiche in seinem Werk Masālik al-abṣār fī mamālik al-amṣār. Edited and translated by Klaus Lech. Wiesbaden: Harrassowitz, 1968. Uray-Köhalmi, Käthe. “Sibirische Parallelen zur Ethnographie der Geheimen Geschichte der Mongolen.” In Mongolian Studies, edited by Louis Ligeti, 247–64. Budapest: Akademiai Kiado, 1970. Ward, L. J. “The Zafar-Namah of Hamdallah Mustau-. and the Il-Khan Dynasty of Iran.” PhD diss., Manchester University, 1983. Waṣṣāf al-Ḥaz̤rat, ʿAbd Allāh ibn Faz̤l Allah. Kitāb-i Waṣṣāf al-Ḥaz̤rat. Bombay: Muḥammad Mahdī Iṣfahānī, 1852/1853. Yuan Shi , edited by Song Lian . Beijing : Zhong hua shu ju , 1976. Zamchī Is-.zārī, Mu‘īn al-Dīn Muḥammad. Rawz̤āt al-jannāt fī awṣāf madīnat Harāt. Edited by Muḥammad Kāẓim Imām. Tehran: Dānishgāh-i Tihrān, 1959.

!"#$ % Culture and the Arts

∵

"#$%&'( ) Music in the Mongol Conquest of Baghdad: Ṣafī al-Dīn Urmawī and the Ilkhanid Circle of Musicians

Michal Biran*

Thinking about music in the context of the Mongol conquest of Baghdad (656/1258) usually brings to mind elegies lamenting the fall of the city, the caliphate, and occasionally the whole of Islamic civilization.* This article, however, examines a di+ferent kind of music; it introduces the story of Ṣafī al-Dīn Urmawī (613–93/1216–94), the master musician and singer of the last ‘Abbasid caliph. By dint of his virtuosity and ingenuity, Urmawī managed to save both himself and his neighbourhood from the conqueror’s wrath and secure an important post in the Ilkhanid regime. In the ./rst section of this work, I will translate and analyze the court musician’s personal account of the takeover, which is preserved in the Mamluk encyclopedia of Shihāb al-Dīn al-ʿUmarī (d. 749/1349). The second part consists of a preliminary examination of the fate of Urmawī and his circle of musicians after the fall.2 Their experiences comprise a case study for assessing the fortunes of ‘Abbasid culture under Mongol rule. At least in this particular ./eld, my ./ndings suggest that the termination of the ‘Abbasid regime neither eviscerated nor triggered a decline in Islamic culture. On the contrary: the Mongols’ ascension led to the continued 34ourishing of the ‘Abbasid musical school and to its further enrichment.

* The research leading to these results has received funding from the European Research Council under the European Union’s Seventh Framework Programme (5%/2007–13) /'(" Grant Agreement n. 312397. I would like to thank Reuven Amitai for his valuable comments on an earlier draft of this article. 6 For poems lamenting the fall of Baghdad, see Hassan, “Loss of Caliphate,” 41–49; and Abū Rukun [sic], “The Fall of Abbasid Baghdad,” 81–101. 9 Ilkhanid musicians were deftly described in both ʿAzzāwī, al-Mūsīqā al-ʻIrāqīyah, 22–49 and Neubauer, “Musik zur Mongolenzeit in Iran,” 233–60, esp. fn. on 245. I will eschew rehashing their contributions, from which I bene./ted greatly. Instead, I will examine the material from a slightly di+ferent vantage point.

Ṣafī al- Dīn al-Urmawī’s AccountB

ʿAbd al-Muʾmin b. Yūsuf b. Fākhir Ṣafī al-Dīn Urmawī was one of the more illustrious musical artists and theoreticians in the Muslim world.D Born in Urmiya (a city in modern-day Azerbaijan province in Iran), he arrived in Baghdad as a young boy. Urmawī launched a career in jurisprudence at the newly established Mustanṣiriyya College (opened 631/1234), which would soon become the city’s most renowned madrasa. Like many scholars of his time, Urmawī was a polymath. In addition to his expertise in Sha./‘ite and comparative law, he was well versed in calligraphy, Arabic, poetry, history, mathematics, and of course music. By the age of 21, he had already completed his magnum opus, Kitāb al-adwār (The book of cycles), a systematic exposition on the modal system, which was one of the most in34uential works on Islamic music theory.F Like its author, who was also an accomplished singer and lute player, the book combines musical theory with practice. At one and the same time, Urmawī made a name for himself in the discipline of calligraphy. When the last ‘Abbasid caliph, al-Mustaʿṣim bi-Allāh (r. 640–656/1242–58), set up his own library and looked for copyists, Ṣafī al-Dīn was hired due to his excellent penmanship. He was elevated to court musician relatively late in al-Mustaʿṣim’s reign, thanks to the recommendation of his former student, the caliph’s songstress Luḥāẓ. Soon after, he became one of the emperor’s closest companions, even tutoring his sons. Moreover, Urmawī had the ear of the Caliph’s top o+./cials. In the process, he acquired considerable wealth, as his musical talents netted him a generous annual pension of 5,000 dinars.I Urmawī’s experiences during the Mongol conquest of Baghdad are recorded in the volume on musicians (vol. 10) of the encyclopedia Masālik al-abṣār fī mamālik al-amṣār (Paths of discernment into the kingdoms of the lands),

J Long after this article was sent to the editor, I found out that this part was translated by G. J. V. Gelder. See his “Sing Me to Sleep,” 1–9. His focus is, however, much narrower than mine, nor was he interested in the Mongol facet of this story. K On Urmawī’s stature in the ./eld of Islamic music, see e.g., Shiloah, Music in the World of Islam, 55–58, 111–23; and Neubauer, E. “Ṣafī al-Dīn Urmawī.” In Encyclopaedia of Islam, Second Edition; online version at http://www.brillonline.nl/subscriber/entry?entry=islam_SIM-6447. L For a recent description and evaluation of sundry versions, see Wright, “The Modal System”; Idem, “A Preliminary Version of the ‘Kitāb al-Adwār,’” 455–78; and below. M Neubauer, “Ṣafī al-Dīn Urmawī,” and the references therein. See above all al-Kutubī, Fawāt al-wafayāt, 2:31–32; al-Ṣafadī, al-Wāfī bi’l-wafayāt, 19:342–43; al-ʿUmarī, Masālik al-abṣār, 10:350–51; Ibn al-Ṭiqṭaqā, al-Fakhrī fī al-ādāb al-sulṭāniyya, 74, 449–50; and al-Fakhrī/ Whitting, 49, 317. PQR;S ;> TUV PW>XWY SW>ZQVRT W[ :=XU\=\ ?@] which was compiled by the Mamluk historian and administrator Shihāb al-Dīn al-ʿUmarī, one of the most knowledgeable Mamluk historians on the subject of the Mongols.^ ʿUmarī’s account is based on the testimony of al-ʿIzz al-Irbilī, alias al-Ḥasan b. Aḥmad b. Ẓafar (d. 726/1326), a physician and teacher who migrated to Damascus from the Ilkhanate. Al-Irbilī was also described as a historian who penned, inter alia, many “unique biographies” (tarājim gharība).a Furthermore, he is cited in Mamluk histories on the Ilkhan Ghazan (r. 695– 704/1295–1304) and his famous vizier, Rashīd al-Dīn (d. 718/1318),b thus he probably left the Ilkhanate after Ghazan’s reign. With this short survey behind us, let us now turn to al-Irbilī’s account of Urmawī’s fortunes in the midst of the Mongol takeover of Baghdad:*c al-ʿIzz al-Irbilī mentioned in his History:

I was sitting with ʿAbd al-Mu’min [i.e., Urmawī] in the Mustanṣiriyya College and [the topic of] the fall of Baghdad (wāqiʿat-Baghdād) came up. He told me that Hülegü summoned the city’s leaders and notables (ru’asā’ al-balad wa-ʿurafā’ahu) and asked them to divide the gated quarters (durūb)** of Baghdad and its neighborhoods (maḥālihā) and the houses of its people of means (buyūt dhawī yasārihā) between the commanders of his dynasty. They divided them and allotted every neighborhood or two or every two markets to a great amir. The quarter (darab) in which I lived was allotted to a commander of 10,000 riders, named

d The whole third volume of ʿUmarī’s encyclopedia is devoted to the Chinggisids. This part had been edited, annotated and translated into German by Lech. See al-ʿUmarī, Das Mongolische Weltreich. e al- Ṣafadī, A‘yān al-‘aṣr wa-a‘wān al-naṣr, 2:188–89; al-Ṣafadī, al-Wāfī, 12:239; Ibn Ḥajar, al-Durar al-kāmina, 2:92; and Ibn Kathīr, al-Bidāya wa-l-nihāya, 14:144. f For example, see al-Ṣafadī, A‘yān, 4: 9, 14, 43; and ʿUmarī, Masālik, 10:193. 6g ʿ Umarī, Masālik, 10:353–56 (or the facsimile edition: ed. F. Sezgin with A. Jolhosha, E. Neubauer, 10:311–15). ʿUmarī’s text is quoted, with minor changes, in al-Ḥamawī, Thamarāt al-awrāq, 461–66 (I would like to thank Prof. Michael Lecker for referring me to this source). The story is cited in ‘Azzāwī, al- Mūsīqā, 27–31; and Nājī Maʿrūf, Tārīkh ʿulamāʾ al-Mustanṣirīya, 1:270–74. A Hebrew summary, based on ‘Azzāwī’s work, is included in Abū Rukun [sic], “The Fall,” 117–20. The latter dates al-ʿIzz al-Irbilī’s death to 660/1262 and claims that the story is found in Urjūzat al-anghām (p. 117 and n. 514)—a treatise of music by another person with a nearly identical name: Badr al-Dīn Irbilī (686–755/1287– 1354). However, the Urjūza, which is included in ‘Azzāwī’s book (106–17), does not contain this story. 66 For a discussion on darab (pl. durūb) and its meanings, see Eickelman, “Is there an Islamic City?,” 274–94, esp. 283. Dr. Nimrod Luz graciously brought this source to my attention. ?@) :;<=>

Bānū Noyan.*2 Hülegü allowed some of the commanders to kill, capture, and loot for three days, some for two days, and others for only one day, according to their ranks. When the commanders entered Baghdad’s ./rst gated neighborhood, namely the one where I was living, quite a few people of means were gathered there. And at my place, around ./fty of the notable singers (a‘yān al-mughānī), who had property and beauty, were assembled. Bānū Noyan stopped*B at the gate of the quarter, which was barricaded*D with wood and earth. They knocked on the gate and said: ‘Open the gate and obey us, and we will give you safe conduct (amān). And if not, we will burn the gate and kill you.’ Accompanying him [Baiju] were the naptha throwers [zarrāqūn], the carpenters [najjārūn],*F and his armed followers. ʿAbd al-Mu’min said: ‘I’ll go out to him in submission and obedience.’ I opened the gate and went out to him on my own, wearing dirty clothes and awaiting death. I kissed the ground in front of him and he told the interpreter: ‘Ask him: Who are you? Are you the leader of the quarter’s residents?’ I said: ‘Yes.’ He said: ‘If you want to save your life, bring us this and that’—and he asked for quite a lot. I kissed the ground for the second time and said: ‘All that the amīr asked for will be brought, and everyone in this neighborhood will be under your rule. Order your armies to loot the other neighborhoods assigned to them and stay [here] so I can host you along with anyone you want from your retinue. I will collect all that you have asked for.’ He [the commander] consulted with his followers and came with thirty people. I brought him to my house and spread out the precious caliphal carpets and the silk covers (sutūr) embroidered with gold. I immediately served him food—fried and roasted dishes and

69 I assume that this is referring to Baiju Noyan (often rendered Bājū in Arabic sources). Baiju (34. 625–57/1228–59) was the Mongol general and military governor in northwestern Iran. When Hülegü advanced westward, Baiju was among the generals under his command. He distinguished himself in the Baghdad campaign, as his forces subdued the western part of the city. See P. Jackson, “Bāyjū” in Encyclopædia Iranica. 6J While the facsimile and Ibn Ḥijja’s version read waqafa, ʿUmarī’s 2010 edition reads waqata. 6K ʿUmarī’s edition reads mudabbas/mudbis, which can be understood as ‘chained’ (deriving from dabbūs or chain); see Piamenta, Dictionary of Post-Classical Yemeni Arabic, 143. Ibn Ḥijja’s version employs the more suitable “mutattaris,” a derivative of mitras: barricade, barrier, or rampart. Hans Wehr’s dictionary, 93, http://ejtaal.net/m/ aa/#HW=111,LL=1_339,LS=2,HA=76. 6L During the campaign, the carpenters probably built siege machines. PQR;S ;> TUV PW>XWY SW>ZQVRT W[ :=XU\=\ ?@h

sweets—and ate a sample [of the dish] in his presence.*I When he ./n- ished eating, I arranged a royal assembly (majlisan mulūkiyan) for him and brought him gilded dishes from Aleppo [and] glass and silver dishes ./lled with distilled beverages (sharāb murawwaq). When the drinking cups had made a round and he got a little tipsy, I brought ten singers, all of them women, each of whom sang a di+ferent delightful tune (malhāt). I gave them an order, and they all sang in unison. The assembly was animated. He [Baiju] was excited by the music and his soul was satis./ed. He hugged a singer he liked and had intercourse with her (wāqaʿahā) during the assembly, while we were watching. His day came to a close in the best possible way. By the time of the evening prayer, his men arrived with the loot and the prisoners [from other neighbourhoods]. Apart from the food (ʿalīq)*^ and the gifts of wine (hibāt al-ʿawāniyya) that were before him, I brought them wonderful presents of gold and silver dishes and coins, cash, and splendid cloths; I apologized for the dearth [of gifts] and told him: ‘The commander came without notice, but tomorrow, God willing, I will invite the commander to a better feast than this.’ He rode and I kissed his mount, whereupon I returned [to the neighbours]. I gathered the rich people of the quarter and told them: ‘Look out for yourselves; this man will be here tomorrow and the day after as well. Each day, I want to double [the volume of the gifts from] the previous day.’ They collected from their houses all kinds of gold, precious cloths, and arms worth 50,000 dinars. He already came to me before sunrise the next day, and what he saw left him astonished. On this day, he [Bānū] arrived with his wives, and I gave him and his wives precious gifts, gold, and cash worth 20,000 dinars. On the third day, I gave him precious pearls, expensive gems, and a beautiful jennet with caliphal gear, saying: ‘This is the caliph’s mount.’ I served all those who were with him and said: ‘This quarter is already under your command; and if you grant its people their lives, he [i.e., Hülegü] will be blameless in the eyes of God and men, for all that is left to them is their souls.’ He said: ‘I know this. From the ./rst day I gave them their souls, and my soul did not tell me to kill or capture them. But before doing anything else, you

6M The purpose behind this action was probably to demonstrate that the food was not poisoned. 6d While the principal meaning of ʿalīq is ‘fodder,’ it is also used as a metaphor for wine or anything eaten; Lane’s dictionary, 10:421 http://ejtaal.net/m/aa/#HW=652,LL=5_421,LS=2, HA=506,HW_HIDE. ?@i :;<=>

should come with me to see the Qan. I already mentioned you to him and I brought him one of the objects you gave me during the ./rst [few] days. He liked it and ordered [me] to summon you.’ I feared for my life and for the people of the quarter, and I said [to myself]: ‘He will take me out of Baghdad, kill me, and loot the quarter.’ I became apprehensive and said: ‘Oh, khūnad [lord], Hülegü is a great king and I am a lowly man, a singer; I am afraid and in awe of him.’ He said: ‘Don’t be scared, only good things will befall you; he [the Qan] is a man who likes the gifted (ahl al-faḍā’il).’ I said: ‘Can you guarantee that nothing bad will happen to me?’ He said: ‘Yes.’ So I told the people of the quarter: ‘Bring your precious things and give me all you have: splendid singers and a substantial amount of money in gold and silver.’ I brought from my place loads of good food, plenty of wine—old and excellent—and beautiful dishes—all of it from gold and inscribed silver. I took with me three of the most beautiful singers and the best [lute] players. I wore a suit of caliphal cloth and rode on a beautiful jennet, which I used to ride on when I went to the caliph. When Bānū Noyan saw me in this [out./t], he said: ‘You’re a minister!’ And I said: ‘Indeed, I am the caliph’s singer and his companion, but so long as I feared you, I wore those tattered and ./lthy clothes. When I became [one of] your subjects, my status was restored and I felt secure. Hülegü is a great king, greater than the caliph, and I can only enter his presence with courtliness and dignity. He liked my response, and I went with him to Hülegü’s camp. He entered his presence, brought me with him, and said to Hülegü: ‘This is the man I mentioned’ and pointed at me. When Hülegü’s eyes fell on me, I kissed the ground and knelt as is the Tatars’ custom. Bā[nū] Noyan told him: ‘This [man] was the caliph’s singer and did so and so for me; he [already] brought you presents.’ He [Hülegü] said: “Raise him!” and they raised me. I kissed the ground for the second time and wished him well. I o+fered him and his retinue the gifts that I took with me. And whenever I gave him something, he asked about it, then left it. Thereafter, he did the same with the food before asking me: ‘You were the singer of the caliph?’ I said: ‘Yes.’ And he asked: ‘What is the best thing you know in the science of music (ʿilm al-ṭarab)?’ I said: ‘The best is a song that I sing which causes the listener to fall asleep.’ He said: ‘Sing to me for a while, until I fall asleep.’ I regretted [my words], saying [to myself]: If I sang for him and he does not fall asleep he would say: ‘This one is a liar,’ and he may kill me. I must accomplish this with a ruse. And I said: ‘Oh Lord, striking the lute’s chords is only good while drinking wine. Let the king drink two or three cups so that the music will fall into place.’ He said: ‘I don’t want the wine, PQR;S ;> TUV PW>XWY SW>ZQVRT W[ :=XU\=\ ?@j

[for] it takes my mind away from the interests of my kingship; I’m impressed with your prophet who forbade it’; whereupon he imbibed three large cups. When his face turned red, I asked for his permission and sang to him. I had with me a singer called Ṣabā. In [all of] Baghdad there was not [a singer] more beautiful than her or with a better voice. She excelled at playing the lute, picking out gentle melodies that lulled him to sleep. She sang, and before I ./nished the piece I saw him drowsing o+f already. I stopped the singing abruptly and struck a strong note, which woke him up. I kissed the ground and said: ‘Did the king fall asleep?’ And he said: ‘You were right, I fell asleep; now what do you wish from me?’ I said: ‘I request that the king give me a sumayka [tri34e].’*a He asked: ‘And what tri34e would you like?’ I said: ‘A garden that belonged to the caliph.’ He smiled and said to his retinue: ‘This poor singer is short-sighted.’ He then said to the interpreter: ‘Why didn’t you ask for a fortress or a town; what is a garden?’ I kissed the ground and said: ‘Oh King, this garden will su+./ce. What me—the governor of a castle or town?!’ So he allotted me the garden and the full bene./ts that I had when the caliph reigned and added a pension that included bread, meat, and two dinars worth of fodder for my mount. Accordingly, he issued a signed and sealed order (+,rmān mukmil al-ʿalāʾim) on my behalf and I departed. Bānū Noyan chose a commander with ./fty riders, who [bore] a black standard that was Hülegü’s special standard, to protect my quarter. The commander sat at the quarter’s gate. He hung the black standard at the top of the quarter’s gate, and it stayed this way until Hülegü left Baghdad. Al-Irbilī asked: ‘How much did you lose in the process?’ He [ʿAbd al- Mu’min] replied: ‘Over 60,000 dinars of gold, the majority of which was brought to my quarter by the rich people and the rest from various luxuries (na‘am) that the caliph had bestowed unto me.’ I [al-Irbilī] asked him about the salary and the garden. He said: ‘The caliph’s children took it [the garden] from me, saying: ‘This was an inheritance from our father. And the pension was cut o+f by Ṣāḥib Shams al-Dīn al-Juwaynī, but he compensated me for it and for the garden with 60,000 dirhams.’

While it is di+./cult to verify the details in Urmawī’s vivid account, its main con- tours seem plausible enough, as they echo Ilkhanid sources; in addition, many of the elements dovetail neatly with other contemporaneous descriptions of Baghdad’s fall. The much less heroic Persian version of this story appears in the history of Waṣṣāf (d. 729/1329), who describes Urmawī as an unequalled

6e Sumayka is literally a small ./sh. ?Ak :;<=> master in the science of music, a second Pythagoras.*b According to Waṣṣāf, as the Mongols were occupying the city, Urmawī appeared on the threshold of Hülegü’s tent and began playing music. Amid the chaos, he performed from morning to evening, but nobody paid any attention. When Hülegü was informed of the situation, he summoned the musician before him, praised his performance and took him under his aegis. More speci./cally, Urmawī was granted an annual pension of 10,000 dinars (twice the amount of his caliphal stipend) from the government revenues of Baghdad. Additionally, the payment would accrue to his progeny.2c Urmawī’s version strengthens the impression made by other sources that the sack of Baghdad was a meticulously organized campaign in which the troops fully adhered to Hülegü’s orders,2* rather than a sudden outburst of barbarism. Moreover, this story o+fers a di+ferent picture from the regular descriptions of total massacre. The ability of both a local commander and Hülegü himself to enjoy a concert before the city was fully secured (according to most descriptions the Caliph was still alive by the time of the city’s pillage), certainly manifest the con./dence of the Mongols in the result of their campaign.22 This is also reminiscent of Chinggis Khan’s conquest of Bukhara in around 1220, when the assailants appropriated singing girls and wine from the defeated

6f Wa ṣṣāf, Tẚrīkh-i Waṣṣāf, 55; Āyatī, Taḥrīr, 32. In the Muslim world, Pythagoras was as renowned for his musical exploits as his mathematical prowess, and music was deemed to be a science that is close to mathematics. 9g Wa ṣṣāf, Tarīkh, 42–43; Āyatī, Taḥrīr, 23; Khwāndamīr, Tẚrīkh-i ḥabīb al-siyar, 3:107; Idem, (Thackston), 3:60. According to the latter, during the massacre and plunder of Baghdad, Urmawī “crept into a corner and one day presented himself in the vicinity of Hülegü’s tent.” Citing Ḥabīb al-siyar as his source, Ḥājjī Khalīfa claims that Urmawī simply presented himself to Hülegü when the Ilkhan triumphantly entered the city. Impressed by his lute playing, Hülegü forbade his troops from touching the musician’s property. Ḥajji Khalīfa, Kashf al-ẓunūn, m:874. 96 E.g. Boyle, “The Death of the Last ʿAbbāsid Caliph,” 160 (retrieving Naṣīr al-Dīn Ṭūsī’s account); Ibn al-Fuwaṭī, Kitāb al-ḥawādith, 360, and Gilli-Ellevy, “Al-Ḥawādit al-ğāmi‘a,” 367; Rashīd al-Dīn, Jāmi‘ al-tawārīkh, ed. B. Karīmī , 2:713 [hereafter Rashīd/Karīmī] and Jamiʿu’t-tawarikh [sic] Compendium of Chronicles, translated by Wheeler M. Thackston, 3:498 [hereafter Rashīd/Thackston]; al-Nuwayrī, Nihāyat al-arab, 23;234; Waṣṣāf, Tẚrīkh, 43; and Āyatī, Taḥrīr, 23. 99 Mustawfī also states that Hülegü arranged a banquet, which included wine, singing girls, and Mongolian music after the capture of Baghdad and before the caliph’s execution (L. J. Ward, Ẓafarnāmah of Mustawfī, 2:122). However, his version of the conquest is so far-fetched that it calls into question the tenability of this particular anecdote as well. Mustawfī also claims that singers and musicians entertained Hülegü in Samarqand, before the army penetrated the Middle East; ibid., 2:118. PQR;S ;> TUV PW>XWY SW>ZQVRT W[ :=XU\=\ ?A? populace.2B As we shall see, the Mongols’ documented a+./nity for vinous feasts that were enlivened with music lends credence to these two stories. Urmawī provides much more information on the looting—a phase that was practically ignored by contemporaneous accounts, but described in apocalyptic terms by other, mostly later, sources.2D To my knowledge, none of the other versions refer to the notables dividing up their city into plunder zones or to Hülegü’s premeditated assignment of di+ferent looting spans for various ranks. In any event, these periods—ranging from one to three days—support the contention that the sack of Baghdad lasted for a week, as opposed to those sources claiming that it dragged on for 30 to 40 days.2F Although the tactic of foisting a ‘pay or die’ proposition on the defeated populace was not alien to the Mongols, Urmawī is the only one to note that it was imposed on the residents of Baghdad after the conquest: if the Mongols proposed such an option it was usually before they attacked the city and only if the city o+fered no opposition.2I That said, certain other Baghdadi groups are known to have received safe conduct (amān) from Hülegü, often in return for exorbitant sums of money. While the details vary from source to source, most of them agree that amān was granted to the city’s Christians, the Shi‘ites from Ḥilla, merchants from Khurasan (who already had relations with the Mongols), and several Muslim notables.2^ The Shi‘ites, merchants, and

9J Juwaynī/Qazwīnī 1:80–81; Juwaynī/Boyle, 207; cited in Allsen, “Command Performances,” 38. 9K For laconic descriptions or complete omissions, see Boyle, “Death,” 160; Bar Hebraeus, The Chronography, 1:431; Kāzarūnī, Mukhtaṣar al-tẚrīkh, 270–74; Rashīd/Karīmī, 1:713 and Rashīd/Thackston, 2:498; Shīrāzī, Akhbār-i Mughūlān, 34; and Nuwayrī, Nihāyat al-arab, 27:383, 23:234–35, among others. For apocalyptic descriptions, see, e.g., al-Subkī, Ṭabaqāt al-sha+,ʿiyya al-kubrā, 8:261–77; Waṣṣāf, Tẚrīkh, 38–39; and Āyatī, Taḥrīr, 20. 9L The sources according to which the plunder lasted a week are, e.g., Boyle, “Death,” 160; Bar Hebraeus, The Chronography, 431; Rashīd/Karīmī, 1:713 and Rashīd/Thackston, 2:498; Nuwayrī, Nihāyat al-arab, 27:383; those citing a period ranging from 30 to 40 days include, among others, Kitāb al-ḥawādith, 359; Ibn al-Sāʿī [Pseudo], Kitāb? mukhtaṣar akhbār al-khulafā, 136; Ibn al-Kathīr, al-Bidāya, 13:236; Waṣṣāf, Tẚrīkh, 42; and Āyatī, Taḥrīr, 23. 9M Further on this issue see ʿUmarī/Lech, 102, and the discussion in Biran, “Violence and Non-Violence.” 9d For example, see Kitāb al-Ḥawādith, 359, 360 and Gilli-Elewy, “Al-Ḥawādit,” 367, 368; Boyle, “Death,” 159. According to Ṭūsī, scholars, shaykhs, and whoever o+fered no resistance to the Mongols were o+fered amān. He also claims that this option was suggested at the early stages of the conquest. Such an option is also mentioned by Rashīd al-Dīn (Rashīd/Karīmī, 1:710, and Rashīd/Thackston, 2:496) as o+fered to Qāḍis (judges), scholars, shaykhs, ʿAlīds, Nestorian priests and “persons who do not combat against us.” Christians were spared according to Bar Hebraeus, Chronography, 1:431, and according to the Arabic version of his work, Christians, Shi‘ites, and scholars avoided the sword: Ibn al-‘Ibrī (Bar ?Ap :;<=> perhaps others indeed secured their lives at considerable cost.2a As in the case of Urmawī’s quarter, Mongol guards were dispatched to guarantee the safety of the Christians and merchants.2b In all these instances, neighbours 34ocked to those who received protection, in the hopes of saving their own lives. Bc Another interesting element of this narrative is the singers’ prosperity. Urmawī’s description of the ./fty singers in his upscale neighbourhood is indeed commensurate with the last ‘Abbasid Caliph’s well-established interest in, or obsession for, music. In fact, his zealous patronage of this art form is often cited as one of the reasons that he neglected his duties and com- promised his ability to cope with the Mongol threat.B* The riches that these singers were able to fork over to their captors following the two-week siege that the city had just endured are quite impressive. Urmawī’s reputation as a hedonist notwithstanding,B2 viewing the luxuries that he amassed in jux- taposition to the oft-repeated descriptions of the unpaid troops of Baghdad before the Mongol attackBB lends credence to the complaints as to the decadence of the city’s upper class.BD It also calls to mind the famous, apocryphal meeting between Hülegü and al-Mustaʿṣim in the defeated sultan’s treasury, at which the former asked his counterpart why he had refrained from using his ample holdings to build a capable army.BF Urmawī’s description indeed highlights the huge economic disparities in the city as well as the lack of city-wide solidarity. While the musician is deeply concerned about his quarter’s fate,

Hebraeus), Tẚrīkh mukhtaṣar al-duwal, 271. Ibn Kathīr notes that the Jews were saved as well; idem, al-Bidāya, 13:235. 9e The following sources refer to the Shi‘ites: Kitāb al-ḥawādith, 360 and Gilly-Elewy, 368; and Ibn Ṭāwūs, Iqbāl al-a‘māl, 63, 65. For the payment exacted from the merchants, see Ibn Kathīr, al-Bidāya, 13:235. 9f Kitāb al-ḥawādith, 359 and Gilli-Elewy, 367. Jg Ibid.; Bar Hebraeus, The Chronography, 1:431. J6 See, above all, Ibn al-Ţiqṭaqā, al-Fakhrī, 63–65; al-Fakhrī/Whitting, 42–43. The most beautiful story cited herein involves a letter that the caliph wrote to Badr al-Dīn Lu’lu’, Atabeg of Mawṣil, asking him for a company of musicians. While writing the letter, Hülegü’s messenger arrived with a request for catapults and siege machines. This coincidence moved Badr al-Dīn to utter the following words: “Look at the two requests, and weep for Islam and its people!” For more on Mustaṣim’s fondness for music, see Ibn Junayd, al-Mustaʻṣim billāh al-‘Abbāsī, 51–53. J9 E.g., Kutubī, Fawāt, 2: 32. JJ E.g., Kitāb al-ḥawādith, 304, 321, 331, 350 and Gilly-Elevy, 359, 361; Subkī, Ṭabaqāt, 8:262. JK E.g., Ibn al-Ṭiqṭaqā, al-Fakhrī, 63; al-Fakhrī/Whitting, 42. JL E.g., Waṣṣāf, Tẚrīkh, 39 and Āyatī, Taḥrīr, 21; Le Strange, “The Story of the Death of the Last Abbasid Caliph,” 293–300, including the references therein. According to some sources, like Marco Polo (cited in Le Strange), the Caliph is left to starve in his treasury as punishment for his malfeasance. PQR;S ;> TUV PW>XWY SW>ZQVRT W[ :=XU\=\ ?A@ he has no compunction about advising the Mongols to loot other neighbourhoods. Similarly, he does not condemn the rape that was perpetrated at his home. In all likelihood, these factors contributed to the city’s defeat. While there is no doubt that Urmawī was terri./ed by the Mongol onslaught, which may very well account for his above-mentioned behaviour, he considered the Mongols’ actions as a hateful but legitimate prerogative of the conquerors, and they did not prevent him from throwing in his lot with them. Moreover, his description falls under the heading of the relatively mundane accounts of the fall of Baghdad. In fact, it stands in stark contradistinction to the more apocalyptic versions that inform some Mamluk sources and many contemporary Arabic works.BI What is more, Hülegü’s interest in ‘gifted’ people, not least his generosity towards the musician in question, are incompatible with the (exceedingly modern) myth that the ‘Tatars’ destroyed Islamic culture.B^ They ./t well, however, with Rashīd al-Dīn’s description of Hülegü as “A great lover of wisdom, [Hülegü] encouraged the learned to debate the basic sciences and rewarded them with stipends and salaries. His court was adorned by the presence of scholars and wise men (‘ulamā’ wa ḥukamā’).”Ba Interestingly, the other ‘man of talent’ known to receive Hülegü’s auspice during the conquest—a commander and scribe (amīr kātib) who excelled in calligraphy, belles-letters and horsemanship, and was famed for his beauty—was also closely connected to the Caliph’s court.Bb It can be argued that Hülegü sought to enhance his kingly reputation by employing representatives of ‘Abbasid glory at his court. At the very least, the Mongols embraced those cultural elements of the ‘Abbasid court that suited their own norms or bolstered their legitimacy. The fate of Urmawī in Ilkhanid Baghdad reinforces this vantage point.

JM See above; Gilly-Elevy, 371 and e.g. al-Sarjānī, Qiṣṣat al-Tatār min al-bidāya ilā ‘Ayn Jālūt, 101–70, 271–76; Manṣūr, Qiṣṣat suqūṭ Baghdād, 96–98. The American conquest of Baghdad in 2003 revived the memory of the Mongol conquest but in a rather distorted way. See Biran, “Violence.” Jd While the cultural rami./cations of Baghdad’s fall are among the most lamented developments in modern Muslim literature, the treatment of this topic in contemporaneous sources is rather scanty. See Biran, “Violence.” Je Rashīd/Karīmī, 2:734 and Rashīd/Thackston, 2:513 (though he translates ʿulamā’ wa ḥukamā’ as “philosophers and scientists”). For more on this issue see Reuven Amitai, “Hülegü and his Wise Men.” For Hülegü as a humanist and patron of scholars see also Lane, Early Mongol Rule. Jf This is Falak al-Dīn Muḥammad b. Sayf al-Dīn Aydamir al-Musta‘ṣimī (639–710/1240– 1310), whom Hülegü appointed “the supervisor (shiḥna) of the wise men (ḥukama’) who found refuge in his court and were dealing with chemistry.” Ibn al-Fuwaṭī, Talkhīṣ majmaʿ al-ādāb, 3:281; and see the discussion in Biran, “Violence.” ?AA :;<=>

Urmawī’s Circle of Musicians in the Mongol World

After the conquest, Urmawī remained in Baghdad, but apparently stayed in touch with Hülegü. The polymath wrote a scroll (darj) on the Ilkhan’s behalf, which the latter was quite pleased with, and it seems as though he paid his sponsor a visit in Azerbaijan. Moreover, Hülegü appointed him inspector of endowments in Iraq, a position he held until Naṣīr al-Dīn Ṭūsī, an even more acclaimed scholar, took over in 665/1267. Urmawī was a highly respected and very popular inspector. He is even said to have attained “the status of Hülegü himself” among the Iraqi population.Dc In Baghdad itself, the polymath became a companion and bene./ciary of the Juwaynī brothers: ʿAlāʾ al-Dīn ʿAṭā Malik (623–81/1226–83), the famous historian and Mongol governor of Baghdad (657–680/1259–81) and Shams al-Dīn (d. 682/1284), the chief minister of Hülegü and his successor, Abaqa (r. 663–681/1265–82). The Juwaynīs appointed Urmawī to head the Baghdad Inshāʾ (chancellery), most likely on account of his calligraphic skills. In addition, he taught music and calligraphy to the sons of the city’s notables, including those of Shams al-Dīn Juwaynī, Bahā’ al-Dīn and Sharaf al-Dīn Hārūn. The latter showed great interest in music and subsequently became his teacher’s patron. In turn, Urmawī dedicated al-Risāla al-Shara+,yya +,’l-nisab al-taʾlī+,yya (The Shara./an treatise on musical proportions, c. 665/1267)—his most acclaimed work on music theory aside from the Kitāb al-adwār—to Sharaf al-Dīn Hārūn (put to death in 685/1285).D* Following the Juwaynīs’ ouster in the early–mid 1280s, Urmawī lost his administrative post and fell into poverty. He therefore tried his luck in Tabriz, where he met al-ʿIzz al-Irbilī in 689/1290.D2 Moreover, he received some ./nancial assistance from Quṭb al-Dīn Shīrāzī, another distinguished Ilkhanid polymath, who wrote extensive commentaries on Ṣafī al-Dīn’s works.DB The musician eventually returned to Baghdad, where he died in 693/1294 while imprisoned for a debt of 300 dirhams.DD Two of his three sons were employed as scribes (kātib) in the Baghdad administration, and the third was a member of the city’s

Kg ʿ Umarī, Masālik, 10:351–52; Ibn al-Fuwaṭī, Talkhīṣ, 3:319–20. K6 Ibid.; Kutubī, Fawāt, 2:31; al-Ṣafadī, al-Wāfī, 19:342–43; see also Waṣṣāf, Tẚrīkh, 66 and Āyatī, Taḥrīr, 36. For more on the Juwaynīs and their support of the arts, see, e.g., Lane, Early Mongol Rule, 177–212. K9 Kutub ī, Fawāt, 2:31; al-Ṣafadī, al-Wāfī, 19:342. KJ Ibn Ḥajar, Durar, 5:108–9. KK Kutub ī, Fawāt, 2:32; al-Ṣafadī, al-Wāfī, 19:343. PQR;S ;> TUV PW>XWY SW>ZQVRT W[ :=XU\=\ ?A] intellectual community.DF Despite his ultimate fall from grace, Urmawī became synonymous with music in both the Muslim and Mongol worlds. Waṣṣāf considered him one of the four leading scholars of the Abaqa era, sharing this honour with his close associates: Naṣīr al-Dīn Ṭūsī, the astrologer and philosopher; the minister Shams al-Dīn Juwaynī, and Yāqūt al-Mustaʿṣimī, the calligrapher.DI Urmawī’s theoretical works were extremely well received in the Ilkhanate and would continue to serve as lynchpins of music scholarship for generations to come. The most popular of these texts was Kitāb al-adwār, which merited an extensive commentary (as did al-Risāla al-Shara+,yya) from Quṭb al-Dīn Shīrāzī.D^ The ruler of Fars, Jamāl al-Dīn Abū Isḥāq Īnjū (r. 742–756/1342–56), commissioned ‘Imād al-Dīn Yaḥyā b. Aḥmad Kāshī to render Kitāb al-adwār into Persian and write an extensive commentary on this treatise. It continued to be translated, interpreted, and disseminated for hundreds of years in the Jalayirid, Timurid, Ottoman, and Safavid realms. In the early eighteenth century, the book was rendered into French by Pétis de la Croix, in what was the ./rst of several Western translations.Da As we can see, the ‘Abbasid theory of music continued to evolve long after the empire’s demise. During his heyday under the Juwaynīs, Ṣafī al-Dīn Urmawī was the centre of a wide circle of students and colleagues. His most renowned protégés (apart from the Juwaynīs) were Yāqūt al-Mustaʿṣimī (ca. 618–98/1221–98), who studied both calligraphy and music with Urmawī at the caliph’s court and earned

KL Ibn al-Fuwaṭī, Talkhīṣ, 1: 260 (ʿIzz al-Dīn ʿAlī), 4:105–6 (Kamāl al-Dīn Aḥmad). For the third son, Jalāl al-Dīn, see Neubauer, “Ṣafī al-Dīn.” KM Wa ṣṣāf, Tarīkh, 55; and Āyatī, Taḥrīr, 32. As discussed below, Yāqūṭ was Urmawī’s student in ‘Abbasid Baghdad. Urmawī enjoyed the patronage of Shams al-Dīn Juwaynī, and in 661/1262 he delivered Ṭūsī’s disquisition (or letter) on the history of Chinggis Khan to Baghdad; Ibn al-Fuwaṭī, Talkhīṣ, 3:319–20. Khwāndamīr also considers Urmawī as one of the leading scholars of Abaqa’s reign, but o+fers a di+ferent and more extensive list of notables; Khwāndamīr, Ḥabib al-siyar, 3:107; Idem (Thackston), 3:60. Kd Ibn Ḥajar, Durar, 5:108–9. Quṭb al-Dīn Shīrāzī’s commentary on Urmawī’s works is discussed in Fallāhzadeh, Persian Writing, 102–7. Ke For the impact of Kitāb al-adwār, see Fallāhzadeh, Persian Writing, 33–35, 102, 107, 133– 34, 145–46, 149, 201–2, 207–10; and Neubauer, “Safī al-Dīn,” among others. An example of Urmawī’s continued popularity is the recent recording of several notations from Kitāb al-adwār along with the publication of attendant French and English commentaries in Lebanon; http://www.youtube.com/watch?v=Migo5Wm4QK8; http://www.youtube.com/ watch?v=G7DY4V9cO_8 and Abou Mrad, Musique, and see its review by Wright in Yearbook, 197–98. Last but not least, Ṣafī al-Dīn has merited a Facebook page: http://www .facebook.com/pages/Sa./-al-Dīn-al-Urmawi/115898865087179, accessed August 2, 2012. ?A) :;<=> lasting fame in Ilkhanid Iran;Db Shams al-Dīn Aḥmad al-Suhrawardī (654– 741/1256–1340), also a noted calligrapher and musician;Fc and al-Jamāl al-Mashriqī (b. 661/1262), the famous singer.F* In addition, a bevy of less acclaimed students are enumerated in the sources.F2 Urmawī was also in contact with a wide range of peers. More speci./cally, his circle was on close terms with the Mawṣil singers (some of whom moved to Baghdad after the passing of the well-known patron of their profession, Badr al-Dīn Lu’lu’, the Atabeg of Mawṣil, in 657/1259);FB and other accomplished musicians from al-Jazīra, Anatolia, and Khurasan. The Juwaynīs’ largesse facilitated artistic and scholarly exchanges, and enabled the invitation of prominent musicians to Baghdad.FD Urmawī’s students also trained a new generation of musicians who then continued to teach and play his works. The renowned singer Kutayla (34. 750–60/ 1350–60 in Mardin and Cairo) is said to have memorized dozens of Urmawī’s suites (nawba), which by some counts numbered 130.FF Furthermore, Niẓām al-Dīn al-Ṭayyārī, the most acclaimed student of al-Suhrawardī, maintained

Kf On Yaqūt, see ʿUmarī, 10:348; al-Dhahabī, Tẚrīkh al-Islām, 60:373–74; Kutubī, 2:592–93; and Canby, “Yāḳūt al-Musta‘ṣimī,” Brill Online, 2012 accessed August 2, 2012, http://referenceworks.brillonline.com/entries/encyclopaedia-of-islam-2/yakut-al-mustasimi-SIM_7972, inter alios. Lg Al- Ṣafadī, Aʿyān, 1: 414–16; Ibn Ḥajar, Durar, 1: 335; ʿUmarī, Masālik, 10:390–94. L6 ʿ Umarī, Masālik, 10:395–406; Ibn Ḥajar, Durar, 3:240. L9 Among these ./gures are the ‘Abbasid Baghdadi singer Luḥāẓ (ʿUmarī, Masālik, 10:356); the 34autist Ḥasan al-Nāy, alias al-Zāmir (ʿUmarī, Masālik, 10:402; and Neubauer, “Musik,” 255); Sa‘d al-Dīn al-Sīlkū (ʿUmarī, Masālik, 10:402); Zaytūn (ʿUmarī, Masālik, 10:402); and Fakhr the musician (Ibn al-Fuwaṭī, Talkhīṣ, 3:62), who might be identical to Fakhr al-Dīn al-Shahrabānī (Ibn al-Fuwaṭī, Talkhīṣ, 3:234). Two others, ʿAlī al-Sitāhī (or Sītā’ī) and Ḥusām al-Dīn Quṭlugh Būqā, are mentioned in ‘Azzāwī and Neubauer’s works on the basis of the works of ʿAbd al-Qādir Marāghī (d. 838/1435), the most accomplished Persian writer in the ./eld of music, who was raised in Baghdad and wrote a lengthy commentary on Kitāb al-adwār. Unfortunately, I did not have access to his work and have thus far been unable to identify the people mentioned therein. (Farmer, “ʿAbd al-Ḳādir b. Ghaybī,” Brill Online, accessed August 3, 2012. http://referenceworks.brillonline.com/ entries/encyclopaedia-of-islam-2/abd-al-kadir-b-ghaybi-SIM_0091); ‘Azzāwī, al-Mūsīqā, 33–34; and Neubauer, “Musik,” 254–55. LJ The most prominent of these singers was Ibn al-Dahān al-Mawṣilī (d. 687/1287), whose fame approached that of Urmawī; ʿUmarī, Masālik, 10:346–48. For others, see ibid., 10:405, 409; Bar Hebraeus, The Chronography, 1:443; and Neubauer, “Musik,” 236. LK Ibn al-Fuwaṭī, Talkhīṣ, 4:299; and ʿUmarī, Masālik, 10:396–406. LL Umar ī, Masālik, 10:377. At the very least, Kutayla was acquainted with Jamal al-Mashriqī and might have been his student (ibid., 380); for a discussion on nawba, see Wright, “Nawba,” Brill Online, accessed August 3, 2012 http://referenceworks.brillonline.com/ entries/encyclopaedia-of-islam-2/nawba-SIM_5859. PQR;S ;> TUV PW>XWY SW>ZQVRT W[ :=XU\=\ ?Ah close ties with Abū Saʿīd, the last Ilkhan (r. 717–736/1317–35), who was a distinguished musician in his own right.FI The 34ourishing of music under the Ilkhans owed a great deal to the Juwaynīs’ patronage, but another factor was the Mongols’ a+./nity for this art form. From the Mongols’ perspective, music was a valued form of entertainment, which complimented another of their favourite pastimes—drinking. Furthermore, it was part and parcel of their Shamanic rites and a royal status symbol. While possessing their own venerable musical traditions, the Mongols consistently displayed a great deal of interest in the entertainment of other cultures.F^ Consequently, expert musicians from various backgrounds were highly appreciated and generously rewarded in Mongol lands. Partaking in their lords’ feasts, the musicians had ample opportunity to forge bonds and in34uence the Mongol elite. A case in point was Ilkhan Öljeitü’s marriage to a singer by the name of Najma Khatun, who, however, was bribed to ‘lobby’ or manipulate her husband into retreating from Raḥba.Fa Some musicians became their lord’s con./dants. For example, in several cases a singer served as the ruler’s envoy to a neighbouring country.Fb These relations also fostered acculturation. Descriptions of, say, Abū Saʿīd studying music and composing his own Persian verses and lyrics certainly suggest a high degree of royal assimilation into local culture.Ic Musicians also ./lled a variety of other duties, especially in all that concerned ceremonies of state: for instance, they performed at events marking the arrival or departure of the ordu (the ruler’s camp), and they entertained at receptions in honour of visiting delegations.I* Ibn Baṭṭūṭa, who accompanied Abū Saʿīd’s ordu en route from Baghdad to Sultaniyya in the 1320s, reports that there were a hundred of the royal musicians—both singers and players—in the entourage; and this number does not include their counterparts who were a+./liated

LM Al-Ṭayyārī was also a distinguished calligrapher; ʿUmarī, Masālik, 10:406–9. After the fall of the Ilkhanate, he temporally migrated to the Mamluk Sultanate. ʿUmarī draws heavily on al-Ṭayyārī by virtue of the latter’s knowledge on the House of Hülegü. For a discussion on Abū Saʿīd’s musical career, see ʿUmarī, Masālik, 10:370–72. While the centre of the music world appears to have shifted from Baghdad to the royal court in the later Ilkhanate, famous singers—both male and female—were still to be found in Baghdad; see, e.g., ʿUmarī, Masālik, 10:381, 382; and Ibn Ḥajar, Durar, 5:65. Ld Allsen, “Entertainers,” 37–38, 41–42, and the references therein. Le Ibn al-Dawādārī, Kanz al-durar wa-jāmi‘ al-ghurar, 9:261, 268; ʿUmarī, Masālik, 3:185. Lf For musicians who served as envoys, see ʿUmarī, Masālik, 10: 403; and al-Ṣafadī, Aʿyān, 5:563. For singers who were on close terms with the non-Mongol rulers of that era, see, e.g., ʿUmarī, Masālik, 10:346, 375. Mg ʿ Umarī, Masālik, 10:371–73. M6 Allsen, “Entertainers,” 41. ?Ai :;<=> with the amīrs, the vizier, and the khatuns (queens).I2 It also bears noting that the ruler customarily provided his princes, amīrs, and even some civilian o+./- cials, such as the naqīb al-ashrāf (the head of the ʿAlids), with drums, as part of their insignia. The drums were often augmented by other instruments for the purpose of helping these notables form bands.IB Musicians also accompanied Mongol troops to the front lines, where string instruments were played and kettledrums beaten to signify the start of battle.ID One contemporaneous report of the Mongol attack on Baghdad notes that the invading force took up its positions “happily with songs and trumpets.”IF Similarly, when the Ilkhanid naqīb al-ashrāf took 34ight from Abū Saʿīd, he sounded his drums and trumpets. As a result, startled villagers who believed that it was a “Tatar raid” 34ed in panic.II In light of the above, it is evident that there was a healthy demand for musicians in the Mongol Empire. These roles and the leadership’s fondness for music were also mirrored in neighbouring lands.I^ In consequence, talented musicians were often wooed by several rulers,Ia but it appears as though the Mongols o+fered them the best conditions: when al-Jamāl al-Mashriqī migrated to Egypt, he complained of

M9 Ibn Baṭṭūṭa, Voyages, 2:125–57; Idem (Gibb), The Travels of Ibn Battuta, 2:342–44. MJ E.g., Ibn Baṭṭūṭa, Voyages, 1:421–22, 2:126–27; Idem, (Gibb), 1:259–60, 2:344; and Ḥā./ẓ Ābrū, Dhayl Jāmi‘ al-tawārīkh, 102. According to Ibn Baṭṭūṭa, it was rather commonplace in Yuan China, Mamluk Egypt, and the Delhi sultanate for commanders to have their own bands; Ibn Baṭṭūṭa, Voyages, 1:138–39, 3:273–77, 4:223; Idem (Gibb), 1:89; 3:686–88; 4:872, inter alios. MK E.g., Polo, The Travels of Marco Polo, 1:337–38. ML Igeret Ya’kov b. Eliyahu (ca. 1263), as cited in Arnon, “The Mongols and the Jews,” 64 [In Hebrew]. MM Ibn Baṭṭūṭa, Voyages, 2:422–23, and Idem (Gibb), 2:260–61. Md As in the Mongol Empire, the primary functions of music in the Muslim world were entertainment and strengthening royal bona ./des; Ibn Khaldūn, al-Muqaddima, in his Kitāb al-ʿibar, 1:763–67; Idem (Rosenthal), The Muqaddimah, 2:401, 404. In Ibn Khaldūn’s estimation, however, music is a product of sedentary civilization. He also notes that it has been immensely popular among Persians from antiquity onwards, and, despite the Arabs’ initial reservations about music, the prospering caliphate welcomed singers from Persia and Byzantium. During the ‘Abbasid period, the Arabs indeed developed a re./ned and glorious musical tradition, which was still admired in the venerated historian’s own time. That said, in citing examples, he referred to the halcyon days of Hārūn al-Rashīd (r. 170–193/786–809), and to singers like Ibrāhīm al-Mawṣilī (125/742–188/804) and his sons, and not the later ‘Abbasids (ibid., 1:765–66, idem (Rosenthal), The Muqaddimah, 2:404). Me E.g., ʿUmarī, Masālik, 10:364, 367, 373, 375, 408; and Ibn Ḥajar, Durar, 3:240. These sources describe how the rulers of India, Egypt, and Yemen sought to attract leading Ilkhanid PQR;S ;> TUV PW>XWY SW>ZQVRT W[ :=XU\=\ ?Aj settling for crumbs after having grown accustomed to bread in the Ilkhanate.Ib It is also worth noting that Chinese musicians performed in Iran during Ghazan’s tenure and perhaps other periods as well. Likewise, many Muslim musicians, some of whom were probably from the Ilkhanate, found their way to Yuan China, the Delhi Sultanate, and Egypt. In fact, the Mamluks welcomed members of Urmawī’s circle, especially after the Ilkhanate’s fall.^c Yet another channel for musical exchange was imperial presents, as singing slave girls were often included in the bountiful gift packages that were exchanged between rulers.^* Since it was common for rulers and amīrs to have top-notch musicians instruct their slave girls,^2 these sorts of exchanges were bound to enhance the transfer of ‘high music’ across the Eurasian continent. The massive 34ow of people that informed the Mongol era naturally led to the dissemination of musical instruments. For example, miniatures attest to the presence of Chinese and European instruments in the Ilkhanate.^B Moreover, Rashīd al-Dīn informs us that the Chinese system of notation also made its way into this realm.^D Since Urmawī does not mention this phenomenon in the Risāla, it is di+./cult to ascertain whether he was aware of these transfers. The second generation of his circle was probably more cognizant of this development, as these musicians often jumped from one court to the next.^F This sort of diversity probably nourished the continued blossoming of music and musicology throughout the borders of the Ilkhanate.

players. For contemporaneous musicians in the Mamluk sultanate, see Neubauer, “Musik,” 238–40. Mf ʿ Umarī, Masālik, 10:396. dg E.g., ʿUmarī, Masālik, 10:408; and Allsen, “Entertainers,” 42–45. d6 E.g., Ibn Baṭṭūṭa mentions that the sultan of Delhi sent Hindu singing girls to the Yuan emperor; Ibn Baṭṭūṭa, Voyages, 4:773; Idem (Gibb), 4:1; ʿUmarī, Masālik, 10:396, 375. For discussion of the gifts exchanged between Mongol and Mamluk dignitaries see Little, “Diplomatic Missions,” 39. d9 E.g., ʿUmarī, Masālik, 10:396; and Ibn Ḥajar, Durar, 3:240. dJ Farmer, “Reciprocal In34uences,” 334–35, 340–42; Allsen, “Entertainers,” 45; and ʿUmarī, Masālik, 10:402. For a discussion of Chinese interest in foreign musical instruments and their medical functions, see Chang De, in Li You, Xishiji, 2:141. dK Rash īd al-Dīn, Tanksūq-nāmah, 38–39. For a full English translation of this passage, see Jahn, “Some Ideas of Rashīd al-Dīn,” 146–47; cited in Allsen, “Entertainers,” 44. dL E.g., ʿUmarī, Masālik, 10:396, 392, 402, 407; Ibn Ḥajar, Durar, 3:240; Ṣafadī, Aʿyān, 1:414–15. Of special interest is al-Jamāl al-Mashriqī’s reference to an urdhul (a sort of reed pipe) that, in his estimation, was used by the Franks; ʿUmarī, Masālik, 10:407. For the in34uence of Chinese and Indian music on their Persian counterpart during the thirteenth and fourteenth centuries, see Fallāhzadeh, Persian Writing, 206, 209, 216. The Chinese notation system, however, was not included in the Islamic theoretical musical literature. ?]k :;<=>

Conclusion

As opposed to those writers who emphasize the catastrophic dimensions of the fall of Baghdad, Urmawī’s personal account reinforces the more mundane perspectives on this watershed event. The fate of this polymath, his oeuvre, and students in the post-conquest Ilkhanate also demonstrates that, at the very least, the Mongol triumph did not close the curtain on ‘Abbasid musical culture. In fact, due to its compatibility with the new regime’s indigenous norms and the open world that the empire created, this musical tradition continued to evolve, spread, and thrive under Mongol rule. What is more, it would have a tremendous impact on the later development of Islamic music in the Arab, Persian, and Turkish world.

Bibliography

Abou Mrad, Nidaa. Musique de l’époque abbaside: Le legs de Ṣa+,y a-d-Dīn al-Urmawī/ Music of the Abbasid Era: The Legacy of Ṣa+,y a-d-Dīn al-Urmawī [sic]. Paris: Maison des cultures du monde, 2005. Abu Rukun, Alaa S. [sic]. “The Fall of Abbasid Baghdad (658H/1258) as Re34ected in Arabic Historiography and the Arab Literature.” PhD diss., Bar Ilan University, Israel, 2010. [In Hebrew] Allsen, Thomas T. “Command Performances: Entertainers in the Mongolian Empire.” Russian History 58 (2001): 37–46. Amitai, Reuven. “Hülegü and his Wise Men: Topos or Reality?” In Politics, Patronage, and the Transmission of Knowledge in 13th–15th Century Tabriz, edited by Judith Pfei+fer, 15–34. Leiden: Brill, 2014. Anon. (Pseudo Ibn al-Fuwaṭī). Kitāb al-ḥawādith. Beirut: Dār al-gharb al-Islāmī, 1997. Arnon, Na’ama. “The Mongols and the Jews: People of the Book and People of the Horses.” MA diss., Bar Ilan University, 2005. [In Hebrew] Āyatī, ʿA-M. Taḥrīr-i ta’rīkh-i Waṣṣāf. Tehran: Chāpkhānah-i ‘ilmī, 1346/1967. ‘Azzāwī, ʿAbbās. al-Mūsīqā al-ʻIrāqīyya fī ʻahd al-Mughūl wa-l-Turkmān. Baghdad: Sharikat al-tijāra, 1951. Bar Hebraeus. The Chronography of Gregory Abu’l-Faraj . . . commonly known as Bar Hebraeus. Translated by Ernest Alfred Wallis Budge. Reprint, London: The Georgian Press, 2003. ———. Ta’rīkh mukhtaṣar al-duwal. Beirut: Dār al-mashriq, 1992. Biran, Michal. “Violence and Non-Violence in the Mongol Conquest of Baghdad.” Paper presented at the conference “Legitimate and Illegitimate Violence in Islamic Thought,” Exeter, uv, September 3–4, 2012. PQR;S ;> TUV PW>XWY SW>ZQVRT W[ :=XU\=\ ?]?

Boyle, John Andrew. “The Death of the Last ‘Abbāsid Caliph: A Contemporary Muslim Account.” Journal of Semitic Studies 6 (1961): 145–61. Canby, Sheila R. “Yāḳūt al-Mustaʿṣimī.” In Encyclopaedia of Islam, Second Edition; online version at http://referenceworks.brillonline.com/entries/encyclopaedia-of- islam-2/yakut-al-mustasimi-SIM_7972. (see earlier remarks) Chang De, in Li You. Xishiji (Record of an Embassy to the West). Translated in Mediaeval Researches from Eastern Asiatic Sources, compiled by Emili V. Bretschneider, 1:122– 56. London: Trübner & Co., 1910. al-Dhahabī, Shams al-Dīn Muḥammad b. Aḥmad. Ta’rīkh al-Islām. Beirut: Dār al-kitāb al-ʿarabī, 2003. Eickelman, Dale F. “Is there an Islamic City? The Making of a Quarter in a Moroccan City.” International Journal of Middle Eastern Studies 5 (1974): 274–94. Fallāhzadeh, Mehrdad. Persian Writing on Music: A Study of Persian Musical Literature from 1000 to 1500 45. Uppsala: Uppsala Universitet, 2005. Farmer, Henry George. “Reciprocal In34uences in Music ’twixt the Far and Middle East.” Journal of the Royal Asiatic Society 66/2 (1934): 327–342. ———. “ʿAbd al-Ḳādir b. Ghaybī.” In Encyclopaedia of Islam, Second Edition; online version at http://referenceworks.brillonline.com/entries/encyclopaedia-of-islam-2/ abd-al-kadir-b-ghaybi-SIM_0091. Gelder, Geert Jan. H. van, “Sing Me to Sleep: Sa./ al-Din al-Urmawi, Hulegu, and the Power of Music,” Quaderni di Studi Arabi n. s. 7 (2012): 1–9. Gilli-Ellevy, Hend. “Al-Ḥawādit al-ğāmi‘a: A Contemporary Account of the Mongol Conquest of Baghdad, 656/1258.” Arabica 58 (2011): 353–71. Ḥā./ẓ Ābrū. Dhayl Jāmiʿ al-tawārīkh. Edited by Khānbābā Bayānī. Tehran: Chāpkhānah-i ‘ilmī, 1317/1938. Ḥājjī Khalīfa. Kashf al-ẓunūn ‘an asāmī al-kutub wa’l-funūn. Istanbul: Maṭabi‘ wikālat al-maʿārif al-jalīlah, 1941. Hassan, Mona F. “Loss of Caliphate: The Trauma and Aftermath of 1258 and 1924.” PhD diss., Princeton University, 2008. Ibn Baṭṭūṭa, Voyages d’Ibn Baṭṭūṭa. Edited and translated by C. Defrémery and B. R. Sanguinetti. Paris: L’Imprimarie Nationale, 1969. ———. The Travels of Ibn Battuta. Translated by H. A. R. Gibb and Charles Buckingham, 4 vols. Cambridge: The Hakluyt Society, 1958–94. Ibn al-Dawādārī. Kanz al-durar wa-Jāmiʿ al-ghurar, vol. 9. Cairo: Deutsches Archäologisches Institut Kairo, 1960. Ibn al-Fuwaṭī, Kamāl al-Dīn Abū al-Faḍl. Talkhīṣ majmaʿ al-ādāb fī mu‘jam al-alqāb. Edited by M. al-Kāẓim. Tehran: Mu’asasat al-ṭibāʿa wa-l-nashr, 1995. Ibn Ḥajar. al-Durar al-kāmina. Cairo: Dār al-kutub al-ḥadītha, 1966. Ibn Ḥijja al-Ḥamawī. Thamarāt al-awrāq. Cairo: Maktabat al-khashānjī, 1971. ?]p :;<=>

Ibn Junayd, Yaḥyā Maḥmūd. al-Mustaʻṣim billāh al-ʿAbbāsī. Beirut: al-Dār al-ʿarabiyya li’l-mawsu‘āt, 2008. Ibn Kathīr. al-Bidāya wa-l-nihāya. Beirut: Dār iḥyā’ al-turāth al- ʿarabī, 1993. Ibn Khaldūn. al-Muqaddima. In his Kitāb al-ʿibar. Beirut: Dār al-kitāb al-lubnānī, 1961. ———. The Muqaddimah: An Introduction to History. Translated by Franz Rosenthal. Princeton: Princeton University Press, 1967. Ibn al-Sā‘ī, ʿAlī b. Anjab [Pseudo]. Kitāb mukhtaṣar akhbār al-khulafā. Cairo: Būlāq, 1891. Ibn Ṭāwūs, ʿAlī b. Mūsā. Iqbāl al-a‘māl. Beirut: Manshūrāt mu’asasat al-a‘lamī li’l-maṭbū‘āt, 1996. Ibn al-Ṭiqṭaqā, Muḥammad b. ʿ Alī. al-Fakhrī fī al-ādāb al-sulṭāniyya wa’l-duwal al-islāmiyya. Paris: Librairie Emile Bouillon, 1895. ———. al-Fakhrī. Translated by C. E. J. Whitting. London: Luzac, 1947. Jackson, Peter. “Bāyjū.” In Encyclopaedia Iranica mw: 1–2 http://www.iranicaonline.org/ articles/bayju-baiju-or-baicu-mongol-general-and-military-governor-in-northwestern-iran-34. Jahn, Karl. “Some Ideas of Rashīd al-Dīn on Chinese Culture.” Central Asiatic Journal 14 (1970): 134–47. Juwaynī, ʿAṭā Malik. Ta’rīkh-i Jahāngushā. Edited by Mīrzā Muḥammad Qazwīnī. London: Luzac, 1912. ———. The History of the World-Conqueror. Translated by John Andrew Boyle. Reprint, Manchester: Manchester University Press, 1997. Kāzarūnī, ʿAlī ibn Muḥammad. Mukhtaṣar al-ta’rīkh, min awwal al-zamān ilā muntahā dawlat Banī al-‘Abbās. Baghdad: Mudīrīyat al-thaqāfa al-‘āmma, 1970. Khwāndamīr. Ta’rīkh-i ḥabīb al-siyar. Tehran: Kitābkhāna-yi khayyām, 1333/1956. ———. Ḥabīb al-siyar. Translated by Wheeler M. Thackston. Vols. 3 and 4. Cambridge, x$: Department of Near Eastern Languages and Civilizations, Harvard University, 1994. al-Kutubī, Muḥammad b. Shākir. Fawāt al-wafayāt. Beirut: Dār al-kutub al-‘ilmiyya, 2000. Lane, George. Early Mongol Rule in Thirteenth Century Iran. London and New York: RoutledgeCurzon, 2003. Le Strange, G. “The Story of the Death of the Last Abbasid Caliph, from the Vatican xy. of Ibn-al-Furāt.” Journal of the Royal Asiatic Society 2 (April 1900): 293–300. Little, Donald P. “Diplomatic Missions and Gift Exchanged by Mamluks and Ilkhans.” In Beyond the Legacy of Genghis Khan, edited by Linda Komaro+f. Leiden: Brill, 2006, 30–42. Manṣūr, Aḥmad. Qiṣṣat suqūṭ Baghdād. Beirut: Dār Ibn Ḥazm, 2003. PQR;S ;> TUV PW>XWY SW>ZQVRT W[ :=XU\=\ ?]@

Ma‘rūf, Nājī. Tārīkh ‘ulamā’ al-Mustanṣirīyya. Cairo: Dār al-sha‘b, 1976. Neubauer, Ekhard. “Musik zur Mongolenzeit in Iran und den angrenzenden Landem.” Der Islam 4 (1969): 233–60. ———. “Ṣafī al-Dīn Urmawī.” In Encyclopaedia of Islam, Second Edition; online version at http://www.brillonline.nl/subscriber/entry?entry=islam_SIM-6447. al-Nuwayrī, Shihāb al-Dīn Aḥmad b. ʿAbd al-Wahhāb. Nihāyat al-arab fī funūn al-adab. Cairo: al-Hay’ah al-Miṣrīyah al-‘āmmah li’l-kitāb, 1984. Piamenta, Moshe. Dictionary of Post-Classical Yemeni Arabic, pt. 1. Leiden: Brill, 1990. Polo, Marco. The Travels of Marco Polo. Edited by Henry Yule and H. Cordier, 2 vols. New York: Dover, 1993. Rashīd al-Dīn. Jāmiʿ al-tawārīkh. Edited by B. Karīmī. 2 vols. Tehran: Iqbāl, 1338/1960. ———. Jami‘u’t-tawarikh [sic] Compendium of Chronicles. Translated by Wheeler M. Thackston. Cambridge, x$: Department of Near Eastern Languages and Civiliza- tions, Harvard University, 1998–99. ———. Tanksūq-nāma-yi ṭibb ahl-i Khitā. Edited by M. Minūwī. Tehran: University of Tehran, 1972. al-Ṣafadī, Ṣalāḥ al-Dīn, Ḥalīl b. Aybak. A‘yān al-‘aṣr wa-a‘wān al-naṣr. Beirut and Damascus: Dār al-./kr, 1998. ———.al-Wāfī bi’l-wafayāt. Beirut, 1987–1991. al-Sarjānī, Rāghib. Qiṣṣat al-Tatār min al-bidāya ilā ʿAyn Jālūt. Cairo: Mu’asasat iqra’, 2006. Shiloah, Amnon. Music in the World of Islam: A Socio-Cultural Study. Detroit: Wayne State University Press, 1995. Shīrāzī, Quṭb al-Dīn. Akhbār-i Mughūlān. Qum: Kitābkhāna-i buzurg-i ḥaz̤rat-i āyatallāh al-‘uẓmá Marʻashī Najafī, 2010. al-Subkī, Tāj al-Dīn Abū Naṣr. Ṭabaqāt al-sha+,ʿiyya al-kubrā. Aleppo: Maṭba‘at ‘Isā al-Bābī al-Ḥalabī, 1964. al-ʿUmarī, Shihāb al-Dīn Ibn Faḍlallāh. Das Mongolische Weltreich: al-‘Umarī’s Darstellung der mongolischen Reiche in seinem Werk Masālik al-abşār fī mamālik al-amşār. Edited and translated by K. Lech. Wiesbaden: Harrassowitz, 1968. ———. Masālik al-abṣār wa-mamālik al-amṣār. Beirut: Dār al-kutub al-‘ilmiyya, 2010. ———. Masālik al-abṣār wa-mamālik al-amṣār. Ed. F. Sezgin with A. Jolhosha, E. Neubauer. Frankfurt am Main: Institute for the History of Arabic Islamic Science at the Johan Wolfgang Goethe University, 1988. Waṣṣāf, ʿAbdallāh b. Faḍlallāh. Tarīkh-i Waṣṣāf (Tajziyat al-amṣār wa-tazjiyat al-a‘mār). Bombay, 1269/1852–1853. Reprint, Tehran, 1338/1959–1960. Ward, L. J. “Ẓafarnāmah of Mustawfī.” PhD diss., Manchester University, 1983. Wright, Owen. The Modal System of Arab and Persian Music: 45 1250–1300. Oxford: Oxford University Press, 1978. ?]A :;<=>

———. “A Preliminary Version of the ‘Kitāb al-Adwār.’” Bulletin of the School of Oriental and African Studies 58/3 (1995): 455–78. ———. Review of Musique de l’époque abbaside: Le legs de Ṣa+,y a-d-Dīn al-Urmawī / Music of the Abbasid Era:The Legacy of Ṣa+,y a-d-Dīn al-Urmawī by Nidaa Abou Mrad. Yearbook for Traditional Music, 39 (2007): 197–98. ———. “Nawba.” In Encyclopaedia of Islam, Second Edition; online version at http:// referenceworks.brillonline.com/entries/encyclopaedia-of-islam-2/nawba-SIM_ 5859. !"#$%&' ( Historical Epic as Mongol Propaganda? Juwaynī’s Motifs and Motives

Judith Kolbas*

Overview

It is often said that history is written by the conquerors. If so, steppe groups are the exception. It is the conquered people of the Middle East who have provided the most information about Mongol origins and successes there. One author, ʿAlāʾ al-Dīn ʿAṭā Malik Juwaynī (1226–83) has been particularly important since he was involved in some of the contemporary events and consciously re./ected on history. As a result, his knowledge and skillful writing have greatly a0fected understanding of the Mongols. His work, Tār"̄kh-i jahān gushā,1 which has been translated into English by John Andrew Boyle as Genghis Khan: the History of the World-Conqueror,2 can be considered not only a history but also a literary masterpiece. As such, it merits critical analysis, which a number of scholars have provided. Melville has carefully reviewed a number of these scholars;3 but, so far, none have approached the work as representing the transformation of two strong cultures. For example, Rypka simply held that “Juvaini’s style alternates between a greater and lesser degree of ornamental rhetoric.”4 On the other hand, Morgan noted that “the Persian sources have remained much the same . . . ,” thereby identifying a good deal of the problem since the sources have been reviewed over and over again. Nevertheless in some respects, as he observed, there has been a shift from an overwhelming emphasis on the Mongol invasions and mass destruction to the demographic,

* I am indebted to several people who have helped me to understand my originally rather inarticulate concerns about the text. They are Bruno De Nicola as editor for his careful and encouraging comments, Barbara Brend for her exploration with me of literary matters and James Clark for his help with rhetoric. Literary analysis is completely removed from my normal 67eld of economic history, so their assistance has been crucial. 8 Juwaynī/Qazwīnī. : Juwaynī/Boyle. ; Melville, “Jahangosha,” 378–82. < Rypka, “Poets and Prose Writers,” 550–625, 623.

© !"#$#!%$&!' ()$%% #*, %'$+'#, ,-./ | +"$ .-.../1/23425561.63,/_552 CDE =>?@AB cultural and political changes the Mongols brought to the Middle East.F A few people have given more detailed comments such as Boyle: “Juvaini was master of what was already the traditional style of Persian prose literature. It is a style which disposes of all the rhetorical devices known . . . Word-plays are indulged in, whenever possible . . . However, Juvaini was a man of taste: he had his rhetoric under some measure of control and could, when the occasion demanded it, tell his tale in the most plain and straightforward language” (xi–xii). More recently, Jackson has provided another evaluation by contrasting the moral and religious positions of Rashīd al-Dīn and Juwaynī, both high-level administrators for the Mongols, and observed that Juwaynī issued almost immediately in “one early passage . . . nothing less than an unabashed plea for political quiescence.”G Jackson’s attention to an emotional and moral tone could be followed much further since Juwaynī’s book, in this author’s opinion, interpreted some immense philosophical and social changes in Iranian culture.H On the other hand, a number of these critical assessments of literary developments during Mongol ruleI would also bene67t from a comparison with the new artistic types of personal and introspective painting, poetry, prose novel and particularly popular theatre that arose in Mongol China.J Thus although the period in Iran saw “the 67nal triumph of New Persian over Arabic,”1K actual styles and methods continue to deserve identi67cation and de67nition. In Juwayni’s case, no study has yet addressed his work as a whole from both the literary and historical angles since his message was full of contradictions with many height- ened images and allusions that go far beyond immediate historical interest. This project would be a challenging task, but perhaps the following discussion can indicate some paths to follow, especially for historians.

L Morgan, “The Mongols in Iran,” 131–36, 132. M Jackson, “Mongol Khans,” 109–22, 110. N Many of the arguments in the following pages have been put forward before, in particular very convincingly by Boyle, Grousset and Jackson in the works cited. This paper can only add another lens to their insights. O For example, Poliakova in “Literary Canon,” 237–56, 244, has reviewed several Persian chroniclers from the eleventh to the 67fteenth centuries and considers that in Juwaynī’s work “literary etiquette predetermines the author’s selection of facts.” Although she makes important points about stylistic aspects, the following discussion suggests that Juwaynī had a much broader agenda on which to base his chronicle than evolving but standard images and phrases. P Rossabi, “The reign of Kubilai Khan,” 414–89, 476–79. 8Q Morgan, “The Mongols in Iran,” 135. "RBS>TRUA? VWRU AB X>YZ>? WT>WAZAY[A? CD(

Textual Development

As di067cult as it is to deal with this complicated material, it can provide signi67- cantly more information to the 67eld of history if it is also considered by way of literary themes, variations, rhetoric, motifs and motives. To do this, literary histories such as Juwaynī’s should 67rst be re-evaluated in light of an historical process in itself, namely the stages during which two completely distinct but extremely sophisticated cultures impose themselves on each other with or without awareness of the process. Such an approach to social and cultural modi67cations was 67rst introduced by Richard Ettinghausen in 1935 for artistic in./uences on the Islamic world.11 He proposed three stages consisting of transfer, adoption and integration. The initial confrontation led to wholesale movements or transfers of the entire form. In Juwaynī’s case, it was the plain and rather detailed reports of Mongol battles, the conquerors’ physical appearance, strange customs and other direct observations without much comment. The second stage, that of adoption, in which one accepts and learns to live with new ideas and social contexts, is re./ected by his comments, for example, that the Mongols were not as pompous as previous Islamic rulers. He also suggested a reversal of the process by providing numerous anecdotes about the Mongols accepting the Persians as the best group in their entire empire. Finally in the integration phase, during which two cultures meld and bene- 67t each other, he made the plea that Jackson noted. He supported that idea throughout the text by giving speci67c incidents to show that he understood a good deal of Mongol culture and that life had improved for both. Presumably, he did not write his book in one period but re-worked passages many times and continued to insert items and delete others as his manuscript took shape. As a result, the three stages often appear in conjunction with each other. An analysis of this three-fold process and his rhetorical style is at a very early stage, but hopefully the following discussion will encourage scholars to open the book again.

88 The topic is discussed in greater detail by Marina Whitman. She examined the images and contexts of Chinese landscapes of contemplation, sages and symbols of spring that moved into Persian culture, noting that often they eventually came to emphasize power and pleasure, themes not at all related to the original scenes. Her study is instructive and suggests that Ettinghausen’s method for assessing artistic activity can also be useful for literature: Marina D. Whitman, “The Scholar,” 255–61, 256. CD\ =>?@AB

Motives and Motifs

To begin, it is important to realize that Juwaynī was not a chronicler, nor did he incorporate decrees or o067cial speeches. Rather, Juwaynī’s style was extremely ./owery but clear enough, and he provided dramatic intensity to otherwise dry facts. As a consummate writer of both prose and poetry, he had some literary devices and favourite themes which in./uenced the accuracy of his statements and the spin on some important events. His 67nal comments of integration, however, appear regularly within his motives, motifs and rhetoric. Two motifs are the most striking and illustrate the incremental development of his message. However, others deserve at least quick recognition as well and will be noted at various points. The salient motifs are contrasts of good and evil, namely the physical destruction of Persian urban culture and its populace and the remarkable bene67cence of Chinggis Khan’s successor, Ögetei. Destruction and kindness are chosen here because Juwaynī expended considerable e0fort on them, carefully selecting his material to make his points more powerful. Moreover, the themes clearly illustrate some of his major rhetorical schemas, tropes and sentence euphony.

Negative Motif To start with the theme of destruction, the reader should take into account cultural di0ferences and Juwaynī’s personal viewpoints. Juwaynī carefully and convincingly covered the Mongol invasion of Khurasan so that the reader is left in no doubt about its terrible e0fect on the area. Nevertheless, it is necessary to remember that this region was his homeland, giving him a personal reaction to as well as invaluable knowledge about events. His eloquent details re./ected his own and his family’s intrinsic and immediate involvement in them. As a result, historians take away from this section the overall impression of total warfare. Due to emotional, social and psychological divergences between traditional Persian and Mongol cultures, Juwaynī did not often present these di0ferences. Rather, he generally transferred the bald facts quickly without any interpretation or explanation. Since he did not provide the motivation or strategy of the Mongols, the many examples that describe the invasion of Khurasan exhibit the initial form of literary development. The text is straightforward; but regardless, it was shaped by the Persian perspective of much structural obliteration accompanied by shocking levels of massacres. On the other hand, to understand it from a balanced point of view, the historian is required to look beyond Juwaynī’s position as a prominent Persian resident of Khurasan. This has been achieved by Grousset, who explained the Mongol behaviour of massacring, "RBS>TRUA? VWRU AB X>YZ>? WT>WAZAY[A? CD] looting and burning as bewilderment about sedentary life.12 The destruction followed nomadic practice on the wide steppe. Parties had raided constantly for centuries, taking away with them moveable material. Dwellings were easily made and carried away whereas permanent cities and urban economies were quite beyond their experience. Therefore, the military treated sedentary civilizations as they had their own in terms of both people and goods. Thus, towns were not sources of continuous and consistent wealth creation but short-term infusions of it. Nevertheless, there is more to the di0ference in Mongol and Persian cultural experience; there was a signi67cant psychological aspect as well. Cities enclosed by constraining walls were forbidding places to Mongols, trapping people within their supposedly defensive walls like sheep. Even today, this author’s extended living with herders in Mongolia provided many dem- onstrations that walls enclose livestock, never people. As a result, the purpose of walls is the opposite for settled and nomadic communities, protection for urban people and imprisonment for nomads. Finally, regarding strategic realities, if an entire group were at war or on the move evading contact with the winner of the last battle, scattered groups could re-gather and often quickly rebound in population numbers.13 If directed by strong leaders, they could regain prominence. That made it necessary to attempt to slaughter enemies completely as the only e0fective way of eliminating reprisal. Therefore, the Mongol method was not wanton cruelty according to them but a traditional defence strategy, which they applied at the initial conquests. These social and strategic factors are not expressed by Juwaynī, only the blanket destruction. Since the basic reports are almost a listing, he presumably wrote this section during the early phase of the composition and added ./ourishes later. The main tone is negative. In contrast with deploring such barbarity, Juwaynī often gave a subtle but consistent impression that he approved of Mongol behaviour. This situation could be considered the second stage, that of adoption in literary terms, since he showed that he understood Mongol culture well and tried to explain it. One example is his observation about the lack of grandiose titles for Mongol rulers: “It is one of their laudable customs that they have closed the doors of ceremony, and preoccupation with titles, and excessive aloofness and inaccessibil- ity; which things are customary with the fortunate and the mighty” (26–27).

8: Grousset, The Empire of the Steppes, 249–52. 8; Herodotus was the 67rst to appreciate this pattern, which more recent observations and statistics have corroborated, suggesting the possibility of up to a 40 per cent growth rate: See Buell, “Central Eurasia.” CE^ =>?@AB

In another illustration of adoption, in this case of the Mongols adapting to Iranian ways, he addressed Persian identity by noting the distinction between his region and that of other Mongol domains. That di0ference consisted not only of religious rectitude but also of artistic and personal qualities. Perhaps he also responded to some uneasiness in Iran of the presence of so many Chinese and Buddhists involved with the Mongols. Few authors expressed it as clearly as he did as a problem of acculturation. He allayed these challenges to his society, for example, with an anecdote about Ögetei: “Qa’an ordered his attendants to fetch from the treasury all sorts of jewels from the lands of Khorasan and the two Iraqs . . . and arms from Bokhara and Tabriz . . . and likewise what was imported from Khitai, being . . . inferior to the others . . . All these things he ordered to be laid side-by-side so that it might be seen how great the di0ference was. And he said: ‘the poorest Moslem has many Khitayan slaves while the great emirs of Khitai have not one Moslem captive. And the reason for this can only be the bene67cence of the Creator, Who knoweth the station and rank of every nation. It is also in conformity with the ancient yasa of Chinggis Khan, according to which the blood-money for a Moslem is forty14 balish (silver ingots) and for a Khitayan a donkey” (207).1F Juwaynī left his readers in no doubt that Persian culture was the best in the empire. As for religious superiority, he cited the fate of Küchlüg (d. 1218), who apostatized and converted to Christianity.1G Küchlüg then persecuted Muslims, which was one reason the Mongols defeated him. In comparison with Küchlüg, the Mongols acquiesced to the local Muslim religion. Juwaynī concluded with the admonition: “And be it remarked that whoever molests the faith and law of Muhammed never triumphs, while whoever fosters it, even though it not be his own religion, advances day by day in prosperity and consideration” (68). He provided the same sentiment to Möngke Khan: “Of all the people and religious communities, he showed most honour and respect to the Moslems. It was upon them that he bestowed the largest amount of gifts and alms and it was they who enjoyed the greatest rights” (600). Juwaynī then proceeded to give an example of a Muslim celebration in the Mongol capital, Qara Qorum, in 1252 (600). There are many other examples

8< The rhetorical use of the number four will be discussed shortly. 8L Rashīd al-Dīn followed Juwaynī’s lead with many similar anecdotes about Ögetei. See Rashīd al-Dīn, Jami‘u’t-tawarikh [sic] Compendium of Chronicles, translated by Wheeler M. Thackston, 334–45. 8M Juwaynī blames Küchlüg’s new wife for converting him. This is another motif, Juwaynī’s aversion to women in politics. He often ascribed dire consequences to their behaviour. His Persian viewpoint is in marked contrast to that of steppe society. An historian who has noted Juwaynī’s bias is Weatherford, Genghis Khan, 161. "RBS>TRUA? VWRU AB X>YZ>? WT>WAZAY[A? CEC of Juwaynī’s attempt to infuse the Mongols with a sophisticated understanding of the religion and culture of Greater Iran. Not only did he reassure his audience but he also re-enforced the adoption process by showing that the Mongols were undergoing the same changes.

Psychological and Philosophical Undercurrents The 67nal stage of his literary development, that of integration, is exhibited by a psychological approach of acceptance to the invasion. In one aspect, it presented itself through his e0fusive discussion of Mongol military e067ciency and success which indicated that there was no alternative to the ultimate winner. As a result, since Iran joined the champion and was active within the Mongol Empire, its status gave it power especially in political contexts. Besides accepting events as they stood, which Juwaynī espoused, there was a highly developed concept in his society of the hierarchy of authority. It was a basic philosophical foundation for the theory of the state at that time, most brilliantly expounded by al-Fārābī (./. about 870–950) but based on Aristotelian concepts. Al-Fārābī stated: “The heart comes to be 67rst and becomes the cause of the existence of the other organs and limbs of the body . . . This applies to all existents. [In particular] the excellent city resembles the perfect and healthy body, all of whose limbs co-operate to make the life of the animal perfect and to preserve it in this state.”1H Juwaynī explained the concept more vividly and practically at the outset of his book: “it is necessary and essential that (if) the threat of the sword . . . were to remain in abeyance and men were content with that which is promised in the next world everything would be confused; the common people, whose feet are bound with What is restrained by authority [Sura _`: 65] would have their hands freed. . . . ”(17). “He should prepare a certain person [Chinggis Khan] and make his nature the receptacle of all manner of power and daring, and ruthlessness, and vengeance, and then by means of praiseworthy qualities and laudable properties bring it into a position of equi- librium”(19). The body politic theory of the state and its subsequent hierarchy was con67rmed by religion. For example, the Qur’an states in Sura ```: 25: “Say, O God: Master of the kingdom. Thou givest the kingdom to whom Thou wilt and seizest the kingdom from whom thou wilt . . . ” The idea was so entrenched in the culture that this statement appeared in 1258 on Hülagü’s initial gold and silver coinage at his conquest of Baghdad.1I Therefore, there were philosophical, political and religious foundations that promoted the view that military power

8N al-F ārābī, al-Fārābī, 231. 8O Kolbas, The Mongols in Iran, 156. CEa =>?@AB provided a strong leader and was in itself a legitimatizing factor for a political system. Juwaynī had good grounds for accepting Mongol rule.

Techniques in Presenting Facts Within this overall context of integrating the Mongols into an Islamic worldview, Juwaynī stressed their God-given superiority by providing impressive details of the conquests; there are names, numbers and precise actions, most of which would have been derived from the transfer phase and re-employed for his later purpose. Nevertheless, even these facts do not tell the full story since they were sometimes telescoped or ignored. At this juncture in the study, Juwaynī’s rhetorical style can also be examined, a style that allowed him to create truly impressive e0fects. An excellent example is his description of the conquests of two major commercial, political and cultural centres of Khurasan, those of Marv and Nishapur. In each case, the city voluntarily submitted at the arrival of the 67rst Mongol contingent. Indeed, Mongol coinage from the inhabitants of both places was struck to pay tribute.1J Juwaynī limited the initial submission almost to a passing remark because he had a di0ferent intention in relating the details, that of stressing the power of the new state. Once the army left, however, the cities ‘revolted.’ Then the Mongols returned and wreaked ‘vengeance.’ Thus, for the 67rst Marv revolt, he wrote: “the Mongols now entered the town and drove all the inhabitants . . . out on to the plain. For four days and nights, the people continued to come out . . . The Mongols ordered that, apart from 400 artisans . . . the whole population, including women and children, should be killed . . . ” with each Mongol soldier responsible for executing 400 people (161–62). Nevertheless, the city revolted again, according to him. Whereupon, “Torbei arrived with 67ve thousand men . . . They took the town within an hour; and putting camel halters on believers, they led them out in strings of ten and twenty and cast them into a trough of blood. In this manner, they martyred a hundred thousand persons” (166). This description, although ghastly, used Persian culture to enhance the pathos since Mongols use nose toggles on camels instead of halters. Nevertheless, the purpose of the hyperbole was not to actually describe the scene but to illustrate the abject humilia- tion and su0fering of people if they revolted. The story continued: the Mongol army then left Marv and regrouped, leaving a small contingent to 67nish killing whoever remained in the city. It took forty-one days to accomplish that, and then the contingent also left. He wrote further: “In the whole town, there remained not four persons alive.” A neighbouring Turkoman chief rallied whoever was left in the villages around the city

8P Kolbas, The Mongols in Iran, 42–43; also Grachev and Petrov, “Klad mednih poserebrennih.” "RBS>TRUA? VWRU AB X>YZ>? WT>WAZAY[A? CEb and repopulated the town: “ten thousand people gathered around him and he was emir for six months . . . ” (167). The Mongols returned again, this time with one hundred thousand men, and again tortured and tormented the townsfolk. They spent forty days this time doing the job. “In the town and the villages, there remained not a hundred souls alive . . . ” (168). It is apparent that the 67rst repression left more than four people alive in Marv since it is inconceiv- able that 10,000 villagers and tribal people could have gathered in the place or region so soon afterwards. The number is undoubtedly exaggerated for e0fect, yet Juwaynī acknowledged the continuing challenge to Mongol rule.2K Meanwhile, as continued in the next chapter, the army pressed on to Nishapur, for which Juwaynī wrote: “It was commanded that the town should be laid waste in such a manner that the site should be ploughed upon . . .” (177). Only 400 craftsmen were saved and marched o0f to Turkestan. The rest, he implied, were slain, with separate piles for the bodies of men and women. As before, a contingent was left behind to round up and kill any who had succeeded in hiding (178). Here, he gave no number of days for the task. However, the event could hardly have taken place as described since Nishapur was a major city ten years later in which Juwaynī’s father was a leading 67gure.

Rhetoric The numbers and sentiment but not the names in the descriptions of Marv and Nishapur match so closely that the real facts are obscured. Instead of taking them at face value, scholars should recognize some distinct rhetorical devices connected with schemas and tropes. A schema alters ordinary sentence structure to give something larger scale, while a trope is a 67gure of speech that enhances the e0fect of a word, such as a simile, metaphor or hyperbole. Here, it was a schema of connection, that of parallelism. That device was a similarity of structure in a pair or series of related words, phrases or clauses which balanced the construction with ornamentation. It can also add coherence to an argument.21 Juwaynī did not dispense with the main events but repeated almost the same exaggerated details about Marv and Nishapur in order to make

:Q Juwaynī was not alone in in./ating 67gures. A near contemporary, Ibn Khaldūn, writing in Egypt in the late fourteenth century, recognized this tendency and thoroughly demol- ished any faith in such reporting: The Muqaddimah, 16, and continuing for many more pages of examples. A more recent disclaimer relates to the second battle of Homs in 1281. Contemporary sources gave 67gures for the Mamluk army at 100,000 or half that at 50,000 men. Amitai-Preiss doubted both 67gures and preferred not to depend on any particular number. Amitai-Priess, Mongols and Mamluks, 193–94. :8 Corbett, Classical Rhetoric, 447. CEd =>?@AB a point. By doing so, he re./ected or transferred in a wholesale manner information about the Mongol war machine and then put a highly emotional twist to it. The narrative related direct if controlled history, but the literary touches propelled the pace and explosion of events. Thus although historians have so far generally discredited the accuracy of the actual numbers, they should also be critical of the reliability of the skewed statements because they stress certain events and almost ignore others. It is necessary, therefore, to be aware that Juwaynī’s agenda dealt with more than the basic actions of the Mongols. He was also imposing his 67nal message, notably that the old order had been smashed. Nevertheless, even though Juwaynī transferred the Mongol message of intimidation with these stories of atrocities, he provided various means for understanding the newcomers. The methods were often simple, but they were the beginning of adoption, the second phase of literary development. In particular, he glossed the motif of utter destruction and other events with a number that seems to have been agreeable to him and his contemporaries. The round number four appeared continuously; four itself, forty and four hundred.22 It is clearly a case of another rhetorical schema, in this situation of polyptoton or the repetition of words derived from the same root. His love of euphony, however, in this instance, apparently overrode his knowledge of Mongol culture. This round number contrasted with the Mongol preference for the odd number three and its multiples. Steppe tradition, going back at least to the seventh century in the Orkhon Valley of Central Mongolia, had exhibited numerous uses of it in ceremonies and building layouts.23 The legacy still exists with many customs such as the circumambulation three times around sacred cairns or processions led by Chinggis Khan’s standard of nine horsetails. Once again, small details indicate that rhetoric and cultural divergence between Iranian and Mongol aesthetics and practices sometimes informed the text more than reality did. It seems that Juwaynī ignored this minor di0ference in order to frame the Mongol invasion in a capsule of eloquent Persian language and cultural norms that could lead to adoption.

Positive Motif In the 67nal step of integration, the most telling and clear case is the positive motif of the glori67cation of the noble ruler: here Ögetei Khan (1229–41) is the speci67c example. In a remarkable series of extended anecdotes, Juwaynī

:: The number appears regularly throughout the text. Another example is his report about the festivities that preceded Ögetei’s quriltai: “When the forty days had come to an end . . .” (186). :; Kolbas, “Khukh Ordung,” 303–28, 318. "RBS>TRUA? VWRU AB X>YZ>? WT>WAZAY[A? CED showed the integration of the two completely diverse cultures. For a full chapter entitled “On the Deeds and Actions of Qa’an,” which runs for 35 pages in the English translation (201–35), the theme is magisterial bene67cence. He introduced his many tales that often included speci67c individuals with the comment: “Those in need that came to him from every side speedily returned with their wishes unexpectedly ful67lled and o067ce-seekers and petitioners straight-away with whatever each of them had desired. . . . Upon those that came from distant and rebellious lands he bestowed presents in the same way as upon those that were from near and subject countries. And no one went away from his presence disappointed or frustrated” (203). Numerous stories follow of lowly individuals highly rewarded 67nancially or honoured by being admitted to Ögetei’s presence. The anecdotes are quite wonderful and rather inventive, but each has a clear message of bene67cence. One expressly had the khan saying: “How long must I ask you (a courtier) not to deny my bounty and begrudge petitioners my property?” (223). Indeed, the apocryphal tales resounded so well that Rashīd al-Dīn repeated similar ones in Part Three of his history of Ögetei.24 The khan received special attention because Juwaynī loved words. He was aware of the Mongol language, in which the in67nitive ög, meaning ‘to give,’ formed the basis of the khan’s name. The prepositional ending of dei or tei indicates ‘with’ so that the word Ögetei can be expressed as ‘generosity.’ The meanings of many other Mongol names are known, such as Möngke, ‘enduring,’2F and Hülagü, ‘a loyal and clever horse’;2G but Juwaynī did not even hint at the meanings of these names. It seems he chose to elaborate so extensively on Ögetei because the name symbolized the Persian as well as the steppe sense of a great ruler, and Ögetei personi67ed just such a ruler. In spite of that, Ögetei had lived in the past, even though the recent past; so the idealization had the motif of a Golden Era. Since modern scholars have not been aware of this word play nor its signi67cance, they have generally depicted Ögetei as a pro./igate at worst and too generous at best. Neither characteristic can be veri67ed from a better understanding of Juwaynī’s word-play. It was part of a positive spin for the integration of two cultures.

:< They require twelve pages in the English translation, 2:334–45. :L The word is usually translated as ‘eternal’ but is better understood as ‘enduring’; Jackson, “Mongol Khans,” 113. :M There has been no general analysis of the culture around Mongol names, especially the manner and point at which they changed. Although most names were probably given at birth, it is obvious that some common ones, for example, Buqa, a good wrestler, were given or granted to the individual as an adult to note his physical prowess. CEE =>?@AB

To further the theme of the just and noble ruler, in the next chapter, Juwaynī dealt with Ögetei’s buildings and palaces for three pages in the English translation (236–38). As with Ögetei’s name, Juwaynī’s treatment was a complete break from descriptions of any other ruler, whether pre-Mongol or Mongol. This chapter established that the Mongols were not rough nomads wander- ing around the steppe or, in fact, wanton destroyers of cities. Indeed, Ögetei’s policy to stabilize the empire had a large component of urbanization. It was implemented across the empire and can be attested by numismatic evidence for revised tax policies.2H In addition, Ögetei established Qaraqorum, which became the ceremonial centre of the empire, especially during Möngke’s reign (1251–59). This theme was not wishful thinking since urbanisation continued even after Juwaynī’s time as other khans also saw bene67ts to urbanization. Qubilai (1260–94) founded Daidu and Arghun (1284–91) Sultaniyya. Moreover, various khans of the Golden Horde, especially Özbeg (1312–41), established a string of cities along the eastern bank of the Volga River, the 67rst time this region had any urban centres. For this early period then, Juwaynī captured in a speci67c chapter a major transition in imperial administration that started in the middle of Ögetei’s reign. Tellingly for historical context and accuracy, these two chapters immediately follow the one on the second quriltai or congress that established the important reforms.2I

Propaganda This brief survey of negative and positive motifs may have created the impression that Juwaynī was helping the Mongols with their propaganda. Most commentators have been more sympathetic by implying that he could hardly have condemned the rulers for whom he worked,2J yet the motifs and reasons for them would have been apparent to Persian readers. In that respect, his work interpreted historical events according to the new fabric of society and daily life. The 67rst motif emphasized that the Mongol conquest did slaughter many,

:N Kolbas, The Mongols in Iran, 97–98; regarding taxation and urbanization. Further, the programme was e0fective since numismatic evidence shows that Khurasan was returning to normal by 1242 at the end of Ögetei’s reign: Kolbas, The Mongols in Iran, 109. :O On Möngke’s accession to the throne, the khan revitalized the empire with strong, centralizing reforms that followed the original outlines set by Ögetei especially with the taxation policies of Yalavach. Kolbas, The Mongols in Iran, 97, 100, 140; and Allsen, Mongol Imperialism, 79–80, and fn. 4. :P For example, Boyle explains that “it was not su067cient to record the good qualities of the Mongol invaders; as an o067cial in that service, Juvaini had to justify the invasion itself. This he has done by representing the Mongols as the instrument of divine will.” Juwaynī/ Boyle, xlv. "RBS>TRUA? VWRU AB X>YZ>? WT>WAZAY[A? CE( but those operations were mainly due to rebellion or resistance. The actual numbers did not matter: the spirit of revolt was crushed. In Juwaynī’s time, Iran was united; and since the Pax Mongolica had general sway across Asia, he seemed to say that there were substantial bene67ts from peace and stability. The second motif suggested that the Mongols reigned with wisdom and bene67cence; at least the ideal could be found in one khan. Juwayni probably re./ected the view of many since, once again, Rashīd al-Dīn followed this stance in his later comments extolling the reforms of Ghazan Khan (1295–1304).3K Thus although the text could be viewed as serving Mongol interests, it seems that he expounded or even assisted in clarifying the prevailing opinions held by an elite Persian audience. That group was well versed in the arts of literature and prevailing philosophical concepts, none of which particularly concerned the Mongols. Instead, their culture used a di0ferent calendar, language, script and symbols of royalty which distinctly separated them from the subject peoples. The internal evidence from Juwaynī’s text, such as his constant allusions to Islamic and recent dynasties like the Khwarazmshahs for whom his grandfather had worked, leads to this interpretation of Persian acceptance and integration. With his compositional development and chosen events, Juwaynī ended by providing a vision of security and good governance;31 and he did so by depicting much of Mongol activity in Persian cultural terms.

Stylistic In!luences Ultimately though, this analysis cannot ignore in./uences Juwaynī received from previous masterpieces.32 To observe these only brie./y, Juwaynī was perhaps inspired thematically by the epic confrontation of tragic 67gures found in the Shāhnāma. Stylistically, the great epic did a0fect his poetry as a number of distiches mirrored the tone and intensity of the equivalent kinds of con./ict in the poem.33 Moreover, in some limited cases such as praising Jalāl al-Dīn, Juwaynī even borrowed a few verses from the Shāhnāma.34 For prose

;Q For example, Rashīd al-Dīn stated in Chapter 20: “These several important matters, which not one of the former caliphs or sultans managed, is easy for the Ruler of Islam, may his reign be eternal. It is so well accepted that it (the reform) is the cause of the general tran- quility.” Kolbas, The Mongols in Iran, 317. ;8 Boyle’s last comment shadows such an interpretation: “their object is clearly to reconcile the author and his readers to the inevitable.” Juwaynī/Boyle, xlviii. ;: Melville has supplied a more detailed and helpful look as well as the analyses of sources and in./uences by others, “Jahangosha”, 380, followed by an extensive bibliography. ;; Melikian-Chirvani, “Conscience du passé,” 135–77. For thematic in./uence, 137; for distiches 139, 141 and following. ;< Juwaynī/Boyle, xliv. CE\ =>?@AB

in./uences, he built on the dramatic but naturalistic style already well developed by Abū al-Faẓl Bayhaqī, who wrote in 1187, an historian of the Ghaznavids. Both authors worked in the chancellery for the new conquerors but wrote exciting episodes and anecdotes for the public. Their primary focus was organizing and interpreting their readers’ perceptions of the impact of warlords and nomads upon urban culture.3F Unlike Bayhaqī, however, Juwaynī also incorporated other literary developments to full advantage. He also wrote in a period of somewhat related developments amongst the Mongols themselves as recognized by Igor de Rachewiltz. He considers that The Secret History was also an epic, especially an epic chronicle rather than a heroic epic.3G Moreover, Juwaynī was not the 67rst to travel this road of amalgamating the strong foundations of one culture into another. Even preceding the works discussed above, the process had also occurred after the Islamic conquest of Iran, 67rst being absorbed slowly but then ingrained at least by the eleventh century. Later during the Seljuq period, for example, new rulers often created their genealogy to re./ect descent from both Muhammad’s circle and that of the mythi- cal forebears of Iranian nobility. This dual integration is especially evident in the genre of the Mirrors for Princes.3H In e0fect then, Juwaynī reacted in much the same way to invaders as had his forbears, only he created a di0ferent and broader genre that included all educated Persians, not just rulers and their courts.

Conclusion

His conclusion was hard-won since he had initially deplored so much loss in the abrupt and devastating clashes of empire-building. Eloquent metaphors underscored his concern: “the land of Khorasan in particular, . . . the location of desirable things and good works, the font of learned men, the rendezvous of the accomplished, the spring-abode of the learned, the meadow of the wise, the thoroughfare of the pro67cient and the drinking place of the inge- nious . . . To-day, I say, the earth hath been divested of the adornment of those clad in the gown of science and those decked in the jewels of learning and letters. . . . ”(6).3I Even this lament with its many rhetorical tropes of metaphor,

;L Hardy, “Review of Marilyn Waldman,” 334–44; 336, 342. ;M de Rachewiltz, The Secret History of the Mongols, 1. ;N Tor, “The Islamisation of Iranian Kingly Ideals in the Persianate fürstenspiegel,” 115–22. ;O He continued to thoroughly lament the state of a0fairs, giving no credit in the slightest to the Mongols, a tirade that could be considered the beginning of his transition to the later stages and chapters. "RBS>TRUA? VWRU AB X>YZ>? WT>WAZAY[A? CE] hyperbole and presumably paronomasia3J in the Persian original was probably meant to encourage others to revive intellectual activity in a stable realm. It was also a call, it seems, to reinstate his own formative training that might have derived from Aristotle’s Rhetoric as the canon for good writing. The training would have stressed the 67ve basic tenets of rhetorical exposition, those of invention, arrangement, style, memory and delivery.4K Juwaynī’s entire work incorporated all these points. Furthermore, a reliable study of the original Persian text would better uncover his full style. He would have exercised his own method of sentence articulation and in particular maximized sentence euphony. That would include rhyming prose or sajʿ.41 Moreover, he probably surrounded important words that had strong stress by ones with weak stress to enhance his choice of words by contrast. Indeed, euphony and rhythm com- municate as much as the formal word.42 It seems that Juwaynī was extremely conscious of the aural cadence, bite or soothing nature of his phrases. As a result, in a variety of ways, Juwaynī employed a literary style that used detailed factual history to create epic prose. His motifs were well-developed, not clichés, and his motives were to explain in Persian terms the new order. He was able to depict seismic shifts in world history and cultural patterns, but historians should be cognizant of his artistic methods and social motives. It is worth considering that he may have been a spokesman for the entire society. The Mongols, too, were adapting and integrating to produce a “strikingly atypical evolution” of a steppe empire.43 Juwaynī created a masterpiece for a Persian memory of the period, and as he constructed his work over the three stages that were later intermingled which interpreted his time, he chose speci67c directions and selected the impact of his narrative. He composed a vast lyrical tapestry full of rhetorical devices and human emotions through de67nite themes and variations, poetic allusions and dramatic intensity. Juwaynī ended with the belief that order could emerge from so much chaos and was able to present his own epic struggle almost forty years after the Mongols 67rst appeared in Iran.

;P Paronomasia is the use of words alike in sound but di0ferent in meaning. ?@AB

Bibliography

Allsen, Thomas T. Mongol Imperialism: The Policies of the Great Qan Mongke in China, Russia and the Islamic Lands, 1251–1259. University of California Press, Berkeley, 1987. Amitai-Priess, Reuven. Mongols and Mamluks: The Mongol—Il-khand War, 1260–1281. Cambridge University Press, Cambridge, 1995. Buell, Paul D. “Central Eurasia: Genocide as a Way of Life?” Unpublished. Corbett, Edward P. J. Classical Rhetoric for the Modern Student. Second Edition. New York, 1971. al-Fārābī, Abū Naṣr. al-Fārābī on the Perfect State: Abu Naṣr al- Fārābī’s Mabādi’ Ārā’ Ahl al-Madīna al-Fadilā. Translated by Richard Walzer: Clarendon Press, Oxford, 1985. Grachev, Anton I. and Pavel N. Petrov. “Klad mednih poserebrennih dirhemov Merva vremeni zavoevaniy Chingiz-khana.” [A hoard of silver-washed copper dirhams from Marw of Chingiz Khan at the time of conquest]. Numismatika Zolotoy Ordi [Numismatics of the Golden Horde] 1 (2011): 124–38. Grousset, René. The Empire of the Steppes: a History of Central Asia. French original edition, 1970. Translated from the French by Naomi Walford. Rutgers University Press, London, 2008. Hardy, Thomas. “Review of Marilyn Waldman, Toward a Theory of Historical Narrative: A Case Study in Perso-Islamicate Historiography.” History and Theory 20 (1981): 334–44. Ibn Khaldūn, Walī al-Dīn. The Muqaddimah: An Introduction to History. Volume 1. Translated by Franz Rosenthal. Princeton University Press, Princeton, 1958. Jackson, Peter. “Mongol Khans and Religious Allegiance: the Problems Confronting a Minister-historian in Ilkhanid Iran.” Iran 47 (2009): 109–22. Juwaynī, ʿAṭā Malik. Ta’rīkh-i Jahāngushā. Edited by Mīrzā Muḥammad Qazwīnī. 3 vols. Leiden and London: Brill and Luzac, 1912, 1916, 1937. ———. The History of the World-Conqueror. Translated by John Andrew Boyle. Reprint, Manchester: Manchester University Press, 1997. Kolbas, Judith. “Khukh Ordung: A Uighur Palace Complex of the Seventh Century”. Journal of the Royal Asiatic Society 15 (2005): 303–28. ———. The Mongols in Iran: Chingiz Khan to Uljaytu 1220–1309. Routledge, London and New York, 2006. Melikian-Chirvani, Assaddullah Souren. “Conscience du passé et résistance culturelle dans l’Iran Mongol.” In L’Iran face à la domination mongole. Edited by Denise Aigle, 135–77. L’Institut Français de Recherche en Iran, Teheran, 1997. Melville, Charles. “Jahāngošā-ye Jovaynī” Encyclopedia Iranica, i`_: 378–82. Routledge Kegan Paul, London, 1985. Morgan, David O. “The Mongols in Iran: A Reappraisal.” Iran 42 (2004): 131–36. "RBS>TRUA? VWRU AB X>YZ>? WT>WAZAY[A? C(C

Poliakova, E. A. “The Development of a Literary Canon in Medieval Persian Chronicles: The Triumph of Etiquette.” Iranian Studies 17 (1984): 237–56. de Rachewiltz, Igor. The Secret History of the Mongols: a Mongolian Epic Chronicle of the Thirteenth Century. Translated with a Historical and Philological Commentary. Volume 1. Brill, Leiden and Boston, 2004. Rashīd al-Dīn. Jami‘u’t-tawarikh [sic] Compendium of Chronicles. Translated by Wheeler M. Thackston. Vol 2. Cambridge, j#: Department of Near Eastern Languages and Civilizations, Harvard University, 1998–99. Rossabi, Morris. “The Reign of Kubilai Khan.” In The Cambridge History of China. Volume 6. Alien Regimes and Border States: 907–1368. Edited by Herbert Franke and Denis Twitchett, 414–89. Cambridge University Press, Cambridge, 1994. Rypka, J. “Poets and Prose Writers of the Late Saljuq and Mongol periods.” In The Cambridge History of Iran. Volume 5. The Saljuq and Mongol Periods. Edited by J. A. Boyle, 550–625. Cambridge University Press, Cambridge, 1968. Tor, D. G. “The Islamisation of Iranian Kingly Ideals in the Persianate fürstenspiegel.” Iran 49 (2011): 115–22. Weatherford, Jack. Genghis Khan and the Making of the Modern World. Three Rivers Press, New York, 2004. Whitman, Marina D. “The Scholar, the Drinker, and the Ceramic Pot-Painter.” In Content and Context of Visual Art in the Islamic World. Edited by Priscilla P. Soucek, 255–61. College Art Association, Pennsylvania State University Press, London, 1988. !"#$%&' ( From the Mongols to the Timurids: Re!"nement and Attrition in Persian Painting

Karin Rührdanz

In 1544/45, the calligrapher Dūst Muḥammad compiled and interpreted an album (Topkapı Sarayı Müzesi, H. 2154) for Prince Bahrām Mīrzā, the brother of Shāh Ṭahmāsp (r. 1524–76). His introduction contains the now famous statement that during the reign of the Ilkhan Abū Saʿīd, “Master Ahmad Musa . . . lifted the veil from the face of depiction, and the [style of] depiction that is now current was invented by him.”0 Starting with Aḥmad Mūsā, Dūst Muḥammad constructed a history of Persian painting, building upon a chain of biographical notes stressing the master–disciple relationship leading up to his time in the mid sixteenth century. His narrative was originally supported by a selection of pictorial examples, some of which are still in the album and allow us to follow his reasoning. A mi‘rāj painting1 that he attributed to Aḥmad Mūsā represents the beginning of the transformation process. A late fourteenth-century miniature2 ascribed by Dūst Muḥammad to ʿAbd al-Ḥayy marks its completion. The latter picture once illustrated a manuscript (British Library, Add. 18113) which contains three of Khwājū-yi Kirmānī’s mathnavīs copied in 1396 for the Jalayirid Sultan Aḥmad in Baghdad. Together with nine more miniatures still preserved in the manuscript they exemplify the degree of re45nement Persian painting had reached within seven decades:

Their densely organized scenes, with subtle rhythms and colors, have a fairy-tale quality that transcends their narrative function. Minute details and delicately rendered 45gures add to the intensity of these illustrations, which depict the turbulent romance of the Persian prince and princess Humay and Humayun.6

7 Thackston, A Century of Princes, 345. 9 H. 2154, fol. 31b, for colour reproduction see Çağman and Tanındı, The Topkapı Saray Museum, no. 45. ; H. 2154, fol. 20b, for colour reproduction see Sims, Peerless Images, 262–63, no. 179; for the identi45cation as part of a Jalayirid manuscript see Prentice, “A Detached Miniature,” 60–66. < Lentz and Lowry, Timur and the Princely Vision, 53.

The Khwājū-yi Kirmānī miniatures perfectly demonstrate the skilful exploitation of the newly developed stylistic means—the emphasis on the vertical combined with a view from above to the horizon; the larger space accom- modating smaller 45gures and the opening to the other world on the margin; and the ideal landscape or an architectural background, in this case a pala- tial structure worthy of paradise—for the visualization of mystical ideas.N In the century that followed, interpretation and paraphrasing of the romantic mathnavī with strong mystical connotations largely determined the development of Persian manuscript illustration.O Analysis of the development of Persian painting from c. 1330 to c. 1400 has mostly focused on the formation of the late Jalayirid style as represented by the miniatures in the Khwājū-yi Kirmānī manuscript and its Timurid continuation in Shiraz and Herat in the early 45fteenth century.P Since very few illustrated manuscripts are preserved from the Jalayirid period, a hypothetical chronology of artistic evolution has been constructed, mainly by using the miniatures cut from manuscripts and mounted in albums.Q Inevitably, this reconstruction had to be carried out with the starting and the end point in view, and, consequently, sidelined those paintings which did not 45t a straight line of stylistic development. Also, research largely focused on those pictures in the Istanbul albums that were identi45ed as having belonged to Kalīla wa Dimna and Shāhnāma manuscripts.R This has created the impression that Persian miniature painting started with a very narrow range of texts to be illustrated which was expanded over time. In fact, the second half of the fourteenth century seems to have seen the most open-minded approach to the visualization of literary subjects.0S Building on the artistic experience acquired during the late Ilkhanid period, painters boldly took on a variety of genres and topics before

T On the technical side, the relationship between text and image was built upon a grid which served the arrangement of the lines of poetry on the written surface as well as the structuring of the painting’s composition, see Porter, “The Illustrations,” 359–74. U Fitzherbert, “Khwājū Kirmānī,” 137–51; Barry, Figurative Art, 121–31. V Most detailed Gray, ed., The Arts of the Book. W Duda, “Die Buchmalerei der Ğala’iriden,” (1972), 28–76 and (1973), 153–220. Y Cowen, Kalila wa Dimna; Grube, “The Istanbul University Library,” 52–77; O’Kane, Early Persian Painting; Atasoy, “Four Istanbul Albums,” 19–49; Idem, “Illustrations Prepared,” 262–72. 7Z Dealing primarily with a di[ferent group of material from the Istanbul albums, the so- called Siyah Qalam works, and attributing it to Jalayirid Tabriz and Baghdad, an article by Bernard O’Kane, “Siyah Qalam,” 2–18, points in the same direction: The ‘window of opportunities’ opened by the Mongol conquest was fully exploited in the later half of the fourteenth century, and then closed step by step. =>\ ']DAM^G_ a streamlining process started which eventually made Persian manuscript illustration the perfect companion to mathnavīs and dīwāns. Since the path of development leading up to the point of re45nement reached in the last decade of the fourteenth and the 45rst decades of the 45fteenth century in Jalayirid and Timurid painting is well known, attention is directed, here, to those promising beginnings that fell victim to the 45ltering. What follows is an attempt to suggest a literary context for several groups of album miniatures that, stylistically, belong to the main stream of artistic development from the late Ilkhanid to the Jalayirid periods but stand out because of their enigmatic subjects. This paper is only the 45rst step of a work in progress and looks for preliminary answers among the material in one of the two relevant groups of albums, the Diez albums Diez A Fol. 70–72, bequeathed to the now Berlin State Library by Freiherr von Diez, Prussian envoy to the Ottomans until 1790.00 In these albums, a number of paintings taken from the Topkapi Saray albums H. 2152 and H. 2153 were rearranged, together with material from other sources.01 Therefore, this research cannot be completed before the preliminary conclusions are checked against information contained in the relevant Istanbul albums.02 One must also take into account that Diez’s personal approach led him to acquire a selection which would not adequately re`aect the original content of the Topkapi Saray albums. One of the Istanbul albums, H. 2152, has been analyzed as a “Timurid work- shop album,”06 and it seems highly possible that H. 2153 (and probably H. 2160) played a similar role, although not in a Timurid but rather in an Āq Qoyūnlū or Ottoman court atelier. The court artists themselves were obviously inclined to preserve manifold material for reference and inspiration even if it were far removed from what they were commissioned to produce at that time. Relying on the scant information available about the current content of the Istanbul albums, it seems that Diez acquired a fair share of the Jalayirid material. In Ipşiroğlu’s catalogue of the Diez albums, it is described under Group ddd: Post-Ilkhanid miniatures.0N Conservation work, which included removing the pictures from the albums and taking apart those miniatures that had been glued together back to back, established that almost none of them had text on the back.0O Although how and

77 Ip şiroğlu, Saray-Alben. 79 On the genesis of the Diez albums, see Roxburgh, “Heinrich Friedrich von Diez,” 112–36. 7; For an overview, see Çağman, “On the Contents of Four Istanbul Albums,” 31–36. 7< Roxburgh, The Persian Album, chapter 3, 93–106 in particular. 7T Ip şiroğlu, Saray-Alben, 35–57. 7U Appel and Dieter, “Die Saray-Alben,” 227–32. @ABC %DE FBGHBIJ %B %DE %KCLAKMJ =>e when the texts ‘disappeared’ remains an enigma, a recent 45nding con45rms that the cut-out pictures were parts of manuscripts.0P Concerning literary genres and topics illustrated by those context-deprived miniatures, Ipşiroğlu mentioned the Shāhnāma, Oriental versions of the Alexander romance, popular literature or adventure and seafaring stories.0Q He did not try to be more speci45c when dealing with the individual miniatures. However, even this general statement is signi45cant: it contains highly unusual genres and topics if compared to the output of illustrated manuscripts over the following two centuries. Referring to the assumed literary context of the miniatures, Ipşiroğlu claimed elsewhere “that cultural life was reduced to a popular level” in post-Ilkhanid times.0R This seems a one-dimensional judgment if we take into account, for instance, Browne’s evaluation of the fourteenth century as “remarkable alike for the quantity and the quality of the poets and writers which it produced.”1S However, the album material testi45es not to the production, but to the con- sumption of literature and, more speci45cally, to that part of literature that was deemed appropriate and worthy to be visualized. As we shall see, this part is barely connected to the literary production of the fourteenth century. Mystical poetry written before and after the Mongol conquest obviously had not yet cast its spell over patrons and ateliers. It was another kind of text which proved most attractive, as the more or less preserved manuscripts of the Shāhnāma and Kalīla wa Dimna show. The following is an attempt to categorize material from Ipşiroğlu’s “Post-Ilkhanid miniatures” group with respect to its connection to certain genres and texts. One miniature, correctly identi45ed for the 45rst time by Nora Titley, is “Bahman visiting the mausoleum of Rustam’s family.”10 She based her identi45- cation on an early Timurid miniature illustrating the Bahman-nāma in a collection of post-Shāhnāma epics.11 This manuscript (British Library, Or. 2780)

7V Sims, “Bilder zu Ferdausis Schahname,” 30–31. Interestingly, two pages (103b and 100b) in the album H. 2153 with text on the miniature side, preserved and to be read continuously (see Khaleghi-Motlagh edition, 2:31), constitute the two sides of the same leaf. See Atasoy, “Four Istanbul Albums,” 45gs. 23 and 24. Should we take into account the splitting of pages at an early stage? 7W Ip şiroğlu, Saray-Alben, 36. 7Y Ipsiroglu, Painting and Culture, 55. 9Z Browne, A History of Persian, 159. 97 Diez A Fol. 72, 29, see Ipşiroğlu, Saray-Alben, ddd/7 (p. 40), Taf. fgddd, Abb. 24; for a colour reproduction see Komaro[f and Carboni, The Legacy of Genghis Khan, 54, 45g. 52; or Gonnella and Rauch, Heroische Zeiten, 117, no. 37. 99 British Library Or. 2780, fol. 171b; see Titley, Persian Miniature, 45–47, pl.3. =>h ']DAM^G_ is dated 1398,12 but was only provided with inserted miniatures about 20 years later, in early Timurid Shiraz.16 Together with an illustrated Garshāsp-nāma, dated 1354,1N this compendium of post-Shāhnāma epics supports the hypothesis that some of the Jalayirid paintings once illustrated post-Shāhnāma epics. These epics belong to the texts that were sidelined in the transformation described earlier. When they later received the attention of painters once again, they were no longer treated as separate texts but as interpolations in the Shāhnāma.1O It may well be that a number of miniatures so far identi- 45ed as Shāhnāma illustrations come, in fact, from those epics, because well- known Shāhnāma topics reappear in them, with other heroes performing the same deeds. On the one hand, the iconography concerning the appearance of Rustam does not appear to have been 45xed for the most part of the fourteenth century. Apart from being dressed in his famous tiger skin, he might be depicted undistinguished or with a leopard skin.1P On the other hand, in the above-mentioned collection of epics transcribed in 1398 and illustrated most probably under the Timurid prince Iskandar Sultan, Garshāsp is twice shown in an attire that resembles that of Rustam.1Q This demonstrates that a leopard, and probably even a tiger skin garment, often combined with the related head- dress, was sometimes used to identify other members of Rustam’s family who became protagonists in the post-Shāhnāma epics. Some album miniatures featuring heroes dressed in this way1R may, therefore, originate from post- Shāhnāma epics. So far, however, neither a review of the Garshāsp-nāma2S nor of the Bahman-nāma20 has produced reliable identi45cations, i.e., descriptions

9; Rieu, Supplement, 133; Titley, Miniatures, 39–40, no. 99. 9< Wright, “Firdausi and More,” 65–84. 9T Topkapi Saray Museum, H. 674, see Ettinghausen, “On Some Mongol Miniatures,” 59–65. 9U Rührdanz, “About a Group of Truncated Shahnamas,” 118–19. 9V In a characteristic Rustam episode, the hero is dressed in leopard skin in the miniature Diez A Fol. 71, p. 26, no. 2, see Gonnella and Rauch, Heroische Zeiten, 124, no. 47. 9W Or. 2780, fol. 18a, see Stchoukine, Les peintures des manuscrits timurides, pl. fd; fol. 29b, see Robinson, Persian Miniatures, pl. dd. An album miniature (Diez A Fol. 71, 46) represents Sām, father of Zāl, dressed like Rustam, see Gonnella and Rauch, Heroische Zeiten, 125–26, no. 48, in the same way as Zāl is set apart from other heroes and courtiers surrounding King Zāv on a miniature from the Great Ilkhanid Shāhnāma, Sackler Gallery, Washington, S1986.107, see Komaro[f and Carboni, The Legacy of Genghis Khan, 41, 45g. 37. 9Y Diez A Fol. 72, p. 26, see Ipşiroğlu, Saray-Alben, ddd/23, p. 45; for a reproduction, see Gonnella and Rauch, Heroische Zeiten, 114, no. 35. ;Z Ṭū sī, Le livre de Gerchāsp. ;7 Irānshāh ibn Abī ’l-Khayr, Bahman-nāma. I would like to thank Shabnam Rahimi- Golkhandan for checking the text in detail. @ABC %DE FBGHBIJ %B %DE %KCLAKMJ =>> of events other than the above-mentioned episode that are su[45ciently speci45c to be 45rmly recognized as illustrated by Diez album miniatures. Two miniatures, which belong to a series of four pictures extant from a story about a man and a female ape, are related to a completely di[ferent genre. While the picture showing the white-bearded man approached by a group of gesticulating monkeys21 is in a Topkapi album together with a second one, Diez acquired the two more interesting miniatures that tell us the story.22 Although not 45tting the pictures in detail, a story in the ʿAjāʾib al-Hind by Buzurg ibn Shahriyār comes close.26 Taking the multiplicity of illustrations in some fourteenth-century manuscripts into account,2N it is not unlikely that this story was provided with four miniatures, although one would prefer to think of a more elaborated text.2O Such a text could have been part of a collection of several seafaring tales. A few miniatures representing adventures at sea in the Diez albums could also have belonged to such tales.2P There is, however, another possibility. A monkey story, as well as stories about journeys or `aights by sea, shipwrecks and encounters with horrible sea monsters are also part of several popular romances. The late sixteenth-century illustrations of the Dārābnāma made for the Mughal emperor Akbar (r. 1556– 1605) include not only a number of 45ghts and adventures at sea, but also the encounter of a heroine’s brother with a female ape.2Q Moreover, even in some post-Shāhnāma epics, journeys by sea play a role.2R The romances, however,

;9 H. 2153, p. 155b, see Ipşiroğlu, Painting and Culture, 58 and 100, no. 15. ;; Diez A Fol. 71, p. 12, and Diez A Fol. 72, p. 19, see Ipşiroğlu, Saray-Alben, ddd/62 and ddd/63 (pp. 52–53, Taf. fgd, Abb. 22). ;< Buzurg ibn Shahriyār, The Book of the Marvels of India, 58–61. In this account, a sailor staying behind on a ship enjoyed visits of a she-monkey and eventually had sex with her. Having noticed her pregnancy he abandoned the ship and left in a small boat. Only much later he heard that his returning comrades met his o[fspring on the ship. ;T This was, for instance, established for the Great Ilkhanid Shāhnāma, see Blair, “On the Track of the ‘Demotte,’ ” 128; and the less sumptuous and smaller in size Kalīla wa Dimna in the BnF, suppl. pers. 913, see O’Kane, Early Persian Painting, 283. ;U Because of their size (for instance, for the published miniature Diez A Fol. 72, p. 19: 24.5 x 24 cm), it does not seem likely that the four paintings were made to accompany an oral presentation of the story. ;V Diez A Fol. 71, pp. 17, 36, 39, and Diez A Fol. 73, p. 29, see Ipşiroğlu, Saray-Alben, ddd/67, ddd/66, ddd/65, and ddd/68 (respectively, pp. 54–55, Taf. fg, Abb. 20, Taf. fgd, Abb. 21). ;W Compare an illustration in the Dārābnāma made for Akbar, now British Library Or. 4615, fol. 41a; see Titley, Miniatures from Persian Manuscripts, 8–11, no. 18. ;Y When the hero goes by ship to confront an enemy, or undertakes a kind of sightseeing journey to remote islands, the battles and adventures are commonly located on land, =>( ']DAM^G_ seem the most promising choice because of the range of subjects they o[fer to narrative painting:

Thematically, popular literature displays elements common to both polite and folk literature. [. . .] the popular romances are derived by and large from the national legend, with exceptions [. . .] In like manner, the romances contain numerous motifs common in folklore, travel literature and popular religion.6S

There are seven pictures in the Diez albums which cannot but derive from popular romances. These miniatures show princes and warriors in action while a strange, usually smaller 45gure, is present, dressed in a short garment, his naked legs in light shoes, and on his head a conical cap or hat (45gs. 8.1– 8.5).60 Ipşiroğlu calls him a jester (“Hofnarr”).61 What part, however, may a jester play on the battle45eld, in connection with the storming of a prison, or in providing a prince with the proper out45t for war? The proximity to the court of this seemingly non-heroic 45gure and the activities he is engaged in lead to the conclusion that he must be an ʿayyār. The multifaceted term ʿayyār is used here in the ‘positive sense,’ meaning “a noble-minded highwayman, or a generous, clever, brave, modest, pious, chaste, hospitable, generally upright person. This last image is found mostly in poetry, in adab and Su45 texts, and popular romances.”62 In the popular romances, it is the principal ʿayyār, sometimes helped by his ʿayyār friends, who, with unfail- ing loyalty, tackles the seemingly hopeless situations the prince had put himself in; who 45nds the way into the secret enemy prison to rescue the abducted princess; who risks his life many times while acting in disguise; and who does all the dirty work to enable the prince to become victorious at the end. He gives advice to the prince, delivers his messages, and spies on the enemy troops, but never 45ghts in battle or engages in single combat with a warrior. Therefore, an ʿayyār would never be represented acting and dressed like a warrior hero.

while the crossing of the sea is only mentioned as a means to get there; see, for instance, the second part of the Garshāsp-nāma: Asadī Ṭūsī, Le livre de Gerchâsp, 2. j

@dkLAE (.= Prince being equipped with arms and armour by the king. lm^^mJnKnIKBmDEo $AELJJKJpDEA qLImLAnEJKm_, rAKEGm^nmEKILGH, sKE_ # @BI. >=, t. uh, GB. =.

His appearance should re`aect his lower social background and his particular function. One way to check whether the peculiar 45gure on those seven album pictures might be an ʿayyār is through visual comparison with such heroes in extant illustrations from popular romances. Some information can be extracted from the only preserved manuscript of the ʿayyār-centred romance Kitāb-i Samak-i ʿAyyār,66 which was transcribed and illustrated in the period immediately preceding the Jalayirid paintings. Its three volumes6N prove that the subject had been deemed appropriate for illustration, at least at the provincial court of the Inju dynasty in Shiraz between 1330 and 1350. The illustrations, however, are

<< Gaillard, “Samak-e ʿAyyār.”

@dkLAE (.u A king on horseback encounters a slain warrior. lm^^mJnKnIKBmDEo $AELJJKJpDEA qLImLAnEJKm_, rAKEGm^nmEKILGH, sKE_ # @BI. >=, t. ??.

not overly helpful. Figure drawing in Inju style does not di[ferentiate much between courtiers and servants, women and beardless men, warriors and ʿayyārs. Nevertheless, Samak is distinguished by his attire, if only slightly. Like some other 45gures, for instance a vizier, he appears in a kind of civil servant dress. In contrast to those other non-warrior 45gures, however, he is characterized by one dress element which identi45es him with some consistency: a conical cap with some turban cloth slung around.6O From descriptions in the text of the Kitāb-i Samak-i ʿAyyār and some other popular romances, Hanaway concluded “that there was more or less a standard

borders of Iran, while the garb develops into a

@dkLAE (.? On a terrace, courtiers take care of a man fainting in front of a ruler. lm^^mJnKnIKBmDEo $AELJJKJpDEA qLImLAnEJKm_, rAKEGm^nmEKILGH, sKE_ # @BI. >=, t. ?e.

veritable uniform. The appearance of courtly shāṭirs seems to partly in`auence the ʿayyār image in the later sixteenth century. On several miniatures in the Ḥamza-nāma (made for Akbar in the third quarter of the sixteenth century), ʿUmar ‘Umayya, the greatest ʿayyār of the age, is dressed in short pants and shirt, with an upper garment that is shorter in front than at the back, has bare legs and wears pointed shoes. He carries a dagger in his belt together with some other objects, including a rope with tas- sels, and holds a battle axe in one hand.N2 His head, however, is sometimes unobtrusively covered by a small turban of the same type as can be seen on other men. Less important ʿayyārs on the Ḥamza-nāma miniatures display fewer elements of the shāṭir’s typical garb.

T; Seyller, The Adventures of Hamza, for instance, 110–11, no. 31; 128–29, no. 38, and 134–35, no. 40. @ABC %DE FBGHBIJ %B %DE %KCLAKMJ =(?

An illustrated manuscript of another popular romance, Dāstān-i Qirān-i Ḥabashī, was made in Bukhara between 1660 and 1680.N6 In its miniatures, painted by artists belonging to the Janid court atelier, the black ʿayyār hero Qirān is usually depicted wearing a short upper garment, and below it another garment which is also shorter than those displayed by other people. When shown running or 45ghting, one can observe the combination of short pants, puttees and shoes.NN His turban is Indian in style and is decorated with a long feather. ʿAyyārs occasionally use a piece of cloth, sometimes seen slung around the shoulder, to serve as a bag and also to defend themselves with while they attack with a long dagger.NO As divergent as the appearances may be in detail, over the centuries the garb of ʿayyārs as well as shāṭirs is characterized by certain constants: a shorter garment which does not cover the legs from above or below the knee; shoes instead of boots; and a special hat, mostly tall and conical in shape. However, the important accessories—dagger, rope and sack—are less often depicted. This makes identi45cation of the less o[45cial 45gure of the ʿayyār di[45cult because elements of his dress are shared with those of other working people.NP Therefore, some album miniatures, where a similar 45gure may also be identi- 45ed as a servant, workman, or thief, are not taken into account here.NQ Nevertheless, the identi45cation of the precise literary context remains a challenge. First, there is a wealth of material to check.NR Second, all these prose texts (not only the romances) are unstable and were mainly transmitted orally.OS This is not to suggest that the miniatures discussed here once served to accompany oral performances, as was obviously the case with some large- format miniatures in the Topkapi albums.O0 Their size speaks against it. With

T< State Library Berlin, Ms. Or. Fol. 3180, see Stchoukine, Flemming, Luft and Sohrweide, Illuminierte islamische Handschriften, 179–85, no. 66. TT Fols. 17a, 51a, 96a, 115b, 125a. TU Fol. 51a, see Rührdanz, “Miniatures of the Bukharan Court,” 393, 45g. 7. TV For such a Jalayirid example, see Diez A Fol. 73, 39, see Ipşiroğlu, Saray-Alben, ddd/68 (p. 55). TW The decapitated 45gure on a lavishly decorated throne makes it likely that the 45gure in short garments `aeeing from the scene while still brandishing one of his two long dag- gers is an ʿayyār (Diez A Fol. 71, p. 5); see Ipşiroğlu, Saray-Alben, ddd/17 (p. 43); for a colour reproduction, see Gray, Arts of the Book, 102, pl. ffgddd. However, on his head sits a turban, not the characteristic conical cap. TY Not to speak of the possibility that the images could belong to hitherto unknown texts. UZ Hanaway, “Variety and Continuity,” 59. U7 Atasoy, “Illustrations Prepared for Display,” 262–72; while Atasoy’s interpretation of the function of these miniatures is very convincing, it is doubtful whether they all depict Shāhnāma subjects. =(\ ']DAM^G_ a width of less than 30 cm and an even smaller vertical extension, they seem to have belonged to large manuscripts in the tradition of the 45rst half of the fourteenth century. However, the transcription may each time have followed the oral presentation by a storyteller, and, thus, may have included variation in details of an episode or even whole stories not contained in published editions.O1 Stylistic similarities between the miniatures aside, since consistent details in its appearance allow the ʿayyār 45gure to be recognized in seven miniatures, it is reasonable to assume that they all belonged to the same manuscript, and, therefore, most likely to the same prose romance. Starting the search, it seemed appropriate to 45rst check the Samak-i ʿAyyār, because the extant Inju manuscript provided additional proof for the popularity of this text. The results, however, were not conclusive.O2 Of course, the Samak-i ʿAyyār contains, for instance, more than one episode in which a hero lifts his opponent from the saddle and throws him through the air, while the story allows explicitly or implicitly for Samak being present at this event.O6 Alas, such a subject is not su[45ciently speci45c. The more particular, even odd, subjects which would 45rmly connect a picture to a text passage remain unidenti45ed. Such, for instance, would be the miniature showing the ʿayyār walking behind a crooked man who is kept upright by a funny construction while meeting an arriving horseman and his entourage (45g. 8.4).ON Another would be the depiction of an attack against dark-skinned warriors aimed at freeing their prisoners (45g. 8.5). Here the ʿayyār, who seems to have ful45lled his part in preparations for the rescue action, is shown leaving the place.OO Individual zangīs (black servants or slaves) mostly feature as prison guards in Samak-i ʿAyyār. One among them turned into a warrior hero. However, their presence as a group in a fortress or castle rather asks for an episode happening in the zangīs’ own lands and points to other romances like the Fīrūzshāh-nāmaOP and the Ḥamza-nāma.OQ

U9 Such an explanation was once put forward by Simpson in respect to textual di[ferences between the small Shāhnāmas which were obviously made at the same time and place, see Simpson, The Illustration of an Epic, 116–19. U; The check was based upon the Russian translation of Faramarz ibn Khudādādh, Samak- ‘aiiar. Since this and other translations omit or summarize part of the romances, they are, naturally, of limited use as a text source. U< Diez A Fol. 72, S. 23 (Ipşiroğlu ddd/24); the best 45tting episode tells about Mardāndukht vanquishing Mihrak, see Faramarz ibn Khudādādh, Samak-‘aiiar, 2:293. UT Diez A Fol. 72, S. 20 (Ipşiroğlu ddd/10). UU Diez A Fol. 72, S. 22 (Ipşiroğlu ddd/16). UV B īghamī, Love and War. UW Seyller, The Adventures of Hamza. For the Iskandar-nāma, see Southgate, Iskandarnamah. A large part of this text is dedicated to Iskandar’s prolonged battle with the zangīs. However, ʿayyārs do not have a part in it. @ABC %DE FBGHBIJ %B %DE %KCLAKMJ =(e

@dkLAE (.\ A king on horseback is greeted by a group of people led by a man held up by a device installed between two buildings. lm^^mJnKnIKBmDEo $AELJJKJpDEA qLImLAnEJKm_, rAKEGm^nmEKILGH, sKE_ # @BI. >u, t. uv.

Mainly because of the nature of the romances and the seafaring tales, etc., as unstable and frequently orally transmitted texts, it may still take a long time, or may even prove impossible, to identify the precise subjects of many illustrations. I hope to have shown, however, that fourteenth-century Persian painting visualized literature of very di[ferent genres, from the heroic epic and the romance to the tales of merchants and sailors. It is time to take this arguably diverse kind of literature into consideration when evaluating the popularity of certain genres and texts during the fourteenth century. They constitute a third main group beside the illustrated Shāhnāma and Kalīla wa Dimna manuscripts. All three groups share a high entertainment value, which must =(h ']DAM^G_

@dkLAE (.e. Zangis in a castle attacked by warriors aiming to free captives. lm^^mJnKnIKBmDEo $AELJJKJpDEA qLImLAnEJKm_, rAKEGm^nmEKILGH, sKE_ # @BI. >u, t. uu.

have made them very popular at the Mongol courts. Without doubt, this quality would also have facilitated the spread of messages contained in the text and directed by the Iranian elite to the Mongol dynasties they were serving. These messages were neither narrow in focus nor uniform. It seems, however, that at least one other common denominator connects the popular romances with the Shāhnāma and the Kalīla wa Dimna: advising the Mongol kings and their men on wise rule, proper behaviour and navigation of life at court. The function of the animal fables (Kalīla wa Dimna) as a ‘mirror of princes’ is well known. They cautioned the ruler about the consequences of his decisions @ABC %DE FBGHBIJ %B %DE %KCLAKMJ =(> and taught everyday wisdom useful to the king as well as to less elevated persons.OR There was more about the pitfalls of a complex society than about positive role models; they were more about “survival rather than virtue.”PS The fables represent a didactic tradition originating from Indo-Iranian sources, and were widely adopted as a warning against the perils of life and a guide to wise decisions and good governance.P0 The Shāhnāma—apart from the dynastic-political aims frequently connected to representative courtly copiesP1—propagated a more speci45c Iranian ideal of kingship and of every- body’s place in society. It stressed dependence of human wellbeing upon legitimate kingship, indisputable obedience to the king by the paladins, and justice as the foundation of rule.P2 With reservations, this pertains to the post- Shāhnāma epics too, particularly to the oldest. The popular romances added to the model of conduct the ideal of javānmardī.P6 “It [. . .] forms the most perva- sive theme in the romances, appearing in the general form of the ideas which govern the actions of the Iranian heroes, and in the speci45c form of ʿayyārī and the adventures of the ʿayyārs.”PN Although religion recedes into the background in the prose romances, we cannot exclude the fact that the connection of javānmardī to mystical ideas that was well established in the fourteenth century helped to have this kind of literature accepted into the group of illustrated

UY Fouchécour, Moralia, 414–20. VZ O’Kane, Early Persian Painting, 23. V7 Because of the popularity of this kind of literature, two texts of a similar kind, the Sindbād- nāma and the Marzbān-nāma, were checked, but could not be connected with miniatures in the Diez albums. For an interesting manuscript containing all three texts and illustrated in the Shiraz style of the second quarter of the 45fteenth century, see Grube, “Ibrahim-Sultan’s ‘Anthology,’ ” 101–17. V9 Rührdanz, “Zu Rolle und Funktion illustrierter Shāhnāmeh,” G-6: 75–84. V; The didactic function of the Shāhnāma has only recently caught the attention of scholars, see Fouchécour, Moralia, 53, and passim. Fouchécour focuses on explicit teaching in the historical part of the Shāhnāma. It seems, however, that the action-45lled earlier parts should be given the same credit for teaching by story rather than discourse. For re`aection on two di[ferent sets of morals in the Shāhnāma, see idem, “Akhlāq-i pahlavānī,” 10–16. See also Meisami, “The Šāh-Nāme as Mirror for Princes,” 265–73. My perception of this function of the Shāhnāma has strongly pro45ted from discussions with Nasrin Askari on her research for her PhD thesis, and I am thankful for her allowing me to read a draft of the 45rst chapter. For the recently published thesis, see Askari, The Medieval Reception of the Shāhnāma as a Mirror for Princes. V< Zakeri, “Javānmardi,” 594–601; Hanaway, Persian Popular Romances, 129–78; Gaillard, Le livre de Samak, 161–68. VT Hanaway, Persian Popular Romances, 129. =(( ']DAM^G_

texts at the court ateliers.PO The disappearance of the prose romances and other popular literature from the range of texts deemed worthy of illustration seems equally connected to the increasing in`auence of Su45sm on elite culture, as it was connected to a growing “taste for the intricate.”PP The mystical re- shaping of the ideal human being, and the ideal ruler in particular, asked for a di[ferent kind of educational impact. Consequently, texts which focused on poetic expression of mystical ideas and which were, therefore, ideally suited to inspire their visualization, gained preference. While the Shāhnāma and Kalīla wa Dimna stayed on the agenda of court ateliers into the second quarter of the 45fteenth century, the popular literature was the 45rst to be eliminated from it. Thus, classical Persian painting of the 45fteenth century emerged from its Ilkhanid and Jalayirid foundations only to consequently develop just one branch of the tree to perfection.

Bibliography

Abdullaeva, Firuza, and Charles Melville. The Persian Book of Kings: Ibrahim Sultan’s Shahnama. Oxford: Bodleian Library, 2008. Appel, Klaus, and Dieter George. “Die Saray-Alben der Staatsbibliothek und ihre Restaurierung.” Jahrbuch Preussischer Kulturbesitz df (1971): 227–32. Asadī Ṭūsī. Le livre de Gerchāsp, poème d’Asadi (Junior) de Toûs. Translated by Clement Huart (vol. 1) and Henri Massé (vol. 2). Paris: Librairie orientaliste, 1926–1951. Askari, Nasrin. The Medieval Reception of the Shāhnāma as a Mirror for Princes. Leiden: Brill 2016 (forthcoming). Atasoy, Nurhan. “Four Istanbul Albums and Some Fragments from Fourteenth-Century Shah-Namehs.” Ars Orientalis gddd (1970): 19–49. ———. “Illustrations Prepared for Display during Shahname Recitations.” In Memorial Volume of the Fifth International Congress of Iranian Art and Archaeology 2, edited by A. Tajvidi and Muhammad Y. Kiyani, 262–72. Tehran: Ministry of Culture and Arts, 1972. Barry, Michael. Figurative Art in Medieval Islam and the Riddle of Bihzād of Herāt (1465– 1535). Paris: Flammarion, 2004. Bīghamī, Muḥammad ibn Aḥmad. Love and War: Adventures from the Firuz Shāh Nāma of Sheikh Bighami. Translated by William L. Hanaway, Jr. Delmar, y.z.: Scholars’ Facsimiles and Reprints, 1974.

VU Gaillard, Le livre de Samak-e ʿAyyār, 168. VV Subtelny, “A Taste for the Intricate,” 56–79. @ABC %DE FBGHBIJ %B %DE %KCLAKMJ =(j

Blair, Sheila S. “On the Track of the ‘Demotte’ Shāhnāma Manuscript.” In Les manuscrits du Moyen-Orient, edited by François Déroche, 125–31. Istanbul: Institut Français d’Études Anatoliennes and Paris: Bibliothèque Nationale, 1989. ———. “A Mongol Envoy.” In The Iconography of Islamic Art: Studies in Honour of Robert Hillenbrand, edited by Bernard O’Kane, 45–60. Edinburgh: Edinburgh University Press, 2005. Browne, Edward G. A History of Persian Literature under Tartar Dominion ()* 1265– 1502). Cambridge: Cambridge University Press, 1920. Cowen, Jill S. Kalila wa Dimna: An Animal Allegory of the Mongol Court. New York, Oxford: Oxford University Press, 1989. Çağman, Filiz. “On the Contents of Four Istanbul Albums H. 2152, 2153, 2154, and 2160.” Colloquies on Art and Archaeology in Asia 10 (1985), 31–36. Çağman, Filiz and Zeren Tanındı. The Topkapı Saray Museum: The Albums and Illustrated Manuscripts. Edited by J. M. Rogers. Boston: Little, Brown, 1986. Duda, Dorothea. “Die Buchmalerei der Ğala’iriden.” Der Islam 48 (1972): 28–76; 49 (1973): 153–220. Ettinghausen, Richard. “On Some Mongol Miniatures.” Kunst des Orients 3 (1959): 59–65. Faramarz ibn Khudādādh, Samak-‘aiiar, Nataliia Kondyreva (vol. 1) and A. M. Mikhalev (vol. 2). Moskow: Nauka, 1984. Fitzherbert, Teresa. “Khwājū Kirmānī (689–753/1290–1352): An Éminence Grise of Fourteenth-Century Persian Painting.” Iran 29 (1991): 137–51. de Fouchécour, Charles-Henri. Moralia: Les notions morales dans la littérature persane du 3+/9+ au 7+/13+ siècle. Paris: Éditions Recherche sur les Civilisations, 1986. ———. “Akhlāq-i pahlavānī va akhlāq-i rasmī dar Shāhnāma-i Firdawsī.” In Tan-i pahlavān va ravān-i khiradmand: Pazhūhishhā-yi tāza dar Shāh-Nāma, edited by Shāhrukh Maskūb, 10–16. Tehran: Ṭarḥ-i nau, 1381/2003. Gaillard, Marina. Le livre de Samak-e ʿAyyār. Structure et idéologie du roman persan médiéval. Paris: !y'l, 1987. ———. “Du ʿayyār au šāter: L’évolution du personnage à travers le roman persan en prose.” In Liber amicorum: Ètudes sur l’Iran médiéval et modern o,fertes à Jean Calmard, edited by Michele Bernadini, Masashi Haneda, and Maria Szuppe, 143–73. Cambridge: Skilliter Centre for Ottoman Studies; Rome: Istituto per l’Oriente C. A. Nallino, 2006 (=Eurasian Studies g/1–2). ———. “Samak-e ʿAyyār.” In Encyclopaedia Iranica; available online at http:// www .iranica.com./articles/samak-eayyar. Gonnella, Julia, and Christoph Rauch, eds. Heroische Zeiten: Tausend Jahre persisches Buch der Könige. Berlin: Museum für Islamische Kunst, 2010. Gray, Basil, ed. The Arts of the Book in Central Asia, 14th–16th centuries. Paris, London: {y&l!r, 1979. =jv ']DAM^G_

Grube, Ernst J. “The Istanbul University Library Kalilah wa Dimnah and other 14th- Century Persian Manuscripts.” In A Mirror for Princes from India, edited by Ernst J. Grube, 52–77. Bombay: Marg, 1991. ———. “Ibrahim-Sultan’s ‘Anthology of Prose Texts.’ ” In Persian Painting from the Mongols to the Qajars: Studies in Honour of Basil W. Robinson, edited by Robert Hillenbrand, 101–17. London and New York: I. B. Tauris, 2000. Hanaway, William L. “Variety and Continuity in Popular Literature in Iran.” In Iran: Continuity and Variety, edited by Peter Chelkowski, 59–75. New York: Center for Near Eastern Studies, 1971. ———. “Persian Popular Romances before the Safavid Period.” PhD diss., Columbia University, New York, 1970. ———. “ʿAyyār dd. ʿAyyār in Persian sources.” In Encyclopaedia Iranica ddd: 161–63. London and New York: Routledge & Kegan Paul, 1988. Ibn Abī ʾl-Khayr, Irānshāh. Bahman-Nāma az Irānshāh b. Abī ʾl-Khayr. Edited by Raḥīm ‘Afīfī. Tehran: Shirkat-i Intishārāt-i ʿilmī va farhangī, 1370/1991. Ibn Shahriyār, Buzurg. The Book of the Marvels of India. Translated by L. Marcel Devic. London: Routledge, 1928. Ipsiroglu, Mazhar S. Painting and Culture of the Mongols. New York: Harry Abrams, 1967. ——— (Ipşiroğlu). Saray-Alben. Diez’sche Klebebände aus den Berliner Sammlungen (=Verzeichnis der orientalischen Handschriften in Deutschland gddd). Wiesbaden: Franz Steiner, 1964. Komaro[f, Linda, and Stefano Carboni. The Legacy of Genghis Khan. New York: Metropolitan Museum of Art, New Haven and London: Yale University Press, 2002. Lentz, Thomas W., and Glenn D. Lowry. Timur and the Princely Vision: Persian Art and Culture in the Fifteenth Century. Los Angeles: Los Angeles County Museum, Washington: Arthur M. Sackler Gallery, 1989. Meisami, Julie. “The Šāh-Nāme as Mirror for Princes.” In Pand-o Sokhan: Mélanges o,ferts à Charles-Henri Fouchécour, edited by Christophe Balaÿ, Claire Kappler, and Živa Vesel, 265–73. Tehran: Institut français de recherche en Iran, 1995. O’Kane, Bernard. Early Persian Painting: “Kalila and Dimna” Manuscripts from the Late Fourteenth Century. London: I. B. Tauris, 2003. ———. “Siyah Qalam: The Jalayirid Connection.” Oriental Art 49/2 (2003): 2–18. Porter, Yves. “The Illustrations of the Three Poems of Khwājū Kirmānī: A Turning Point in the Composition of Persian Painting.” In Écrit et culture en Asie centrale et dans le monde turco-iranien, -+–-.-+ siècles/Writing and Culture in Central Asia and the Turko-Iranian World, 10th–19th Centuries, edited by Francis Richard et Maria Szuppe, 359–74. Paris: Association pour l’avancement des études iraniennes, 2009. Prentice, Verna. “A Detached Miniature from the Masnavis of Khwaju Kermani.” Oriental Art 27/1 (1981): 60–66. @ABC %DE FBGHBIJ %B %DE %KCLAKMJ =j=

Qubādī, Shigufa (Ghobadi, Chokofeh). Mathnawī-yi Jamāl-u Jalāl tasnīf-i Muḥammad Nazl-Ābād dar sāl 808q. Tehran, 1382/2004. Rieu, Charles. Supplement to the Catalogue of the Persian Manuscripts in the British Museum. London: British Museum, 1895. Robinson, Basil W. Persian Miniatures. Oxford: Cassier, 1957. ———. A Descriptive Catalogue of the Persian Paintings in the Bodleian Library. Oxford: Oxford University Press, 1958. Roxburgh, David J. “Heinrich Friedrich von Diez and His Eponymous Albums Mss. Diez A Fols. 70–74.” Muqarnas 12 (1995): 112–36. ———. The Persian Album 1400–1600: From Dispersal to Collection. New Haven and London: Yale University Press, 2005. ———, ed. Turks: A Journey of a Thousand Years, 600–1600. London: Royal Academy of Arts, 2005. Rührdanz, Karin. “Zu Rolle und Funktion illustrierter Shāhnāmeh-Handschriften im 14. und 15. Jahrhundert.” Wissenschaftliche Zeitschrift der Martin-Luther-Universität Halle-Wittenberg fffg (1986), G-6: 75–84. ———. “About a Group of Truncated Shahnamas: A Case Study in the Commercial Production of Illustrated Manuscripts in the Second Part of the Sixteenth Century.” Muqarnas 14 (1997): 118–34. ———. “Miniatures of the Bukharan Court Atelier in a copy of Khwājū Kirmānī’s Khamsa dated 1078/1667–68.” In Écrit et culture en Asie centrale et dans le monde turco-iranien, -+--.-+ siècles/Writing and Culture in Central Asia and the Turko- Iranian World, 10th–19th Centuries, edited by Francis Richard and Maria Szuppe, 375–402. Paris: Association pour l’avancement des études iraniennes, 2009. Seyller, John. The Adventures of Hamza: Painting and storytelling in Mughal India. Washington: Smithsonian Institution, London: Azimuth, 2002. Sims, Eleanor. “Ibrāhīm-Sultān’s Illustrated Ẓafar-nāmeh of 898/1436.” Islamic Art 4 (1991): 175–217. ———. Peerless Images: Persian Painting and its Sources. New Haven and London: Yale University Press, 2002. ———. “Bilder zu Ferdausis Schahname in den Diez-Alben der Staatsbibliothek zu Berlin.” In Heroische Zeiten: Tausend Jahre persisches Buch der Könige, edited by Julia Gonnella and Christoph Rauch, 28–35. Berlin: Museum für Islamische Kunst, 2010. Simpson, Marianna S. The Illustration of an Epic: The Earliest Shahnama Manuscripts. New York, London: Garland, 1979. Southgate, Minoo S. trans. Iskandarnamah: A Persian Medieval Alexander Romance. New York: Columbia University Press, 1978. Stchoukine, Ivan. Les peintures des manuscrits timurides. Paris: Geuthner, 1954. ———. Flemming, Barbara, Paul Luft, and Hanna Sohrweide. Illuminierte islamische Handschriften (=Verzeichnis der Orientalischen Handschriften in Deutschland fgd). Wiesbaden: Franz Steiner, 1971. =ju ']DAM^G_

Stockland, William. “The Kitab-i Samak ‘Ayyar.” Persica 15 (1993–1995): 143–82. Subtelny, Maria. “A Taste for the Intricate: The Persian Poetry of the Late Timurid Period.” Zeitschrift der Deutschen Morgenländischen Gesellschaft 136 (1986): 56–79. Thackston, Wheeler M. A Century of Princes: Sources on Timurid History and Art. Cambridge, Mass.: Aga Khan Program for Islamic Architecture, 1989. Titley, Norah M. Miniatures from Persian Manuscripts: A Catalogue and Subject Index of Paintings from Persia, India and Turkey in the British Library and the British Museum. London: British Library, 1977. ———. Persian Miniature Painting and its In1luence on the Art of Turkey and India. London: British Library, 1983. Welch, Stuart C. Persian Painting: Five Royal Safavid Manuscripts of the Sixteenth Century. New York: Braziller, 1976. Wright, Elaine. “Firdausi and More: A Timurid Anthology of Epic Tales.” In Shahnama: The Visual Language of the Persian Book of Kings, edited by Robert Hillenbrand, 65–84. Aldershot: Ashgate, 2004. Zakeri, Mohsen. “Javānmardi.” In Encyclopaedia Iranica fdg: 594–601. !"#$ % Relationships with Neighbouring Actors

∵

"#$%&'( ) Champions of the Persian Language: The Mongols or the Turks?

Aptin Khanbaghi

The fact that Persian continued (and indeed continues) to be a major language in Western Asia after the Arab conquest in the seventh century is generally deemed to be a fact requiring no explanation. Persian has certainly had a long and glorious history, and it was once the language of rulers whose empire stretched across all of Western Asia, brie*+y even vaunting provinces in other continents. But past glory did not safeguard the languages of the Pharaohs, the Incas, or the Lydian Empire, and per se there is no reason why it should have done so for Persian. Indeed, since the Arab conquest in the seventh century brought with it the extinction of several major languages, the di,ferent fate of Persian, which went on to *+ourish in medieval times, stands out as an oddity requiring scrutiny and, indeed, explanation. This article will study the role played by the Mongols in the changing fates of the Persian language.- To do so, we will have to look further a./eld than just Iran: in particular we will focus on Anatolia. Moreover, our examination of the status of Persian in the Turkic and Mongol periods will begin by taking a step back, to the decades following the advent of Islam. For it is in the twists and turns of this period that important seeds were sown, resulting eventually in the resurgence of the Persian language and the birth of classics such as the Shāhnāma.

Before the Mongols

Iran under the Muslim Arab and Iranian Dynasties (c. 632–999 !") The Arab rulers of Iran in the eight and ninth centuries fostered the di,fusion of Arabic. Not only was support o,fered to those who wrote in it ex novo:

0 Existing surveys include: Morgan, Medieval Persia, 2–3; Elgood, A Medical History of Persia, 233–2333; Browne, A Literary History of Persia, 355.

© !"#$#!%$&!' ()$%% #*, %'$+'#, ,-./ | +"$ .-.../1/23425561.63,/_5.. ;)< 45678695: the literary and scienti./c treasures of Ancient Iran were also translated into Arabic, so that knowledge of their original language grew less necessary.= In the new political and cultural order, Persian was losing its purpose. So much so, that the language was believed to be doomed even by the Persian- speaking elite of the time. An example is Balādhūr@̄, who reports that o,./cial records, previously kept in Persian, were now written in Arabic:B his narration suggests that the Arab rulers were hoping to stamp Persian out.C As it turned out, however, the Arab period was not the end-point of Persian, only a chapter in its eventful history. For it was to resurface a couple of centuries later, in a new script. And in fact, the Arabs had never succeeded in extinguishing Persian completely: it continued to be used in the Caspian provinces, even for o,./cial records.D In part, of course, Persian owed its survival during the Arab period to the Iranians’ volition to preserve their language. This aspect is stressed by Shahrokh Meskoob,E but it is not the whole story—there were also broader historical factors. One of these is that Persian continued to be employed in Central Asia as a lingua franca between the various ethnically Iranian groups for whom Persian was not a ./rst language and the Arabs.F Another is that the Zoroastrian communities, although their population was declining rapidly, were still important

G Danner, “Arabic Literature in Iran,” 570–71, 582: The most famous men of culture traced down at this point are of Iranian descent, but writing in Arabic: e.g. Ibn al-Muqa,faʿ, Ibn Khurdādhbih, D@̄nawar@̄, et al.; Gabrieli, “Ibn al-Muḳa,fa‘,” 883; Ibn al-Muqa,faʿ attached himself to al-Manṣūr’s uncle. K Balādhūr@̄, Futūḥ al-buldān, 466. L Some scholars see evidence of a di,ferent attitude in the fusion of Persian and Arabic which led to the formation of Classical Persian. See for example Frye, The Golden Age of Persia, 172–74 (also pointing to the opportunities which the Arab conquest provided for the spread of the Persian language in the East) and Spuler, “The Evolution of Persian Historiography,” 127–28, 132. Such stances are not wholly persuasive, however. The Arabs in the eight and ninth centuries were hardly campaigning for the preservation of the Persian language. M Rypka, History of Iranian Literature, 66: Rypka refers to the bilingual scripts on the towers of Lajim and Resget and also Mil-i Radkan, which are in Arabic and Pahlavi. They date back to the eleventh century; Juwaynī/Qazwīnī, 1:73: Up to the thirteenth century, some of the rulers of the Caspian regions were reputed to belong to ancient Iranian aristocracy. They had particular interest in the Persian culture. Mustawf@̄, Tār$̄kh-i guz$̄da, Muqadamma, 413: Mustawf@̄ mentions the Ziyarids in the late tenth century. O Meskoob, Iranian Nationality and the Persian Language, 29–31, 35–37, 42; Mahmoodi- Bakhtiari, “Planning the Persian Language in the Samanid Period,” 260. Meskoob links Iranian nationality and the Persian language, and observes the Iranians’ resistance to Arabicization as an important factor in the survival of Persian. P Meisami, Persian Historiography, 17: this is regarding the usage of Persian in Central Asia. Q56RS:T7U TV W5X SXYU:67 Z679[69X ;)\ during this period, and their priests continued to use Persian, albeit in the old Pahlavi script. A third, and no less important factor, was the emergence of the Iranian Sa,farid and Samanid dynasties in the ninth century. The Sa,farid and Samanid rulers were kind to Persian. Since they had embraced Islam, and did not intend to challenge the spiritual rule of the Caliphate, an atmosphere was born in which a renaissance of Persian could happen without it having connotations of political resistance or insurrection. Historically, Persian resistance against the Arabs was expressed through religious ideology and failed in the eight and ninth centuries.] The Persian language survived precisely because it was not used as a political tool, but mainly as a cultural one. E,fectively, Arabic was maintained for most o,./cial purposes in the ninth century, and remained highly regarded as the language of science. Ninth- and tenth-century Persian scholars continued to write in Arabic. This period saw prominent ./gures from the Persianate world like Muḥammad al-Bukhārī (d. 870), Abū Ḥanīfa D@̄nawar@̄ (d. 895), Aḥmad Balādhūr@̄ (d. 892), Muḥammad b. Jar@̄r al-Ṭabar@̄ (d. 923), and Abū Bakr Muḥammad ibn Zakariyyā al-Rāz@̄ (d. 925) contribute to the ./elds of science and the humanities in Arabic. On the other hand, Persian speaking poets were henceforth supported at court. As a result, this period fostered the rise of classic Persian literature, primarily in the shape of poetry.a An important innovation of the Sa,farid-Samanid period is that Persian was now written in the Arabic script. This made the language much more learner-friendly, since the old Pahlavi script had been rendered unnecessarily complicated, perhaps in a conscious drive to maintain literacy as the preserve of a learned few.-b Since increasing numbers of people of Iranian stock were studying Arabic, and since Persian was a vernacular language, the expression of Persian in the Arabic script made the language accessible to the masses in

c Khanbaghi, “De-Zoroastrianization and Islamization,” 211; D. Sajana, ed., Dinkard, 62–63. d The earliest documents available in Classical Persian are in the Hebrew script; however, it can be surmised it was only used by the Jewry and therefore played a marginal role in the resurgence of the Persian language. See Fischel, “Israel in Iran (A Survey of Judaeo- Persian Literature),” 823–24. 0e To summarize, in the Pahlavi version of the Aramaic scripts not only vowels were not represented like in most Semitic scripts but the number of letters representing consonants were reduced as well, making the identi./cation of the sounds of the letters unclear. Many Aramaic words were used as logograms to represent Pahlavi words, making the orthogra- phy ambiguous, for example the Pahlavi word for ‘mother’ was written om as in Aramaic but read as mādar. ;)f 45678695: written form. The usage of the Arabic script preserved the Persian language and served the purpose of literati and their patrons. The Sa,farid and Samanid dynasties, which maintained Arabic as one of their o,./cial languages while promoting Persian as the language of literature and scholarship, built a multicultural environment to which the Turkic groups could develop a sense of belonging. This was to prove important in the history of Persian when Turkic dynasties assumed control of Iran.

Iran under the Early Turkic Dynasties (c. 963–1256) From the late tenth to the mid thirteenth centuries, until the conquest by the Mongols, Iran was ruled by dynasties of Turkic origin:-- they had taken the reins of power from Iranian rulers and inherited territories that were culturally Iranian. The changing hands of power occurred at a time when Persian was *+ourishing among the nobility and people of culture. The various Shāhnāmas and other Iranian epics were all produced in greater Khurasan,-= which was the cradle of the Irano-Turkic Muslim dynasties. Persian was therefore not associated solely with Iranians but regarded as the language of art and much used at local courts. Indeed, the Turkic rulers looked very positively on Persian language and culture.-B Like members of the Iranian dynasties which in the Islamic period traced their lineage back to Sassanian kings, Turkic rulers such as the Ghaznavids (998–1040) and the Seljuqs (1055–1157) invented Iranian ancestors for themselves.-C

00 Turkic groups had been settling among Iranians since Antiquity and were not considered completely foreign to the Iranian world. Ibn al-Nad@̄m, Kitāb al-Fihrist, 823–24; Grignaschi, “Quelques spécimens de la littérature sassanide (Karnamag),” 19–20: Khusraw Anūshīrvān encouraged the conversion of Turks to Zoroastrianism in Iran; Hwui Li, The Life of Hiuen-Tsiang, 45: “The Turks worship ./re.” In another Central Asian principality “The king and people do not believe in the law of Buddha, but their religion consists in sacri./cing to ./re,” see Khanbaghi, “De-Zoroastrianization and Islamization,” 203. 0G Meskoob, Iranian Nationality and the Persian Language, 35, cites the names of the poets stemming from Khurasan; Rypka, History of Iranian Literature, 113, notes that until the reign of the Seljuqs, the most important poets of the Persianate world stem from Transoxania. 0K Rypka, History of Iranian Literature, 115: Rypka agrees that the Turkic rulers in Medieval Central Asia regarded Persian as their own heritage; Gardet, “Religion and Culture,” 601, refers to the acceptance of Persian in the early Turkic empires. 0L Meskoob, Iranian Nationality and the Persian Language, 36–37. Q56RS:T7U TV W5X SXYU:67 Z679[69X ;))

In this favourable setting, Persian was able to *+ourish under the Turkic rulers of Western and Central Asia, even before the arrival of the Mongols. This *+ourishing went beyond poetry and literary art to embrace other highly varied genres. In part this was probably due to the adoption of the Arabic script for writing Persian: any literate Persian who knew Arabic could now venture to use the Arabic script to write in their own mother tongue.-D It seems that from the early eleventh century onwards, the pressure to write in the Arabic language, even for scienti./c and scholarly domains, had considerably lessened. For example, among the historiographers, Gardīzī (d. 1061) wrote his Zayn al-akhbār in Persian for the Ghaznavid Sultan Zayn al-Milla ʿAbd al-Rashīd (d. 1053); Abū al-Faḍl Bayhaqī (d. 1077) wrote the Tār$̄kh-i Masʿūdī in Persian and dedicated it to Sultan Masʿūd (d. 1041);-E and the major part of the Tār$̄kh-i S$̄stān (author unknown) was written in Persian around the same period. Not long afterwards, Kaykavus b. Iskandar (d. 1098) wrote his Qābūsnāma (also called Andarznāma), one of the earliest specimens of andarz (collections of advice to posterity), in Persian.-F We also have two important works by Nāṣir Khusraw: in his Safarnāma (after 1052), a pioneering travel journal in Persian, he recorded his travels across the Islamic world; while his Gushāyish va Rahāyish is one of the earliest philosophical treatises in Persian.-] During this period, the ./rst Persian manual of rhetoric was written by Muḥammad ibn ʻUmar al-Rādūyān@̄.-a Bahrāmshāh (d. 1152), who was one of the last Ghaznavid sultans, commissioned the Arabic-Persian translation of the collection of fables entitled Kal$̄la wa Dimna, which three centuries earlier had been translated by Ibn al-Muqa,faʿ from Middle Persian to Arabic.=b One of the most important pieces on governance ever written in the medieval world was produced, in Persian, by Niẓām al-Mulk at the behest of the Seljuq ruler Malik Shāh (d. 1092).=- Before attaching himself to the Seljuq rulers, Niẓām al-Mulk had begun his political career in the service of the Ghaznavids

0M Ibid. 0O Piemontese, Storia della Letteratura Persiana, 3, 50, 52–53. 0P Kay Kā’ūs b. Iskandar, Kitāb-i Naṣ$̄ḥatnāma: Qābūsnāma, Introduction; Bosworth, “Kay Ka’us b. Iskandar” In Encyclopaedia of Islam, Second Edition 32:815; Kay Kā’ūs b. Iskandar, Qābūsnāma, 50. 0c Piemontese, Storia della Letteratura Persiana, 3, 66. 0d Pourjavady, “Literary debates in the Sa./na-yi Tabriz,” 133. Ge Rypka, History of Iranian Literature, 223. G0 Lambton, “The Dilemma of Government in Islamic Persia,” 55. jkk 45678695: and was therefore an experienced administrator whose Sıȳ āsatnāma became a valuable source of reference for the rulers who followed. The most important medical treatise known from this period was written in Persian by Shaykh Kamāl al-D@̄n Bad@̄‘ al-Zamān Abū Faḍl Hubaysh al-Ghaznaw@̄ (d. 1183), titled the Kitāb al-Kifāyah f$̄ al-ṭibb.== In his Arabic works he is also known as al-Ti*+@̄s@̄ (from Tbilisi, in Georgia), and it seems that he composed his Persian sections in Ghazna,=B but his work is dedicated to the Anatolian Seljuq Sultan ʿIzz al-Dīn Qilij Arslān b. Masʿūd (d. 1192).=C It is worth noting, however, that the author of these volumes was active at the peripheries of the Persianate world where Persian was not necessarily the local language. The status of Persian was thus being consolidated throughout the reign of the Turkic rulers, although important treatises continued to be written in Arabic. The most celebrated scholars of this period who stemmed from the Iranian world, such as the philosopher Miskawayh (d. 1030), the polymath Ibn Sīnā (d. 1037), the scientist Bīrūnī (d. 1048) and the theologian Ghazālī (d. 1111), couched their masterpieces in Arabic. The mystic Suhrawardī (d. 1191), who appeared on the scene slightly later, wrote in both Arabic and Persian, a clear sign that both languages were in use among the elite. The coexistence and parallel use of Arabic and Persian at this time re*+ects the receptive attitude of the Turkic rulers towards both. The administration of the Ghaznavids vacillated between Arabic and Persian at the discretion of the viziers, demonstrating that these two languages had become an inherent part of the political and cultural system of Iran and Central Asia.=D Persian literati *+ocked to the areas of Seljuq rule from Eastern Iran, especially Khurasan. As mentioned by Julie Meisami and Claude Cahen, with reference to the Seljuq vizier Anūshīrvān ibn Khālid’s (d. 1159) work Nafthat al-maṣdūr=E and the Tār$̄kh al-wuzarā by an uncertain author,=F this can be well exempli./ed by considering the movements of Persian historians. Another good example of

GG Izgi,l “Hubeyş et-Ti*+isi,” 268–70; Muḥammad Amīn Riyāḥī, Nufūz̲-i zabān va adabiyāt-i Fārs$̄ dar qalamruv-i ‘Usmānī, 39. GK Storey, Persian Literature, 2: part 2, 213–14. GL Storey, Persian Literature, 3: part 1, 87. GM ‘Utb @̄, al-Tār$̄kh al-Yam$̄n$̄, 2: 170–71; Meisami, Persian Historiography, 51. GO Storey, Persian Literature, 1: 1089: the full title of the book cited above is Nafthat al-maṣdūr fī ṣudūr zamān al-futūr wa-futūr zamān al-ṣudūr. See also, “Anōsharwān b. Khālid” in First Encyclopaedia of Islam 3: 357: Anūshīrvān ibn Khālid’s memoirs were translated into Arabic by ‘Imād al-Dīn al-Iṣfahānī, but the original Persian was lost. GP Meisami, Persian Historiography, 144; Cahen, “Historiography of the Seljoukide Period,” 60–61. Q56RS:T7U TV W5X SXYU:67 Z679[69X jk; a historian migrating from East to West is Ẓāhir al-Dīn Nīshapūrī (d. 1187), who moved from Khurasan to Hamadan to serve the Seljuqs. An e,fect of these movements and expansions was that, under the Turkic rulers, literary and scholarly activities declined in Khurasan while they thrived in other areas—not only Fars and northwest Iran, but even Azerbaijan and India.=] For, as the Turkic rulers’ sphere of in*+uence expanded beyond Iran, so they enlarged the reach of Persian language and culture. The Seljuqs conquered formerly Byzantine territories in Anatolia, and in this connection we shall have more to say about them below. The Ghaznavids and Ghurids extended their rule as far as the Indian subcontinent, taking both Persian and Arabic with them.=a The Ghaznavids made Lahore a centre for Persian scholars:Bb Abū al-Ḥasan Hujwirī al-Ghaznawī produced the Kashf al-maḥjūb (ca. 1073), considered one of the most important early treatises on Su./sm;B- and Minhāj Sirāj Jūzjānī (thirteenth century) wrote the Ṭabaqāt-i Nāṣirī in Lahore for the Mamluk Sultan of Delhi, Nāṣir al-Dīn Fīrūz Shāh (d. 1266). While Persian was eventually embraced by Indians of di,ferent faiths, Arabic remained in use only among Muslims.B= Arabic was identi./ed with a faith, while Persian was associated with the rulers and their traditions, regardless of their religious a,./liation. The shift in cultural activity from East to West which occurred under the Seljuqs presaged the expansion of Persian high culture into Anatolia, which was to occur when the Seljuqs conquered this area.

Anatolia before the Mongols With the exception of scholars in present-day Turkey, the dissemination of Persian in Anatolia has not been as widely studied as the spread of the Persian language in the Subcontinent. Nonetheless, it is an important strand in the complex tapestry which is the history of the Persian language. We will focus on the Seljuq period, which began in 1071, when Alp Arslan defeated the Byzantine Emperor Romanus Diogenes at the battle of Manzikert. The Seljuqs were an Iranian-based dynasty of Turkic origin. Anatolia as they found it was a land of many languages. According to the Safarnāma of Nāṣir-i Khusraw (d. 1088), the town of Akhlat in Eastern Anatolia was trilingual, with

Gc Bah ār, Sabk-shinās$̄, ya Tar$̄kh-i taṭavvur-i nasr-i Fārs$̄, 2:359. Gd A dynasty with a Turko-Persian heritage, which rose to power in the eleventh century in Central Asia. Ke Sh ūshtar@̄, Tār$̄kh-i zabān va adab$̄yāt-i Ir, ān dar khārij az Ir, ān, 230. K0 Ibid., 69. KG Qutbuddin, “Arabic in India,” 316. jkj 45678695:

Arabic, Persian and Armenian.BB Thus, right from the start, the role of Persian in the territories ruled by the Anatolian Seljuqs was not con./ned to culture and administration: it was also in use as a vernacular tongue, albeit on a much smaller scale, however. Paradoxically, the Anatolian Seljuqs seem to have been even more heavily Persianized than their counterparts on the Iranian plateau. Sixteen of the Anatolian Seljuq kings bore old Iranian names such as Kaykhusraw, Kayqubād and Kaykāvūs.BC While some scholars maintain that, initially, the Anatolian Seljuqs took over the existing Byzantine administration,BD others hold they that brought their own Persian system from the East.BE It is di,./cult to assess the merits of the two positions, since from the reign of the early Seljuq rulers of Anatolia the primary sources do not allude to the Seljuq administration. Hence, it is di,./cult to detect the origins of the bureaucrats; whereas from the eve of the Mongol conquests and thereafter there are many references to o,./cials who ran the kingdom, many of whom came from the Iranian plateau. Despite the fact that the Seljuqs were Turks, they hardly promoted writing in Turkish.BF Indeed, even before the arrival of the Mongols, Persian was the language they used for o,./cial matters. They further supported learned men well versed in Persian, and Charles Melville asserts that, under their rule, historical writing in Persian was already displacing Arabic.B] The Seljuqs of Anatolia took a personal interest in Persian language, culture and literature. Ibn B@̄b@̄ relates many instances where they themselves used Persian. For example, Sultan ʿIzz al-Dīn Qilij Arslān (d. 1192) addressed his courtiers in Persian (bi-zabān-i pārsī) while still a child, which suggests that the Sultans were bilingual; perhaps also, that they were taught to use Persian in a formal setting.Ba Ibn B@̄b@̄ also reproduces poems composed by Seljuq princes, such as that which Quṭb al-Dīn Malik Shāh of Sivas recited for his brother, the Sultan Rukn al-Dīn Sulaymān Shāh 33 (d. 1204), and a stanza composed by Sultan ʿIzz al-Dīn Kaykāvūs (d. 1220) which he sent to his Parvāna (court

KK Khusraw, Safarnāma, 7; Riyāḥ@̄, Nufūz-i zabān va adabiyāt-i Fārs$̄, 7. KL Riy āḥ@̄, Zabān va adab-i Fārs$̄, 5. KM Peacock, “Saljuqs—Saljuqs of Rum” asserts that the Seljuqs maintained the Greek administration when they arrived in Anatolia. KO Riy āḥ@̄, Zabān va adab-i Fārs$̄, 10, 23. Cf Ibn B@̄b@̄, Mukhtaṣar-i Saljūqnāma; and Āqsarā’ī, Musāmarat al-akhbār wa musāyirat al-akhyār. KP Gandjeï, “Turkish in Pre-Mongol Persian Poetry,” 70–71. Kc Melville, “From the Saljuqs to the Aq Qoyunlu (ca. 1000–1500 "'),” 474. Kd Ibn B@̄b@̄, al-Awāmir al-ʻalāʼ$̄ya, 25. Q56RS:T7U TV W5X SXYU:67 Z679[69X jkr representative) Ẓahīr al-Dīn ‘Illī.Cb According to Tourkhan Gandjeï, Rukn al-Dīn Sulaymān Qilij Arslān (r. 1196–1204) and Ghiyāth al-Dīn Kaykhusraw (r. 1204–11) likewise composed poems in Persian.C- The sultans’ high regard for, and use of, Persian is also evident in their many commissions for Arabic works, which they could not necessarily read, to be translated into Persian.C= The Seljuqs surrounded themselves with courtiers well versed in Persian. The sons of viziers, employed in the Seljuq apparatus like their fathers, would dedicate their poems to the sultans, to wit those produced by Niẓām al-Dīn Aḥmad Arsanjānī for ʿ Izz al-Dīn Kaykāvūs (d. 1220). Ibn B@̄b@̄ recounts that Niẓām al-Dīn Arsanjānī was generously compensated for his poems.CB Ibn B@̄b@̄ further records Persian poems produced by other public ./gures, such as the Su./ Shaykh Majd al-Dīn Isḥāq (d. 1234), who eulogized Seljuq sultans (for instance Ghiyāth al-Dīn Kaykhusraw).CC Perhaps in consequence of its receptiveness to Persian culture, Seljuq Anatolia acted as a magnet for leading Iranian literary and scholarly ./gures, who moved there from Greater Khurasan (continuing the brain drain from this area which had already set in earlier, as noted in the previous section). Examples include the poet Ẓahīr al-Dīn Faryābī (d. 1201)CD and the historian Muḥammad ibn ʿAlī Rāwand@̄ (d. after 1205). Rāwand@̄ stemmed from Kashan and ended up in Konya, as evidenced by the dedication of his Rāḥat al-ṣudūr wa āyāt al-surūr to the local Seljuq Sultan, Ghiyāth al-Dīn Kaykhusraw.CE The Seljuqs’ westward push had a huge e,fect in propelling Persian culture across the length and breadth of Anatolia. This e,fect was going to be continued, even intensi./ed, with the arrival of the Mongols.

The Mongols

Introduction As is well known, the Mongols generated a favourable environment in which the Persian language and culture could thrive. This is sometimes attributed, especially among modern historians of the Arab world, to a supposedly anti-Arab

Le Ibid., 40–41. L0 Gandjeï, “Turkish in Pre-Mongol Persian Poetry,” 70. LG Riy āḥ@̄, Zabān va adab-i Fārs$̄, 24–25. LK Ibn B@̄b@̄, al-awāmir al-ʻalāʼ$̄ya, 45, 83, 151. LL Ibid., 18, 29–32. LM Ibid., 18. LO R āwand@̄, Rāḥat al-ṣudūr, 372–74; Meisami, Persian Historiography, 229, 237–38. jks 45678695: feeling which the Mongols are thought to have entertained in the earlier part of their rule, before their conversion to Islam.CF Whether this was really so is open to question. We have already seen that the Anatolian Seljuqs were capable of supporting both Persian and Arabic, so per se there is no evidence of anti-Arabic feeling in the Mongols’ acceptance of Persian in non-Arabic speaking territories to the East and North of Mesopotamia. As for the tradition of attributing anti-Arabic feelings to the Mongols by medieval and other early Muslim writers, this attribution probably rests on their killing the Caliph, and on the havoc and destruction which the Mongols brought to Baghdad, the heart of the Caliphate, in 1258. Given the central position of the Baghdad Caliphate in the medieval Islamic world, one can certainly see, with Meskoob, that by capturing and de-sanctifying it the Mongols weakened the position of Arabic—not just in Iran, but in the rest of the Muslim world.C] But this was the secondary e,fect of the Mongols’ indi,ference to religion, which resulted in a more liberal atmosphere—there was no anti-Arabic policy per se. Indeed, under Mongol rule Arabic maintained a level of prestige as a sacred language, not only in Iran, but throughout the territories where Persian was used. After the Mongols’ conversion to Islam, we—unsurpris- ingly—./nd Arabic texts from the Mongol period.Ca To set aside the question of the Mongols’ attitudes to Arabic, and return to the main theme of our paper, the Mongols’ role in determining the fates of Persian should perhaps be thought of in terms of two separate phenomena: their causing movements of refugees away from the areas they conquered and what they did in their new-found dominions.

The Mongols as a Cause of Migrations We already observed a westward shift in the centre of gravity of Persian culture under the Turkic dynasties. The Mongols intensi./ed this, though in the opposite way to the Turkic rulers: while the latter had attracted Persian speakers

LP For example, see Sahid, “Arabic Literature,” 667–68: He argues that the Seljuqs and the Mongols were averse to Arabic. Lc Meskoob, Iranian Nationality and the Persian Language, 126; Zarr@̄nkūb, Murūr$̄ bar nasr-i Fārs$̄, sayr$̄ dar shiʻr-i Fārs$̄, 164: Zarr@̄nkūb shares Meskoob’s view. Ld Danner, “Arabic Language: Arabic Literature in Iran,” 2:240–43: Some Persian authors have been mentioned along with their work written in Arabic under the Mongols, such as the famous manual, ‘Awārif al-maʿārif (The gifts of gnosis) by Shihāb al-Dīn Suhrawardī (d. 632/1234); and Tajrīd al-kalām on philosophical and religious issues written by Khwāja Naṣīr al-Dīn Ṭūs@̄ (d. 1273); and Anwār al-tanzīl wa asrār al-ta’wīl (The lights of revelation and the secrets of interpretation), a commentary on the Qur’an written by ʿAbdallāh b. ʿUmar Baīḍawī (d. 716/1316). Q56RS:T7U TV W5X SXYU:67 Z679[69X jkt with their receptive and positive attitude, the Mongol armies drove Persian and Turkic peoples westwards, either because they were *+eeing for their lives or, as in the case of Turks, they were being carried with the Mongol hordes to new pastures.Db In this way, Mongol rule resulted in a greater density of Turkic settlement in northwest Iran. Though their vernacular languages were Turkic, they studied Persian for cultural purposes and accorded it high prestige. As a result of their movements, Persian culture spread even more organically throughout the Turkic world, for the Turkic rulers conquered more territories after the arrival of the Mongols and carried with them practices they regarded as their own heritage. As paradoxical as it may seem, the decline of Persian as a vernacular language in northwest Iran did not hamper the *+ourishing of Persian high culture in the same area. Tabriz became the centre of Persian culture under the IlkhansD- at the same time as its population was being Turki./ed.D= Scholars have tended to pay little attention to smaller Turkic dynasties when discussing the patronage of Persian literature in the Mongol period. However, Persian literature owes much to the Salghurids, who through astute diplomacy saved the province of Fars and its capital Shiraz from Mongol pillage. The treatises on prosody and poetic art by Shams-i Qays (thirteenth century) con./rm that Fars had become a safe haven for men of letters *+eeing the Mongols’ devastating arrival in Transoxania. Shams himself left Bukhara in 1204, seeking refuge at the court of the Salghurids.DB The Salghurid ruler, Abū Bakr ibn Saʿd (d. 1260) owes his good reputation in Persian tradition to one of the most celebrated Persian poets, Saʿdī (d. 1291), who praised him in his treasured compila- tion, the Gulistān.DC Movements of people *+eeing the Mongols resulted in the further Persianization of the Indian administration. The Mongols did not advance south of the Himalayas, and therefore those *+eeing Iran felt safe to settle per- manently in India. Since the Persian speaking community in India *+ourished during the Mongol rule of Iran, more and more people from the Iranian world

Me The purpose of this paper is not to discuss in what capacity the Turks were migrating with the Mongols. There have been articles mentioning that they were enrolled in the Mongol army, but the subject needs to be investigated further in order to see what civilian groups were taken to the Iranian plateau during this period. M0 Seyed-Gohrab and McGlinn, The Treasury of Tabriz, 9. MG Golden, “Migrations, Ethnogenesis,” 111–12. MK Shams al-Dīn Muḥammad ibn Qays al-Rāzī, al-Muʻjam /0 maʻāy$̄r ashʻār al-ʻAjam, Introduction, 2–3: Shams-i Qays refers to the Mongols as Tatars. ML Lane, Early Mongol Rule in Thirteenth-Century Iran, 122–24. jk< 45678695:

*+ocked to the Subcontinent even after the thirteenth and fourteenth centuries. India was regarded as a haven from political and economic hardships in Iran.DD For example, Muḥammad Qāsim Hindūshāh (b. 1560), the author of the Tār$̄kh-i Firishta, who was born over two centuries after the Mongol conquests, narrates that the princes and scholars from Khurasan took refuge in the Mamluk courts of India in order to escape their invaders.DE So intense was India’s Persianization that by the ./fteenth century, under Sikandar Lodi (d. 1517), Persian was adopted as the o,./cial language of Northern India.DF

Persian in the Areas under Mongol Rule In the early stages of their conquests, the Mongols seem to have been less interested in Persian than in Turkic languages.D] Thus ʿAṭā Malik Juwayn@̄ (1226–83) alludes to the Mongols’ high regard for Uighur and their using it for o,./cial treatises in Khurasan under Chinggis Khan, in 1221.Da That said, they became more interested in Persian once they settled among Persian-speaking peoples. For, whether owing to pressure from the courtiers, or to their own practical choice, the Mongol rulers ended up choosing functionaries for their administration in a region from among its local inhabitants.Eb Examples of Iranian administrators include Naṣīr al-Dīn Ṭūs@̄, who had served in Alamut; the famous Juwayn@̄ brothers, who were from an ancient family of bureaucrats; and the historiographer Rashīd al-Dīn (1247–1318).E- Once administration was delegated to local people, de facto the Ilkhans embraced the language most widely extant in the region. In the Iranian Plateau and Central Asia, this meant Persian, which was thus used as an o,./cial language.

MM Am @̄r@̄, Zabān va adab-i Fārs$̄ dar Hind, 5, 11. MO Ibid., 10. Cf. Shaykh Navāzish ʿAlī, Yak daura-yi nāshinākhta dar tārīkh-i adabīyāt-i Fārsī, 439. MP Qutbuddin, “Arabic in India,” 317; Am@̄r@̄, Zabān va adab-i Fārs$̄ dar Hind, 14–15. Mc Morgan, “Who Ran the Mongol Empire,” 131. Md Juwaynī/Qazwīnī, 1:4, 114, 136; 2:260: The Mongols used the Uighur script to write their own language. Oe Khazanov, “Muhammad and Jenghiz Khan Compared,” 469. Cf: Chan, “Liu Ping-Chung (1216–74),” 98–146: “It is possible to create an empire on horseback, but it is impossible to rule it from that position.” O0 Juwaynī/Qazwīnī, 1: Author’s Introduction: Explanations on Juwayn@̄’s background and his family’s participation in the Seljuq and Khwarazmshah governments; 1:4: Juwayn@̄ attests that his father was already at the service of the Mongols; 2:243–44: Juwayn@̄ refers to the new recruits at the service of Arghun; 2:255–56: and the reappointed local governors; Martinez, “Institutional Development, Revenues and Trade,” 90. Q56RS:T7U TV W5X SXYU:67 Z679[69X jk\

It also deserves mention that the Mongols grew interested in Turkish as a result of contact with the Turkic populations of Central Asia, many of whom were absorbed into the Mongol army.E= These Turkic populations stemmed from regions which were the cradle of Iranian civilization and regarded Persian language and culture as part of their own heritage. The Mongols thus came to Persian culture through Turkic culture, which they held in high esteem.EB By encouraging the use of Persian, the early (non-Muslim) Mongol rulers of Iran prompted even Christians to write in Persian instead of their traditional language (Syriac).EC Thus Iwannis ʿIzz al-Dīn Muḥammad b. Muẓa,far translated the Diatessaron of Tatian into Persian and sent copies to eastern Iran.ED Even some Nestorian travellers wrote their accounts in Persian.EE However, this trend stopped shortly after the Mongols’ conversion to Islam: the Christian Church of Iran was no longer supported by the rulers, and the turmoil which ensued in the region along with Temerlane’s onslaught isolated the Christians from the surrounding Persian culture. Thereafter, there is no record of Christians in the region recording their thoughts and history in Persian. The end of Christian recording in Persian suggests the rapid decline of Nestorians in territories where Persian was common currency. Several major pieces of historical writing in Persian were produced under the Mongols. Notable historians included ʿAṭā Malik Juwayn@̄, who wrote the Tār$̄kh-i Jahāngushā; the vizier Rashīd al-Dīn, who wrote the Jāmiʿ al-tawārīkh; ʿAbd Allāh b. Faḍl Allāh Waṣṣāf (d. 1323), the author of the Tār$̄kh-i Waṣṣāf; and Ḥamd Allāh Mustawf@̄ (d. 1349), who wrote the Tār$̄kh-i guz$̄da. Many literary ./gures dedicated their work to the Juwayn@̄ family; Naṣīr al-Dīn Ṭūs@̄, for example, dedicated his book on saints, Awṣāf al-ashrāf to Shams al-Din Juwayn@̄ (Ṣaḥib al-Dīwān), and Qāḍī Niẓām al-Dīn Iṣfahānī dedicated his collection of poems to both Juwayn@̄ brothers (ʿAṭā Malik and Shams al-Dīn).EF Thus, Persian bene./ted from the support of the Persian notables, who were

OG Juwaynī/Qazwīnī, 1:30; 2:218; Smith Jr., “Mongol Manpower and Persian Population,” 275; Langer and Blake, “The Rise of the Ottoman Turks and its Historical Background,” 486–97. OK Juwayn ī/Qazwīnī, 3, 114: The Mongol yarlighs (ordinances) were promulgated in Uighur at the onset of Iran’s conquest and were carried on stamps to Nishapur in 1221; Morgan, “Who Ran the Mongol Empire,” 124, 128: “Chingiz had the Uighur Turkish script adapted for Mongolian”; Golden, “Migrations, Ethnogenesis,” 111, discusses Turkic as the lingua franca in the Chinggisid states. OL Khanbaghi, The Fire, the Star and the Cross, 84–85. OM Fiey, Chrétiens syriaques sous les Mongols, 75; Thomas and Vahman, “Bible—Persian Translations,” 210. OO Bar Sauma, The Monks of Kublai Khan, 5, 187. OP Browne, “Introduction,” 48–54. jkf 45678695: part of the Mongol apparatus. Their writings had the purpose of consolidating the Ilkhans’ rule, especially when they were recording historical events and, e,fectively, of legitimizing the new Iranian state created by them. Indeed, Waṣṣāf and Mustawf@̄ were in the service of the Ilkhans as tax administrators.E] They have therefore all been criticized on grounds of tendentiousness.Ea For this very reason, the Mongols’ employment of them is evidence of how important Persian was seen to be by the Mongols of Iran in the creation of political consensus; however, fortuitously the task of propagating the language was left to the Persians and the Turks.

Anatolia in the Mongol Period The Persianization of Seljuq Turkey intensi./ed after the Mongol conquest, which prompted the migration of Persian speakers to the region. Thus ʻAbd al-Ḥusayn Zarr@̄nkūb notes that under the Mongols Persian continued to bloom in Anatolia, thanks to the patronage of local Seljuq rulers.Fb We ./nd hints in Ibn B@̄b@̄’s al-Awāmir al-ʻalāʼ$̄ya f$̄ al-umūr al-ʻalāʼ$̄ya and Āqsarāʼ@̄’s Musāmarat al-akhbār wa musāyarat al-akhyār that many administrators in Seljuq Anatolia during the Mongol period were from Iran,F- and this is consistent with names found in the sources, such as Ṣaḥib Shams al-Dīn Iṣfahanī, who functioned under ʿIzz al-Dīn Kaykāvūs (d. 1260).F= The special Seljuq o,./ce of Parvāna, or representative of the sultan, was generally held by Persians, e.g. Niẓām al-Dīn and Mu‘īn al-Dīn Sulaymān Daylamī. The latter was e,fectively in charge of Anatolia at the time of Abaqa (d. 1282), the Ilkhan, who put him to death for collaborating with the Mamluk sultan.FB The vizier Shams al-Dīn Juwayn@̄ installed his son Sharaf al-Dīn Hārūn as his representative, which might explain Ibn B@̄b@̄’s eulogies to the Juwayn@̄ family.FC

Oc Mustawf @̄, Tār$̄kh-i guz$̄da, Muqaddima, 3. Od Zarr @̄nkūb, Murūr$̄ bar nasr-i Fārs$̄, 164. Pe Zarr @̄nkūb, Murūr$̄ bar nasr-i Fārs$̄, 353; Rypka, History of Iranian Literature, 183: Rypka also says that the Seljuqs did not stress their Turkish origins and adopted a Perso-Muslim identity; Melville, “The Early Persian Historiography of Anatolia,” 135–36: Melville avers that under the Mongols Anatolia was integrated further in the Persian cultural sphere with Seljuq rulers remaining as ./gureheads. P0 Ā qsarāʼ@̄, Musāmarāt al-akhbār wa musāyarāt al-akhyār, 408, 418, 432; Anonymous, al-A,thār al-Mawlaw$̄ya /0 al-’Adwār al-Saljūq$̄ya, 350: They refer also to Shams al-Dīn Iṣfahān@̄. Melville, “The Early Persian Historiography of Anatolia,” 138–41. PG Ibn B@̄b@̄, al-Awāmir al-ʻalāʼ$̄ya, 83. PK Ā qsarāʼ@̄, Musāmarat al-akhbār, 409–10, 425, 431; Boyle, “Dynastic and Political History of the Il-Khans,” 361. PL Riy āḥ@̄, Zabān va adab-i Fārs$̄, 78. Q56RS:T7U TV W5X SXYU:67 Z679[69X jk)

The Anatolian Seljuqs’ interest in Persian culture is clearly demonstrated by illustrations of the Shāhnāma found in one of the Halls of Kubadabad, the palace built by ʿAlāʾ al-Dīn Kayqubād 3 (d. 1237). The same palace vaunted mosaics with Iranian themes such as Bahrām Gūr and Laili.FD Like India, Anatolia experienced a large in*+ux of learned Persians during the disruptions wrought by the arrival of the Mongols in Khurasan. This trend is reported in the writings of Muslim travellers.FE The most famous migrant from Khurasan to Anatolia in this period is the acclaimed poet Jalāl al-D@̄n Balkh@̄, better known as Rūmī (d. 1273).FF His father left Balkh presumably due to the political turmoil which followed the Mongol expansion into Central Asia. Likewise, Aḥmad b. Muḥammad Ṭūs@̄, known as Qanā’ī, *+ed Tus for India and ended up in Konya under Kayqubād 3 (d. 1237) and turned the Kal$̄la wa Dimna into poetry for Kaykāvūs 33 (d. 1260).F] Observing that the Mongol era witnessed a *+ourishing of Persian literature in the lands newly conquered by the Anatolian Seljuqs,Fa Riyāḥ@̄ claims that Persian superseded Arabic in Anatolia, and together with Meskoob and Zarr@̄nkūb he concludes that the increased di,fusion of the Persian language was due to the Mongols.]b It is, however, important to stress that the Mongols themselves did not directly support the Persian language in Anatolia. Instead, this was done by the previously established bureaucrats and Turkic princes, who regarded Persian as their own language and heritage, just as their ancestors had already done before the onset of the Mongol period. Persian speakers at the service of the Seljuqs could well have been at the service of the Mongols. Ibn B@̄b@̄, who served the Seljuqs, dedicated his work to the famous historiographer ʿAṭā Malik Juwayn@̄ and praised his brother the vizier Shams al-Dīn Juwayn@̄ (d. 1285), who were at the service of the Mongols.]-

PM Mashk ūr, Akhbār-i Salājiqa-yi Rūm, Illustrations section; Riyāḥ@̄, Nufūz̲-i zabān va adabiyāt-i Fārs$̄, 8. PO Langer and Blake, 485–86: al-Idrīsī (d. 1166), who travelled through Anatolia in about 1117, uses the ancient names of places in his records, whereas Ibn Baṭṭūṭa (d. 1369), who visited the region around 1330, uses Turkish names. PP Walad, Waladnāma—Mathnaw@̄-i Walad@̄, 515. Pc Ṭ ūs@̄, Kal$̄la va Dimna-yi manẓūm, Introduction. Mashkūr, Akhbār-i Salājiqa-yi Rūm, 22. Pd Riy āḥ@̄, Zabān va adab-i Fārs$̄, 30. Cf. Najm al-Dīn Rāzī, Mirṣād al-‘ibād, 19. As evidence, Riyāḥ@̄ cites a list of Su./ authors from this era who took refuge in Anatolia. It includes names such as Najm al-Dīn Rāzī (d. 1256), Fakhr al-Dīn ‘Irāqī (d. 1289), and Sa‘īd Farghānī (d. 1300). ce Riy āḥ@̄, Nufūz-i zabān va adabiyāt-i Fārs$̄, 12: on the rise of Persian in Anatolia cf. Ibn B@̄b@̄, 14–16. c0 Mashk ūr, Akhbār-i Salājiqa-yi Rūm, Introduction, 20. j;k 45678695:

This shining example displays the Mongols’ political role in bringing Anatolia closer to Persia. So profound was the Persianization of Anatolia in the Mongol period that it lasted into Ottoman times. Indeed, for a while Persian remained the language of diplomacy and the court even after the Ottoman conquest of Istanbul.]= More generally, early Ottoman men of letters and administrators emulated Persian models of adab and style in their functions as administrators of the empire.]B There also seems to be evidence for emulation by the Ottomans of the Persian system of government, as applied by the Seljuqs and Ilkhans.]C Shinder traces the Ottoman governmental system back to Rashīd al-Dīn’s administration under Ghazan (d. 1304), the Ilkhan of Persia. He also reports that the libraries of Istanbul are ./lled with Ilkhanid and Timurid administrative hand- books. Moreover, Persian terms were used in Ottoman records along with the Persian solar calendar for tax purposes.]D Halil Inalcik holds that Sultan Bayezid 3 applied the Turkish-Islamic system inherited from the Mongol rulers of Iran to his empire.]E Ottoman Turkish includes a huge stock of Persian vocabulary, which chimes with the Ottomans’ deep interest in literature written in Persian.]F Indeed, much Ottoman literature had roots in Persian and Arabic literature.]] Some very popular stories made their way to the Ottoman court all the way from Central Asia. For example, the Sindbadnāma was originally written in Persian by Muḥammad b. ʿAlī al-Samarqandī in 1160 and dedicated to the Qarakhanid ruler Alp Kutlu Tuna Bilge Masʿūd b. Ḥasan. It eventually reached the Ottoman court in a Turkish translation produced in the thirteenth century.]a

Gathering the Threads

It is di,./cult to assess the Mongols’ role in ensuring the di,fusion of the Persian language, for it is hard to disentangle a number of interrelated factors.

cG Riy āḥ@̄, Zabān va adab-i Fārs$̄, 203–5: on letters sent in Persian under the later Ottoman rulers. cK Shinder, “Early Ottoman Administration,” 506. cL Ibid., 510. cM Ibid., 509. cO Inalcik, “The Emergence of the Ottomans,” 280. cP Ibid., 507. cc Shinder, “Early Ottoman Administration,” 507. cd Renda, “Sindbadnāma,” 311. Q56RS:T7U TV W5X SXYU:67 Z679[69X j;;

Certainly, the Mongols created an Iranian state, in which Persian transcended linguistic boundaries (to wit the Christians early in the period). Yet it does not seem that the Mongols had any special emotional investment in Persian per se: it was the language of the peoples (both Iranian and Turkic) they ruled over in certain areas, and so they upheld it in their administration and allowed it to *+ourish in the cultural domain—for reasons of practical convenience. In this respect, the contribution of the Mongols in expanding the borders of Persian has been sometimes overstated.ab The Mongols arrived in territories where Persian was common currency, and the Turkic rulers who governed these territories for centuries before them had already relayed the language to India and Anatolia. What the Mongols did, in a period shorter than a century, was to generate new opportunities for the Turkic groups to export their heavily Persianized culture to new areas. Indeed, the further the Mongols moved westwards the more Turki./ed they became.a- Thus, while politically the Iranian and Turkic groups were in thrall to the Mongols, at the cultural level the Mongols ful./lled long-held Turko-Persian objectives: they were tremendous catalysts of pre-existing processes. At their zenith, the Mongols blurred the cultural and political borders between Anatolia and Iran. The result was the adoption of the Ilkhanid system by the early Ottomans. If the Ghaznavids and Seljuqs had already been spread- ing the seeds of Persian to the south and the west of Khurasan, the Mongols catapulted Persian out of its cradle, making ultimately India and Anatolia much more stimulating centres of Persian literature and art which endured three centuries after the Ilkhanate’s demise. In the end, it seems that the credit for the ‘Persian renaissance’ of the Middle Ages must be shared jointly by the Turkic groups and the Mongols— though not without considerable assistance from the Iranians themselves, of course!

Bibliography

Am@̄r@̄, Kayūmars. Zabān va adab-i Fārs$̄ dar Hind. Tehran: Shūrā-yi Gustarish-i Zabān va Adab@̄yāt-i Fārs@̄, 1995. Anonymous. al-Axthār al- Mawlaw@̄yah ./ al- Adwār al-Saljūqiyah. In Akhbār-i Salājiqa-yi Rūm, edited by Muḥammad Javād Mashkūr, 341–71. Tehran: Kitābfurūsh@̄-yi Tehran, 1972.

de Lane, Early Mongol Rule in the Thirteenth Century Iran, ix. d0 di Cosmo, Frank and Golden, “Introduction,” 4. j;j 45678695:

Axqsarāʼ@̄, Maḥmūd al-Kar@̄m. “Musāmarat al-akhbār wa musāyarat al-akhyār.” In Akhbār-i Salājiqa-yi Rūm, edited by Muḥammad Javād Mashkūr, 401–97. Tehran: Kitābfurūsh@̄-yi Tehran, 1351/1972. Bahār, Muḥammad Taq@̄. Sabk-shinās$̄, ya Tar$̄kh-i taṭavvur-i nasr-i Fārs$̄. 2 vols. Tehran: Vizārat-i Farhang, 1942. Balādhūr@̄, Aḥmad ibn Yaḥya. Futūḥ al-buldān. Translated by P. K. Hitti (The Origins of the Islamic State). New York: $yz Press, 1916. Bar Sauma, Rabban. The Monks of Kublai Khan. Translated by Ernest A. Wallis Budge. London: Religious Tract Society, 1928. Bosworth, Cli,ford E. “Kay Ka’us b. Iskandar.” In Encyclopaedia of Islam, Second Edition 4: 815–16 Leiden: Brill, 1997. Boyle, John A. “Dynastic and Political History of the Il-Khans.” In The Cambridge History of Iran, vol. 5, The Saljuq and Mongol Periods, edited by John A. Boyle, 303– 421. Cambridge: Cambridge University Press, 2001. Browne, Edward G. “Introduction.” In Ta’r$̄kh-i-Jahān-gushā of ‘Alā’u’d-D$̄n ‘Atā Malik-i- Juwayn$̄, edited by Mīrzā Muḥammad ibn ʻAbd al-Wahhāb Qazw@̄n@̄. London: Luzac, 1912. ———. A Literary History of Persia. Cambridge: Cambridge University Press, 1928. Cahen, Claude. “Historiography of the Seljuqid Period.” In Historians of the Middle East, edited by Peter M. Holt and Bernard Lewis, 59–78. London: Oxford University Press, 1962. Chan, Hok-lam. “Liu Ping-Chung (1216–74). A Buddhist-Taoist Statesman at the Court of Khubilai Khan.” T’oung-pao 53 (1967): 98–146. Danner, Victor. “Arabic Literature in Iran.” In The Cambridge History of Iran, vol. 4, 566– 94. Cambridge: Cambridge University Press, 1975. ———. “Arabic Language: Arabic Literature in Iran.” In Encyclopaedia Iranica, 33: 240– 43 (1982). di Cosmo, Nicola, Allen Frank, and Peter B. Golden. “Introduction.” In The Cambridge History of Inner Asia, The Chinggisid Age, edited by Nicola di Cosmo, Allen Frank, and Peter B. Golden, 1–6. Cambridge: Cambridge University Press, 2009. Dinkard. Ed. Dastur Behramjee Sajana. Bombay: Dutfur Ashkara Press, 1888. Elgood, Cyril. A Medical History of Persia. Cambridge: Cambridge University Press, 1951. Fiey, Jean-Maurice. Chrétiens syriaques sous les Mongols (Il-Khanat de Perse 23334–2354 siècles). Louvain: "z"{, 1975. Fischel, Walter J. “Israel in Iran (A Survey of Judeo-Persian Literature).” In The Jews, their History, Culture and Religion, edited by Louis Finkelstein, 817–57. Philadelphia: The Jewish Publication Society of America, 1949. Frye, Richard N. The Golden Age of Persia, the Arabs in the East. London: Weidenfeld and Nicolson, 1975. Q56RS:T7U TV W5X SXYU:67 Z679[69X j;r

Gabrieli, F. “Ibn al-Muķa,fa’.” In Encyclopaedia of Islam, Second Edition 333: 883–85 Leiden: Brill, 1997. Gandjeï, Tourkhan. “Turkish in Pre-Mongol Persian Poetry.” Bulletin of the School of Oriental and African Studies 49/1 (1986): 67–75. Gardet, Louis. “Religion and Culture.” In The Cambridge History of Islam, 2b, edited by Peter M. Holt, Ann K. S. Lambton and Bernard Lewis, 569–603. Cambridge: Cambridge University Press, 1977. Golden, Peter B. “Migrations, Ethnogenesis.” In The Cambridge History of Inner Asia, the Chinggisid Age, edited by Nicola di Cosmo et al., 109–19. Cambridge: Cambridge University Press, 2009. Grignaschi, Mario. “Quelques spécimens de la littérature sassanide (Karnamag).” Journal Asiatique 254 (1966): 19–20. Hwui Li, Shaman. The Life of Hiuen-Tsiang by the Shaman Hwui Li. Translated by Samuel Beal. London: Kegan Paul, 1911. Ibn al-Nad@̄m. Kitāb al-Fihrist (The Fihrist of Ibn al-Nadim). Translated by Bayard Dodge. New York: Columbia University Press, 1970. Ibn B@̄b@̄. Mukhtaṣar-i Saljūqnāma. In Akhbār-i Salājiqa-yi Rūm, edited by Muḥammad Javād Mashkūr 2–337. Tehran: Kitābfurūsh@̄-yi Tehran, 1972. Inalcik, Halil. “The Emergence of the Ottomans.” In The Cambridge History of Islam, vol. 1, edited by Peter M. Holt, Ann K. S. Lambton and Bernard Lewis, 293–323. Cambridge: Cambridge University Press, 1970. Izgi,l Cevat. “Hubeyş et-Ti*+isi.” Türkiye Diyanet Vak/0 Islam6 Ansiklopedisi, Istanbul: Türkiye Diyanet Vakfı vol. 18 (1999): 268–70. Juwaynī, ʿAṭā Malik. Ta’rīkh-i Jahāngushā. Edited by Mīrzā Muḥammad Qazwīnī. 2 vols. Leiden: Brill, 1912–37. Kay Kā’ūs b. Iskandar. Kitāb-i Naṣ$̄ḥatnāma: Qābūsnāma. Edited by Saʻ@̄d Naf@̄s@̄. Tehran: Kitābfurush@̄-yi Furūgh@̄, 1963. ———. Qābūsnāma. Edited by Ghulām Ḥusayn Yūsuf@̄. Tehran: Bungāh-i Tarjumah va Nashr-i Kitāb, 1352/1973. Khanbaghi, Aptin. The Fire, the Star and the Cross: Minority Religions in Medieval and Early Modern Iran. London: I. B. Tauris, 2006. ———. “De-Zoroastrianization and Islamization: The Two Phases of Iran’s Religious Transition, 747–837 "'.” In Comparative Studies of South Asia, Africa and the Middle East 29/2 (2009): 201–12. Khazanov, Anatoly M. “Muhammad and Jenghiz Khan Compared: The Religious Factor in World Empire Building.” Comparative Studies in Society and History 35/3 (July 1993): 461–79. Khusraw, Nāṣir. Safarnāma. Edited by M. Dab@̄r Siyāq@̄. Tehran: Intishārāt-i Zavvār, 1335/1957. j;s 45678695:

Lambton, Ann K. S. “The Dilemma of Government in Islamic Persia: The ‘Siyasat-nama’ of Nizam al-Mulk.” Iran 22 (1984): 55–66. Lane, George. Early Mongol Rule in Thirteenth-Century Iran: A Persian Renaissance. London: RoutledgeCurzon, 2003. Langer, William L., and Robert P. Blake. “The Rise of the Ottoman Turks and its Historical Background.” The American Historical Review 37/3 (Apr. 1932): 486–97. Mahmoodi-Bakhtiari, Behrooz. “Planning the Persian Language in the Samanid Period.” In Iran and the Caucasus 7/1–2 (2003): 251–60. Martinez, Aresnio Peter. “Institutional Development, Revenues and Trade.” In The Cambridge History of Inner Asia, the Chinggisid Age, edited by Nicola di Cosmo et al., 89–108. Cambridge: Cambridge University Press, 2009. Mashkūr, Muḥammad Javād, ed. Akhbār-i Salājiqa-yi Rūm. Tehran: Kitābfurūsh@̄-yi Tehran, 1972. Meisami, Julie. Persian Historiography to the End of the Twelfth Century. Edinburgh: Edinburgh University Press, 1999. Melville, Charles. “From the Saljuqs to the Aq Qoyunlu (ca. 1000–1500 "').” Iranian Studies 31/3–4 (1998): 473–82. ———. “The Early Persian Historiography of Anatolia.” In History and Historiography of Post-Mongol Central Asia and the Middle East, edited by Judith Pfei,fer, Sholeh Alysia Quinn, 135–62. Wiesbaden: Harassovwitz Verlag, 2006. Meskoob, Shahrokh. Iranian Nationality and the Persian Language. Washington }": Mage Publishers, 1992. Morgan, David. “Who Ran the Mongol Empire?,” The Journal of the Royal Asiatic Society of Great Britain and Ireland 1 (1982): 124–36. ———. Medieval Persia 1040–1797. London: Longman, 1988. Mustawf@̄ Qazv@̄n@̄, Ḥamd Allāh. Tār$̄kh-i guz$̄da. Edited by ʻAbd al-Ḥusayn Navāʼ@̄. Tehran: Am@̄r Kab@̄r, 1984. Peacock, Andrew. “Saljuqs—Saljuqs of Rum.” In Encyclopaedia Iranica; online version at http://www.iranicaonline.org/articles/saljuqs-iii. Piemontese, Angelo Michele. Storia della Letteratura Persiana, vol. 1. Milan: Fratelli Gabbri Editori, 1970. Pourjavady, N. “Literary debates in the Sa./na-yi Tabriz.” In The Treasury of Tabriz, The Great Il-Khanid Compendium, edited by Asghar A. Seyed-Gohrab and Sen McGlinn, 114–138. Amsterdam: Rozenberg Publishers, 2007. Qutbuddin, Tahera. “Arabic in India: A Survey and Classi./cation of its Uses, Compared with Persian.” Journal of the American Oriental Society (Jul.–Sep. 2007): 315–38. Rāwand@̄, Muḥammad. Rāḥat al-ṣudūr wa āyat al-surūr. In Akhbār-i Salājiqa-yi Rūm, edited by Muḥammad Javād Mashkūr, 372–76. Tehran: Kitābfurūsh@̄-yi Tehran, 1972: 32–34. Q56RS:T7U TV W5X SXYU:67 Z679[69X j;t al-Rāzī, Shams al-D@̄n Muḥammad ibn Qays. al-Muʻjam fī maʻāy$̄r ashʻār al-ʻAjam. Edited by Mīrzā Muḥammad ibn ʻAbd al-Wahhāb al- Qazw@̄n@̄ and E. G. Browne. London: Luzac, 1909. Renda, Gunsel. “Sindbadnama: An Early Ottoman Illustrated Manuscript Unique in Iconography and Style.” Muqarnas: An Annual on the Visual Culture of the Islamic World 21 (2004): 311–22. Riyāḥ@̄, Muḥammad Am@̄n. Nufūz̲-i zabān va adabiyāt-i Fārs$̄ dar qalamrav-i ʻUsmān$̄. Tehran: Intishārāt-i Am@̄r Kab@̄r, 1350/1972. ———. Zabān va adab-i Fārs$̄ dar qalamrav-i ʻUsmān$̄. Tehran: Pazhang, 1369/1991. Rypka, Jan. History of Iranian Literature. Dordrecht: D. Riedel Publishing Company, 1968. Sahid, Irfan. “Arabic Literature.” The Cambridge History of Islam, vol. 2b, edited by Peter M. Holt, Ann K. S. Lambton and Bernard Lewis, 657–71. Cambridge: Cambridge University Press, 1970. Seyed-Gohrab, Asghar A., and S. McGlinn (eds). The Treasury of Tabriz: The Great Il-Khanid Compendium. Amsterdam: Rozenberg Publishers, 2007. Shinder, Joel. “Early Ottoman Administration in the Wilderness: Some Limits on Comparison.” International Journal of Middle East Studies 9 (Nov. 1978): 497–517. Shūshtar@̄, ʻAbbās Mihr@̄n. Tār$̄kh-i zabān va adab$̄yāt-i Ir, ān dar khārij az Ir, ān. Tehran: Mān@̄, 1352/1974. Smith Jr., John Masson. “Mongol Manpower and Persian Population.” Journal of the Economic and Social History of the Orient 18/3 (Oct. 1975): 271–99. Spuler, Bertold. “The Evolution of Persian Historiography.” In Historians of the Middle East, edited by Bernard Lewis and Peter Holt, 127–32. London: Oxford University Press, 1962. Storey, Charles A. Persian Literature: A Bio-Bibliographical Survey, vol. 1. London: Royal Asiatic Society, 1970. ———. Persian Literature: A Bibliographical Survey, vol. 2, part 2. London: Luzac, 1971. ———. Persian Literature: A Bibliographical Survey, vol. 3, part 1. Leiden: Brill, 1984. Thomas, Kenneth J., and Fereydun Vahman. “Bible—Persian Translations.” In Encyclopaedia Iranica, 32: 210. Ṭūs@̄, Aḥmad ibn Muḥammad. Kal$̄la va Dimna-yi manẓūm. Tehran: Bunyād-i Farhang-i Irx ān, 1979. Unknown. “Anōsharwān b. Khālid.” In E. J. Brill’s First Encyclopaedia of Islam, 3:357. Leiden: Brill, 1913–1936. ‘Utb@̄, Abū Nasr Muḥammad. al-Ta’r$̄kh al-Yam$̄n$̄. 2 vols. Cairo: no publisher, 1869. Walad, Bahā al-Dīn. Waladnāma—Mathnaw@̄-yi Walad@̄. In Akhbār-i Salājiqa-yi Rūm, edited by Muḥammad Javād Mashkūr, 515–17. Tehran: Kitābfurūsh@̄-yi Tehran, 1972: 515. Zarr@̄nkūb, ʻAbd al-Ḥusayn. Murūr$̄ bar nasr-i Fārs$̄, sayr$̄ dar shiʻr-i Fārs$̄. Tehran: Intishārāt-i Bayn al-Milal@̄ al-Hudā, 1375/1997. !"#$%&' () Darughachi in Armenia

Bayarsaikhan Dashdondog

The formation of the Mongol Empire brought the sedentary civilizations, such as China and Iran, under the steppe people for over century. The Mongol success story in creating the largest land empire was due to their ability to exploit human and material sources through the realms they controlled. The establishment of a military and administrative institution known as darughachi* was one of the transformative organizations made by the Mongols to achieve this exploitation. Along with devastation, depopulation and ravages, as Michal Biran states, Chinggis Khan shaped the governance of the emerging empire.+ The governors, also known as darughachis, were the means by which a military apparatus could be transformed into a civil authority. The function of the Mongol administrative machine in Iran, as well as in China- has got its wide resonance in scholarship. However, the activities of Mongol governance in the Caucasus, particularly in Armenia, are less well known. As a part of Mongol expansion to the Middle East, Mongol governors began to rule Greater Armenia as early as the period from 1220 to 1245, which shows that this governing system was vital when the time of plunder and rich booty was over. The need for economic sustainability became a vital concern for the Mongols, and this was re./ected in their expansion into northern China, Transoxiana, Khurasan, and now into the Caucasus. Controlling and ruling these lands necessitated the establishment of a governing method that was perhaps borrowed from the Qara Khitai o012ce of basqaq or darughachi, a governor who was appointed to control and super- vise submitted sedentary people and oversee the tax collection and requisition

3 Darughachi is a Mongolian term that literally means the one who presses. 5 Biran, Chinggis Khan, 64. 6 Aigle, Persia under Mongol Domination, 65–78; Aubin, “Émirs mongols et viziers persans;” Kolbas, The Mongols in Iran; Lambton, Continuity and Change in Medieval Persia; Lambton, “Mongol Fiscal Administration,” 1: 79–99; Idem, 2: 97–123. Morgan, “Mongol or Persian;” Petrushevsky, “The Socio-Economic Condition of Iran;” Lane, “Arghun Aqa,” 459–82. 7 Buell, “Sino-Khitan Administration in Mongol Bukhara,” 121–51; Endicott-West, “Merchant Associations in Yüan China,” 127–54; Ostrowski, “The ‘tamma,’ ” 262–77. Matsuda, “On Tan- ma-chi in China,” 307–14; Farquhar, The Government of China under Mongolian Rule.

© !"#$#!%$&!' ()$%% #*, %'$+'#, ,-./ | +"$ .-.../1/23425561.63,/_5., !"#$%&"'&( 9: ;<=>:9; ?(@ of auxiliary forces.A The complexity of this administrative o012ce lies in the mastering of semi-sedentary and sedentary societies, which was, in fact, a new challenge for the Mongols themselves that is traceable in the Secret History of the Mongols:

After conquering the Sarta’ul [Muslim] people, Chinggis Khan issued a decree placing resident commanders (darughachis) in all di0ferent cities. From the city of Ürünggechi [Gurgānj] came a father and son, called Yalawachi [Yalavach] and Masqut [Mas‛ūd] of the Qurumchi clan of the Sarta’uls. They told Chinggis Khan about the custom and law of the bal- ghasun [city].E

Based on their experience in dealing with the Chinese cities like Nanjing and Jingdu from 1214,F and facing new practices in governing the cities of Central Asia, the Mongols were deeply concerned to learn about sedentary culture. They put forward the idea of establishing an administrative body, which could evolve from a military machine into a resident o012ce to ensure that the conquered region and town would remain under Mongol rule by being ready to provide manpower as well as paying various forms of taxes. This paper considers the system of darughachis in Armenia, the function of which surely has varied in each country, depending on local conditions.G The administrative division of the Mongol Empire among the sons of Chinggis Khan into separate successor states called uluses guaranteed their autonomy while also ensuring that they remained subject to central imperial control. Chinggis Khan gave all lands to the west of the Irtysh, the Aral Sea and the Amu Darya to the ulus of Jochi.H However, Georgia, Armenia, Arran,

I Morgan, The Mongols, 129–30; Endicott-West, “Merchant Associations,” 45; 151, n.55. On the etymology of these terms, see further below. J Mongolyn Nuuts Tovchoon (KL%) [The Secret History of the Mongols], 92, §263. M Ibid., 85–86, §252–53. N Buell considers that, though it is certain that the 12rst darughachis were appointed by the Mongols for China (seemingly as early as 1214), the area of their empire in which the darughachi system received its 12rst full development was Turkestan; see Buell, “Sino-Khitan Administration,” 133. Turkestan, according to medieval geography, it is the land that extended between Muslim possessions and China, inhabited by Turkish and Mongol nomads; see Barthold, Turkestan down to the Mongol Invasion, 64. On the function of darughachis in the Golden Horde, see Vásáry, “The Golden Horde,” 187–97. O The Ulus of Jochi or later the Golden Horde was the inheritance of Chinggis Khan’s son Jochi and grandson Batu, with its key base in the Pontic and Caspian steppe and its capital at Sarai. In the quriltai (great assembly) of 1235, it was decided to launch an expedition to Russia and ?(U P;QRPS:PST

Shirvan and Azerbaijan (Atrpatakan) were not o012cially a012liated with the ulus of Jochi, despite the continuous attempts of the Golden Horde khans from 1236 to the establishment of the Ilkhanate in 1256, and even later, these countries represented separate regions ruled by a military commander.*V The 12rst Mongol military commander in the Middle East was Chormaqan, who was appointed by Ögetei Khan (r. 1229–41) as governor ‘to the Fourth Clime.’** Consequently, from the 1230s onwards, Georgia and the territories of the Zakʿarids, Greater Armenia (southern and western Armenia), Shirvan, Arran and Mughan, and Azerbaijan were 12rst known as the 12ve vilayets or regions commanded by Chormaqan. As contemporary Armenian historian Grigor Aknertsʿi points out, at the local quriltai (assembly) called by Chormaqan, the Mongol noyans (commanders) divided Armenia into three parts: the northern or the Zakʿarids’ principality; inner or western Armenia, and southern Armenia.*+ Georgia and the Zakʿarid holdings, including Airarat, Siwnikʿ and Artsʿakh, were known as Gurjistan vilayet, and had been divided into eight

then to Eastern Europe in pursuit of the Kipchaks. In 1236, Batu began his conquest of Russia. The Golden Horde ruled over Russia from 1237 to 1480. 3X )*+, 96, §274. Military governor that is glossed in the Secret History of the Mongols (classical text, ed. D. Tserensodnom et al.) as tamghachi [tamgha+ chi] in Cyrillic is debatable, though; the general governor and holder of the state tamgha (seal) is di0ferent from the tamghachi, a tax collector, cf. Doerfer, Türkische und mongolische Elemente, 2: 565–66. Apparently, in Chinese characters this term has been glossed as tammachi; see Kuribayashi, “Word-Index to the Secret History of the Mongols,” [http://www.cneas .tohoku.ac.jp/sta0f/hkuri/articles]. A recent study shows that tammachi is a Mongolian term from tamahu (to seize) and the tammachi was in charge of the military formation called tanma/tammā, which operated as security forces in the occupied lands along the borders; see Bazarsuren, “Is it Tamachin or Tamghachin?” 23–31; also Ostrowski, “The ‘tamma,’” 264 and May, The Mongol Art of War, 22 and 40. On the tammachi in China during the Mongol period, where the function of the o012ce evolved from a borderland scouting or controlling force into more settled formation, see Matsuda, “On Tan-ma-chi in China during the Mongol Period,” 307–14. 33 Juwaynī/Boyle, 482–83. In Rashīd al-Dīn, Chormaqan was appointed as lashkar-tamā along with an army of 40,000 soldiers; Rashīd al-Dīn, Сборник летописей (Compenduim of Chronicles), Z, part 1: 98–99. The lashkar-tamā was the head of a special garrison of soldiers who were sent out from the main body of the imperial army for permanent residence in certain places; ibid., 99. Another de12nition of this term states that lashkar-tamā is an auxiliary force consisting of various nationalities, with only the commanders being Mongols; see Doerfer, Türkische und mongolische Elemente, 1: 255–56. 35 Grigor Aknertsʿi, History of the Nation of the Archers, 34. !"#$%&"'&( 9: ;<=>:9; ?([ tumans, 12ve of which belonged to the land of the Georgians.*, The remaining three tumans were in the Armenian Zakʿarid territories of Ani and Kars; the territories of the Awagids in Siwnikʿ and Artsʿakh; and those of the Vagramids in Gag and Shamkor, including surrounding areas.*- The governing model was not the same in each vilayet; however, the Mongols tried to use local, already existing, social formations.*A Thus, from 1236 until 1246, the Mongols did not intervene in the governing structure of the Zakʿarids. After the escape of Queen Ŕusudan to Swanetia, Eastern Georgia and Armenia were ruled by the Zakʿarid princes, ishkhans.*E Apparently, they were appointed as heads of tumans.*F Only in 1246 did the Mongols bring in David, the son of the Georgian King Lasha, from Caesarea/Kayseri; David’s reign lasted until 1270.*G Therefore, Eastern Georgia and Zakʿarid Armenia were doubly vassals of the Georgian king and of the Great Khan.*H By the second half of the thirteenth century, the Mongol system of ownership had been introduced in southern and western Armenia, known as the vilayet of Greater Armenia.+V The Mongols used this land as their summer and winter base. Local governors were dismissed and the Mongol nobility owned the land as injü, a land allotted to the ruling family.+* However, within the vilayet of Greater Armenia, some principalities of the Taurus Mountains,

36 Ibid., 34. Tuman as an administrative unit, derived from its basic military form, was able to mobilize ten thousand soldiers. 37 Qazv īnī, The Geographical Part of Nuzhat al-Qulūb, 75–96. Babayan, Социально- экономическая и политическая история Армении в DEEE–DEF веках (Socio-Economic and Political History of Armenia in the 13th–14th Centuries), 120. 3I This certainly contradicts Manandian, who in accordance with his Marxist approach, holds that the Mongols, being nomads, in terms of social development stood on a lower level than Armenia and Georgia, and were therefore unable to change the existing formation in all conquered lands; Manandian, Kʿnnakan Tesut’yun Hay Zhoghovrdi Patmutʿyan, 245. 3J Gandzakets ʿi, History of the Armenians, 263. 3M According to the local terminology, ḥākīms and maliks; see Babayan, Социально- экономическая и политическая история Армении в DEEE–DEF веках, 120. 3N Grigor Aknertsʿi, History of the Nation of the Archers, 48; Orbelian, History of Siwnik Province, 420–21. 3O Only from 1256 until 1344 was Armenia a part of Hülegü’s appanage. 5X Babayan, Социально-экономическая и политическая история Армении в DEEE–DEF веках, 121. 53 Ibid., 121; injü land was crown land. It was allotted to the members of the ruling family and was given over into the possession of the princes. The revenue from this land went to meet the expenses of the ruler and his family, and probably for the upkeep of the army. For more details on injü, see Lambton, Landlord and Peasant in Persia, 78; Scherbak, Ранние ??) P;QRPS:PST particularly the Tʿoŕnik branch of the Mamikoneans in Sasun, and the Vaspurakan branch of the Artsʿrunids in Mokk and Ŕshtunik, as well as some mountainous Kurdish and Seljuq families, saved their lands by recognizing the Mongol rulers and entering into direct vassalage. The Artsʿrunid island of Akhtʿamar, the cities of Erzinjan/Erznka and Khlat also preserved their autonomy. The centre of this vilayet was the city of Karin (Erzurum).++

Darughachi-basqaq

In order to mobilize the wealth of all subjugated countries, the Mongol administration aimed to achieve two goals: to provide the army with new recruits and to collect taxes. With the purpose of identifying the ability of the population to pay taxes, the Mongols conducted a census,+, which was implemented by the newly established institution of the darughachis. As was noted above, Chormaqan exercised full power with a wide range of functions: political, military, civil, administrative, as well as 12nancial. According to Armenian historian Kirakos Gandzaketsʿi, he was also a judge or yarghuchi.+- Until the establishment of the Ilkhanate, both governor general and yarghuchi were appointed by the central government; they were in charge of monitoring the administration and justice.+A Local governors or darughachis operated in the conquered towns and districts under the general governor; in our case it was Chormaqan,+E to whom Ögetei Khan issued a decree (yarligh) where the darughachis, and basqaqs should accompany the levy and render assistance to him.+F In some cases, the term darughachi is glossed as ‘an o012cial who is entrusted with a seal.’+G At this point, it is di012cult to de12ne the level of superiority between the functions of

Тюрко-Монгольские языковые связи, FEEE–DEF вв (Early Turco-Mongolian Linguistic Ties, 8–14th c.), 194–95; Doerfer, Türkische und mongolische Elemente, 2:220–25. 55 Babayan, Социально-экономическая и политическая история Армении в DEEE–DEF веках, 121. 56 The identi12cation of natural resources was also the purpose of a census; see Allsen, “Mongols and North Caucasia,” 32. 57 Gandzakets ʿi, History of the Armenians, 275. Yaruga or jaruga in Mongolian means law- suit, litigation, and yarghuchi [yarugha+chi] is judge; see Lessing, Mongolian-English Dictionary, 1037. Cf. Doerfer, Türkische und mongolische Elemente, 4: 64–66. 5I Ratchnevsky, Genghis Khan, his Life and Legacy, 176. 5J )*+, 96, §273–274, 281. 5M Juwaynī/Boyle, 482–83. 5N Ratchnevsky, Genghis Khan, 259, n. 177. !"#$%&"'&( 9: ;<=>:9; ??( general governor and darughachi, since both were entrusted by seals; nevertheless, general governor seems to refer more to a Mongol o012cer whose function, military by nature, was to incorporate newly conquered lands to Mongol Empire, while darughachi relates to the administrative o012ce created by the Mongols to interact with local society in the conquered countries.+H There was no obvious trace of any similar institution functioning in the Mongol homeland, so presumably, this o012ce was intended to deal with non-Mongols in non-Mongol lands. The darughachis were the most active o012cials, exercising great power in the area to which they were assigned; they were responsible for conducting the census and collecting taxes, and were in charge of the courier service and of levying troops from among the local people. Nonetheless, the darughachis were mainly interpreted as oppressors,V in subjugated lands; and it is not unreasonable, due to its linguistic root daru in Mongolian which means to press or suppress, and could be semantically equivalent to Turkish basqaq; the Arabo-Persian version is shaḥna.,* In Armenian sources, it appears as ostikan (governor): ‘They [the Mongols] left evil ostikans in charge of the land’;,+ or as shaḥna: ‘They took Erznkay 12rst of all and left a shaḥna.’,, Since this darughachi controlled the soldiers under his command even though the o012ce was in a transitional stage from a military to a civil one, in the Armenian sources the governors are also referred to as glkhavorʿ (chief- tain): ‘They left a glkhavor called Ghara Bugha.’,- In Orbelian awagatsʿ awag (chief of chiefs) the governor is depicted as a darughachi: ‘Bugha, who was called awagatsʿ awag’;,A and as a pasghag (basqaq): ‘They have gathered in Ti./is to Arghun, who was vazir and pasghaq [overseer] over all lands and was appointed by the great Khan.’,E The o012cial function of darughachi and basqaq was obscure. Some scholars claim that darughachi and basqaq are recorded separately,,F and the rank

5O Buell’s hypothesis on the Khitan origin of darughachis is unproven historically and philo- logically; see Endicott-West, Mongolian Rule in China, 151, n. 55. 6X Cleaves, “Darugha and Gerege,” 237–55; Cf. Doerfer, Türkische und mongolische Elemente, 1: 319–23. 63 Boyle in Juwaynī/Boyle, 44, n. 3, and 105, n. 24; Morgan, The Mongols, 108–9. 65 Gandzakets ʿi, History of the Armenians, 361. 66 Grigor Aknertsʿi, History of the Nation of the Archers, 42. 67 Ibid., 28. 6I Orbelian, History of Siwnik Province, 430. 6J Ibid., 412. 6M Endicott-West, in her study of the o012ce of darughachi (ta-lu-hua-ch’ih) in China, believes that the functions of these two o012ces varied. Endicott-West, Mongolian Rule in China, 18–19, 35. ??? P;QRPS:PST of basqaq was subordinate to that of darughachi.,G However, this view has been based more on which nationalities held this o012ce, and how close they were ethnically to the Mongols, rather than on the basis of function. Some explain that darughachis were Mongol persons or persons regarded as equivalent to the Mongols,,H while basqaqs were local o012cials.-V Others say that the darughachi was often a Uighur and sometimes even a Persian.-* It appears that the darughachi possessed the authority of decision making, while the routine work of administration was delegated to the basqaq, which was, according to some scholars, a new post, introduced at the end of the 1240s, whose function was as local revenue o012cer.-+ According to Chinese sources, the darughachis were appointed by Ögetei Khan. Since the taxation system in China was based on the homestead principle,-, controlling both the local situation and tax collectors by appointing special darughachis was essential for the Mongol central court.

Darughachis and the Census

As far as the sources relate, as early as 1214, the Mongol darughachis were taking control of conquered lands by the use of inventories.-- Later on, Ögetei Khan (r. 1229–41) carried out 12scal reform in China in 1236, which had some e0fect on the situation in Central Asia.-A In Iran in the 1230s, the governors were engaged with the population census and with tax collection, and consequently, Ögetei Khan sent two missions or darughachis to support Chormaqan in governing the vast territories in the Caucasus, Mesopotamia and Iran.-E The 12rst mission was headed by basqaq Chïn-Temür (d. 633/1235–36), a Qara-Khitayan of origin,

6N Ratchnevsky, Genghis Khan, 138–39. 6O Initially, at the time of setting up the Ilkhanate, Oirat Arghun, known as Arghun Aqa and Uigur Körgüz were entrusted with great power, cf. Morgan, The Mongols, 109–10. Lane, “Arghun Aqa,” 459–82. For Chinese examples, see Endicott-West, Mongolian Rule in China, 86–87. 7X Ratchnevsky, Genghis Khan, 138. 73 Grousset, The Empire of the Steppes, 243. 75 Lambton, Landlord and Peasant, 424. Cf. Vásáry, “The Origin of the Institution of Basqaqs,” 201–6. 76 Munkuev, Китайский источник о первых Монгольских ханах (The Chinese Source on the First Mongol Khans), 47. 77 )*+, 85, §252. 7I Schurmann, “Mongolian Tributary Practices of the Thirteenth Century,” 369. 7J Juwaynī/Boyle, 482–83. !"#$%&"'&( 9: ;<=>:9; ??^ who was appointed in 1233 as a governor of Khurasan and Mazandaran.-F The second mission was led by Körgüz (1235–42), a clever and ambitious Uighur Turk. Körgüz, having started to hold a census and to reassess taxes, was called back to Mongolia to answer charges against him by the family of Chïn-Temür.-G Ögetei Khan nominated Arghun Aqa to investigate the case of Edgü-Temür, the son of Chïn-Temür, who wanted his father’s o012ce to fall to him, and not to Körgüz. At court, Arghun Aqa assisted and supported Körgüz, and thus Khurasan and Iraq were entrusted to Körgüz.-H The Mongol census, taken throughout the 1250s in all parts of the Mongol Empire, was a decisive moment for the regulation of the economy in the conquered lands because it was designed to control and mobilize human and non- human resources. It had importance for Armenia as well, because it signi12ed a new stage in the history of the Armenian people, the process of their a012lia- tion to the Mongol Empire. Before the actual censuses, the Georgio-Armenian princes were obliged to pay tax and to provide all necessities, including horses, guides, dwellings, food and carts to the Mongol governors, who held the paiza (tablet of authority).AV However, among the censuses taken in Armenia in the 1250s,A* the most important census was that of 1254, which was intended to put Armenia under the central registration of the Mongol Empire. Although the Armenian sources do not give any reliable data about the size of the population in that period, some details, based on the Mongol census, were preserved in the sources. Thus, according to Manandian, the number of peasants during the reign of King David _ZZ Ulu (1247–70), the son of Lasha, was about one million, and the whole population of Georgia, including Armenia, was four to 12ve million.A+ In terms of the sex or age of people entered in the Mongol registers, the youngest was elevenA, or 12fteen,A- and the oldest was sixty years old.AA The registers initially covered the whole of the male

7M Ibid., 482–83. 7N Boyle, The Cambridge History of Iran, 336–37. 7O Juwaynī/Boyle, 505–6. IX Manandian, K’nnakan Tesut’yun Hay Zhoghovrdi Patmut’yan, 246. I3 According to Vardan Arevelts‘i, there was a Mongol census-taking in Armenia in 1243/44; Arevelts‘i, Chronicle, 147–48. This information seems plausible, since there was a census taken in Iran in 1240; see Allsen, Mongol Imperialism, 130. I5 Manandian, Kʿnnakan Tesutʿyun Hay Zhoghovrdi Patmutʿyan, 251. I6 Gandzakets‘i, History of the Armenians, 362. I7 Grigor Aknerts‘i, History of the Tatars/History of the Nation of the Archers, 37. II Gandzakets‘i, History of the Armenians, 362; Grigor Aknerts‘i, History of the Tatars/History of the Nation of the Archers, 37. ??` P;QRPS:PST population and excluded women,AE but in later periods taxes were levied on women and children as well.AF The entry of women and children into the Mongol rolls can be explained by the fact that the registration of the whole population was carried out separately from that of the male population, in other words, separately from the utilization of manpower for military service. An anonymous Georgian chronicler, relating the census of 1254, says that Arghun had to calculate the number of cavalry men and soldiers who could participate in raids along with Mongol noyans, registering one out of every ten peasants who had good land.AG The other anonymous Georgian source states that the Mongols were taking one man from every ten peasant houses for military service.AH It is mentioned that Northern or Zak‘arid Armenia was obliged to provide 30,000 horsemen. From this account, it is estimated that the population of Zak‘arid Armenia was around 270,000 peasant houses.EV During the process of conducting the census, in the key cities, the Mongol darughachis set up so-called dīwāns, the chancery where davtars, or registry books, were kept.E*

Arghun Aqa-darughachiE+

The darughachi’s o012ce at the beginning of Mongol rule was not yet e0fective in terms of 12scal authority.E, After Ögetei Khan, Güyük Khan (r. 1246–48), taking control of the great kingdom, decided to bring the taxation system into order. He sent tax collectors to his troops in the various lands that had been subdued. In Armenia, they took “one tenth of all property of the troops as well as taxes from the districts and kingdoms conquered by them; from the Iranians, Tajiks,

IJ Gandzakets‘i, History of the Armenians, 362. IM Sebastats‘i, “Annals of Anonymous Sebastats‘i,” 141; Vardan Arevelts‘i, Chronicle, 148. IN Manandian, Kʿnnakan Tesutʿyun Hay Zhoghovrdi Patmutʿyan, 251. IO Anonymous Georgian source quoted in Manandian, Kʿnnakan Tesutʿyun Hay Zhoghovrdi Patmutʿyan, 250–51. For common 12gures in Mongol raids, see Amitai, “Mongol Raids into Palestine (#a 1260 and 1300),” 236–55. JX Grekov, Очерки истории TTTU, период феодализма (Sketches of the History of the bcc', Period of Feudalism), 682. J3 Orbelian, History of Siwnik Province, 411, 427. J5 During the course of events in Ilkhanid history there were two Arghuns; one was a darughachi, whom I call Amir Arghun (d. 1275), or Arghun Aqa; the other one was Ilkhan Arghun, who is referred to here as Arghun Khan (r. 1284–91). J6 Lambton, “Mongol Fiscal Administration in Persia,” 80. !"#$%&"'&( 9: ;<=>:9; ??d

Turks, Armenians, Georgians, Albanians, and from all people under them.”E- Apparently, this was the poll tax imposed in Armenia prior to the establishment of the Ilkhanate. Later, Möngke Khan (r. 1251–58) continued this policy. The person who facilitated the development of a systematic administration was Arghun Aqa.EA Tracing the origin of Arghun Aqa, we 12nd that he belonged to the Oirat tribe.EE From Juwaynī’s biography of him, one can see that Arghun had a prosperous career at the Mongol court, being recognized by four successive Khans. After mastering the Uighur script, though young, he went to the Court of Ögetei Khan and was appointed a bitikchi (scribe/secretary). Showing promise in the chancery o012ce in 1243/4, Arghun was sent to Khurasan, where “he read the yarlighs [royal decree] and brought the a0fairs of the country in order.”EF Upon arriving in Tabriz, he restored to order the administrative a0fairs of that region which had been disturbed by the proximity of the great generals such as Chormaqan, Baiju and others, who regarded this territory as their own property. This was a starting point to reconsider a status of the Caucasus as a part of Great Khan’s territory. After the death of Ögetei Khan, which occurred in 1241, the various princes began to impose their rights, bringing disorder to the Empire’s a0fairs. Arghun was therefore entrusted to collect every paiza and yarligh issued by them. Güyük showed Arghun favour and con12rmed him in the administration of all the territories he held in the East, giving him a tiger-headed paiza and a yarligh, and transferred the a0fairs of all maliks and ministers to him.EG In 1251, in the quriltai (Great Assembly) summoned on the occasion of the succession of Möngke Khan to the throne, Arghun gave an oral report on the chaotic condition of the Empire’s 12nances and of the de12cit in tax payments, a situation caused by the constant succession of unlawful assignments and the stream of harsh elchis (envoys) and tax collectors. He acknowledged and admitted the shortcomings arising from the disorderly state of a0fairs.EH

J7 Gandzakets ʿi, History of the Armenians, 312. JI Lane, “Arghun Aqa: Mongol Bureaucrat,” 478–81. JJ According to Juwaynī, Arghun’s father Taichu was the commander of a thousand men; see Juwaynī/Boyle, 505. Rashīd al-Dīn states that Arghun’s father, whom he does not name, in time of famine, sold his son in exchange for a leg of beef to a certain Kadan of Jalayir; see Rashīd al-Dīn, Сборник летописей, 1, part 1: 95. JM Juwaynī/Boyle, 506–7. JN Ibid., 509. JO Ibid., 514–16. ??e P;QRPS:PST

When Arghun arrived in Khurasan, in August–September 1253, he nominated Amirs and secretaries to organize the census and levy taxes in Iran. Then he set out for the Court of Batu in Russia. On the way, he travelled to Derbent through Georgia, Armenia, Arran and Azerbaijan, and completed the work of holding a census and imposing and assessing taxes.FV The name of Arghun appears in almost all contemporary Armenian sources; he was seen as such a central person due to his association with the census in Armenia in 1254 as “vazir and basqaq (pasghagh), who was instituted by the Great Khan to every land; the commander of all; the master of royal taxation and great divan.”F* According to Sebastatsʿi and Vardan Areveltsʿi, in 703 (1254), Möngke Khan issued a decree to conduct the census in the conquered countries and to levy taxes not only on men but also on women and children.F+ The assessment was delegated to Arghun. Mkhit’ar Ayrivanetsʿi recorded that Arghun conducted a census from east to west.F, Stepʿannos Episkopos says that in 703 (1254), “Arghun [al-Dīn], the destroyer, came, took a census of all lands, and levied tax on every head.”F-

Taxation

In 1259–60, Arghun was again in Georgia and Armenia, introducing a tax called qubchur.FA Among the taxes and levies that the Mongols introduced in Armenia was a typical Mongolian tribute called qalan and a levy called qubchur. Later, under Hülegü, a tax called taghar was brought into Armenia.FE As mentioned above, the census mostly covered the male population, although there were cases that included women, children and the elderly.FF This registration obliged the Georgian and Armenian princes, willingly or unwillingly, to give tribute to the Mongols in order to provide cavalry troops

MX Ibid., 519–21. M3 Orbelian, History of Siwnik Province, 412–13. M5 Sebastatsʿi, “Annals of Anonymous Sebastats‘i,” 141; Areveltsʿi, Chronicle, 148. M6 Ayrivanets ʿi, History of the Armenians, 67. M7 Episkopos, “Annals of Step‘annos Episkopos,” 42. MI On qubchur, tamgha, qalan and taghar, see Lambton, “Mongol Fiscal Administration in Persia,” 90–96. MJ Qupchur is khap‘chur in Gandzakets‘i, History of the Armenians, 374. MM Sebastats‘i, “Annals of Anonymous Sebastats‘i,” 141; Arevelts‘i, Chronicle, 148. Supposedly, this act of imposing tax on women and children in Armenia was questioned at the central court by Arghun Aqa; see Allsen, Mongol Imperialism, 167. !"#$%&"'&( 9: ;<=>:9; ??@ for Mongol military action.FG This service was called qalan or khalan in the Armenian sources.FH Qalan in accounts by Grigor Aknerts‘i and Rashīd al-Dīn was rendered as a tribute of military service or a payment for exemption from Mongol military service.GV The root qala in Mongolian perhaps meant and today means ‘replacement,’ and is usually used for the recruitment of soldiers.G* Therefore, qalan was the tribute paid to recruit soldiers on an occasional basis as was practiced everywhere, including Georgia, Armenia and Iran. As was noted in Armenian sources, two men out of every ten were obliged to serve in the Mongol army.G+ According to this obligation, the Georgio-Armenian army participated in Hülegü’s conquests of Baghdad in 1258 and of Syria in 1260. Qalan may correspond to the Armenian tribute known as hetsel, the cavalry vassal obligations.G, Another Mongolian tax introduced in Armenia was qubchur, originally a herd tax.G- Qubchur in Mongolian meant a tribute on ./ocks and herds, and the root γubči along with other meanings refers to taxation in modern Mongolian.GA Originally, Ögetei Khan had decreed that each military tuman must contribute one three-year-old sheep from every herd to the royal court and one sheep out of every hundred to the poor and needy every year.GE Therefore, initially, the rate of qubchur in all conquered lands was one per cent.GF This tax grew more

MN Grigor Aknerts‘i, History of the Tatars/History of the Nation of the Archers, 26. MO Ibid., 26, 35. NX However, according to Petrushevsky, qalan and qubchur in Iran were used instead of kharāj; Petrushevsky, “The Socio-Economic Condition of Iran,” 532. In 1254, the thirteenth- century Persian poet Pūr-i Bahā, the panegyrist of the Juwaynī family, wrote a ‘Mongol’ ode that is an account of historical events that occurred on the frontiers of Khurasan with a vast number of Mongol technical terms, including taxes in Iran under Mongol dominion; see Minorsky, “Pūr-i Bahā’s Mongol Ode,” 274–305. N3 Tsevel, 647; Lessing, Mongolian-English Dictionary, 916. Qalan is also associated with another Mongol tribute called alban; Smith, “Mongol Campaign Rations,” 48; Lambton, Continuity and Change, 200. N5 Grigor Aknerts‘i, History of the Tatars/History of the Nation of the Archers, 49; Grekov, 682. N6 The term hetsel is found in one of the undated inscriptions of Ani; Kostaneants, 221; Babayan, Социально-экономическая и политическая история Армении в DEEE–DEF веках, 251. N7 For qubchur or alba qubchur, see Schurmann, “Mongolian Tributary Practices,” 325–30; Doerfer, Türkische und mongolische Elemente, Z: 387–91; Lambton, Continuity and Change, 199. NI Tsevel, Mongol Khelnii Tovch Tailbar Tol’, 202; Lessing, Mongolian-English Dictionary, 363. NJ )*+, §279, 99. NM Petrushevsky, “The Socio-Economic Condition of Iran,” 530. ??U P;QRPS:PST complicated during the Ilkhanid period, when it became a levy or additional tax imposed on the conquered population. Later, it stood as a general term for tax, in some cases meaning poll tax.GG Qubchur was collected many times a year if the original levy did not provide su012cient funds—even twenty or thirty times, if we believe the Persian source.GH It was levied in coins, in accordance with the wealth of the person and his ability to pay.HV In the case of Armenia, it was sixty spitak (silver coins) per head.H* It is worth noting that revenue collection in the form of money was preferred within the Mongol Empire, because it was economically feasible to ship cash over great distances and convert it into some other type of goods or services.H+ Taghar was a levy on food and provisions. In west Asian sources, it also meant an extraordinary levy to sustain the imperial army while on campaign.H, This tax was levied widely in the territories of the Mongol Empire, where the economy was based on agriculture. The highest point of this tax was reached in China during the period of the Yuan Dynasty, where the taghar as a levy on crops was taken from each male head and from each piece of land.H- The following describes taghar in Armenia:

Hulegu commanded that the tax called t‘aghar be collected from each individual listed in the royal register. From such they demanded one hundred litrs [pounds] of grain, 12fty litrs of wine, two litrs of rice and husks, three sacks, two cords, one spitak [silver coin], one arrow, let alone the other bribes; and one in every twenty animals plus twenty spitaks.HA

Juwaynī relates that the dispatch of taghars of ./our for provisioning the army covered an area from “Armenia to Yezd [Yazd] and from the land of Kurds to Jurjān,” which was procured by the elchis (envoys) who were in charge of their

NN Allsen, Mongol Imperialism, 153; Lambton, Continuity and Change, 199. NO Rashīd al-Dīn, Sbornik Letopisei, 3:248. OX Juwaynī/Boyle, 517. A hexagon-shaped coinage for the imperial tax qupchur was minted in Tabriz; Kolbas, The Mongols in Iran, 196. O3 Grigor Aknerts‘i, History of the Tatars/History of the Nation of the Archers, 37. O5 Allsen, Mongol Imperialism, 172. O6 Ibid., 186. Taγar in Mongolian is a sack or bag, used mostly for provisions; Doerfer, Türkische und mongolische Elemente, 2:512–19; Tsevel, Mongol Khelnii Tovch Tailbar Tol’, 703; Lessing, Mongolian-English Dictionary, 764. O7 Yüan Shih, chapter 93, in Dalai, Mongol Ulsyn Tuukh, 219. OI Gandzakets‘i, History of the Armenians, 374–75. 1 litr of wine is equal to 2/5 litre, 1 litr of rice is 0.5 kg; Manandian, Kʿnnakan Tesutʿyun Hay Zhoghovrdi Patmutʿyan, 277. !"#$%&"'&( 9: ;<=>:9; ??[ transportation.HE In 1256, Armenians played a part in the Mongol war with the Isma’ilis by providing most of the provisions for the Mongol armies.HF If all the food levies of ulagh (ulaγ-a) and taghar were imposed with no exception, it seems that Greater Armenia faced an economic crisis in 1256. The repercussions of these taxes brought famine to Armenia.HG Besides these three major taxes, there were other taxes. One was called the tamgha, a toll on commercial goods as well as a license for commercial acts.HH The collector of this tax was also called the tamghachi, as he put a tamγa (seal) on goods for sale, similar to a customs stamp.*VV In the Armenian inscriptions of the city of Ani, dated 1270 (719 Arm.), the tamgha was used together with the baj, the local Armenian term for the custom or duty tax.*V* The precise rate of the tamgha is not known. It was levied at varying rates and a substantial part of the revenue of the Empire was derived from it.*V+ Initially, it was levied at ten per cent of the value of each commercial transaction.*V, In 1257, the Genoese frequently traded through Kars, involving Zak‘arid Armenia in the Black Sea trade.*V- Caravan routes passing through Trebizond, and the cities of Erzurum, Berkri, and Khoi in Greater Armenia, were connected with the Italian trading republics.*VA It seems this was a very pro12table business for the Mongols if we consider that Hülegü’s conquests were 12nanced under the tamgha tax.*VE Trade was usually centralized near the royal court, and therefore the tamgha

OJ Juwaynī/Boyle, 621. For the provisioning of large armies campaigning over extraordinary distances, see Smith, “Mongol Campaign Rations,” 223–28. OM Rashīd al-Dīn, Сборник летописей, 3:30. ON Sebastats‘i, “Annals of Anonymous Sebastats‘i,” 142. OO For the tamgha, see Doerfer, Türkische und mongolische Elemente, 2:564–65. 3XX Gandzakets‘i, History of the Armenians, 277. 3X3 Kostaneants‘, The Annals of the Inscription, 221;. The Armenian baj may derive from the Persian term bāj, a tribute on trade ships; Rashīd al-Dīn, Сборник летописей, 3:25. 3X5 Lambton, “Mongol Fiscal Administration,” 84 (citing Nuzhat al-qulūb). Qazvīnī mentioned twenty cities in Iran where the tamgha was levied; Le Strange, Palestine under the Moslems, 113. 3X6 Petrushevsky, “The Socio-Economic Condition of Iran,” 532. 3X7 Manandian, The Trade and Cities of Armenia in Relation to Ancient World Trade, 187. Armenia had had trade relations with Genoa since the beginning of the thirteenth century; Alishan, Sisakan: Teghagrut‘iwn Siwneats‘ Ashkharhi, 437–38. 3XI Manandian, Kʿnnakan Tesutʿyun Hay Zhoghovrdi Patmutʿyan, 260–63. 3XJ Kolbas, The Mongols in Iran, 201. Hülegü’s Syrian campaign was subsidized by Anatolia as well, where the annual tribute reached the sum of 20 tumans in cash, 3,000 gold bars, 1,000 horses and mules, and 500 rugs and satin textiles; Āqsarā’ī, Musāmarat al-akhbār wa musāyarat al-akhyār, 62–63, 73. ?^) P;QRPS:PST maintained the Ilkhanid court, army and civil administration.*VF However, the burden of the tamgha caused the Il-Khan Ghazan (r. 1295–1304) to cut this tax by half, from ten to 12ve percent of the value of each transaction in towns.*VG The basic tax terms that were used and introduced by the Mongols, namely qubchur, qalan, taghar and tamgha, had di0ferent meanings in di0ferent places. Thus, qubchur in Iran meant also a poll tax levied on the subject population.*VH On every occasion that a census was held, the qubchur was reassessed. Being 12xed at the rate of 70 dīnārs per 10 persons (7 dīnārs per person) in 1258, based on a new census, the qubchur ranged in scale from 500 dīnārs for a wealthy person to one dīnār for the poor.**V Thereafter, the imposition of the census and poll tax qubchur seems to have become a regular procedure. In 1259–60, Arghun Aqa was in Georgia and Armenia, not only introducing a tax, but also conducting a military operation against rebel princes.*** In this regard, Kirakos Gandzaketsʿi says that ostikan Arghun set o0f in pursuit the Georgian King David, who ./ed to Abkhazia from the heavy burden of taxes along with other impoverished princes of the provinces; however, Arghun failed to catch the king, and therefore mercilessly ransacked many Georgian provinces.**+ According to Mkhit’ar Ayrivanetsʿi, Arghun destroyed Georgia in 1261.**, Arghun’s activity was backed up by the central court of the Mongols. He was promoted, and under the Ilkhans Hülegü (r. 1256–65) and Abaqa (r. 1265–81) Arghun held the o012ce of Tax-Farmer General (muqāṭi‛i mamālik). Moreover, he participated in the war with Baraq, the ruler of Transoxiana in 1270.**- According to Stepʿannos Episkopos, in 1273, Arghun once more conducted a census in Georgia and Armenia.**A After the death of Arghun Aqa in 1275, the o012ce of darughachi in Armenia was carried on by the Jalayirid Bugha, about whom Frik, the Armenian poet of the thirteenth century, wrote a poem About Arghun Khan and Bugha, where he described the Mongol darughachi as a cause of great su0ferings and hardships throughout Armenia.**E

3XM Boyle, The Cambridge History of Iran, 508. 3XN Ibid., 532. 3XO Juwaynī/Boyle, 551. 33X Ibid., 524. 333 The rebellion is known as the second Georgian revolt of 1259–61, in which the Armenian princes took part; for some of them, the new tax meant that they had to mortgage their estates in order to pay taxes, Gandzaketsʿi, History of the Armenians, 389. The 12rst rebellion was in 1249. 335 Ibid., 389–90. 336 Ayrivanetsʿi, History of the Armenians, 68. 337 Boyle, The Cambridge History of Iran, 340. 33I Episkopos, “Annals of Step‘annos Episkopos,” 44. 33J Frik, Divan, 38–43. !"#$%&"'&( 9: ;<=>:9; ?^(

Conclusions

The establishment of the institution of darughachi had far-reaching consequences. The main duty of this o012ce was to collect taxes on a formal basis in all conquered lands of the empire. In order to achieve this goal, the darughachis started to conduct a census of the population, and a registry of land, livestock and possessions. This was the starting point for the permanent utilization of the conquered territories. Traditional nomadic models of exploitation applied to sedentary societies evolved into speci12c 12scal principles that di0fered in each ulus. The introduction of a Mongol 12scal administration in./uenced the political status of Armenia, resulting in its evolution from a vilayet into an administrative province of the Ilkhanate. The general scheme of the Mongol administration in Zak’arid Armenia before its annexation to the Ilkhanate (1236–56/58) shows that the region was under a double vassalage of Great Khan and Georgian King or Princes (nakharars, who were heads of tumans). In Greater Armenia (1248–56/58) it was the Great Khan and the Governor General successively who commissioned the darughachi. In the successor states (1256/58–1344) it was still the Great Khan but in the name of a blood prince (ruler of ulus), who commissioned the darughachi or basqaq. The actual number of darughachis in the Caucasus is unknown. In some areas of Mongol Empire, the number of o012cials was high. For example, in 1231, when the Korean capital Kaesong had been taken by the Mongols, 72 darughachis were placed in the Mongol protectorate to run it.**F Therefore, it is worth noting that a whole institution of darughachis was set up to maintain resident Mongolian military forces, since the military meaning was never lost, but was simply implemented on a more constant and stable basis in conjunction with a civil o012ce for the far more e012cient and larger scale use of population and natural resources. Darughachis played crucial role in the adaptation of local rulers to Mongol command. It was impossible to rule over the biggest ever land-empire without some degree of local support in the conquered lands. The paradox is that the Mongol Empire, which emerged primarily as a great military power with a re12ned military organization, was quick to establish the civil institution of the darughachis. The darughachis exercised both direct powers, for example where taxation was concerned, and also indirect powers, whereby they shared authority with local princes and governors. This alone, apart from other factors, can explain the often willing servitude of the Georgian and Armenian princes to the Mongol Khans.

33M Grousset, The Empire of the Steppes, 259. ?^? P;QRPS:PST

It is important to note that the application of the darughachi system in Georgia and Armenia was set up right from the beginning of the conquest, based on the Mongols’ prior experience of this institution in China and Transoxiana. Armenia, as a part of the region, was incorporated into the Mongol Empire through the governor system, and Armenia remained under Mongol lordship for more than a hundred years.

Bibliography

Aigle, D. “Persia under Mongol Domination. The E0fectiveness and Failings of a Dual Administrative System.” Bulletin d’ études Orientales 57 (2006): 65–78. Alishan, Gh. Sisakan: Teghagrut‘iwn Siwneats‘ Ashkharhi, Venice: St. Ghazar, 1893. Allsen, Thomas T. “Mongol Census Taking in Rus’, 1245–1275.” Harvard Ukrainian Studies 5/1 (1981): 32–53. ———. “Guard and Government in the Reign of the Grand Qan Möngke, 1251–59.” Harvard Journal of Asiatic Studies 46/2 (1986): 495–521. ———. Mongol Imperialism: The Policies of the Grand Qan Möngke in China, Russia and the Islamic Lands, 1251–1259. Berkeley: University of California Press, 1987. ———. “Mongols and North Caucasia.” Archivum Eurasiae Medii Aevi 7 (1987–1991): 5–40. Amitai, Reuven. “Mongol Raids into Palestine (#a 1260 and 1300).” Journal of the Royal Asiatic Society 2 (1987): 236–55. Āqsarā’ī, Karīm al-Dīn (ca. 723/1323). Musāmarat al-akhbār wa musāyarat al-akhyār, edited by Osman Turan, Ankara: Türk Tarih Kurumu Basımevi, 1944. Arevelts‘i, Vardan. Universal History, completed in 1267. Edited by G. B. T‘osunyan. Erevan: Erevan University Press, 2001. ———. Chronicle. Introduction by R. W. Thomson. New York: Caravan Books, 1991. Aubin, J. “Émirs mongols et vizirs persans dans les remous de l’ acculturation,” Studia Iranica Cahier 15. Paris : Association pour l’avancement des etudes iraniennes, 1995. Ayrivanets‘i, Mkhit‘ar. History of the Armenians. Edited by M. Emin. Moscow: Lazar Institute Press, 1860. Babayan, L. Социально- есономическая и политическая история Армении в DEEE– DEF веках. (Socio-Economic and Political History of Armenia in the 13th–14th Centuries) Moscow: Nauka, 1969. ———. “Armenia in the 12rst Period of Mongol Dominion.” In The History of the Armenian People, edited by Ts. Agayan et al., 597–627. Erevan: Academy of Sciences Press, ZZZ, 1976. !"#$%&"'&( 9: ;<=>:9; ?^^

Barthold, V. Сочинения, ZZ (Works, v.2). Moscow: Nauka, 1964. ———. Сочинения, V. (Works, v.5). Moscow: Nauka, 1968. ———. Turkestan Down to the Mongol Invasion. London: E. J. W. Gibb Memorial Trust, 1958. Reprint, 1977. Bazarsuren, J. “Is it Tamachin or Tamghachin?,” Historia Mongolarum 34 (2013): 23–31. Biran, M. Chinggis Khan. Oxford: Oneworld, 2007. Boyle, J. A. ed. The Cambridge History of Iran. vol. 5. The Saljuq and Mongol Periods. Cambridge: Cambridge University Press, 1968. ———, The Mongol World Empire, 1206–1370. London: Variorum Reprints, 1977. Buell, P. “Sino-Khitan Administration in Mongol Bukhara.” Journal of Asian History 13/2 (1979): 121–51. Cleaves, F. “A Chancellery Practice of the Mongols in the 13th and 14th Centuries.” Harvard Journal of Asiatic Studies 14/3–4 (1951): 493–526. ———. “The Mongolian Documents in the Musée de Téhéran.” Harvard Journal of Asiatic Studies 16/1–2 (1953): 1–107. ———. “Darugha and Gerege.” Harvard Journal of Asiatic Studies 16/1–2 (1953): 237–59. Dalai, Ch. and Ishdorj, Ts., eds. Mongol Ulsyn Tuukh (The History of Mongolia, the 12–14th Century). 5 vols. Vol. ZZ. Ulaanbaatar: Admon, 2003. Doerfer, G. Türkische und mongolische Elemente im Neupersischen, 4 vols. Wiesbaden: Franz Steiner Verlag GmbH, 1963–75. Endicott-West, E. “Merchant Associations in Yüan China: The Ortog.” Asia Major 3rd series 2:2 (1989): 127–54. ———. Mongolian Rule in China, Local Administration in the Yüan Dynasty. Cambridge, Mass.: Harvard University Press, 1989. Episkopos, Step‘annos. “Annals of Step‘annos Episkopos.” In Minor Chronicles of the 13th–18th centuries, 1: 32–64, edited by V. A. Hakobyan, Erevan: Academy of Sciences Press, 1951. ———. “Annals of Step‘annos Episkopos.” In Армянские источники о Монголах (Armenian Sources on the Mongols), edited by A. G. Galstyan, 33–43. Moscow: Oriental Literature Press, 1962. Farquhar, D. M. The Government of China under Mongolian Rule. Stuttgart: Franz Steiner, 1990. Frik, (./. thirteenth century). Poems. Erevan: Academy of Sciences Press, 1941. ———. Divan. Edited by T. Melik‘ Mushkambarean. New York: Melkonian Fund, 1952. Gandzakets‘i, Kirakos. History of the Armenians. Edited by Melik‘-Ohanjanyan. Erevan: Academy of Sciences Press, 1961. Grekov Grekov, B., ed. Очерки истории TTTU, Период феодализма (Sketches of the History of the bcc', Period of Feudalism). 2 vols. Vol. ZZ. Moscow: Academy of Sciences Press, 1953. ?^` P;QRPS:PST

Grigor Aknerts‘i/Grigor of Akanc‘ (1250–1335). History of the Tatars/History of the Nation of the Archers. Jerusalem: St. Jacob Press, 1974. ———. History of the Nation of the Archers (1271/73). Edited and translated by R. P. Frye and R. N. Blake. Cambridge, Mass.: Harvard University Press, 1954. Grousset, R. The Empire of the Steppes. Translated by Naomi Walford. New Brunswick, Lu: Rutgers University Press, 1970. Hakobyan, V. A. ed. Minor Chronicles of the 13th–18th centuries. 2 vols. Erevan: Academy of Sciences Press, 1951 and 1956. Juwaynī, ʿAṭā Malik. Ta’rīkh-i Jahāngushā. Edited by Mīrzā Muḥammad Qazwīnī. 3 vols. Tehran: Pik Irān, 1962. ———. Tasliyat al-Ikhvān. Edited by ‘Abbās Māhyār. Tehran: Ābād, 1982. ———. The History of the World-Conqueror. Translated by John Andrew Boyle. Reprint, Manchester: Manchester University Press, 1997. Kolbas, J. The Mongols in Iran: Chingiz Khan to Uljaitu 1220–1309. London and New York: Routledge, 2006. Kostaneants‘, K., ed. The Annals of the Inscription: Collection of Armenian Inscriptions. St. Petersburg: Imperial Academy of Sciences Press, 1913. Kuribayashi, H. (comp.). “Word-Index to the Secret History of the Mongols with Chinese Transcriptions and Glosses.” ]*^_` Monograph Series 33 (2009): 531–35. http://www.cneas.tohoku.ac.jp/sta0f/hkuri/articles/A46Sosho33.pdf (last visited 6 June 2015). Lambton, A. K. Landlord and Peasant in Persia. London: Oxford University Press, 1953. ———. State and Government in Medieval Islam. London: Oxford University Press, 1981. ———. “Mongol Fiscal Administration in Persia.” Studia Islamica 64 (1986): 79–99; 65 (1987): 97–123. ———. Continuity and Change in Medieval Persia: Aspects of Administrative, Economic and Social History, 11th–14th Century. Albany: Persian Heritage Foundation, 1988. Lane, George. “Arghun Aqa: Mongol Bureaucrat,” Iranian Studies 32/4 (1999): 459–82. Lessing, F., ed. Mongolian-English Dictionary. Bloomington, ZL: The Mongolian Society, 1973. May, Timothy. The Mongol Art of War. Barnsley: Pen and Sword. 2007. Manandian, H. Kʿnnakan Tesutʿyun Hay Zhoghovrdi Patmutʿyan (Critical View of the History of the Armenian People). Vol. ZZZ. Erevan: Academy of Sciences Press, 1952. ———. The Trade and Cities of Armenia in Relation to Ancient World Trade, translated by N. G. Garsoïan, Lisbon: Livraria Bertrand, 1965. Matsuda, K. “On Tan-ma-chi in China during the Mongol Period.” In Proceedings of the conference Chinggis Khan and Globalization, 307–14. Ulaanbaatar, 2014. !"#$%&"'&( 9: ;<=>:9; ?^d

Minorsky, V., “Pūr-i Bahā’s Mongol Ode,” Iranica. Twenty Articles. Tehran: University Press, 1964: 274–91. Mongolyn Nuuts Tovchoo (The Secret History of the Mongols). Translated into modern Mongolian by Ts. Damdinsuren. Ulaanbaatar: State Press House, 1957. ———. The Secret History of the Mongols. Translated by F. W. Cleaves. Cambridge, Mass.: Harvard University Press, 1982. ———. The Secret History of the Mongols: The Life and Times of Chinggis Khan. Translated by Onon Urgunge. Richmond: Curzon, 2001. ———. The Secret History of the Mongols. A Mongolian Epic Chronicle of the Thirteenth Century. English translation with a historical and philological commentary by I. de Rachewiltz, 2 vols. Leiden: Brill, 2004. ———. [KL%] classical text in Cirillic Mongolian, edited by Tserensodnom D., Ulaanbaatar: Monkhiin Useg, 2004. Morgan, D. The Mongols. Oxford: Blackwell, 1990. ———. “Mongol or Persian?: The Government of Il-Khanid Iran,” Harvard Middle- Eastern and Islamic Review 3 (1996): 62–76. Munkuev, N. Китайский источник о первых Монгольских ханах (The Chinese Source on the First Mongol Khans). Moscow: Nauka, 1965. Mustawfī Qazvīnī, Ḥamd Allāh. Tārīkh-i guzīda or “Selected History” (730/1329–30). Edited by E. G. Browne, 2 vols. Leiden: Brill, 1910–13. ———. The geographical part of Nuzhat al-Qulūb. Edited and translated by G. M. Le Strange, 2 vols. Leiden: Brill, 1915, 1919. ———. Nuzhat al-qulūb (740/1340). Edited by Mohammad Dabirsiyaqi. Tehran: Tahoori, 1958. Orbelian, Step‘annos. History of Siwnik Province. Edited by K. Shahnazareants‘. Paris: 1860; Ti./is: 1910. Ostrowski, D. “The ‘tamma’ and the Dual-Administrative Structure of the Mongol empire.” Bulletin of the School of Oriental and African Studies 61/2 (1998): 262–77. Petrushevsky, L. “К истории сельского поселения и сельской общины в Иране в ZZZ–_ вв.”(On the history of rural settlement and rural community in Iran in the 13–15th centuries) Учёные записки Института Востоковедения 16 (1958): 31–51. ———. Земледелие и аграрные отношения в Иране DEEE–DEF веков (Agriculture and Agrarian Relations in Iran in the 13–14th centuries). Moscow-Leningrad: Academy of Sciences Press, 1960. ———. “The Socio-Economic Condition of Iran under the Īl-Khāns.” In The Cambridge History of Iran. vol. 5. The Saljuq and Mongol Periods, edited by J. A. Boyle, 483–537. Cambridge: Cambridge University Press, 1968. Ratchnevsky, P. Genghis Khan, His Life and Legacy. Edited and translated by T. N. Haining. Oxford: Blackwell, 1991. ?^e P;QRPS:PST

Rashīd al-Dīn. Сборник летописей (Compenduim of Chronicles). Edited by A. A. Semyonov, I. P. Petrushevskii, and A. A. Romaskevich; translated by L. A. Khetagurov, O. I. Smirnova, Y. P. Verkhovsky, and A. K. Arends. 3 vols. Moscow and Leningrad: Academy of Sciences Press, 1946–60. ———. Jāmiʿ al-tawārīkh. Edited by B. Karīmī. Tehran: Iqbāl, 1954. ———. Jami‘u’t-tawarikh [sic] Compendium of Chronicles. Translated by Wheeler M. Thackston. 3 vols. Cambridge, K#: Department of Near Eastern Languages and Civilizations, Harvard University, 1998–99. Scherbak, A. Ранние Тюрко-Монгольские языковые связи FEEE–DEF вв. (Early Turco- Mongol Linguistic Connections, 8th–14th Century). St. Petersburg: ZZ '#L, 1997. Schurmann, H. “Mongolian Tributary Practices of the Thirteenth Century.” Harvard Journal of Asiatic Studies 19/3–4 (1956): 304–89. Sebastats‘i. “Annals of Anonymous Sebastats‘i.” In Minor Chronicles of the 13th–18th centuries, vol. 2, edited by V. A. Hakobyan, 115–51. Erevan: Academy of Sciences Press, 1956. ———. “Annals of Anonymous Sebastats‘i.” In Армянские источники о Монголах (Armenian Sources on the Mongols), edited by Galstyan, 23–33. Moscow,1962. Smith, J. M. “Mongol Campaign Rations: Milk, Marmots, and Blood?.” Journal of Turkish Studies 8 (1984): 223–28. Le Strange, G., trans. Palestine under the Moslems, London: Alexander P. Watt, 1890. Tsevel, Ya., Mongol Khelnii Tovch Tailbar Tol’ (Explanatory Dictionary of Mongolian Language). Edited by Ts. Shagdarsuren. Ulaanbaatar: Monkhiin Useg, 2013. Vásáry, István. “The Golden Horde Term Daruġa and its Survival in Russia.” Acta Orientalia Academiae Scientiarum Hungaricae 30/2 (1976): 187–97. ———. “The Origin of the Institution of Basqaqs.” Acta Orientalia Academiae Scientiarum Hungaricae 32/2 (1978): 201–6. Yokkaichi, Yasuhiro. “The Role of jarγuči in the Structure of the Distribution System under the Mongol Empire.” Bull. of the Japan Society for the Promotion of Science 4 (2006): 91–114. !"#$%&' (( The Phoenix Mosque

George Lane*

Hangzhou’s History

The Phoenix Mosque, also known as The Phoenix Temple, is an historical testament to the rise of the political and economic fortunes of the Persians in late thirteenth–early fourteenth-century east Asia. Built in 1281 in Hangzhou, it served the growing Persian Muslim community who had moved to the coastal region as part of the Chinggisid absorption of the former Song lands. In its history, two episodes had drastically changed the political and cultural fortunes of the city. Sometime after 1127, the Southern Song made it their temporary capital (Xingzai) following their expulsion from the north by the Jurchen (Jin) dynasty, and in 1276, the city fell to the Chinggisids who, under the Mongol general Bayan Noyan, proceeded peacefully to occupy the city and open its gates to an in)*ux of westerners:

In the Yuan period the Hui-hui (from Samarqand) spread over the whole of China . . . By the Yuan period the Muslims had spread to the four corners (of China), all preserving their religion without change. (Ming Shi),

Considering the extent and impact of this contained population explosion, surprisingly few artefacts and still fewer buildings or other traces from the Yuan period remain today. At least two religious buildings were constructed by foreigners in the -.rst decade of Yuan rule. Almost all trace of one, the Nestorian church constructed in 1281 by Mar Sargis, has disappeared. However, the Phoenix Temple on Zhongshanlu Street still stands today. It was built in 1281 and remains a monument to the western, mainly Persian-speaking Muslims who )*ocked to Quinsai (Khansai-Hangzhou) in response to the needs of the

* I would like to acknowledge the help I received from British Academy Mid-Career Fellowships in completing the research for this paper. The -.nal account of the Phoenix Mosque project by George Lane with full translations of the tombstones and steles by Alexander Morton and Florence Hodous and contributions by Qing Chen and Clarence Eng will be published 2016/17 by Gingko Publications, London. / Cited in Leslie, Islam in Traditional China, 79.

© !"#$#!%$&!' ()$%% #*, %'$+'#, ,-./ | +"$ .-.../1/23425561.63,/_5.1 567 1234 new Chinggisid administration. The -.ve Chinese steles housed in the Phoenix Temple complex unanimously con-.rm the mosque’s location:

Situated to the south of the Occidental Quarter of Silk Textiles (the Westerners’ Wen-Jin Fang ). Facing east, it is very high. (Hong Zhi 6th year, Ming Dynasty, 1493) Since the construction of the temple of True Religion in the Wen-jin Fang (embroidered silk quarter) of the walled city of Hangzhou, a long time has gone by. (1743) Then they [the Uighurs] chose the westerners’ Wen-jin Fang [quarter of brocade/embroidered silk] for their establishment, and wrote on its pediment: Fenghuang, the Phoenix (1892)

A local gazetteer a8-.rms the address of the mosque but also refers to the existence of a nearby Uighur mosque and the long-gone Nestorian church.9 Before the recent reconstruction of Imperial Street and in particular the rebuilding of the Phoenix Mosque’s imposing Persian-style gateway, the mosque was not a building easily discovered without clear directions. These various historical references underline the continued importance of the mosque throughout the centuries.

The Shi Fang: Temple is west of the Chunxi Bridge. It was founded by Nestorian Christians [Yelikewen Shi] during the Yuan dynasty. The Temple of Worship [Li Bai Ssŭ ] is south of the west Wen Jin Fang. It is the place where the Muslim [Huihui] masters of the Yuan dynasty called out to Buddha [presumably God is meant]. The Uyghur Temple [Weiwuer Si] is west of the Qufu: Bridge. It was founded by Uighurs during the Yuan dynasty. Now the placard above its entrance reads: ‘Longevity of the Spirit’ [Ling Shou ‘Immortality of the Soul’].;

Over the centuries, the mosque has been given a variety of very di8ferent names. It is referred to as the Li Bai Ssŭ (Temple of Ritual Salutations) on some steles

< On the Nestorian church also built in 1281 see Sturton, “The Site of the Nestorian Monastery at Hangchow,” 82–85. = Shi Fang is used in Chinese Buddhism to mean ‘the ten directions’ (eight compass points plus up and down). It is possible that it is used here in error, instead of‘Shi Zi’ = ‘Cross’. > Xia Shizheng (1475), Chenghua Hangzhoufu Zhi, juan 47, 13a; unpublished translation by Florence Hodous. ?@4 A@B43CD EBFGH4 56I and as Wu-lin Yuan/Gardens on a thirteenth-century street map, while Li Bai Ssŭ/Temple is thought to be the oldest designation. Names such as the Hui Hui fang (the Muslim hall) and Temple of the True Believers or Temple of the True Teachings would appear to be purely functional and descriptive titles. A stele dated 1493, housed in the mosque complex, claims that “it is tradition to call these temples Li Bai Ssŭ, or temples of ritual salutations.” A magni-.cent scroll map created for the Qing Emperor in the Qianlong period (1711–99) depicts the Phoenix Mosque as one of the Emperor’s special sites in the city.J In fact, it was used by his beloved consort, the Fragrant Concubine (Chinese: ; pinyin: Xiāngfēi; Uyghur: Ipārḵān / Ипархан) whenever she visited Hangzhou. The claim that the mosque complex sustained considerable damage towards the end of the Song era, leaving the site derelict before being rebuilt in 1281, would appear to be supported by a circa 1274 street map. In the space where the mosque should stand is a site marked simply ‘Wu-lin Yuan/Garden,’ possibly indicating a derelict plot. Earlier steles and gazetteers make no mention of an older building and claim 1281 as the date of the Mosque’s founding.T

The mosque had its origins under the T’ang (618–907) and was destroyed by -.re towards the end of the Song (960–1276).U Some of them were damaged by -.re, others by -.ghting and arms. The halls dedicated to worship and those of the schools were destroyed and reconstructed several times.V At that time the Phoenix Temple had already been destroyed by warfare.W

X This map is on show in Hangzhou’s West Lake Museum. Y Stele dated 1493 in the Phoenix Mosque annex, French translation by Vissière; see Vissière, “L’Islamisme à Hang-tcheou,” 61. Z This is from the stele dated 1670, and was composed by the Chinese Muslim, Liu Zhi, who reproduced the text in his Life of Mahomet or Tianfang Zhi Sheng Shilu Nianpu, Book [[. The stele is kept in an outhouse in the Phoenix Mosque complex; see Vissière, “L’Islamisme à Hang-tcheou,” 61. \ This inscription refers in a general sense to the destruction of the steles and buildings within the complex under the Song, Yuan, and Ming eras. It implies that the mosque was founded in the T’ang era, though it is not speci-.c. The stele is dated 1892 and was composed by Ma Zhaolong; see Vissière, “L’Islamisme à Hang-tcheou,” 71–76. ] This is from an inscribed plaque standing by the side of the pavilion in which is housed the sarcophagus of the Chinese poet, Ding Henian. 5^_ 1234

Imperial Street Mosque

The Song city maps,` show the Wu-lin Yuan )*anked by two sites, the Middle (Zheng) Wazi and the San Yuan Lou, neither a very suitable neighbour for a mosque. The Wazi was an entertainment centre of dubious respectability with sing-song girls and troupes of actors, while the Meng Liang Lu describes the San Yuan Lou as a drinking establishment with prostitutes operating within its premises. The Wu-Lin Yuan would have operated as an open-air beer garden, an extension of the Wazi and the San Yuan Lou, which would explain its absence in some maps and would also make the presence here of a mosque highly unlikely. In later maps, where the mosque is clearly indicated, there is no sign of either the Wazi or the San Yuan Lou.

The Wu Lin Yuan is in front of the Middle (Zheng) Wazi. Opposite is the San Yuan Building. The Kang and Shen families have bars in it [San Yuan Bldg]. The entrance is decorated above with colourful painting, and there are red and green poles with scarlet and green curtains; gold and red gauze gardenia-fruit lanterns are hung up to decorate the rooms and corridors, with luxuriant )*owers and bushes, to make the bar congenial. On entering this establishment, there is a hall some ten or twenty paces long, which divides into two corridors, to north and south: these all lead to clean salons, where it is comfortable to sit. Towards evening lanterns and candles shine brightly, above and below. Several tens of heavily made-up bar-girls gather by the entrance pillars of the main hall, waiting for clients to call for them; gazing at them, they seem out of this world. Then there is Cook Wang’s bar in the Xichun Building at the South Wazi, and Cook Shi’s bar in the Flower Moon Building at the New Street Lane junction. . . .,,

Many must have been inconvenienced by the building of the mosque complex. The newly arrived Muslim community appears to have invited reaction and cultivated a high pro-.le, and yet, other than some infamous examples of ill-will towards Muslim foreigners such as the words of Tao Zongyi, reaction would seem to have been favourable and welcoming. Such a conclusion poses questions which must -.nd solutions in further research.

/a Copies of the four original maps are reproduced in Moule, Quinsai, 12–13; adapted and translated versions of these maps can be found in Polo, The Travels of Marco Polo, 2:193, 197, 213. // Wu Zimu, Meng Liang Lu, 141. Unpublished translation by Florence Hodous. ?@4 A@B43CD EBFGH4 5^(

Bayan Noyan spent his youth in Persian Turkistan and, around late 1255, at the age of eighteen, he accompanied his father to Iran with Hülegü Khan’s entourage. He married and sired a son, Noqai, before being assigned to Qubilai Khan’s service and leaving Iran in 1265.,9 It is possible that the sudden rise in the Persian population of Hangzhou and the province of Zhejiang could be directly or indirectly attributed to Bayan’s early association with Iran. In 1276, after the peaceful conquest of the Song capital and faced with the formidable task of establishing an administration in Hangzhou to replace the former Song o8-.cials, who had either gone north with the empress-dowager, or had simply disappeared, Bayan would naturally have used his connections with Iran and its vast pool of expertise and opportunist adventurers who would be very happy to seek their fortunes in the prosperous east of their newly adopted empire. The Persians who made that journey very quickly formed a small but very powerful and in)*uential community in the former Song capital, which remained an economic and cultural urban powerhouse. The Persian community built their mosque on Imperial Street, the imposing thoroughfare which divided the city in two, from the Ho-ning gate, the entrance to the royal palace precincts, stretching north alongside a busy canal to the northern Wu-lin Gate. South, the Imperial Street skirted the foothills of Phoenix Mountain to the southernmost of the city’s gates, Phoenix Hill Gate, overlooked by a large representation of Lamaist Buddhism’s God of War, Mahagela, fashioned from the rock in 1322 for the Yuan emperor. Today, both the sculpture and the water gate, constructed in the late Yuan era and part of the original city walls, remain standing. New gateways have recently appeared along Imperial Street, giving the impression of the medieval walled city within a city upon which foreign visitors such as the fourteenth-century North African traveller, Ibn Baṭṭūṭa, remarked, “We entered the city, which is really six cities each with its own walls, while the whole is surrounded by one wall.”,: The Muslim cemetery, the former Jujing gardens, including the still standing pavilion of the Muslim poet, Ding Henian, used to stand outside the city walls due west of the Li Bai Temple, as it was originally called. In 1271, an imperial edict had ordered the resurfacing of the Great Street or Imperial Street, both inside the palace precinct and outside as far as the Ch’ao-t’ien Gate (Drum Tower), which is just south of the Phoenix Mosque.

/< Rash īd al-Dīn, Jāmi‘ al-tawārīkh, ed. M. Rawshan and M. Mūsawī, 199; Rashīd al-Dīn, Jami‘u’t-tawarikh [sic] Compendium of Chronicles, translated by Wheeler M. Thackston, 105; Hsiao refers to a wife; see Hsiao, “Bayan,” 585. /= Ibn Baṭṭūṭa, The Travels of Ibn Baṭṭūṭa, 4, 900–2. Justi-.able doubts have been raised about his account of Hangzhou. 5^5 1234 Annotated map of Qinsai based on locally produced Song dynasty maps from 1270s. In classical Chinese fashion, In classical 1270s. maps from map of Song dynasty Annotated produced on locally Qinsai based with north at the bottom. the map is south orientated ?@4 2H?@Bh . kl '4AhBiHj?CB3

efgHh4 (( . ( ?@4 A@B43CD EBFGH4 5^6

20,000 slabs were replaced to accommodate “the continuous stream of car- riages moved over the ground like )*owing water”,; on this central thoroughfare. A few years later, it was the good state of the street which most struck Marco Polo. He noted that

all the streets of the city are paved with stone or brick [and that] . . . [t]he pavement of the main street . . . is laid out in two parallel ways of ten paces in width on either side, leaving a space in the middle laid with -.ne gravel, under which are vaulted drains which convey the rain water into the canals; and thus the road is kept ever dry.,J

Today, the Imperial Street has been replaced by the lively though considerably less regal Zhongshanlu Street, which follows approximately the same route as the Song thoroughfare. Of the mediaeval sites, however, only the re-built Drum Tower and the mosque remain in situ. As of mid 2009, the southern stretch of Imperial Street has undergone extensive and dramatic renovation, and the pedestrian-only areas enclosed within gateways bearing the name of their fang exude wealth and prosperity. Some of the canals remain as Marco Polo would have witnessed them, and detailed records of the city before and after the Mongol invasion provide a vivid picture of Quinsai in mediaeval times.,T One of the Li Bai mosque’s steles describes its location at the centre of the walled city as standing like an isolated mountain, reaching a great height. The stele portrays the mosque as dominating the river and the sea to the east, while to the west it is ‘re)*ected by the lake and hills.’ For Ding Peng, an important government functionary at the Ministry of Religious Rites, who composed this 1670 inscription, the mosque was among the wonders of East Asia, so imposing was its beauty. That impression has again become obvious with the reconstruction work carried out in 2009–10 which recreated Imperial Street and the mosque’s original Persian-inspired gateway. The dramatic renovations which the old city is now undergoing demonstrate vividly how the mosque was at the heart of the city’s commercial district, within a short walk of the still standing Drum Tower and the long gone palace gates. The Persians had their restaurants nearby in the very centre of the business area on Imperial Street, the site of a Muslim eatery right up to the last century. They lived in the southern hills overlooking the busy centre of the city and neighbouring the royal palace hills, in an a8)*uent area known as Strangers’

/> Moule, Quinsai, 22. /X Polo, The Travels of Marco Polo, 189. /Y See Moule, Quinsai. 5^^ 1234

Hill, with views of West Lake and to the south, the wide sweep of the River Wu. However, most indicative of the status that they now commanded in the city was their cemetery, the sprawling Jujing Yuan, which occupied the -.nest real estate in the city, lying on the rich verdant land between the hills and the lakes, formerly part of the royal estates.,U

Persian Sources

While the Chinese literary sources are not too forthcoming as regards Hangzhou’s foreign community, a wide variety of Persian and Arab sources, plus a sprinkling of European commentaries and of course Marco Polo, paint an interesting picture of the city. Rashīd al-Dīn, Banākatī, Mustawfī and Waṣṣāf are important sources of information about the Phoenix Mosque and the capital city of Machin (Manzi or South China). Rashīd al-Dīn refers to Khinsāī in his Tārīkh-i-Manzī where three mosques are mentioned.

According to the report of Pulad Chingsang the country of Machin pays to the sovereign a yearly sum of 900 tumans. The capital is named Khinsāī; its wall is 11 parsangs in diameter. In the town there are three post stages. The houses are of three storeys. Khinsāī contains three large mosques of the -.rst rank which on Fridays are -.lled with Muslims. The inhabitants are so many that for the most part they (Muslims) do not know each other.,V

Mustawfī, whose relatives played such a crucial role in the assimilation of Iran into the Toluid polity,,W recognized not only the importance of Khinsāī but the in)*uential role of Muslims in its governance:

Māchīn: A great and extensive kingdom which the Mongols know as Nankiyās [Nankiyas is a Mongol corruption of the Chinese word for barbarian]. It is of the First and Second Climes, and its capital is the city of Khansāy, which some call Siyāhān [corruption of Si-Hu or West Lake by which name the city’s dominating great lake is known]. They say that in all the habitable world there is no greater city than this, or at any rate

/Z Zhou Mi, Gui Xin Za Shi, 142–43. /\ Rashīd al-Dīn, The History of the Khutāy Kings’ Family, 2. /] See Lane, Whose Secret Intent?; Mustawfī, The Ta’r&́kh-i-Guz&́da, 231–33, 236; Persian text ed. Navā’ī, 840, 848. ?@4 A@B43CD EBFGH4 5^n

that in the regions of the east there is no larger town. There is a lake in the midst of the city, six leagues in circumference, and the houses of the town stand round its borders. The climate is warm, and both the sugar- cane and the rice crop produce abundantly; but dates are so rare, and di8-.cult to come by, that one mann-weight of these is bartered for ten manns of sugar. Most of their meat is -.sh, but beef is eaten, and the mut- ton is excellent, being exceedingly expensive. The population is so great that they have several thousand—some say ten thousand—watchmen and guards to oversee the city. Most of the people are In-.dels, yet the Moslems though so few in number have the power in their hands.9`

Waṣṣāf Shīrāzī, the historian and prominent administrator under Abū Saʿīd (r. 1316–35), the last Ilkhan of Iran, while not speci-.cally mentioning the city’s Muslim community, devotes enough space to describing the prosperity and wealth of Khinsāī to signify its importance to Iran:

p.21, Line 11 Description of the country of Chīn: Khunzāī [Hangzhou] is the greatest city of the regions of Chin, “a paradise as broad as the heav- ens.” It is elongated in shape such that its circumference measures about twenty-four parasangs. Its streets are paved with burnt brick and with stone. The public buildings and the houses are built of wood, and decorated with an abundance of paintings of exquisite elegance. Between one end of the city and the other are positioned in three places, ‘Yams’ (post- stations). The length of the main market has been described as three parasangs, comprising sixty-four quadrangles (market squares) similar to each other in structure, and with parallel lines of columns. Line 15 The salt duty brings in seven hundred ‘balish’ in paper-money [bālesh chāw] every day. The number of craftsmen is so great that in the goldsmith’s trade alone thirty-two thousand craftsmen have been counted; as for the rest “estimation will inform you.” There are 700,000 soldiers in the city and 700,000 citizens [raiyat], whose number is recorded in the o8-.ce of the census and the pages of the register. In addition, there are seven hundred churches/temples [kalisa] resembling fortresses, and every one of them over)*owing with clerics without faith, and monks without religion [kashīshīn-i bī kīsh, va rahābīn-i bī dīn], line 17 as well as other o8-.cials, wardens, attendants and idol-worshippers, with followers and folk, whose names are not entered in the listings and census and who are exempt from dues and taxes [qolānāt]. Forty thousand soldiers make up the ‘peo-

ple of the watch’ [ahl-i-ḥerāsat] line 19 and the night patrol. When [night falls] in groups they take up their accustomed positions at the gates, in the neighbourhoods, and in the alleys [kūchahā] streets and corners with the utmost care and [allow people to sleep easily]. Within the city there are 360 bridges line 23 built over canals as ample as the Tigris, which are rami-.cations of the sea of Chin; and various types of vessels and ferry- boats, proportionate to the needs of so many people, ply upon the waters [bar āb ravān] line 24 [in such numbers as to be beyond counting] . . . The throng of all kinds of people from the corners of the four quarters of the world who have come and gathered, for trade and the various needs in a kingdom like this, is clearly evident to the faculty of reason and the aptitude of the mind. These details concern the original capital city p.22 line 1 but there are four hundred other lofty and extensive cities among its territories and dependencies and the least of which of those settlements is greater than Shiraz or Baghdad. And of these Lankin-fu and Zaytūn and Chin-i-Kalān, like Khunzai, are called Shang, meaning Great City, in the registers of the Supreme Diwan.9,

While a number of steles, most of which are today found in a specially built annex in the mosque complex itself, provide some detailed speci-.cs concerning the mosque’s history, only one of them is inscribed in Arab-Persian script, and this early stele relates little solid information concerning the mosque’s past. Rather than provide a date or a name, the inscription tells us that the mosque “is founded upon piety,” and assures the readers that “Whoever buildeth God’s house, God shall built his house in this world and the next,” a few lines later repeating that “Whoso buildeth a mosque for God most high, God buildeth for him a house in paradise.” A verse then follows, invoking the memory and glory of Iran’s legendary kings and heroes, Khusraw and Rustam, and calling upon the majesty and reign of Dày Ming Khan99 to continue with God’s protection and blessing, “May his kingdom abide and his authority be maintained.”9: The text continues to explain that “provisions for the future” are guaranteed for all those who ensure the well-being of the people of Islam, especially by the building of such a special abode, “a building which like the Kaʿba has neither like nor equal,” inside of which “there are gardens of Paradise,” from which “every evening borrows brightness from its lamps” and “the holy angels sprinkle water from their eyes on the surface of its courtyard . . . [while/and] the Houris sweep

The Tombstones and the Mosque

A number of the twenty-one extant tombstones rescued from what was once the lakeside Muslim cemetery, the Jujing Yuan, and now housed alongside the steles in the mosque complex, are written in a mixture of highly stylized Arabic and Persian. The headstones date from the Yuan period and record the deaths of a mixture of military and religious -.gures and merchants, who include two amirs by the names Bakhtiyar and Badr al-Dīn. Many, like Khwāja ibn Arslān Khānbāliqī, the merchant Simnānī, and the preacher Tāj al-Dīn were considered martyrs, signifying that they had died away from their homeland. Other than Khānbāliqī and a certain Khwāja Ḥusām al-Dīn ibn Yaghān Ṭughril Bak, whose names evoke their Turkish backgrounds, the nisbas of the other owners of the tombstones indicate Greater Iran as their homeland. Simnan, Isfahan, Bukhara, Aleppo, and Khurasan are thus indicated, while travels to the two Iraqs, Syria, Arabia, and Najd are referenced, and performance of the Hajj acknowledged. The tombstones clearly indicate that these men were well known and moved in elite circles with Maḥmūd al-Simnānī, “known in the cities and known to the kings of the regions of al-‘Irāq.” At least one of the

<> Ibid., lines 13–19. Islamic calendar, including the Turkic animal calendar and the Chinese calendar, indicating the more cosmopolitan attitude characteristic of the Toluid Mongol world,9T found also, for example, in Tabriz circa 1283 in a short chronicle on the early Ilkhanate.9U These were men who travelled and interacted and were comfortable with people from a variety of cultural backgrounds. The style of the inscriptions suggests the work of a skilled Persian artisan and, although many of the inscribed compositions are Qur’anic, there are also present at least seven di8ferent verses. Four of these poems might possibly have been composed locally, but three are elegies written speci-.cally for named individuals, and these must therefore have been especially composed in honour and commemoration of the respected -.gures buried in the Jujing cemetery. This suggests that the community could a8ford to maintain not only highly skilled calligraphic engravers but also their own poets, examples of whose work has sadly only been preserved on these unique headstones. From a total of twenty-six lines of verse dedicated to a certain Khwāja Jalāl, whose full name and details are no longer decipherable, only twenty-three lines survive. However, those twenty-three lines are su8-.cient to demonstrate a poet fully cognizant with the rules and conventions of Persian poetry, as well as being familiar with traditional Persian poetic imagery, Qur’anic motifs, astrology, and Islamic history. The few examples found on these headstones reveal their com- posers to have been sophisticated and well-educated wordsmiths.9V Today, the Phoenix Temple complex itself houses most of the historical relics related to the mosque, which as well as the tombstones include a most impressive wooden mihrab. The Li Bai prayer hall consisting of three interconnected rooms is a solid, brick structure and contains in the central chamber this magni-.cent mihrab, made from red wood, gilded and inscribed in verses from the Qur’an. The beautifully preserved mihrab, which is set upon a heavy, ornately carved stone altar, has been dated to 1451. It is similar in design to the mihrab adorning the wall of the main prayer hall of the 1341 Songjiang or Dahuading mosque. Above, decorating the interior of the central dome, is an ornate painting depicting )*owers, mountains, birds, animals and rivers contained within the shape of an outlined symmetrical )*ower, also dating from the Ming period and in a remarkably good state of preservation. The dome on the southern side of the central dome likewise has a similar ornate )*owery design in somewhat better condition and more complete than that above the central chamber;

Dating the Mosque

Records and dates concerning the construction of the mosque prior to and during the Yuan dynasty (1276–1368) are sketchy and vague. The 1493, 1743, and 1892 steles state that the mosque was built south of the West Wen-Jin- Fang area—the embroidered silk textiles and brocades quarter (see above). Only the 1670 stele claims that a mosque existed before 1281. “The mosque had its origin under the Tang (618 to 907) and was destroyed by -.re towards the

<] Yang Xinping, “A Chronological Study.” 5n_ 1234 end of the Song (960 to 1279).”:` Certainly, the three-domed prayer hall standing today dates from 1281. Various sources,:, both literary and archaeological, provide the evidence and background for the mosque’s long history. In more recent years, the mosque complex su8fered destruction at the hands of city planners during the 1920s, when the imposing Wang Yue Lao Diwan (Watching Moon Gate-house), built during the Ming dynasty, was bulldozed. Fortunately, plans and grainy black and white photographs of the magni-.cent structure survived and, in 2009, a replica of the Persian-style building was faithfully and painstakingly reproduced. ʿAlāʾ al-Dīn’s role in the mosque’s construction is attested in the archaeological and literary sources. The stele dated 1670 states that the “Grand Master A-lao-ding [ʿAlāʾ al-Dīn] from the “Western regions” [Xi Yu] rebuilt the mosque from its foundations creating a magni-.cent edi-.ce, a view expressed in various other sources.:9 The stele celebrating the poet Ding Henian, whose pavilion stands amongst the trees and bushes of the gardens decorating the south- eastern shoreline of West Lake, explains that in the late 1270s, the poet’s great grandfather, A-lao-ding, arrived from the west and, moved by the spirit of charity, constructed a mosque for the growing Muslim, mainly Persian, community, a story also recounted by the poet’s friend and fellow writer, Dai Liang. A Ming commentator, Tian Rucheng, claims that he restored a former mosque that had fallen into disrepair.:: According to the Ming Shi and contemporary records, ʿAlāʾ al-Dīn and his brother, both rich businessmen, contributed generously to Qubilai Khan’s war e8fort in subduing the west, and, in recognition of their -.nancial assistance, government positions, land, and property were pro8fered. The younger brother entered Qubilai’s administration while ʿAlāʾ al-Dīn accepted land, feeling too old to enter government service. Since the Muslim population of the city was on the rise, ʿAlāʾ al-Dīn oversaw the construction of the mosque in its central and imposing location and his generosity has been recorded in various sources since that time.:; The growing in)*uence of the Muslims of Hangzhou is attested to by Ḥamd-Allāh Mustaw-., who claimed that though few in number the Muslims held “power in their hands.” Rashīd al-Dīn claims that there were three mosques in the city and that Muslims were so numerous they rarely

=a Vissière, “L’Islamisme à Hang-tcheou,” 61. =/ Vissière lists the steles, and also gazatteers, inscriptions often unpublished; see Lane, “Phoenix Mosque,” in Encyclopædia Iranica. =< Vissière, “L’Islamisme à Hang-tcheou,” 61; Dai Liang, Jiu-ling-shan fang-ji, juan 11, 2a. == Tian Rucheng, Xi Hu you lan zhi yu [West lake travel record], chapter 18. => See Yuan Shi on Tian Rucheng. ?@4 A@B43CD EBFGH4 5n( An early 20th century depiction of the Phoenix Mosque complex before the implementation of the Town Planner’s Planner’s depiction of of century the implementation 20th before An early Town the Phoenix Mosque complex the modernisation scheme. ?@4 2H?@Bh . $@B?B kl

efgHh4 (( . 5 5n5 1234

“recognized one and other.”:J Ibn Baṭṭūṭa, who spent more time in Zaytun, also mentions that the Muslim community of Hangzhou was large, prosperous and in)*uential.:T Although the steles suggest that the original Muslim settlers came overland from the west, the subsequent expansion of the Muslim population of the coastal cities of eastern China was from the sea, with the three cities of Hangzhou (Quinsai), Quanzhou (Zaytun), Guangzhou (Canton) playing an important role as gateways to the Islamic world. The stele dated 1493 states that the Muslims came from the west and built mosques wherever they settled, and the stele dated 1892 speci-.es that the Uighurs, after serving in the emperor’s armies, were rewarded with honours and employment and were invited to settle in China, and that the emperor himself, Suzong of the Tang dynasty, ordered the construction of mosques to encourage the integration of his Muslim mercenaries. With the establishment of the Ilkhanate in Iran in the late 1250s and its strong links with the east, Iranians were again attracted to China. There is evidence that Hülegü Khan (d. 1265) dispatched military engi- neers to assist his elder brother in his wars against Ariḡ Buqā and Qaidu.:U The appointment of Persian governors to a number of Chinese provinces at this time would have presumably entailed their travelling with their own retinue of administrators and soldiers, further bolstering the communities of Manzi and Khitai. The Phoenix Temple of Hangzhou stands in quiet testimony to the city’s Islamic heritage and to the brief period in a glorious history when Muslims united East and West. Though Quanzhou has undergone considerable archaeological study, much works remains to be carried out in Hangzhou.

The Jujing Yuan

The Jujing Yuan has been transformed into modern Hangzhou’s beautiful public gardens, the Orioles Singing in the Willows. The splendour of the modern gardens must explain the attraction the cemetery held for generations of the Yuan dynasty’s Muslims. Information on the nature and even the exact location of the Jujing Yuan, which became the site for the -.nal resting place for mediaeval Hangzhou’s Muslim community, is often di8-.cult to -.nd and not

=X Mustawf ī, Nuzhat al-qulūb, 261, tr. 254; Rashīd al-Dīn, The History of the Khutāy Kings’ Family, 2. =Y Ibn Baṭṭūṭa, The Travels of Ibn Baṭṭūṭa, 4, 901–2. =Z Bretschneider, Mediæval Researches, f: 273–74; Moule, Quinsai, 85–86. ?@4 A@B43CD EBFGH4 5n6 always easy to interpret or understand. References to the gardens are written by commentators who are rarely sympathetic to the presence of foreigners on Chinese soil and some of the best-known chroniclers from the Yuan and early Ming period were also strongly nationalistic, hardly noted for their objective and balanced commentaries. As late as 1920, during the Republican period, an observer noticed that there was a stone plaque beside the road to the Hupao Temple, near the south-east corner of the West Lake, which bore four Chinese characters meaning ‘Old Graveyard of the Muslims.’ This was probably a stone set up by more or less contemporary Muslims in Hangzhou to mark the boundary of the graveyard. The discovery of the tombstones and the location of the old cemetery occurred in two phases in the last century, the 1920s and the 1950s. Demolition and construction were taking place during the 1920s including the dreadful demolition of the imposing Phoenix Mosque’s entrance Wang Yue Lao Diwan and what remained of the city’s ancient walls. One hundred or so stones are believed to have been unearthed in 1921 and o8-.cially recorded at a discussion in Shanghai. The historian, Bai Shouyi, conducted fuller research in 1936 and his work forms the basis of all subsequent research, but it was not until the 1990s and the appearance of Guo Chengmeii that evidence of in-depth research on the Phoenix Mosque tombstones could be found. The discovery of the stones provided evidence for the location of the Jujing Yuan cemetery. However, the priority at that time was the building of the lakeside road. Later construction of the extensive public gardens and the discovery of a large number of ‘foreign’ tombstones was not seen as urgent, so much so that the fate of at least eighty of the tombstones remains unexplained to this day. It is possible that they might have been re-buried then or later to avoid the destruction of the Cultural Revolution. Their fate is unknown. Whether the cemetery came into use before the mosque or whether the Jujing Yuan (, gardens) outside the Qing-bo Gate was converted to a Muslim burial ground after the Persian community had become established remains uncertain. However, references to the Jujing gardens themselves can be found in pre-Yuan sources. In 1268, Qian Shuo-you compiled the Lin’an Gazetteer in which he described the gardens thus:

The Jujing Yuan are outside the Qingbo Gate. The Emperor Xiaozong [reigned 1162–89] lavished attention on them . . . ’ other emperors also frequented them down to the time of Emperor Ningzong [reigned 1194–1224] and the Empress Chengsui . ‘Now of the old buildings all that remain are the hall called the Lan Yuan and the pavilion called the Hua Guang . There is also a pavilion where red 5n^ 1234

Prunus mume trees are planted . . . ’ and two bridges, old pine trees, etc. In summer people go in boats to view the lotus )*owers.:V

In the Song Shi, there are about twenty short references to the gardens which report that the emperor and empress dowager blessed the Jujing Yuan with their presence (or words to that e8fect). Clearly, the garden was a kind of imperial summer palace, actually used as a residence by senior members of the imperial family. Considering the size of the imperial retinue, it seems likely that the garden must have been on a grand scale. In the West Lake Travel Record (Xi-Hu you-lan) of the commentator Tian Ru-cheng, the gardens, which stretched from the Qing-bo Gate and the Nan-shan (South Mountain) Road to the lakeside, encompassed ‘a scented meeting hall, an ocean spring, a look- out, and fragrant, )*owery halls’ and included the emperor Xiao-zong’s ‘lucky walk.’ It was said that during Xiao-zong’s time (1163–89) the park was very lively.:W The art collector and social commentator Zhou Mi (1232–98) claimed that ‘Noble women go to the Jujing gardens to avoid the summer heat.’ Today, on the northern site of the gardens, the banks of the shoreline are lined with willow trees, whose long weeping tresses sway in the breeze ‘like green waves churning the air.’;` The area is known as “Orioles singing in the willows” and is one of the ten poetically named scenic places of West Lake. For unexplained reasons, the park fell into a period of disuse under the emperor Ning-zong (1195–1224). Zhou Mi (1232–98) described the gardens as desolate () during these years.;, No mention is made of the period from 1224 to 1291, the year that Zhou Mi records the use of the Jujing gardens as a Muslim cemetery. However, on the maps originally executed in 1268, the gardens appear to be marked, though the eighteenth-century copy- ist Lu Wenzhao has indicated their location with square blanks, signifying his inability to decipher the characters. That the square blanks are substitutes for the characters (Jujing yuan) is veri-.ed by some later copies of the original maps executed by the Ming scholar, Tian Ru-cheng. His reproduction of the Song maps include characters for most of the blanks found in the Lu Wenzhao reproductions, even though the quality of his cartography and calligraphy are far inferior to the excellent 1268 maps from the Xian-chun Lin’an Zhi.;9

=\ Paraphrased and translated by Florence Hodous, unpublished. =] Guo Chengmei and Guo Qunmei, “Textual Research,” 66. >a From a lakeside plaque describing the area. >/ Zhou Mi, Wu Lin Jiu Shi, 4th juan cited in Guo Chengmei and Guo Qunmei, “Textual Research,” 66. >< These maps of Song Lin’an are taken from the Xi Hu You Lan Zhi by Tian Rucheng , printed in the 26th year of the Jiajing reign period of the Ming dynasty ?@4 A@B43CD EBFGH4 5nn

Zhou Mi, a dissident and Song Loyalist, counted at least one Muslim and westerner amongst his very select friends and intimates. Gao Kegong (1248– 1310) was a fellow -.ne art collector and, as General Secretary to the Censorate, was an important reformer in the Yuan administration. His family and descendants founded the Muslim community of Dahuading (Songjiang) and its mosque, built in 1343, which still stands in Shanghai’s southern suburbs today. Gao Kegong’s presence among Hangzhou’s elite intellectual and artistic circle, whose illustrious members included Qubilai Khan’s favourite painter and calligrapher, Zhao Mengfu (1254–1322), is yet more evidence that the Muslim community of Hangzhou had integrated at the highest levels and in-.ltrated a previously closed clique. Zhou Mi discusses Muslim burial practices in practical terms and appears very knowledgeable on the subject, obviously speaking from personal experience. In his Random Jottings, he describes in some detail the preparation and burial of the dead by the local Muslim community, suggesting that he must have been on intimate enough terms with the participants to have been invited to observe and possibly partake in these rites.

The Huihui custom is that every time someone dies, there is a special person who, pouring water from a large copper pitcher, cleans their stomach and abdomen in order to rid them completely of bad ‘qi’ (). Further, [he] washes [them] from the crown of their head to their feet, and when the washing is -.nished, he wipes them dry with a cloth. Using cloth woven from ramie;: or [plainly-woven] silk, or cotton cloth, they make it into a sack and put [the body] in it naked. Only then do they put [the body] into a co8-.n. For the co8-.ns they use thin planks of pine wood, into which the body barely -.ts, and the corpse is not accompanied [into death] by any object. The dirty water from washing the body is collected in a great pit at the bottom of the room, which is covered with a stone, and is called “beckoning the soul.” They put a table on top of this pit and make an o8fering of food every four days, and after 40 days they stop. The co8-.n on that very day is taken out and interred in the Jujingyuan, a garden which is in the care of a Muslim (Huihui). Anyone who rents a lot pays a -.xed price, and the caretaker has bricks, mortar,

(1547); facsimile edition, Taibei : Chengwen Chubanshe , 1983. The originals can be found in: Qian Shuo-you , Xianchun Lin’an Zhi [Gazetteer of Lin’an [compiled during] the Xian-chun reign-period], 3 vols. (Photographic reprint of the edition of the tenth year of the Daoguang reign period of the Qing dynasty [1830].) >= 5np 1234

and craftsmen at his disposal, which people pay him for specially. When the time for mourning the dead comes, their family members all cut their faces, [tears of blood],;; tear their hair and rend their upper garment; they beat their breasts and loudly wail and sob, so it can be heard [lit. vibration] near and far. When they take out the co8-.n, if [the dead person] was rich then beggars will hold candles and scatter fruit along the road, but if [the deceased] was a poor man, this does not happen. Then everyone in order, whether young or old, kneel as is their custom, and the mourners then tap their boots as music and they comfort each other, and when their feelings have been calmed they direct the crowd to return to reciting their scriptures. Three days later, they again return to the place of burial, where the rich sacri-.ce many oxen and horses and host a banquet for their people, including the poor and beggars of their neighbourhood/ near them/in their vicinity. Sometimes one hears it said that when they arrive at the place of burial, they take the body out of the co8-.n, and bury the body alone in the hole, with the face facing the West. [dated ‘spring of the Xin Mao year’—1291].;J

There is a common practice among Turkic peoples and their neighbours, at least in the pre-Islamic period, of lacerating their faces at funerals. The references to ‘tears of blood’ in the Quanzhou inscriptions;T and elsewhere could well refer to this at one time common practice of the laceration or slashing of the face when mourning the dead, described by Dennis Sinor as ‘an ordinary sign of grief, loyalty or respect for the dead’ among the Uighurs.;U On Strangers’ Hill, on the grounds of the 500-year-old Wansong Academy;V , founded in 1498 during the Ming dynasty, at least three sarcophagi have been uncovered. Whether this represents the southern border of the Jujing cemetery lands or whether these sarcophagi had merely been mislaid is not known, and the answers to such questions will remain elusive until the stones themselves are properly examined.

>> This appears in a tombstone inscription for Naina Muḥammad b. ʿAbd Allāh d. 1305, unearthed in Quangzhou. See Chen Dasheng, 34. >X Zhou Mi , Gui Xin Za Shi , 142–3. Unpublished translation by Florence Hodous. >Y Chen Dasheng, Islamic Inscriptions, 34. >Z Sinor, The Cambridge History of Early Inner Asia, 304, 327. >\ Address: 76, Wansongling Road. Bus: 102. Opening hours: 7.30 am–5 pm. Admission fee: 10 yuan. Tel: +86 571 86079490. ?@4 A@B43CD EBFGH4 5nr

Though some Persian and Muslim merchants lived alongside the other wealthy notables of Hangzhou in the exclusive Strangers’ Hill region overlooking both the lake and the wide, sweeping Che River to the south, it was the area to the immediate east of Imperial Street that was dominated by Muslims. There was a common adage among the Chinese, made famous by the Tang poet Li Shangyin in his “Miscellaneous notes” under “incompatibilities,” which contains the following: “Poor Persians, sick doctors, thin Sumo wrestlers, fat brides.” The Huihui Xin Bridge () still stands today, though no doubt the actual bridge has been rebuilt many times. It stands between the often mentioned Feng-le Bridge () and the Jian Bridge (, , ) strad- dling the Zhong-he canal (Middle Canal, ). Tao Zongyi, writing circa 1366, describes the area between the bridge and the Phoenix Mosque:

There were eight high houses at the head of the Jian (Ch’ien) Bridge, and were popularly called Eight Pavilions. All of them were inhabited by rich Huihui people.;W

The Muslims of Hangzhou

The writings of the commentator Tao Zongyi, who was based in Hangzhou, reveal a xenophobic attitude not uncommon at the time. He continued his piece on Hangzhou with an anti-Muslim tirade, ridiculing a local tragedy while lampooning Muslim names before concluding his tasteless tale with a reference to the Jujing cemetery.

Once there was a wedding ceremony which was totally di8ferent from that of Chinese; inter-marriages between uncle and niece and between cousins were even allowed. Neighbours crowded there and peeped at them. Some of them (neighbours) even climbed to the overhanging eaves, railings and windows, so the building collapsed causing the death of all the hosts, guests, bride and groom. It was a very strange thing Wang Mei-ku, an o8-.cial of the area, wrote a lampoon, the Baia—huo wen [With clever punning on the personal names “A lao-wa, Tao-la-she, Pieh-tu-ting, Mu-hsieh-fei” all Hui-hui personal names, says Tao’s commentary, the poem goes on] Their (Muslim) clothes and headgear are covered with dust, their elephant noses are now )*at, their cat’s eyes no longer shining.

>] Tao Zongyi, Nancun Chuogenglu, juan 28, unpublished translation by Liu Yangsheng; Leslie, Islam in Traditional China, 93. 5n7 1234

Alas, in one day, all their hopes for a long life are gone . . . The kitchen is moved to the Jujing gardens, their cemetery. The cry “Allah” is not to be heard any more. Alas! The tree has fallen, and the monkey grandchildren of the monkey Hu have dispersed.J`

With Muslims occupying the exclusive plots on Strangers’ Hill and prime urban space in central Hangzhou, and their cemetery covering what were once royal gardens over-looking the lake, resentment against these foreigners, who held such prestige and power and were so close to their Mongol masters, was often evident. This was especially apparent amongst the Song loyalists, some of whom remained uncompromising in their opposition to cooperation with the Yuan elite. However, as Jennifer Jay has shown in her studies of the nature of Song Loyalism, opposition was neither as widespread nor as united as has often been reported.J, Cooperation with the Mongols had occurred at various levels ever since hostilities had broken out against the Jurchen in the second decade of the thirteenth century. As the Chinggisids inexorably wore away at the Song resistance, high-level defections occurred at a regular rate and Chinese o8-.cials ‘joined the queue’ behind Semu and the Jin to be assigned positions in the Toluid bureaucracy. Seen in this light, resentment against the newcomers might have been tempered slightly with Persians and Muslims seen as rivals rather than oppressors. The loathing expressed by these loyalists for the Muslims has a sense of des- peration, though it was by no means universal:

Hui-hui is also Hui-hu [a former name for the Uighurs]. Their custom is not to eat pig. It is popularly related that this is because the ancestors of the Hui-hui are descended from pigs. The Tartars are the present Yuan bandits . . . Even when they bathe, the Hui-hui still stink . . . The Hui-hui also serve Buddha. They built a tower called the Buddha Tower which is very high. Once a man making a solemn oath climbed to the top of the tower and called to the Buddha in a loud voice unceasingly. Confused and in a frenzy, he suddenly heard from the air the Buddha answering him. He took a knife in his hand and cut o8f his male member and threw it to the ground. Finally he threw his whole body down from the top of the tower and was smashed to pieces, and died. All because he felt that the Buddha had responded to his worship. The Hui-hui fought over his member to use as

Xa Tao Zongyi, quoted in Leslie, Islam in Traditional China, 93. X/ Jay, see A Change in Dynasties. ?@4 A@B43CD EBFGH4 5nI

medicine, and preserved it in a box and handed it down as a precious object.J9

Despite these negative sentiments, integration was evident at many levels. There had been a history of trade relations between Iran and China stretching back over the centuries and merchants had been plying the sea routes and overland caravan trails for generations, establishing strong business networks which would sometimes hibernate but never die. Hangzhou ceased to be the political capital of the Chinese state, but it maintained its paramount position as an administrative centre. In particular, as far as south-eastern maritime trade was concerned, while Quanzhou continued as the hub of this lucrative sea trade, Hangzhou became the administrative powerhouse of the industry re)*ected in the wealthy gardens where Muslim merchants and notables laid out their dead and departed. Apart from the tombstones, which provide names along with some very basic biographical details, very little information exists concerning other residents of the cemetery. This is surprising, since the Phoenix Mosque was well known and references to its existence remain scattered throughout China’s rich literary history. Most famously, the emperor Qianlong (25 September 1711–7 February 1799) awarded his fragrant concubine, Iparhan, use of the mosque and its gardens and included the sanctuary as an honoured place on his famous map of Hangzhou’s West Lake, which depicts his favourite locations in the region.J: Today, as mentioned earlier, only two graves remain in situ in what is now a public park, though three sarcophagi have recently come to light in the grounds of the 500-year-old Wansong Academy . One is a small pavilion set back from the shores of the lake that provides a tranquil setting in which to contemplate the words of Hangzhou’s most celebrated poet, Ding Henian, whose remains are buried there. Nearby, close to a memorial stele to the Qing-bo Gate, is a small enclosure containing three sarcophagi commemo- rating the Amir Bakhtiyārī.

Ding Henian

Ding Henian died at 89. In 1424, at his speci-.c request, he was buried in the Jujing Yuan in a leafy grove by the water’s edge, which suggests that he had

X< Cheng Ssu-hsiao (Cheng Sue-nan), 1239–1316, T’ieh ham Hsin-shih, quoted in Leslie, Islam in Traditional China, 92. X= The map is now on display in Hangzhou’s West Lake Museum. 5p_ 1234 some special attachment to the gardens or to the city or West Lake. A memorial plaque to the poet con-.rms details of the mosque’s construction and Ding Henian’s connection to ʿAlāʾ al-Dīn. The plaque claims that the founder of the mosque is one and the same as the poet’s great-grandfather, ʿAlāʾ al-Dīn, a link nevertheless not universally accepted as valid.J; Such linkage purely on the evidence of commonly held names has often occurred with regard to ‘western’ players, despite such Muslim names being in extremely common usage. The plaque beside the memorial pavilion praises Ding Henian for his devotion to his mother, who had disappeared in the confusion of war. He had had a vision in a dream that his mother was dead and, in memory of his love for her, he built her a tomb and then arranged that his remains should be buried along side her when he died. For this act of devotion and -.lial piety, he has been honoured with the construction of the stone pavilion and a respected position in the city’s cultural heritage. He owed his education to another in)*u- ential woman in his life, his sister Yue’e , whose biography appears in the Ming-shi. She earned her entry because of her martyrdom, which is recorded as having occurred as a result of bandits attacking and occupying Yuzhang. She was already a woman of some in)*uence in the city, so when she chose death instead of surrender to rape by the bandits and drowned herself with her baby daughter in her arms, her example was followed by other women present, nine in all. The elders, who were consulted, decided, ‘As the ten martyrs were of a single mind and died together, we cannot bury them in separate graves.’ The Grave of the Ten Martyrs, along with a small stele inscribed by her brother, the poet Ding Henian, was erected south of Huang-ch’ih, the village where they had lived.JJ An early glimpse of Ding Henian’s tomb shows that the actual pavilion has changed little since the time of Vissière’s photograph in the -.rst decade of the twentieth century or possibly earlier, but its surroundings are unrecognisable. Apart from the riotous vegetation, two edi-.ces are no longer existent, namely Hangzhou’s walls, discernable to the right in the background of the photograph, and also the porticos of a small mosque to the left, set back from the pavilion.

Sharaf al-Dīn

Not all Muslims buried in the Jujing cemetery were from the Persian elite; some rose to prominence within the Chinggisid ranks as artisans. The life of

X> Liu Yingsheng, in private conversations and emails with the author. XX Ch’en Yüan, Western and Central Asians, 282–83. ?@4 A@B43CD EBFGH4 5p(

Sharaf al-Dīn, the -.rst darughachi of Shanghai county, is an exemplary success story not to be confused with the life of another Sharaf al-Dīn, the darughachi of Quanzhou/Zaytun and a member of the powerful Qunduzi family who had Nanhai and Sengge connections. Sharaf al-Dīn of Shanghai county was the son of an artisan whose family were probably rounded up in the -.rst wave of the Chinggisid conquest of Khwarazm. Sharaf al-Dīn, who, at his own request, was buried in Hangzhou’s Muslim cemetery, was greatly honoured in his own lifetime. His achievements were celebrated on a stele commissioned by his sons Naṣīr al-Dīn [Nasuluding] and Mubārak-Shāh [Mubalasha] in 1324. His father, Kamāl al-Dīn, is recorded as having been a craftsman who later became chief artisan with a corresponding administrative position. The absence of any nisba or reference to a hometown or any other indication of family origins or ancestral ties suggests that his father was just one among the many artisans who were transferred from Khwarazm to the various parts of China to work on one of the royal princes’ appanages. The prevalence of fully Islamic names in the family and the absence of Chinese, Mongolian or even Turkish names suggests the family’s possibly Persian background. Sharāf al-Dīn attracted the attention of a very ‘great one,’ Bayan Noyan, conqueror of Hangzhou. “In the year 1275, he [was appointed] to the Central Secretariat and was recommended to the king by Bayan, who much appreciated his talent.”JT

Dao Wu

Another -.gure of some importance is also recorded in the literary sources as having been buried in the city’s new Muslim cemetery. Dao Wu, also known as Shanchu, hailed from ‘A-lu-wen’ (possibly Hulvan between Kermanshah and Baghdad) in the ‘Western Regions,’ and held various in)*uential positions, including Grand O8-.cer or Daifu, Associate Director of the General Administration of Wenzhou. He -.nally became the Assistant Administrator of the Hangzhou Sub-prefecture, after which, according to the Ming historian Song-lian [1310–81], ‘he died, and was buried in the Jujing garden left of the Ling-zhi Temple in Hangzhou City. His son La-zhe-jun was buried with him.’JU

XY Guo Xiao-hang, “The Study and Interpretation of the First Shanghai Da Lu Hua Chi—She La Fu D ing in Yuan Dynasty.” XZ Quoted in Guo Chengmei and Guo Qunmei, Textual Research, 66. 5p5 1234

Habashī

A further -.gure buried in the Jujing cemetery and illustrative of the calibre of people who chose the gardens as their -.nal resting place is a certain Ding Wen-yuan , also known as Habashī, who was born in 1284 and died of an unspeci-.ed pestilence forty-seven years later on his way to take up a new post in the north, having left his family in Hangzhou. The family was originally from Khotan, east of Kashgar. Habashī died aged 47. His son, Muxie, is recorded as having expressed his father’s wish to be buried by West Lake and accordingly his wish was granted. “The Western Hills of Hangzhou were loved by the people in the past, so we may follow [the former practice of burial by the lake in the Muslim cemetery.]”JV

The Bakhtiyārīs

Today, erected on the site of the old Qing-bo Gate, can be found the sarcophagus of a revered holy man of the Bakhtiyārī family and his two companions. Two reconstructed steles commemorate Bakhtiyārī’s presence in Hangzhou. As previously mentioned, there is some confusion concerning the identity of Bakhtiyārī, with a local legend claiming him to be an early missionary who travelled from the west during the T’ang or Song period. However, Guo Chengmei has discounted this popular legend and assumes that Bakhtiyārī was a Persian merchant, though whether his companions were his sons or assistants remains uncertain, as does the connection with the headstone of the illustrious Amir Bakhtiyār Nīlūsiya-Nikūnal, son of Amir Abū Bakr Dawūsiya-Nikūnal, son of the almost legendary Sayyid ‘Ajall ʿUmar al-Bukhārī, governor of Yunnan until his death in 1279. Amir Bakhtiyār died in August 1330. In Quanzhou, an Amir Sayyid ‘Ajall Toghan-shāh, apparently from the same illustrious family, is recorded as dying in October 1302.JW

Other Tombstones

Other tombstones now housed in the Phoenix Mosque paint a cross-section of the Muslim community of Khunsāī, dominated by Persians. The inscrip-

X\ Cited in Guo Chengmei and Guo Qunmei, Textual Research, 67; Bai Shouyi, Hui zu ren wu zhi, 135–38, 342–50; Chen Hanzhang, “Zhongguo huijiao shi,” 145. X] Chen Dasheng, Islamic Inscriptions, 33–34, 81–82. ?@4 A@B43CD EBFGH4 5p6 tions are for the most part clearly legible, although, unfortunately, some crucial information, such as names and dates, often located at the bottom of the headstone and forming the -.nal lines of the inscription, have been obscured or irretrievably damaged. The legibility of the inscriptions has been enhanced by the Chinese art of ‘rubbing.’ Amongst those buried in the Jujing cemetery whose headstones were exca- vated in the 1920s, merchants are well represented. The merchants not only maintained their positions of in)*uence but developed and expanded their roles. The case of ʿAlāʾ al-Dīn, the benefactor of the Phoenix Mosque, is one example of a merchant using his money to further his own aims and that of his community. The respect and acclaim that individual merchants could achieve is re)*ected in the inscription on the tombstone (A7)T` of al-Simnānī, unearthed from the Jujing cemetery. As one who died far from home, he is of course honoured as a martyr: “The death of the exile is martyrdom.”      (      )T, But the inscription also lauds his travels:

he had travelled in the lands and visited the righteous, devoted servants of God, and gone to the West of the Earth and gone to the East, visited Syria and Iraq, performed the lesser pilgrimage and been to Najd, honoured the poor and is spoken of for bene-.cence.T9

The praise continues with the awarding of the title of “Shaykh, the most noble, the great, the generous, . . . ” eventually arriving at his second ‘title’ of “pride of the merchants, ornament of the good and the noble, famous in the cities, re-.ned of character, pure of manner, familiar among princes of the regions of the coasts,” until it arrives at the climax, “Splendour of Islam and the Muslims, privileged with the solicitude of the Lord of the Worlds,” and ends with his full name. This was evidently a very important and well-travelled businessman who turned to religion in his later years and died revered as a Shaykh and Su-.. Whether he would have known or even been a follower of the famous Su-. poet ʿAlāʾ al-Dawla Simnānī (1261–1336) must remain speculation. Similarly, another tombstone from the cemetery (A9) celebrates Shams al-Dīn Isfahānī, who died on 24 September 1316, duly noting that “the death of

Ya The classi-.cation of the tombstones corresponds to Guo Chengmei’s listing in his article on the Hangzhou Mosque and headstones; see Guo Chengmei and Guo Qunmei, “Textual Research,” 67–68. Y/ This saying attributed to Muḥammad does not appear in these words in Wensinck’s concordance of Prophetic Ḥadīth. Y< Unpublished translation by Alexander Morton. 5p^ 1234 the exile is martyrdom.” This individual is also lauded as a Su-. and honoured as a merchant, “the Shaykh, the most distinguished . . . the pride of the merchants, famous in the cities, patron of the learned and the strangers, . . . known among kings of al-‘Iraq [‘irāq can also mean the regions of the coast].” On the reverse, there is a Persian elegy to ‘Khwāja Mohammad, Jewel of Isfahan,’ which notes that Shams al-Dīn Isfahānī left for China from the kingdom of Iraq. Another headstone appears to be that of his son, Khwāja ʿAlāʾ al-Dīn, who died eleven years later, in 1327, though the verses in Persian and Arabic give no personal details. Verses which are attributed to ʿAlī on other headstones do not necessarily indicate Shi‘ite leanings. That the Muslim community greatly valued its businessmen and clergy is plainly evident from the laudatory tone of other headstones celebrating the lives of its leading citizens. One tombstone (A1) commemorates a Turkish or possibly Mongol merchant, Khwāja Ḥusām al-Dīn son of Khwāja Yaghān? Ṭughril Bak? Yeke Wali? (#xyz{f), referred to as a young man (shābb) and a martyr, with verses recording his travels and contact with the ‘righteous.’ His death is recorded as 29 Rabīʿ ff, 707/5 November 1307. The tombstone labelled A3 shares many of the features found among the other extant inscriptions, though the lower section containing personal details and dates is missing. However, from the available information, it is evident the deceased was a young man and a merchant. He had travelled and visited (made pilgrimage to) the righteous. He had been to the East and the West, and travelled in Syria, Iraq, Najd and performed the lesser pilgrimage (ʿumra). Like others, he is famed in the cities, with much of the wording identical with the corresponding part of the epitaph of Maḥmūd Simnānī (A7). Whether this signi-.es some connection between the two, possibly as travelling merchants, or merely a shared epitaph composer or inscriber, remains conjecture. Once again the stone intones, ‘The death of the exile is martyrdom.’ One stone (A5) celebrates the life of a preacher, himself the son of a preacher (wā’iẓ), Imām Tāj al-Dīn Yaḥyā bin Imām Mawlānā Burhān al-Dīn; unfortunately, it lacks a date. The son is described as a young man (shābb) who died aged forty-one. The front of the stone is composed of Arabic verses while the reverse has little more than predictable pieties. Though the frequent references to ʿAlī do not necessarily denote Shi‘ism, the clear comparison on the tombstone labelled A6 between the virtues of the -.rst four Caliphs and the merchant, Shihāb al-Dīn Aḥmad b. ʿAbdullāh, of the great family of Muḥammad Ḥalabī, points unequivocally to an adherence to Sunnism. The deceased’s name and honori-.c, khwāja, often awarded to merchants, survive in the Persian verse on the reverse side of the tombstone, but the top and bottom portions of the front face are missing. ?@4 A@B43CD EBFGH4 5pn

The fame and attraction of the Jujing cemetery has already been observed from documentary evidence, and this is underlined by the presence of amirs in the gardens. The inscription (A13) for Amir Badr al-Dīn, son of a Ṣadr, is composed of Qur’anic verses, and some lines belonging to al-Iṣfahānī are also found on the headstone:

[Line 1.] Every soul will taste death. [Line 2.] This is the resting-place of the amir, the most illustrious, the noble, the great, lacking peer [Line 3.] or match, source of good fortune, centre of bene-.cence, patron of the masters [Line 4.] of religion, supporter of the possessors of true knowledge, the elect of the beloved of the time, [Line 5.] Master of generosity and benevolence, Amir Badr al-Dīn, son of the Sad[r]

The other amir represented among this collection of tombstones is Amir Bakhtiyār of the Bukhārī family (A11), whose founder, an army commander under the Khwarazmshah, joined the Chinggisid forces following his defeat c. 1220 and enrolled his sons in the keshig, where they prospered. The most illustrious of the clan was Shams al-Dīn Sayyid ‘Ajall, the governor of Dali (Yunnan province), whose progeny subsequently appeared in positions of power and in)*uence in lands in the east and west, with even the Safavid historian, Khwāndamīr, claiming descent.T: A descendant of Sayyid ‘Ajall is mentioned in the stele dated 1670:

During the years hongwou (1368–1398), there was a descendent in the 7th generation of Sayyid ‘Ajall, prince of Xianyang, named Hazhi (the Hajji) who went to the administrator of the imperial palace to hear proclaimed there the orders of the Sovereign. Permission was given for the construction of mosques (Li Bai Ssŭ) in all the provinces and, during the following epochs, decrees which conferred favours were pronounced as if they were to be a permanent law.T;

Both Amir Bakhtiyār and his father Amir Abū Bakr held either Mongol or Chinese honori-.cs: Amir Bakhtiyār Nīlūsiyā-Nikūnal and his father, Amir Abū Bakr Dawūsiya-Nikūnal. On the reverse side of Amir Bakhtiyār’s headstone,

Y= See Khwāndamīr, Habibu’s-siyar, f, Introduction, ix. Y> Unpublished translation by Florence Hodous. 5pp 1234 an image has been carefully inscribed into the surface, executed in intaglio rather than in relief. No annotations or verses decorate the picture. The design depicts a smoking incense burner placed on a table and )*anked by two )*ower vases. In common with the other tombstones, the borders of the central design and sometimes the actual sides of headstones are composed of intricate patterns, either abstract arabesques or entwined leaves and plants which bear close resemblance to the blue-and-white ceramic themes, at that time a major export item to the west. This is one of the few tombstones from Hangzhou that had previously come under close scrutiny. This has proved extremely for- tuitous since the headstone was wantonly attacked during China’s Cultural Revolution and the resulting damage has obscured parts of the text as well as spoiling the image of the incense burners. Without the rubbing from 1936, it would be impossible to date the stone. The -.nal two lines, now either completely destroyed or buried beneath restorative concrete, read as follows:

[Line 9.] . . . And he died on the twenty--.rst of Shawwāl in the year seven hundred and thirty of the Hijra [7th August 1330]; [Line 10.] in the Cathayan fashion, year . . . [?] of Zhishun; the twenty- third of month seven of the year of the sheep in the Turkish fashion.

Though not an amir, Khwāja Muḥammad b. Arslān al- Khānbāliqī might well have been a minister as suggested by his title, Khwāja. His tombstone (A15) indicates his death in March 1317 and his unusual nisba, al-Khānbāliqī, suggests a signi-.cant attachment to Qubilai Khan’s new capital founded in 1272. Another Khānbāliqī, recorded as Shaykh ʿUmar,TJ was buried in Zaytun (Quanzhou) in 1302, but no other information has yet been forthcoming. Khwāja Muḥammad’s father’s Turkish name, Arslān, suggests roots in Turkistan, which better explains the use of the term ‘martyr,’ since, if Khānbāliqī had been his ‘ancestral’ home, Hangzhou, a relatively short distance to the southeast, could hardly have been considered an exile. The nisba ‘al-Khānbāliqī’ could well have been bestowed upon Khwāja Muḥammad and possibly his family if he had been appointed to an in)*uential and central position in the new administration, or if the family had served as high-ranking o8-.cials. Perhaps pertinent to this o8-.cial is the existence of a commemorative stele dated 742/1341–2 found in Qaraqorum. It lists the names and contributions made by a number of pious Muslim individuals, most of whose nisbas suggest their Central Asian origins. This remarkable stele is signed by a certain Sharaf al-Dīn bin Muḥammad Khānbāliqī.TT

YX Chen and Kalus, no. 52, 137–38, pl.xxxiv; Chen Dasheng, -.g. 39, 34, 52. YY Ibid. ?@4 A@B43CD EBFGH4 5pr

The Phoenix Mosque Steles

The following is a list of the tombstones to be found in the Phoenix Mosque.TU Numbers A1 to A14 stand independently with both sides clearly visible and accessible. Tombstones A15 to A20 are considerably smaller and are set into the wall, with only the front visible and accessible. The following list brie)*y describes each tombstone and the similarities in wording will be maintained.

A1 29 Rabīʿ ff 707/5 November 1307, Khwāja Ḥusām al-Dīn son of Khwāja Yaghān? Ṭughril Bek, #xyz{x? The presence of the verse “The death of the exile is martyrdom”TV and the reference to travel suggests that this shābb or young man was a merchant. The date occurs on both sides, though the texts are di8ferent, with Arabic verse appearing at the bottom of the reverse side. A2 The lower portion of the inscription is missing. No personal details or date, though there are references to ʿAlī on the front side. The reverse contains both Arabic and Persian verse. A3 This is a tombstone for another merchant. Once again, the lower portion of the inscription is missing and the text lacks a name and date. From the inscription, it is clear that the deceased was a young man and evidently a merchant who had travelled and visited (made pilgrimage to) the righteous. He had journeyed to the East and the West and speci-.cally had travelled in Syria, Iraq, and Najd, and performed the lesser pilgrimage. He is famed in the cities. Much of the wording of this epitaph is identical to that found in corresponding part of the epitaph of Maḥmūd Simnānī (A7), which could point to a connection between the two as either business partners or family members who travelled together. The common      verse, “The death of the exile is martyrdom” (      ) underlines the deceased’s status as an expatriate. On the reverse side of the tombstone, the inscription consists of one Arabic verse and an incomplete Persian qaṣīda, opening with an invocation to God. The qaṣīda ends with a prayer pleading for God to

YZ I am indebted to Guo Chengmei and Guo Qunmei for the classi-.cation of the tombstones and to the late Sandy Morton, who was able to transcribe and translate them. Y\ This and similar sayings attributed to Muḥammad do not appear in these words in Wensinck’s concordance of Prophetic Ḥadīth, but clearly appealed to the expatriate Muslims of Yuan China. They occur on several of the Hangzhou tombstones (for example, Number 3), and on several more in the neighbouring province of Fujian. For the concept of types of ‘martyrdom’ other than that of death in righteous battle see the Encyclopaedia of Islam (Alexander Morton). 5p7 1234

grant mercy on ‘me,’ presumably referring to the poet or the reciter of the poem. A4 With only the top of the inscription surviving, no personal details are discernable. However, the reverse has the beginning of the same Persian poem as appears on A3. A5 The tombstone of Imām Tāj al-Dīn Yaḥyā the preacher (al-wāʿiẓ), son of the Imām Mawlānā Burhān al-Dīn the preacher, unfortunately lacks any dates. Both son and father are de-.ned as preachers (wāʿiẓ) and the deceased is described as a young man (shābb) who died aged forty-one. While some Arabic verses appear on the front, the reverse side consists solely of pieties. A6 Damage to the top and bottom of this tombstone has eliminated any dates concerning the deceased, Shihāb al-Dīn Aḥmad bin ʿAbdullāh, descendant of the great family of Muḥammad Ḥalabī. The deceased’s name only survives in the Persian poem on the back of the tombstone, where his virtues are compared with those of the -.rst four Caliphs, indicating that he is Sunni and that his virtues entitle him to the status of Khwāja. The two sides of the tombstone are strikingly di8ferent. Not only do the calligraphic styles of the script on the front and back di8fer, but the intricate and distinctive border designs clearly visible on the reverse side of the tombstone are absent from the front. The border design on the reverse consists not only of the usual )*oral arabesque but of an inner border of a more unusual plain geometric pattern. Both borders enclose a far less formal and stylistic script that that found on the front of the tombstone. A7 Only the initial ‘7’ of the year remains of the date appearing on the tombstone of the Su-. merchant Maḥmūd b. Muḥammad b. Aḥmad, known as Tāj Malīḥ (?) al-Simnānī. Damage to the bottom of the inscription has obscured the -.nal line/s of the inscription containing the date. Though the title shaykh might have been awarded in recognition of his age and status, the text makes it plain that he was a practising and respected Su-. as well as a very widely travelled merchant. The inscriptions on the reverse of the tombstone record that he had travelled to the East and West, visited Syria, ‘Iraq and Najd, performed the lesser pilgrimage (ʿumra), is known in the cities and known to the kings of the regions of al-‘Irāq. He was a contemporary of the well-known Su-. poet, ʿAlāʾ al-Dawla Simnānī (1261–1336), who had once been a boon companion of Arghun Khan (1258–91, r. 1284–91) and whose family were courtiers and ministers at the Ilkhanid court. Coming from the same comparatively small, provincial town, Simnan, the families, if not related, would have been acquainted; ?@4 A@B43CD EBFGH4 5pI

this would explain Tāj Malīḥ al-Simnānī’s prestigious connections and the respect he commanded (see also above).

[Line 8.] and he is the Shaykh, the most noble, the great, the generous, the distinguished, the long-lived, the eminent, the honoured, dear to the great and the noble, source of favour and justice, [Line 9.] pride of the merchants, ornament [?] of the good and the noble, famous in the cities, re-.ned of character, pure of manner, familiar among the princes of the regions of the coasts, [Line 10.] . . . Splendour of Islam and the Muslims, privileged with the solicitude of the Lord of the Worlds,TW

Like his neighbours in the Jujing cemetery, Tāj Malīḥ al-Simnānī achieved martyrdom in death, as suggested by the same Qur’anic phrase (“The death of the exile is martyrdom”) as that included on several of the other stones. It is interesting to note that the front of the tombstone (A7a) contains verses attributed to ʿAlī ibn Abī Ṭālib.

A8 Neither side of this tombstone has any decipherable inscriptions. Floral decoration adorns the front side (A8a), while the other side is completely indecipherable—if indeed the apparent markings on its surface were originally inscriptions rather than abstract or )*oral designs. A9 The tombstone is clearly dated 6 Rajab 716/2 October 1316, and the name Khwāja Shams al-Dīn Muḥammad b. Aḥmad b. Abī Naṣr al-Iṣfahānī, a merchant, is easily discernable. The tombstone remains in excellent condition and a full translation and commentary has already been published by Alexander Morton.U` The khwāja had been given the title of ‘shaykh’ possibly out of respect for his age and position in the community rather than for his spiritual achievements. The reverse side of the stone, containing the few biographical details, has an elegy in Persian on the deceased, who is referred to as Khwāja Muḥammad, Jewel of Isfahan, “the pride of the merchants, famous in the cities, . . . known to the kings of al-‘Irāq.”U, It is further stated that he came to China from the kingdom of Iraq.

Y] From Alexander (Sandy) Morton’s notes and unpublished translation of the tombstones. Za Morton, “Muslim Gravestones in the Phoenix Mosque in Hangzhou,” 197–200. Z/ Morton, “Muslim Gravestones,” 200. 5r_ 1234

A10 Both sides of this headstone have been damaged and the top lines are missing. However, the year and month, Rabīʿ f, 752/ May 1351, are legible, as is the name of the deceased, Maḥmūd b. Muḥammad b. Jamāl al-Dīn al-Khurāsānī. Deciphering is more di8-.cult because the text was carved in relief and the resulting ‘rubbings’ are far from distinct. A11 The headstone of Amir Bakhtiyār b. Abī Bakr bin ʿUmar al-Bukhārī has already been discussed in detail above, and a translation into French appeared in 1945.U9 Signi-.cantly, a photograph and transcription of the tombstone was published revealing a -.nal line of text which has since been lost after the stone was badly damaged. Once again, Alexander Morton’s work has revealed these lost details and he concludes that three dates are given: the Islamic date of 21 Shawwāl 730/7th August 1330, the Turkish date, 23 of month 7 of the year of the sheep, and the year in the Chinese fashion, using the Emperor Toqtemür’s era name.U: The reverse side is devoted solely to the illustration of an incense burner surrounded by )*oral arabesques. A12 Khwāja ʿAlāʾ al-Dīn bin Khwāja Shams al-Dīn al-Iṣfahānī died on 23 Jumādā ff, 727/24 May 1327. It is probable that he is the son of Khwāja Shams al-Dīn Muḥammad (A9), whose headstone is of far better quality that that of his o8fspring. In the list of epithets, he is called al-shahīd, or martyr. No personal details are given in the verses in Arabic and Persian on the back. The Persian verses are introduced by the heading tarjama (‘translation’), but it is only the last two lines corresponding to the -.nal line of the Arabic couplet, attributed to ʿAlī on other tombstones, which are translated. Like his father, Khwāja ʿAlāʾ al-Dīn was a merchant, but— judging by the quality of the calligraphy—not as successful as his father. A13 The status of the military o8-.cer Amir Badr al-Dīn, son of al-Ṣadr, is re)*ected in the high quality of the calligraphy of this beautifully executed tombstone. Unfortunately, damage to the base of the stone has not only obliterated any dates but also the name of the Amir’s father, leaving only the honori-.c ‘al-Ṣadr,’ usually awarded to civil o8-.cials and other lay dignitaries. The actual script is large and generously adorned with punctuation and vocalization symbols; this is particularly the case with the bold, imposing Qur’anic quotations which head the texts and domi- nate both sides of the stone.

Z< Yugong banyuekan (The Evolution of Chinese Geography) August 1936; Répertoire chronologique d’épigraphie arabe, xiv (Cairo, 1954), 272. (Here the tombstone is mistakenly said to be in Pekin, the -.nal line of the inscription is omitted and the reference is wrongly dated 1935).      . ’, transliterated in Persian characters؟ ؟ ؟  ؟،“ Z= ‘ }" Sh ~ " x ?@4 A@B43CD EBFGH4 5r(

A14 The name Khwāja Jalāl, in whose honour this tombstone was constructed, appears in a later part of the poem composed in Persian, of which twenty-three lines of the original twenty-six survive. The -.rst twenty lines celebrate the power of God as creator and the wonder of His miracles, as recorded in Muslim tradition. In the view of Alexander Morton, the poem demonstrates the competence of Hangzhou’s Persian wordsmiths, an ability considered equal to that of their fellow Persians in the West. A 15 Dated 5 Muḥarram 717/19 March 1317, the tombstone of Muḥammad b. Arslān al-Khānbāliqī has already been commented upon and the signi-.- cance of the nisba emphasized. This Turkish family was highly in)*uential.

efgHh4 ((.6 The tombstone belonging to Khwāja Muḥammad bin Arslān Khānbāliqi, d. 1317. $@B?B kl ?@4 2H?@Bh. 5r5 1234

A16, A17, A19. These small headstones, which are also set into the wall, are mainly or completely illegible. Though no text or traces of script appear on the surface of A17, a )*oral border design is still clearly visible, while A16 contains some religious verses. A18 This small tombstone, now set in the wall of the mosque outhouse, commemorates an ʿālim, one learned in Islamic law. However, it is probable that the two fragments of text are probably independent of each other; the lower fragment, for instance, has been arbitrarily set within the missing portion of the headstone and cemented into place. Both fragments contain sayings attributed to the Prophet expressing God’s grief at the loss of an ʿālim. A20, A21. Both contain script, but the words are not clearly discernible and more work is needed on both these tombstones. A20 is small and set into the wall of the mosque storage room, while A21, a double-sided stone comparable in size to the other tombstones A1–A15, is housed nearby in the Archaeological Institute, Hangzhou. The textual inscription on A20 appears to have been deliberately obscured manually. What is particularly intriguing is that the script which has been scratched away is unlikely to be Persian, Arabic or Chinese. No tails or dots, characteristic of Arabic and Persian, are discernible, and the direction of the writing is horizontal, ruling out Chinese script. Why this particular small tombstone was targeted is a mystery. More investigation will be needed to determine what the original script might have been.

Conclusions

It would appear that in a comparatively short period of time, the newly arrived Persian community had managed to establish themselves at the heart of Hangzhou and to have gained the good will of their Chinese hosts. They had been given a corner of the former royal gardens, a peaceful and picturesque stretch of land between the south hills and the lake, to bury their dead. But while their cemetery was idyllic, it was also secluded and out of town. The new mosque, however, could not have been more imposing and it was situated in the heart of the populous, downtown, commercial district. The construction of their mosque fronted by the towering Wang Yue Lao gatehouse would have necessitated the demolition of two popular centres of entertainment. On one side of the derelict site where the mosque was built was the San Yuan Lou, a bar whose bar girls and customers made use of the neighbouring ‘gardens,’ and on the other side there was the Zheng (Middle) Wazi, a popular place of theatrical entertainment. Their demolition to make way for a foreigners’ place ?@4 A@B43CD EBFGH4 5r6 of worship would presumably have angered certain segments of the local community. The Persians were rich merchants so it must be assumed that they brought merchandise or services with them which were highly desirable for and much prized by the local Chinese. One attribute the newcomers would most certainly have had would have been contacts. The Yuan administration was rich in ‘Westerners’; Muslims were found at all levels of government and held ministerial, state and local positions. The in)*ux of sophisticated Iranians after 1260 would have had a profound e8fect on Hangzhou society, as these worldly Persian merchants sought local partners with whom to establish trading links and partnerships. This wealthy and worldly community, who could a8ford to import not only their own calligraphers but also their own poets, impressed their new neighbours, and despite the vitriol of a minority, gained acceptance and a welcome from the local people. Though little remains today of that mediaeval Persian community’s presence, the short time that they were in Hangzhou must have strengthened good relations that had existed elsewhere in the country. The small army of tombstones that has survived for seven centuries by Hangzhou’s fabled West Lake presents the face of a proud, con-.dent and relaxed community.

Bibliography

Bai Shouyi . Hui zu ren wu zhi [Muslim Personalities]. Yinchuan Shi: Ningxia ren min chu ban she, 1985. Brechneider, Emil. Mediæval Researches from Eastern Asiatic Sources: Fragments towards the knowledge of the geography and history of central and western Asia from the 13th to the 17th century. 2 vols. New York: Elibron Classics, 2002 (Facsimile reprint of London: Trübner & Co., 1888). Chen Dasheng. Islamic Inscriptions in Quanzhou (Zaitun). Fujian: Ningxia People’s Publishing House, 1984. Chen Dasheng and Ludvik Kalus. Corpus d’Inscriptions Arabes et Persanes en Chine. Paris: Libraire Orientaliste Paul Geuthner, 1991. Chen Hanzhang. “Zhongguo huijiao shi.” Shixue yu dixue (1926) (reprinted in Bai Shouyi, Zhongguo Yisilan shi cungao, 1982. Ch’en Yüan. Western and Central Asians in China under the Mongols. Translated and annotated by Ch’ien Hsing-hai and L. Carrington Goodrich. Nettetal: Steyler Verlag- Wort und Werk, 1989. Cleaves, Francis Woodman. “The Biography of Bayan of the Barin in the Yuan Shih.” Harvard Journal of Asiatic Studies 19/3–4 (1956): 185–303. Dai Liang . Jiu-ling-shan fang-ji . Shanghai: Shangwu yinshuguan, , 1922. 5r^ 1234

Guo Chengmei and Guo Qunmei. “Textual Research on Hangzhou Ancient Islamic Tombstones written in Arabic and Persian.” (In Chinese). Researches On The Hui 1. China Academic Journal Electronic Publishing House, 1997. Guo Xiao-hang. “The Study and Interpretation of the First Shanghai Da Lu Hua Chi— She La Fu D ing in Yuan Dynasty.” Historical Review 4 (2007). http://en.cnki.com.cn/ Article_en/CJFDTOTAL-LWBI200704021.htm. Hsiao, C. C. “34. Bayan.” In In the Service of the Khan, edited by Igor de Ratchewiltz, 584–607. Weisbaden: Harrassowitz Verlag, 1993. Ibn Baṭṭūṭa. The Travels of Ibn Baṭṭūṭa -. 1325–1354. 4 vols. Edited and translated by H. A. R. Gibb & C. F. Beckingham. London: Hakluyt Society, 1994. Institut français d’archéologie orientale du Caire. Répertoire Chronologique D’épigraphie Arabe, xiv. Le Caire: Impr. de l’Institut français d’archéologie oriental, Cairo, 1954. Jay, Jennifer. A Change in Dynasties: Loyalism in Thirteenth-Century China. Washington: Western Washington University Center, 1991. Juwaynī, ʿAṭā Malik. The History of the World-Conqueror. Translated by John Andrew Boyle. Reprint, Manchester: Manchester University Press, 1997. ———. Ta’rīkh-i Jahāngushā. Edited by Mīrzā Muḥammad Qazwīnī. Vol. 3. London: Luzac, 1937. Khwāndamīr, Habibu’s-siyar. The Reign of the Mongol and the Turk Genghis Khan— Amir Temur. 2 vols. Edited and translated by W. M. Thackston. Cambridge, Mass: The Dept. of Near Eastern Languages and Civilizations, Harvard University, 1994. Komaro8f, Linda, ed. Beyond the Legacy of Genghis Khan. Leiden and Boston: Brill, 2006. Lane, George. “Whose Secret Intent?” In Eurasian In/luences on Yuan China, edited by Morris Rossabi, 1–40. Singapore: National University of Singapore Press, 2012. ———. “Phoenix Mosque.” In Encyclopaedia Iranica; online version at http://www .iranicaonline.org/articles/phoenix-mosque. ———. “Mongol News: The Akhbār-i Moghulān dar Anbāneh Qutb by Quṭb al-Dīn Maḥmūd ibn Masʿūd Shīrāzī.” Journal of the Royal Asiatic Society 22/3–4 (October 2012): 541–59. ———. Early Mongol Rule in 13th Century Iran. London: RoutledgeCurzon, 2003. ———. “Sayyid Ajall.” In Encyclopaedia Iranica; online version at http://www.iranicaonline.org/articles/sayyed-ajall. Leslie, Donald D. Islam in Traditional China: a Short History until 1800. Belconnen #!%: Canberra College of Advanced Education, 1986. Liu Yingsheng and Peter Jackson. “Chinese-Iranian Relations: iii.” In Encyclopaedia Iranica; online version at http://www.iranicaonline.org/articles/chinese-iranian-iii. Masson Smith jnr, John. “Hulegu Moves West: High Living and Heartbreak on the Road to Baghdad.” In Beyond the Legacy of Genghis Khan, edited by Linda Komaro8f, 111–34. Leiden and Boston: Brill, 2006. ?@4 A@B43CD EBFGH4 5rn

Melville, Charles. “From Adam to Abaqa: Qadi Baidawi’s rearrangement of history (part f).” Studia Iranica 30/i (2001): 67–86. ———. “From Adam to Abaqa: Qadi Baidawi’s rearrangement of history (part ff).” Studia Iranica 36/i (2007): 7–64. Morton, A. “Persian Poetry in Hangzhou under the Yuan.” In Quinsay—The City Marco Polo Experienced (), Yuan dynasty Hangzhou research collection, edited by Lizhian and Songtao , 235–38. Hangzhou: Hangzhou Press (), 2012. ———. “Muslim gravestones in the Phoenix Mosque in Hangzhou.” , , In Qinghua Yuanshi , no. 1, edn. f, edited by Prof. Yao Dali, 193–214. Nanking: Qinghua Academy of Chinese Learning, Qinghua University, Dec. 2011. Moule, A. C. Quinsai and Other Notes on Marco Polo. Cambridge: Cambridge University Press, 1956. ———. “Hangchou to Shangtu in 1276.” T’oung Pao 16/1–5, (1915): 339–419. Mustawf̄ Qazw̄n̄, Ḥamd Allāh. The Ta’r&́kh-i-Guz&́da: or “Select History” of Hamduʾlláh Mustawf&́-i-Qazw&́n&́. Edited and translated by E. G. Browne. 2 vols. Leiden: Brill, 1910–13. ———. Nuzhat al-qulūb, edited and translated Guy Le Strange. Leiden and London: Luzac, Brill, 1915, 1919. ———. Tārīkh-i guzīda, edited by ʿAbdul al-Ḥusayn Navā’ī, Tehran: Intishārāt Amīr Kabīr, 1362/1983. ———. Ẓafarnāma, edited by Naṣr Allāh Pūrjawād̄ and Nuṣrat Allāh Rastgār. Tehran: Iran University Press, 1377/1999. (Facsimile of British Library Or. 2833). Polo, Marco. The Travels of Marco Polo. Edited and translated by Henry Yule & Henri Cordier. 3rd ed., 2 vols. New York: Dover Publications Inc., 1993. Nai Te-weng. “Tu ch’eng chi sheng; The Wonders of the City.” Edited and translated by A. C. Moule. The New China Review 3/1 (1921): 12–17, 356–67. Rachewiltz, Igor de, H.-LChan, c.-c. Hsiao Ch’i-ch’in, P. W. Geier, eds., with the assistance of M. Wang. In the Service of the Khan: Eminent Personalities of the Early Mongol-Yüan (1200–1300). Wiesbaden: Otto Harrassowitz Verlag, 1993. Rashīd al-Dīn. Jāmiʿ al-tawārīkh. Edited by M. Rawshan and M. Mūsawī. 4 vols. Tehran: Nashr-i Alburz, 1373/1994. ———. Jāmiʿ al-tawārīkh: The History of the Khutāy Kings’ Family. Edited and annotated by Muḥammad Rushan. Tehran: Mīrās-i Maktūb, 2006. ———. Jami‘u’t-tawarikh [sic] Compendium of Chronicles. Translated by Wheeler M. Thackston. Cambridge, #: Sources of Oriental Languages and Literature 45, Central Asian Sources, Harvard University, 1998. Shīrāzī, Quṭb al-Dīn. Akhbār-i-Monghulān. Edited by Iraj Afshar. Qom: Library of Ayatollah Mar‘ashi Naja-., 2009. 5rp 1234

Sinor, Dennis, ed. The Cambridge History of Early Inner Asia. Cambridge: Cambridge University Press, 1990. Smith Lewis, Agnes. An Inscription Recording the Restoration of a Mosque at Hangchow. Translated by E. G. Browne. Cambridge: Cambridge University Press, 1911. Sturton, Stephen D. “The Site of the Monastery at Hangchow.” The Asiatic Review 41 (1945): 82–85. Tao Zongyi. Nancun Cuogenglu, juan 28, Beijing: Xin shi ji wan you wen ku [no date available]. Tian, Rucheng. Xi Hu you lan zhi yu / [Ming] Tian Rucheng zhuan, jin shi 1526, Hangzhou: Zhejiang ren min chu ban she, 1980; facsimile edition, Taibei : Chengwen Chubanshe , 1983. Thackston, Wheeler, ed. and trans. Classical Writings of the Medieval Islamic World, 3 vols of Mirza Haydar Dughlat, Khwandamir, Rashiduddin Fazlullah, London: I. B. Tauris, 2012. Qian Shuo-you . Compiled 1268. Xianchun Lin’an Zhi [Gazetteer of Lin’an compiled during the Xian-chun reign-period], 3 vols. Taibei : Chengwen Chubanshe , 1970. Vissière, A. “L’Islamisme à Hang-tcheou.” Revue de Monde Musulman 22 (1913): 1–84. Ward, L. G., “The Ẓafar-Nāmah of Ḥamdallāh Mustaufī and the Il-Khān Dynasty of Iran.” PhD dissertation. Manchester: Manchester University, 1983. Waṣṣāf, ʿAbd Allāh ibn Faḍl Allāh Shirāzī. Tajziyat al-amṣār wa tazjiyat al-a‘ṣār, Tārīkh-i Waṣṣāf. Edited by M. M. Isfahānī, Tehran, 1853. Reprint, Tehran: Ibn Sīnā, 1338/1959–60. ———. Taḥrīr-i Tārīkh-i Waṣṣāf. edited by ʿAbd al-Muḥammad Āyatī, Tehran: Muʾassasa-yi muṭāla‘āt wa taḥqīqāt-i farhangī, 1372/1993. ———. unpublished tr. Alexander Morton. Wu Zimu. Meng Liang Lu, juan 16. Hangzhou: Zhejiang Renmin Chubanshe, 1980. Xia Shizheng (1475) Chenghua Hangzhoufu Zhi, juan 47, Ji’nan : Qi Lu shu she, 1996. Yang Xinping . “A Chronological Study of the Beginnings and Rebuilding of the Hangzhou Mosque.” (Hangzhou zhenjiaosi chuangshi, chongjian niandaikao ). The Journal of the Hangzhou Teachers’ College (Social Sciences edition). Issue 3 (1987). Yuan Shi , 880–05 Beijing: Zhongguo guo ji guang bo chu ban she, 2009. Yugong xuehui , Yugong Banyuekan, (The Evolution of Chinese Geography), Taibei Shi: Taiwan Datong shuju , August 1936. Zhou Mi. Gui Xin Za Shi [xujishang ]. Beijing: Zhonghua Shuju , 1988. ———. Wu Lin Jiu Shi: 10 Juan. Xianggang: Di zhi wen hua chu ban you xian gong si, 2006. ———. Zhou Mi’s Record of Clouds and Mist Passing before One’s Eyes: An Annotated Translation. Edited and translated Ankeney Weitz. Leiden: Brill, 2002. !"#$%&' () Mamluk and Mongol Peripheral Politics: Asserting Sovereignty in the Middle East’s ‘Kurdish Zone’ (1260–1330)

Boris James*

Scholars of the medieval Islamic world, as well as those who focus exclusively on the history of the Kurds, have paid scant attention to the impact of border geo-political transformation in the fourteenth century on both the ethnic and ethno-territorial identity of the Kurdish people.* It is the contention of this article that this omission is a grave error, and that the emergence of military and political rivalries between the Mamluks of Egypt and the Ilkhanids of Iran over the Middle East’s ‘Kurdish zone’ constitute a key historical moment in the formation of the Kurdish identity.+ In historical terms, conceptions of Kurdish territoriality most certainly pre-date the fourteenth century; nevertheless, faced with the overawing power of the Mongol war machine and in order to o,fset their military inferiority outside Egypt and Syria, the Mamluks adopted a relatively novel set of strategies and policies towards the Kurdish populated highlands of western Asia. These policies in turn had a signi-.cant impact on Kurdish ‘ethnicity,’ not only raising its political signi-.cance but also territorializing it./

* We gratefully acknowledge the contribution of Andrea Stanton, who read and commented on the -.rst draft of this article. Special thanks go to Mr Djene Bajalan, who has spent a great deal of energy and time struggling with our linguistic defects. 0 Two exceptions can be noted, the works of Ismet Cherif Vanly, “Le déplacement du pays Kurde,” and of Vladimir Minorsky, “Kurds and Kurdistan.” 2 It is worth noting that, although it does not focus speci-.cally on the Kurdish aspects of the issue, Amitai’s work on the Mamluk–Ilkhanid confrontation sheds light on many questions that underlie the present article and provides its historical framework. See, for example, Amitai-Preiss, Mongols and Mamluks. 3 Here, ‘ethnicity’ means a process that binds individuals together and separates them from the rest of the society according to their alleged belonging to an ethnic group. This phenomenon has been successfully described and analysed by Fredrik Barth as resulting from the constitution of ethnic boundaries created by the manipulation of cultural traits by individuals, groups and states. Martiniello, L’ethnicité dans les sciences sociales contemporaines; Barth, Ethnic Groups and Boundaries. On the medieval Kurds, see James, “Arab Ethnonyms (ʿAjam, ʿArab, badū and Turk . . .).”

We will commence our study on the Kurds in this period by examining the formation of notions of a ‘Kurdish zone’ in the centuries preceding the emergence of the Mongol and Mamluk polities, highlighting the often prob- lematic nature of relations between the Kurdish tribes and imperial powers. This study will continue by investigating the ways in which both empires sought to project their power into the mountainous domains of the Kurds. In particular, it will look at the ways in which the Mamluk failure to transform their essentially defensive victories over the Mongols in Syria into o,fensive military successes prompted Cairo to adopt a novel set of political practices on the Mesopotamian frontier. These practices entailed the co-option rather than domination of Kurdish groups resident in this region. The study will conclude by discussing some of the broader implications of these policies on the evolution of the Kurdish community. Arabic and Persian sources of the thirteenth and fourteenth century (chronicles, geographic encyclopaedias, biographical dictionaries) will provide most of the material that supports our demonstration.

The Evolution of a ‘Kurdish Zone’: A Lawless Zone?

Medieval Arabic and Persian sources state that the main Kurdish groups lived anywhere between the Fars region in the east, and the Syrian Jazira in the west, and from Georgia in the north, to Khuzistan in the south.> These same sources also noted that the Kurds inhabited a multitude of environments from warm steppe-like deserts to high snow-capped mountains. However, after an overall survey of the sources between the eighth and the fourteenth centuries, the image that emerges is that of an agro-pastoral population essentially residing in mountainous zones.? Indeed, pre-Mamluk sources used a number of designations to describe the tribal@ lands of the Kurds, including Bilād al-Akrād, Zūzān

A Minorsky, “Kurds and Kurdistan.” B James, “Une ethnographie succincte.” C This article does not intend to address in details the question of tribal organization of the Kurds. The notion of tribe as the notions of qabīla, ashīra or khayl in Arabic and Persian is a very fuzzy concept which usually refers to originally non-urban societies whose members share one or several putative common ancestors. This familial representation partially de-.nes the social and political organization of said groups. The remote nature of the Kurdish territory allowed the formation of many of these tribal societies. That is why we are referring here to a tribal Kurdish territory although we do not consider the tribal organization as being the exclusive social and political inscription of local populations. See James, Les Kurdes dans l’Orient mamelouk et mongol de 1250 à 1340, 187–333. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; )

‘I d es / Sea Ku e Ke ve egion Cities and fortresse Ri Lak R rmia

Shahrazur

Lak

U ā

e t

a i Baghda r ā t e

‘I b

REFERENCES P

Irbil a b

n Z a Dvin r

Akr

Tikrit s

s Kirkuk

Urmi p Va p e

U L is e e gr t Ti an Arjish V Lak Cizr Siir Ahlat Mosul

Bitlis s e t Sinjar a r eyf ph Eu n Mardin Amid Hasank

Erzinca Malaty al-Sham Bilad Aleppo Beirut Damascus Cilicia a Cairo Map of Kurdish dominated territories in the 13th century. in the 13th Map of territories dominated Kurdish anean Se

Mediterr RSTEL: () . ( )=U 789:; al-Akrād and Jibāl al-Akrād. Compared to the nomadic Turkic populations, it seems that the zone of Kurdish tribal movement was relatively restricted. They occupied an intermediary position, living in the territories between the Arabian deserts and the Iranian and Anatolian plateaus. The stability of this state of a,fairs seems to have been disrupted, at least temporarily, by the arrival of Turkic tribesmen in the region during the eleventh century. This resulted in incessant feuding between the two groups over pastoral lands until the middle of the thirteenth century.V Yet, despite this conWXict, Kurds were never entirely displaced from the region, although some regions of Fars and Azerbaijan, which had previously been associated with the Kurds, ceased to be linked to them after the twelfth century. At the same time, notions of Kurdish tribal territory gradually shifted towards the south- west. For example, the region of al-Zūzān (the highlands between modern-day Ahlat and Cizre), in the works of al-Balādhurī and Ibn Ḥawqal in the ninth and tenth centuries, was perceived as an exclusively Armenian territory.\ However, Ibn al-Athīr, writing in the thirteenth century, describes the region as Zūzān al-Akrād (The Zūzān of the Kurds) which suggests, if not a demographic majority, a signi-.cant and inWXuential Kurdish population in that region.] Despite its reputation as a highly destructive event in Islamic history, the Mongol invasions and their domination of Upper Mesopotamia and the Jibal region in the fourteenth century do not appear to have disturbed the transhumant economy of the Kurds. According to John Masson Smith Jr., the northern regions through which the Kurds travelled during the summer, namely the area between the western edge of the Caspian Sea and Diyār Bakr, were considered by the Mongols as part of their lands of Azerbaijan.*^ Indeed, Diyār Bakr and the region of the Zarīna Rūd, to the south of Lake Urmia, were both winter pastures (qishlāq) for the Mongol horde. However, due to the latitudinal and seasonal organization of Kurdish herding practices, as the Kurds moved north- wards so did the Mongols. The Mongols’ summer pastures (yaylāq) were situated to the north in Alātāq and beyond, as well as to the south-east in Sughurluk and Siyāhkūh near Qum.** By no means could the Kurds have competed for these pastures with the new masters of the region, yet the Mongols did not

_ For example, the conWXict that erupted in 1186 between Kurds and Turkomen in Jazīra, Diyār Bakr, Ahlat, Syria, Shahrazūr and Azerbaijan. Ibn al-Athīr, al-Kāmil fī l-ta’rīkh, 10: 136; Ibn Shaddād (Bahā’ al-Dīn), al-Nawādir, 63. ` Ibn Ḥawqal, Kitāb ṣūrat al-arḍ, S, 348; al-Balādhurī, Futūḥ, 176. a Ibn al-Athīr, al-Kāmil fī l-ta’rīkh, 2:379; al-Ḥamawī, al-Buldān, 3:158. 0b Masson Smith, “Mongol Nomadism,” 45–47. 00 Ibid. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; )=( seem inclined to move into Kurdish grazing grounds, which were probably too uneven to maintain a compact army of tens of thousands of horsemen. Indeed, the Kurdish populated regions south of Lake Van, east of Lake Urmia and in Diyār Bakr were not used by the Mongols for grazing, at least, according to one informant, until 1293.*+ Thus, in the fourteenth century, it is meaningful to speak of a Kurdish zone. This is not to suggest that the link between the Kurds and certain territories is ‘primordial’; rather we should regard such links as the result of speci-.c historical processes. At this point, it is worth brieWXy discussing the political and sociological nature of this Kurdish zone. It is important to note that it was not populated by Kurdish entirely nomadic groupings. In his geographic encyclopaedia the Masālik al-abṣār fī mamālik al-amṣār, the Mamluk scholar al-ʿUmarī*/ completely confuses geographic designations and the anthroponomy of the tribes by assigning tribal names to geographic landmarks. The author or his informant travelled across the region of Jibāl and enumerated the names of the Kurdish tribes to be found on his route. Passing from one tribe to another, that is to say, from one area to another and moving closer to the Jazīra, he writes: “After these [tribe A] we -.nd the [tribe B].” The vocabulary used is an indication of a lack of movement of certain tribes. These tribes’ members lived in (yaskunūna), owned (ladayhim, biyadihim . . . ), governed (yaḥkumūna) or were residents (muqīmūn) of this or that region. Nevertheless, the settlement of the tribes in a territory did not imply the total absence of movement and was often linked with a transhumant lifestyle. There are abundant examples of Kurdish transhumance during the Mamluk period as well.*> The sources, especially al-ʿUmarī, describe livestock farming as one of the characteristics of the Kurdish lifestyle, with sheep farming being the most widespread.*? In addition to more peaceful economic activities, the Kurdish population of the Pre-Ottoman period was also associated with banditry. Indeed, although Kurds pursued various economic activities such as pastoralism, agriculture and trade, the image the Arabic and Persian sources transmit of them is essentially that of bandits and raiders. The raider economy was a ‘savage’ mode through which many Kurdish groups sustained themselves; and in a broader sense, it was a way of life associated not only with the Kurds but also other ‘bed- ouin’ peoples (in the Khaldunian sense).*@ Such practices are, of course, only

02 Ibn al-Munshī, Nuzhat al-naẓīr, 19b. 03 al- ʿUmarī, Masālik al-abṣār, 3:124–35. 0A Ibid., 126. 0B Ibid., 129, 134, 127. 0C Martinez-Gros, Ibn Khaldoun et les sept vies de l’Islam. )=) 789:; possible if other groups provide goods, and, in the long run, this is not viable in a closed economy. At a local level, the di,fusion of such a practice results in the WXight of all the productive elements such as craftsmen, peasants, merchants and livestock farmers. The ‘geographic bottleneck’ that existed to the south of Lake Urmia was one of the few ways for the caravans and merchants to cross from Iran to Mesopotamia and Anatolia. The narrow and uneven nature of this territory allowed the Kurds to seize loot and taxes without fear of retaliation from the regional powers, which were either uninterested in the problem or unable to solve it.*V Without accepting uncritically the widespread topoi of the Kurds as brig- ands and looters, we can envisage, from a macro-geographic perspective, the reasons for the perpetuation of plundering as a speci-.c Kurdish trait in the Middle Ages. Many factors were favourable to the rise of a raiding economy. It appears that the presence of numerous trade routes crossing the Kurdish territory o,fered a tempting target for potential bandits. At the same time, the speci-.c political and geographical con-.guration also helped promote banditry. More precisely, as a territory remote from the regional powers, consisting of a series of valleys and peaks and constituting both a barrier and gateway between two macro-regions, raiders could carry out their activities with relatively little risk. Under the Ayyubids (twelfth and thirteenth centuries), the local domination of the Begtikinid Turkoman dynasty seems to have tempered banditry among the Kurds, even though Ibn Jubayr (d. 1217) and Yāqūt al-Ḥamawī (thirteenth century) point out the risks incurred by the travellers who dared to pass through the Jazīra and Shahrazūr.*\ Both the overthrow of the Ayyubids in Egypt and Syria and the ongoing conWXict between the Mamluks and the Ilkhans helped create a political vacuum which probably triggered the resumption of banditry and plundering at the end of the thirteenth and the beginning of the fourteenth century. Al-Qazwīnī (d. 1282), in the Āthār al-bilād wa-akhbār al-ʿibād, considers the inhabitants of the Shahrazūr succinctly as ‘Kurdish road cutters’ (quṭṭāʿ al-ṭarīq), namely highwaymen, a representation that followed the Kurds through the medieval historiography and beyond. In the fourteenth century, Ḥamd Allāh Mustawfī Qazwīnī, in his Persian work Nuzhat al-qulūb, describes

0_ Percy Kemp’s comment on Khayrallāh al-ʿUmarī’s account in the seventeenth century depicts a situation that is similar in many ways to that of the Kurds in the fourteenth century. The Mosuli writer depicts the insecurity and economic instability in the Kurdish and Yezidi environment of Mosul’s hinterland, portraying the area as covered with predatory enclaves. Kemp, Territoire d’Islam. 0` Ibn Jubayr, Riḥla, 240; Yāqūt al-Ḥamawī, Muʿjam al-buldān, 8:375–76, “Shahrazūr.” 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; )=c the inhabitants of Darband Zangī near Zāb, in the region of Kurdistan, in the same way: “They are robbers highwaymen [rāhzan = quṭṭāʿ al-ṭarīq] and base people.” Al-ʿUmarī refers to the exactions of the Kurdish tribes during the same period. He notes that the Bābīriyya and Lūsiyya tribes, who had left Shahrazūr because of the Mongol invasion, caused great destruction in Syria. However, quoting the Sūrat al-raʿd, the author suggests that the misfortune that descended upon them was merely a just divine punishment: “every end has a divine decree [likulli ajalin kitābun].”*] It should be emphasized here that these remarks signal a negative judgment against the Kurds in general; the view of a member of an urban elite looking up and down on the periphery of his world. It also reWXects the wide di,fusion of a social, economic and political order at the core of which were military activities. The conception of the Kurdish bandit indicates neither a situation of absolute disorder nor their predilection for informal practices, but a social economy of war in which the imperial states or local powers were not central players. Indeed, medieval authors often highlighted the fact that Kurdish citadels were remote, inaccessible and impregnable (at least to the imperial state). In addition, many examples are to be found in Arabic sources showing that the tribes were able to evade and resist the power of the surrounding states, an ability which was aided by geographical factors.+^ Thus, in a general sense, we are presented with an image of the Kurdish zone as essentially lawless, as it was simply not possible for the states to impose ‘order’ on the region. Again, it should be reiterated that such a view was very much informed by the biases of particular authors and did not necessarily reWXect the actual state of a,fairs within the Kurdish zone. While we are presented with a picture of a lawless region, it was only lawless from the perspective of the states that were unable to impose their order on the region. Since the beginning of Arabic historiography, Kurdish tribes have been depicted as -.ghting each other and, indeed, conWXict might almost be described as the central structural function of Kurdish society at that time. However, this scheme could not be perpetuated

0a The appearance of the term qāṭiʿ al-ṭarīq is not insigni-.cant. It was in fact a juridical category. The qaṭʿ al-ṭarīq (banditry) is one of the -.ve breaches of Islamic law punishable by the ḥadd. It is a “crime against the religion” as it was mentioned in the Quran. The culprit was to have his foot, hand or head cut o,f, or was to be cruci-.ed. It is noteworthy that al-ʿUmarī never uses such a term to qualify the Kurdish groups, since his aim was some- how to rehabilitate the Kurds. al-Qazwīnī, Kitāb āthār al-bilād, 266; Mustawfī, Nuzhat al-qulūb, 107; al-ʿUmarī, Masālik al-abṣār, 3:126; Schacht, Introduction au droit musulman, 147–51. 2b al- Ḥamawī, Muʿjam al-buldān, 3:375–76, 4: 278; al-ʿUmarī, Masālik al-abṣār, 3:130, 131. )=d 789:; without a mechanism of conWXict resolution that re-establishes the balance, namely a law of war and peace. The war cannot be total as its raison d’être lies only in its resolution.+* When two tribes were at war they were bound to come to terms.++ The one that was considered to have lost asked for mercy (amān), displaying its submission to the other without disappearing and without losing face. Alliances between Kurdish tribes were also possible, especially in the situation of a struggle against external enemies such as the Turkish Ghuzz (Oğuz) during the eleventh century.+/ Cordial neighbourly relations between tribes could also exist during the Mamluk period between the Markawān on the one side, and the Zarzāriyya and the Jūlmarkiyya on the other. The Jūlmarkiyya, who were not of Kurdish origin, “mingled with the local people at the moment of wars (‘inda isti‘māl al-ba’s) trying to establish peace with their enemies. [ . . . ] They adopted the Kurdish lifestyle (inkharaṭū fī suluk al-Akrād) and thus lived in peace.”+> Through al-ʿUmarī’s text we can see that the Jūlmarkiyya survived by accepting the local social order and consequently became Kurds. The cycle of tribal war can also be explained by the three core tribal values: 1) Solidarity within the tribe (based upon the genealogical principle);+? 2) Hospitality; 3) Generosity with the external world. Long before Ibn Khaldūn, authors used the term ʿaṣabiyya—often translated as ‘esprit de corps’—to designate this principle of solidarity. At -.rst glance, it seems to have pejorative connotations. It evokes the idea of solidarity; a solidarity between seditious people and against universality. Yet, at the same time, it is valorised by its frequent association with the term murūʼa, a term which resembles the normative conception of ‘chivalry,’ a social convention (i.e. an unwritten law) encouraging generosity, courage and uprightness.+@ These human qualities are the values and ethos which are established as rules within

20 It is possible here to draw a parallel with medieval Europe. Dominique Barthelemy has shown how the negative depiction of “feudality” (the Knight society) as a savage and unorganized violent society tempered only by the Church’s intervention is a misleading conception. Based on a wide range of medieval primary sources, his study deconstructs what he calls the “black legend” of the knights and demonstrates the existence of an order and a balance in the exercise of violence based on revenge, “a permanent ideological construction” called the knightly feud (la faide chevaleresque). Barthelemy, Chevaliers et miracles, 15. 22 al- ʿUmarī, Masālik al-abṣār, 3:133. 23 Ibn al-Athīr, al-Kāmil fī l-ta’rīkh, 8:177–78. 2A al- ʿUmarī, Masālik al-abṣār, 3:131, 132. 2B On nasab see, Szombathy, “Genealogy in Medieval Muslim Societies.” 2C al-Qazw īnī, Kitāb āthār al-bilād, 289. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; )=f these tribal societies. The solidarity within the tribes implies the participation of all the members in a lex talionis ethos. Moreover, any failure by a particular tribe or individual to respect the unwritten code of hospitality and generosity would lead to the violent intervention of other tribes and individuals.+V These principles suggest a fairly sophisticated system of social norms and political practices, symbolic frameworks governing the form of social organisation amongst the Kurds and restricting the possible modi-.cation of its foundations. These unwritten rules were also a shield against outside intervention. They are the guarantor of the tribes’ autonomy. In short, the Kurdish zone was far from ‘lawless’; it was merely beyond the purview of the order of imperial states. Hence, the image of outlaws that the Kurds bore was precisely due to the fact that they had their own laws which were in contradiction to the Mamluk and Mongol orders, even when they were formally subjected to it. This leaves us with the question of how the Mamluk and Mongol imperial states sought to project their power into the Kurdish zone. Here, ‘imperial state’ is de-.ned as all those political actors that possessed economic, military, symbolic and political power that can lead to either the radical modi-.cation or the tenacious maintenance of existing social and political structures. Speci-.cally, it refers to the Seljuqid, Ayyubid, Mamluk and Ilkhanid sultanates, which constituted distinct political and territorial units and which claimed ‘a monopoly of legitimate violence.’+\ These policies were able to militarily subjugate the tribes, to co-opt local leaders, to assign them to speci-.c roles (such as tax collectors) and facilitate tribal restructuring. In this sense, the two main imperial actors of the fourteenth century were the Mamluks and the Mongols, both of whom possessed di,ferent attitudes towards the Kurdish tribes and thus pursued dif- ferent strategies in their dealings with them.

2_ These mechanisms have been described many times by European ethnographers. Among them, Marcel Mauss brilliantly depicts the practice of “gift and counter-gift” in south Paci-.c societies. See Mauss, Essai sur le don. 2` With the term ‘imperial state’ I do not refer to the ‘modern state’ as de-.ned by the existence of -.xed laws binding ‘governed individuals as well as rulers’ in citizenship and consequently suppressing the form of personal obedience; rather I refer to the State as a structure of legitimate violence according to Weber’s and Pierre Clastres’ conception. This assertion might look a bit anachronistic. However, Ayyubids, Mamluks and Ilkhanids created or reinvested institutions (administrative o,-.ces, judicial courts etc.) that aimed at legitimizing the exclusive exercise of their violence. Moreover the existence of an o,-.- cial literature reinforced this process. See Clastres, La société contre l’Etat, 173; and Colliot- Thélène, Le désenchantement de l’Etat, “Introduction.” )=g 789:;

Mongol Policies: From Conquest to Economic Integration

In historical terms, on the eve of the Mongol conquests, ‘imperial’ power in the Kurdish zone was waning. The Turkoman dynasty of the Begtikinids, who had ruled over Irbil, disappeared in 1233, and its lands were inherited by the distant Caliph of Baghdad. Simultaneously, Badr al-Dīn Lu’lu’ and his family replaced the Zengids of Mosul and Sinjār. However, during the Mongol invasion, the dynasty of Badr al-Dīn Lu’lu’+] and the Caliph were both overthrown and by 1262, there was very little chance that these dynasties would be able to a,fect a restoration of their power./^ Subsequently, the Mongols sought to subjugate the Kurdish zone. These initial campaigns of conquest were the historical moment in which the most violent confrontations between Mongols and Kurds occurred. According to Ibn al-ʿAmīd:

In 1256 . . . Hülegü, took over the Ismaeli territories in Iran; He conquered the fortress of Alamūt and killed all its occupants including the master of this place who was also their master and the chief of their calling (daʿwa). [ . . . ] Then he decided to rid the country of Kurds, Turcomen and Shahrazūriyya. He sent Kitbūghā to the Kurds land where they have entrenched themselves in mountains and caves. [ . . . ] As for Kitbūghā [the envoy of Hülegü], he took the Land of the Kurds (Bilād al-Akrād) and their citadels which he destroyed while most of them WXed to Syria./*

It is worth emphasising that, from 1260 to 1320, the Kurdish zone was almost exclusively under the inWXuence of the Mongols. The Mamluks led only a few unsuccessful raids into these remote regions, despite the fact that from al- Malik al-Nāṣir b. Qalāwūn’s (1309) assumption of power in Egypt onwards, the Ilkhanids experienced a relative decline in these areas. From the death of Hülegü in 1264 until the beginning of Ghazan Khan’s campaign in Syria in 1293, the Mongols had only formal control of Diyār Bakr. This state of a,fairs bene-.ted the Seljuqs of Rūm, who gained some ground in the region. However, after Ghazan subjugated the Kurdish tribes as well as the other local polities, Mongol groups established themselves more -.rmly in the region. Once again, the encounter with the local Kurdish polities was violent. The Kurds and the Ayyubid principality of Ḥiṣn Kayfā (Hasankeyf) had to wait

2a Patton, “Badr al-Din Lu’lu’.” 3b Ibid., 99. 30 Ibn al-ʿAmīd, al-Makīn, al-Majmūʿ al-mubārak, 98. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; )=< for the takeover of Öljeitü Khudābanda in 1303, for more conciliatory and integrating policies from the Ilkhanids./+ In broad terms, Mongol policies towards the Kurdish zone underwent considerable change between the reigns of Hülegü and Abū Saʿīd (the latter r. 1316– 35). This evolution most likely reWXected the economic integration of Kurdish lands and tribes into the Ilkhanid Empire’s trading zone. In this respect, the presence of important interregional trade routes passing through or by the Kurdish zone is signi-.cant. The Silk Road caravans that linked Aleppo, Damascus, Mosul and Bagdad to Central Asia skirted around the Zagros Mountains to the north and through foothills to the south of Lake Urmia, towards either the Taurus Mountains or Mesopotamia, right in the middle of the Kurdish regions. According to Stefan Heidemann, thanks to the stable political situation created from the Seljuq period until the end of the Ayyubid period, the merchants frequently took the Silk Road between Aleppo, other areas of Syria and Iraq.// R. Amitai shows that trade and travel were a,fected and fell during the conWXict between Mongols and Mamluks (1260–81)./> In this period, merchants avoided crossing the Euphrates valley and travelled via Cilicia in order to reach the Mediterranean coast. Three factors minimized the impact of these changes for the Kurdish regions. First, the period in which trade exchanges between the two empires fell did not last long and it seems, according to R. Amitai, that under the rule of Qalāwūn (1278–88) and thanks to the initiative of the Ilkhan Aḥmad Tegüder (1282–84), trans-frontier trade restarted with renewed vigour. Second, in avoid- ing the Euphrates, the caravans still had to cross Kurdish lands on their way to Cilicia, and third, even if the merchants could not reach Mamluk territory from Iran, they were providing the markets of Upper Mesopotamia, controlled by the Mongols, with a huge amount of goods, especially the WXourishing city of Mosul. Intra-Ilkhanid trade was signi-.cant and was bound to inWXuence the Kurdish regions. The Baghdad–Maragha and Baghdad–Hamadan roads were known and had been frequently traversed since at least the Seljuq period. The presence of these networks had required the creation of forti-.ed anchorage points in the Kurdish region, for instance the two Khuftiyān (Rawanduz and Qala Diza). During the Ilkhanid era, these citadels were under Kurdish Zarzārī control./? It is noteworthy that, at least during the reigns of two Ilkhanid Monarchs, Ghazan

32 Ibn al-Munshī, Nuzhat al-naẓīr, 17b, 18, 19b, 23a. 33 Heidemann, Das Aleppiner Kalifat, 35; Heidemann, Die renaissance der Städte, 338–39. 3A Amitai, Mongols and Mamluks, 207–13. 3B al- ʿUmarī, Masālik al-abṣār, 3:129–30. )== 789:;

(694–703/1294–1304) and Öljeitü (703–16/1304–16), the centre of gravity of the state was situated in northwest Iran in a Greater Azerbaijan. The most important cities (Maragha, Tabriz), including other political capitals (Sultaniyya, Ujan), were located there. Thus they were separated from the big cities of the former Muslim potentates (Mosul, Baghdad) by the Kurdish zone, which some- how became an articulation point of the western Empire. For instance, Charles Melville has shown the use that Öljeitü made of the road from Sultaniyya to Baghdad which passed through Hamadan and Chamchamal, and another road that went to Mosul from Maragha./@ Thanks to the numismatic works of Judith Kolbas, we know the importance of the monetary dimension of these regions’ economies. Mosul, Mardin, Irbil and to a lesser extent Sinjar, Mayyafariqin (Silvan), Ḥiṣn Kayfā (Hasankeyf), Jazīrat Ibn ʿUmar (Cizre) and Siʿird (Siirt) were, at least during the second half of the thirteenth century, important centres within the Mongol Empire for the minting of gold, copper and silver coins./V Judith Kolbas ends her study in 1309, right at the moment of the decline of the Mongols’ inWXuence in the region. Indeed, in order to demonstrate the progress made by the Mamluk sultan al- Malik al-Nāṣir b. Qalāwūn (1309–41) towards ‘restoring’ Mamluk inWXuence in the east, the Egyptian historian al-Maqrīzī (1364–1442) indicates that “in 741 (1341), before his death, coins were minted in his name, in Baghdad [. . .] in the land of Ibn Qaramān (Central Anatolia), in the Jibāl al-Akrād (the Mountains of the Kurds) and in many eastern countries.”/\ This is not to suggest that a speci-.cally ‘Kurdish’ coinage could have existed, but rather that the Kurdish tribes and polities linked to the cities were operating within a strongly monetized environment even in the places that pre-Ottoman historians would have considered less ‘civilized.’ As mentioned above, it was these economic changes which framed the evolution of Mongol policies in the Kurdish zone. This implies di,ferent phases in the tax collection policies of the Mongols. It is di,-.cult to know which speci-.c taxes were enforced during Hülegü’s domination. An interesting account is to be found in Ibn al-Munshī’s chronicle./] According to this author, Hülegü’s taxation system resembled extortion or direct exploitation. It resulted in violent depredation of the lands as well as the seizure of goods and tributes from the local polities who submitted to the Khan. The Kurdish zone was a space to be conquered and re-conquered. After Hülegü’s death (663/1263), predation

3C Melville, “The itineraries of Sultan Öljeitü.” 3_ Kolbas, The Mongols in Iran, 131. 3` al-Maqr īzī, Khiṭaṭ, 4:212. 3a Ibn al-Munshī, Nuzhat al-naẓīr, 18–23. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; )=Q became more common, as the Mongol troops left Diyār Bakr. Ghazan’s reign opened a new period of conquests and devastation. In 1293, he led a campaign against Syria in which he passed through Diyār Bakr. Soon after Ghazan’s defeat by the Mamluks, the Mongols reoccupied Diyār Bakr. Each Mongol group stationed in Diyār Bakr was assigned a Shiḥnakiya (prefecture) over a region. To punish the country’s people for their failing to cooperate with the Ilkhanids and their collaboration with the Anatolian Seljuqids, these groups plundered Diyār Bakr. Nevertheless, the level of violence and depredation probably decreased with the establishment of a niyāba in Diyār Bakr, an authority responsible for the collection of Mongol taxes in the region. During Ghazan’s campaign in Syria, a Kurdish Dunbulī (Bukhtī) Amīr, Abū Bakr, contacted the Khan, who handed him the niyāba as he had provided good advice and knew well the internal a,fairs of Diyār Bakr. This meant he had the power to collect taxes from the potentates (ḥukkām), a -.xed amount of money and goods he had to bring every Nowrūz (New Year) to the horde in Tabriz. He was assisted in his task by a group of Mongols and Turkomans who took their share and caused trouble by interfering in local a,fairs. These troubles seem to have stopped at Ghazan’s death. Öljeitü o,fered to -.x the amount that every potentate had to pay, and let them bring the taxes to Tabriz without sending either the shiḥnas or the nā’ib. His successor, Abū Saʿīd, perpetuated the deal. These conciliatory and stabiliz- ing policies probably resulted in a greater integration of the Kurdish zone in the Ilkhanid Empire. Written at the end of this process of integration, between 1335 and 1340, Ḥamd Allāh Mustawfī’s Nuzhat al-qulūb gives an interesting picture of the -.scal situation of the Kurdish lands.>^ For instance, the revenue collected from the province of Fars>* (a non-Kurdish territory) amounted to 3,000,000 dīnārs. This can be compared with the amounts extracted from heavily Kurdish-populated regions such as the provinces of Kurdistan (Hamadan and Shahrazur), ʿImadiyya (Amedi), ʿAqr (Akre), Irbil, Nihavend, Abhar and Sinjar, which furnished in total 509,900 dīnārs>+ on the one hand, and Diyār Bakr and Diyār Rabī‘a, which furnished 1,920,000 dīnārs on the other. Unfortunately, due to the lack of data concerning the population in any of these provinces, it is di,-.cult to know whether Kurdish-populated regions contributed more

Ab Mustawf ī, Nuzhat al-qulūb, 107. A0 Aigle, Le Fârs sous la domination mongole (,---.–,-/. S.), 77. A2 Mustawf ī, Nuzhat al-qulūb, 59–60, 74, 102–3, 105. )QU 789:; or less than average to the imperial treasury.>/ Certainly, Diyār Bakr, probably the most integrated Kurdish region within the empire, was contributing large amounts of currency; and, more generally, the Ilkhanids were able to extract signi-.cant amounts of revenue from the Kurdish zone. In summation, the usage of coercive force against the Kurdish tribes was at -.rst mainly the result of the Mongols’ conception of power, which did not tolerate any other kind of sovereignty. In theory, any autonomous tribe should be destroyed.>> However, despite this, the Kurdish tribes seem to have weath- ered the initial Mongol storm and, moreover, as Mongol power stabilized were integrated into the Ilkhanid political and economic structure. They also greatly bene-.ted from the military struggle between the Mamluks and the Ilkhanids, using the conWXict to play one side o,f against the other. For instance, al-ʿUmarī writes: “In a place called Bayn al-Jabalayn, in the district of Irbil, there is a group (qawm) employed by both regimes and that fawns on the two sides. In winter, they are deferential to the Mongols and during the summer, they help the Syrian cavalry in their military campaigns.”>? Clearly, the geo-political clash between the Middle East’s two most powerful empires left open a political space which allowed the Kurdish tribes to maintain a certain level of autonomy.

Mamluks and Kurds: The Emergence of a ‘Kurdish Policy’

During the Mamluk-Ilkhanid period and within the context of the warrior economy (i.e. banditry) that we described earlier, the collection of protection taxes (khafāra, ḥimāya) became extremely signi-.cant. It had long been paid to tribes by unarmed sections of the population in order to ensure the security of a region—especially its roads and cultivated -.elds. The right to collect this tax was granted by larger political entities to local tribes who controlled certain areas.>@ With regard to Kurdish regions, during the Mamluk period, travellers,

A3 Although one might note that a calculation taking into account the area of each province may yield meaningful results. AA It is worth emphasizing that the image of intransigence against local Kurdish powers that the Mongols bear is due to the complexion of Iranian pro-Mongol historiography and Egyptian pro-Mamluk literature rather than actual discriminatory policies. Mamluk chroniclers insist on the anti-Islamic actions of the Mongols and the Ilkhanid chroniclers tend to assert the overarching power of a compact Iranian empire at the head of which only lies the Gengiskhanid family. AB al- ʿUmarī, Masālik al-abṣār, 3:127. AC Cahen, “Ḥimāya,” 394, and “Khafāra,” 911; Eddé, La Principauté ayyoubide d’Alep, 334. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; )Q( peasants and merchants had to cross a series of mountain passes or bridges (darband) held by diverse Kurdish groups, who collected the khafāra.>V In this speci-.c case, the khafāra appears to have been a tax collected by the tribes in exchange for the protection of individuals passing through a rugged and dangerous region. For instance, al-ʿUmarī describes the khafāra collection by the Kurdish Sahriyya:

Their darband is situated in between two high mountains irrigated by the Upper Zāb. In a frightening roar, three bridges are brought down [ . . . ] The darband stands at two hundred cubits from the surface of the water. Between the two mountains, the bridges are -.fty cubits long and two cubits large. [ . . . ] Flocks, horses and men cross them. [The rickety bridges] jump up or sag and both scare and play with the mind of the traveller who crosses them. [The Sahriyya] collect the khafāra there, and let pass whoever they want.>\

The speci-.c geology of Irbil’s hinterland reveals, for instance, gorges that follow the WXow of the Rūbar-i Rawanduz. We might ask exactly what distinguished banditry from the collection of protection taxes? In the article on the ḥimāya, the Encyclopaedia of Islam suggests that this tax was sometimes conceded by an imperial state and at other times ‘usurped’ by a tribal group. It is most likely that the khafāra, once usurped, ceased to be a tax, becoming mere banditry in the eyes of the Arab geographers. In contrast to banditry, the khafāra was an entirely legal and legitimate tax collection mode established in accordance with Islamic law. Al-ʿUmarī does not indicate whether or not the collection rights that the Kurdish tribes possessed had been usurped. However, interestingly, he notes that the tribes collecting the khafāra were only doing so within the area under the formal inWXuence of the Mongol Empire and against imperial will. Thus, neither the Mamluk state nor the Mongol state seem to have formally granted such tax collection rights to the Kurds. However, in another case, ʿIzz al-Dīn Ibn Shaddād evokes the role of the Qarāwil (also called Qarāghul), a group led by a Kurdish Hadhbānī Amīr, around the year 1260, who were in charge of collecting the khafāra (yukha012rūn = they were protecting ?) on the roads of the Ilkhan’s territory.>] Signi-.cantly, none of this information is mentioned in ‘pro-Mongolian’ Persian sources.

A_ al- ʿUmarī, Masālik al-abṣār, 3:127, 130, 131, 134. A` Ibid., 129. Aa Ibn Shaddād, ʿIzz al-dīn, Ta’rīkh al-Malik al-Ẓāhir, 333 (238a). )Q) 789:;

In practice, the notion of protection is a euphemism, because the tribes who collected this ‘tax’ were ‘protecting the taxpayer’ from their own forcible exactions. Hence it was a form of ‘highwayman taxation’ that has been sys- tematically presented by Arab and Iranian authors as banditry but resembles a local form of tax collection which was neither as informal nor as random as it has been presented. Indeed, the khafāra constitutes a legitimization of banditry and a form of tax collection on the outer edges of imperial civilization. The question of khafāra points to a major paradox within the Kurdish society of that time. It was a sign that a form of autonomous military organization likely to create the conditions for autonomous political organization existed. However, at the same time, the narratives produced by Mamluk historiography, and often written by Mamluk administrators, described the Kurds as a force integrated into the Mamluk imperium as part of their struggle against the Mongol ‘unbelievers.’ Although the reality of Kurdish extractions was much more complicated, the predilection for Mamluk authors to use the term khafāra when it came to these ‘taxes’ is part of the narrative depicting a general Muslim front against the Ilkhans, as well as a rhetorical device which depicted the Mamluk sovereignty over the Kurdish regions expressed in the form of sovereignty over its tax revenues. The khafāra also had a practical reality distinct from its use in Mamluk propaganda. This was a far more localized matter which was connected to the relationship between the petty kingdoms and fortress principalities in the Kurdish zone and certain tribal groupings. For example, the only time the khafāra seems to have been usurped is when the Kurdish Karādila disobeyed the Ayyubid prince of Ḥiṣn Kayfā (Hasankeyf), al-Malik al-Ṣāliḥ (1302–26). According to the local historian and chronicler, Ibn al-Munshī, the Karādila of al-Buqʿa “were standing on the roads at the edge of the city and were collecting for themselves the khafāra and toll taxes (ʿabr al-durūb) from the caravans.” Al-Malik al-Ṣāliḥ annihilated these usurpers in 1312. However, two years later, both sides’ attitudes were quite di,ferent. When al-Ṣāliḥ summoned the Karādila of Zīḥ,?^ they heeded his call; and when he asked them to obey his orders, they did. Consequently, the sultan granted them the right to collect khafāra and toll taxes (ʿabr al-durūb). The Karādila of Zīḥ thus entered the regular army of the sultanate and received a regular income. Later, al-Ṣāliḥ forbade the collection of khafāra and ‘abr. However, he transformed the village of Zīḥ into an iqṭāʿ (-.scal unit) which permitted the remuneration of the Karādila, who had been integrated into the army. Here the khafāra surfaces as

Bb Ibn al-Munshī, Nuzhat al-naẓīr, 23, 25a. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; )Qc an intermediary tax collection system before the -.rmer integration of groups into the state apparatus.?* As we have seen above, while crushing the Kurdish tribes, the Mongol authorities sought to integrate some Kurdish leaders into their political system on an individual basis. This was done, most probably, in order to diminish the inWXuence of the tribesmen. This was the case for Abū Bakr al-Dunbulī. Some other tribal chiefs were instituted as Ilkhanid agents either when the Mongols were unable to impose their will or when they needed to have certain policies, such as tax collection, implemented. The rulers of both the Māzanjāniyya- Humaydiyya and the Jūlmarkiyya-Hakkāriyya principalities held a high o,-.cial

B0 A whole section could have been dedicated to the issue of iqṭāʿ, as it raises the question of territorial control (by the state or by local polities) and tax collection, as well as resource distribution. However, here, we shall only emphasize some aspects that seem to con-.rm Amitai’s hypothesis on the subject. The iqṭāʿ is always granted by a central state and it is obvious that Kurdish amirs received iqṭāʿs from Mongols, Mamluks and Ayyubids, whatever this might imply in terms of political prerogatives or tax collection. First, although e,fective iqṭāʿs were given to Kurdish amirs in Syria, Mamluk iqṭāʿs granted to Kurdish amirs in Kurdish lands appear to have been mere promises of authority over a territory to be conquered. This is the case with Sayf al-Dīn Mankalān, the Jūlmarkī Amīr who was granted Irbil as an iqṭāʿ by Baybars when he took refuge in Cairo, long before the Mamluk armies could reach this region. On the other hand, al-ʿUmarī mentions the iqṭāʿs granted by the Mongols to Kurdish amirs (Jūlmarkiyya and Humaydiyya). The authority granted and the sharing of revenues allocated to amirs and the Mongol Horde remain unclear. However, it does not seem that Jūlmarkiyya or Ḥumaydiyya were ever o,-.cial elements in the Ilkhanid army. Most likely they were local (autonomous) polities that the state needed to co-opt and integrate, what they did by conceding o,-.cial iqṭāʿs. This also appears to be the case for the iqṭāʿs given to some Kurds during Tegüder’s reign as highlighted by Amitai-Preiss. On the other hand, on one occasion, iqṭāʿs were granted to a Kurdish amīr, an agent of Ghazan: Abū Bakr al-Dunbulī, the Shihna of Diyār Bakr. Also, it is noteworthy that no iqṭāʿ were granted to any Mongol soldier in these regions whether before or after Ghazan’s death. This tends to con-.rm Amitai’s hypothesis that these were not classic iqṭāʿs, that the Ilkhanids generally maintained a system of pasture land distribution, and that such pasture land distribution was not applied in Kurdish regions since these lands were not used as grazing land by the Mongols. Another aspect of the issue is the fact that the phenomenon is not reported in Persian literature. Does this mean that the Mamluk authors transposed some very speci-.c Mongol administrative practices into Arabic terminology (‘iqṭāʿ ’) in order to be understood by their audience? This question requires further investigation. On an iqṭāʿ to Sayf al-Dīn al-Jūlmarkī see Ibn Shaddād, Ta’rīkh al-Malik al-Ẓāhir, 332–33 (237–38). On iqṭāʿs to Jūlmarkiyya and Ḥumaydiyya see al-ʿUmarī, Masālik al-abṣār, 3:128, 132. On Abū Bakr al-Dunbulī see Ibn al-Munshī, Nuzhat al-naẓīr, 19b. On iqṭaʿs to Kurds by Tegüder see Amitai, “Turko-Mongolian Nomads and the iqṭāʿ System,” 157. )Qd 789:; position in the Ilkhanid state apparatus. For instance, after the Mongols gave up their e,forts to overthrow Mubāriz al-Dīn Kak, the Māzanjānī-Ḥumaydī paramount chief, the latter was handed a pā’iza (an investiture document) which made him (or recognized him as) the chief of the region of Irbil and ʿAqr (Akre), where he was obliged to collect the taxes on behalf of the Ilkhanids during the second half of the thirteenth century.?+ Signi-.cantly, the Mongols also granted a pā’iza to a minor chief, al-Asad b. Matkākīn, who was responsible for collecting taxes in the main cities of the region. This seems to have been an attempt to dilute the authority of Mubāriz al-Dīn Kak. However, despite Mongol e,forts to enact a policy of ‘divide and rule,’ he and his descendants remained in charge of the region. As for the Jūlmarkiyya and its chief, Asad al-Dīn Mūsā b. Mujallī b. Mūsā b. Mankalān, whose stronghold was situated to the far north and west of the Mazānjāniyya region in the country of Jūlmark-Hakkārī south of Lake Van, it was also o,-.cially recognized by the Ilkhanids.?/ The strategy adopted here was to co-opt intractable local forces. By co-option we mean providing -.nancial and military assistance as well as a title to support individuals who were to represent the central Ilkhanid power. The advantages of this strategy for the Mongols was twofold. On the one hand, the Mongols avoided the destruction of taxable resources and major military expenditures. On the other, they ensured that they were able to harness local resources and asserted their claims of sovereignty over regions within the Kurdish zone. Such policies were, of course, not unproblematic, as they facilitated the autonomy of Kurdish tribes and principalities. The fact was that fear of violence and general mistrust undergirded the relationship between Kurdish tribes and the Ilkhanid imperium. Moreover, when Aḥmad Tegüder (r. 1282–84) endeavoured to pursue more conciliatory policies towards the Kurds,?> this approach was unpopular among the Mongol military elite and was one factor in Tegüder’s assassination. Such risks were less important for the Mamluks, as they had nothing to lose in co-opting local polities in a region where they had very little military or political inWXuence. For instance, the fact that the Mamluk ruler, al-Malik al-Ẓāhir Baybars (r. 1259–77), put administrators, soldiers and a royal court at the disposal of the chief of the Jūlmarkiyya, Sayf al-Dīn Mankalān, the grandfather of Asad al-Dīn mentioned earlier, certainly inWXuenced the social and political

B2 al- ʿUmarī, Masālik al-abṣār, 3:128. B3 al-Qalqashand ī, Ṣubḥ al-a‘shā, 7:283. BA Amitai, “Turko-Mongolian Nomads and the iqṭāʿ System,” 157; and Ibn al-Dawādārī, Kanz al-durar, 8:264. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; )Qf structures of this tribe and was a low-cost investment in the Kurdish zone.?? Mamluk administrative activities also reinforced the role of the Egyptian state apparatus in shaping the social and political organization of the Kurdish territory. In order to implement policies aiming to ensure its permanence and the extension of its sovereignty, the Mamluk state had to record, classify, as well as appoint administrators and rulers. This was the function of the dīwāns and especially the dīwān al-inshā’ (chancellery), one of the administrative authorities that played the role of communication channel between the sultanate, administrators and the local leaders. During the -.rst half of the fourteenth century, -.fty Kurdish amirs within the Kurdish zone were subject to an o,-.cial correspondence with the Mamluk chancellery and received manāshīr (property noti-.cations).?@ This state of a,fairs had both practical and symbolic implications. Certain Kurdish groups were aided by Mamluk military support; at the same time, the symbolism of being appointed by an imperial state institutionalized local leaders, who began to look less and less ‘tribal.’ Although the Mamluk state did not create these tribes ex nihilo as some functionalist theoreticians suggest,?V it contributed to the rise of its favoured individuals and groups. In e,fect, the Mamluks were creating an interface with the tribal world, which in turn, promoted the integration of Kurdish groups in their Empire. Although, at -.rst glance, the policies of the Ilkhanids and the Mamluks seem quite similar—except in the area of tax collection—the co-option of individuals and tribes in each empire had di,ferent implications. Particularly signi-.cant was the Mamluks’ co-option of Kurdish groups—not on an individual basis as in the case of Mongol policies—but as a people. These policies were more than the mere implementation of local powers; they reWXected the Mamluks’ desire to create a powerful coalition against the Mongols through reinforcing the notion of the Kurds as a distinct category, while at the same time territorializing it. In this sense, Mamluk policies were somewhat paradoxical; they both promoted integration and di,ferentiation. This paradox reminds

BB Ibn ʿAbd al-Ẓāhir, al-Rawḍ al-zāhir, 87. BC Ibn Nāẓir al-Jaysh, Kitāb tathqīf, 74–81; and al-Qalqashandī, Ṣubḥ al-aʿshā, 7:285–89. B_ Morton Fried believes that tribes are the result of contacts of some isolated societies with ‘complex’ social organisations that can be de-.ned as a state. Van Bruinessen and Paul used Morton Fried’s conceptions on the tribe. We would suggest, like Khazanov, that peripheral societies and state inWXuence each other rather than one overwhelmingly dominating the other. See Paul, “Perspectives nomades, état et structures militaires,” 1072. Bruinessen, Agha, Shaykh and State, 135; Fried, On the Evolution of Social Strati12cation and the State; and Khazanov, “Nomads and Oases in Central Asia,” 69–91. )Qg 789:; us of the crucial role of the state in shaping a geographical space, simultaneously resisting or enforcing the local political con-.gurations. More concretely, not only does the state endeavour to appoint local leaders in order to project its inWXuence and sovereignty, it also names the lands over which it seeks to rule. Indeed, this naming is, in itself, an overtly political act. The historical evolution of the Middle East’s Kurdish zone has already been examined; hence the issue at this point is not the actual location of this zone but rather how to deal with the spatial representations at stake in the sources of the fourteenth century. The only toponyms indicating a Kurdish presence in particular regions to be found in Persian pro-Mongol historiography are Kurdistān and Vilāyat-i Akrād (The Province of the Kurds). According to Mustawfī, the province of Kurdistan had been created in 1152 by the Seljuq sultan Sanjār, in the region formerly ruled by the Kurdish Hasanwayhids (10th century) from Nihavend to Shahrazūr. It was located in the Zagros mountains and extended from Hamadan to Hulwan.?\ This source is controversial as it was written long after Sanjār’s reign. Nevertheless, these administrative appellations were most likely born out of a certain ‘ethnic reality,’ namely the fact that within these regions one could -.nd many Kurds. Especially during the thirteenth and fourteenth centuries, categories such as Vilāyat-i Akrād were used by Iranian authors to describe the same area, although sometimes, Oshnaviyeh and Arjīsh were also referred to in such a way. However, the Persianate conception of an administrative Kurdistan was relatively restricted. It did not include all the areas where Kurds could be found or all the areas which Arabic (and especially Mamluk) geographers described as being ‘Kurdish.’?] For instance, it never included Diyār Bakr. As already noted, Arabic-speaking authors used several categories to designate the places where the Kurds lived: Zūzān al-Akrād, Jibāl al-Akrād and Bilād al-Akrād. These categories mainly applied to the region of Diyār Bakr, Lake Van, the hinterland of Mosul and further southeast to the region of Shahrazūr up to Hamādān . In this sense, the expression Bilād al-Akrād, which is extremely similar to the Persianate term Kurdistān, is a relatively late category that probably -.rst appears during the twelfth century. Although there is no precise description of this region, it seems to generally refer to a western Zagros zone

B` Mustawf ī, Nuzhat al-qulūb, 59–60. Ba Contrary to the Mamlūk sources, Persian sources generally ignore the Kurdish presence. For instance, it is quite surprising that Mustawfī does not report the presence of Kurdish inhabitants in Irbil, al-ʿImādiyya (Imādiyeh), ʿAqr, Ḥiṣn Kayfā (Hasankeyf) or Ushnuh/ Ushnaviyyeh, even though they had been noticed for centuries by Arab historiographers. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; )Q< overlapping the Zūzān (the region South of both Lake Van and the Hakkārī mountains).@^ We should not forget that the Kurdish territory and the area described as Jibāl were o,-.cially under Mongol sovereignty during the period that concerns us. However, the Mamluk state intended to give it an administrative substance, that is to say an o,-.cial complexion. Al-ʿUmarī, a Dīwān al-Inshā’ administrator and son of an administrator, refers to Jibāl as a speci-.cally Kurdish zone. In his geographic encyclopaedia, as well as in the chapter concerning the mamālik (countries) in the Ṣubḥ al-aʿshā’ of al-Qalqashandī, Kurds appear in a separate section, namely in the sections relating to Jibāl. Similarly, Muḥī al-Dīn Ibn ʿAbd al-Ẓāhir (d. 1293) reproduces an excerpt of al-Mansūr Qalāwūn’s act of succession to his sons. Amongst the many dependencies over which the sultan exerted his sovereignty and which his sons inherited, there was the “forti-.ed and mountainous realm of the Kurds” (al-Mamlaka al-Akrādiyya al-ḥaṣīna al- jabaliyya) associated with its “potential conquests” (futūḥātuhā).@* In reality, actual Mamluk inWXuence over this region was extremely limited. Yet, what is signi-.cant is that the Mamluks put their hopes in the Kurds as a whole, that is, as an abstract category of people—a people who would settle and control a speci-.c territory on behalf of the Mamluks.

The Place of the Kurds in the Mamluk Imperium

The Ayyubid period (1169–1250) represented the climax of the Kurdish integration in Egypt and Syria.@+ Conversely, when the Mamluks came to power in Egypt, the Kurds ceased to be both a central military force within the Egyptian army and a major political faction within the state’s institutions. Nevertheless, in the military context, they continued to be recognized as a distinct category. In treaties, administrative manuals and, possibly, within the ‘order of battle’ they were often associated with Turkomen and Arab soldiers.@/ Such an

Cb James, Saladin et les Kurdes; Ibn al-ʿImād, Shadharāt al-dhahab, 7:423; Abū Shāma, Tarājim, 204; Baybars al-Manṣūrī, Zubdat al-12kra, 329, 352. C0 Ibn ʿAbd al-Ẓāhir, Tashrīf al-ayyām, 202. C2 James, Saladin et les Kurdes. C3 Such a state of a,fairs was reported by Ibn ʿAbd al-Ẓāhir (d. 1293), Faḍl Allāh al-ʿUmarī (d. 1348–49), Ibn Nāẓir al-Jaysh (d. 1384), al-Qalaqashandī (d. 1418). Although these authors did not write in the same period, they do describe the perpetuation of the situation in Mamluk administrative discourse over the years. All of them served as secretaries in the Mamluk chancellery. Some examples of Kurds associated with Turkoman groups can be )Q= 789:; association was a clear sign of their relegation outside the core military oligar- chy. For instance, the administrative manuals were constituted in such a way that a hierarchy becomes apparent in which the Kurds, along with the other free-born military groups such as the Turkoman and Arab tribesmen, held the lowest ranks, while Mamluks and Mongols occupied the highest ranks. At the same time, the classi-.cation of individuals and groups made it possible to pursue speci-.c policies towards them, i.e. assigning them to higher or lower military roles and payments. For instance, it seems that the Kurdish and Turkoman Ṭabalkhāna received less favourable treatment from the Dīwān al-Inshā’ than their Mamluk counterparts, as they received manāshīr (edicts) on a paper one third the size of regular manāshīr. At the same time, they were not granted the honorary letter yā’ at the end of their title formula. An example found in Ibn ʿAbd al-Ẓāhir’s (d. 1293) al-Rawḍ al-zāhir fī sīrat al-Malik al-Ẓāhir, highlights the material necessity for classi-.cation. In 1265, Egypt was struck by high levels of inWXation. In response, Sulṭān Baybars conducted a census of all the poor people of Cairo and asked the Dīwān al-Jaysh to count all the amirs and give each of them the responsibility of supporting a group of indigents. He also withdrew a -.scal district (nāḥiyya) from each Kurdish or Turkoman amīr.@> These two examples do not indicate a clear and systematic discrimination against the Kurds per se, yet they do reveal the new political con-.guration in which Mamluk slave elites were favoured over freeborn tribesmen. This relegation of the Kurds out of the organs of the Egyptian imperial state did not mean that the Kurds were unimportant within the Mamluk polity as a whole. Rather, it indicated that Kurds were to be assigned a new role, one in which they would assist the Mamluks in their struggle against the Mongols on the eastern edges of their empire. As we have seen, some Mamluk geographical works emphasized the Kurdish aspect of the region of Jibāl. Whether it was called Jibāl, Jibāl al-Akrād or simply al-Akrād, this region was the subject of serious attention from the Mamluk administration in terms of its population, military potential and economic resources. Indeed, the administrative manuals of al-ʿUmarī and Ibn Nāẓir al-Jaysh devote whole sections to the Kurdish amirs of the eastern regions, including the procedure that was to be followed in corresponding with them.@?

found in the following works: Ibn ʿAbd al-Ẓāhir, Tashrīf al-ayyām, 96, 105, 158, 202; Ibn Nāẓir al-Jaysh, Tathqīf al-taʿrīf, 140, 157. CA Ibn ʿAbd al-Ẓāhir, al-Rawḍ al-zāhir, 188–89. CB al- ʿUmarī, al-Taʿrīf, 42–50; Ibn Nāẓir al-Jaysh, Tathqīf al-taʿrīf, 74–81; al-Qalqashandī, Ṣubḥ al-aʿshā, 7:285–89. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; )QQ

The Kurdish amirs with whom the Mamluk chancellery corresponded were supposed to exercise their coercive and political power within the Jibāl, a region that was o,-.cially under Mongol sovereignty. It is noteworthy that Mamluk texts seem eager to show that some Kurdish amirs mentioned in the chancellery’s lists held high ranks (makāna) within the Mongol Empire. This seems peculiar, and we might ask to whom they were loyal. The answer is probably both and neither; yet geo-political realities precluded the establishment of a sovereign Kurdish entity. Al-ʿUmarī explains the reasons for this in his Taʿrīf:

[The Kurds] are countless. If the sword of discord was not cutting their growing sprout and was not preventing their eruption, they would pour into the lands and would seize many goods. However, they are inclined toward disagreement and dissension. The sword stands drawn between them, blood stays shed, order remains scorned, eyes are wet and spat- tered with blood . . . @@

Thus there is no reason for the Mamluks not to claim sovereignty over the Kurdish regions. That is why “The Inaccessible and Mountainous Realm of the Kurds and its Conquests” (al-Mamlaka al-ḥaṣīna al-Akrādiyya al-jabaliyya wa futūḥātuhā)@V is said to be under Mamluk sovereignty. One can sense in these examples the nature of the Mamluks’ ‘Kurdish policy,’ namely their desire to use a potentially strong but WXawed ‘esprit de corps’ (ʿaṣabiyya), as a weapon in order to extend the Mamluk realm against the Mongols. This policy is reWXected in the creation of the position of ‘Commander of the Kurds’ (Muqaddam al-Akrād). In this regard, the ‘directive’ (waṣiyya) that was sent to the Muqaddam al-Akrād is extremely interesting:

He must gather these scattered groups. He must unite the whole groups that have split. He must win the heart of the leaders among them who have WXed [their lands]. He must ensure the disputes amongst them are laid to rest in order to [be able to] use their violence against the in-.dels. They shall cease oppressing each other and begin beating the heathen. He must maintain the idea that our goodwill is not bogus and that the least territory that we grant them is better for them in the eyes of God than the ones that they own in the extremity of the ʿAjam lands and in Shahrazūr. [ . . . ] Surround them with words of agreement. And wait

CC al- ʿUmarī, al-Taʿrīf, 47; al-Qalqashandī, Ṣubḥ al-aʿshā, 7:283. C_ Ibn ʿAbd al-Ẓāhir, Tashrīf al-ayyām, 202. cUU 789:;

with them as we depart for Jihād. Put on your breast-plate and be ready for battle.@\

It is clear that this military position was meant as one which would mobilize and unite the Kurdish tribes in the struggle against the Mongols. In a broader sense, the Mamluk state, contrary to the policies of earlier medieval Middle Eastern polities as well as those of the Mongols, attempted to reinforce and bene-.t from the Kurdish ʿaṣabiyya through the creation of a title denot- ing leadership over the community. In Ibn Khaldūn’s view, the ʿaṣabiyya was a product of the harsh conditions of the borderlands. From its birthplace on the periphery, it bursts into the ‘civilized’ world, overthrows the existing order and results in the establishment of a new a dynasty/regime (dawla).@] The Mamluks obviously had no intention of allowing the Kurdish ʿaṣabiyya the opportunity to emerge from the borderlands and overthrow their imperial state. Even if not explicitly expressed in the sources, Kurdishness was most likely regarded as providing the ʿaṣabiyya of the Mamluks’ predecessors, the Ayyubids, just as much as Turkishness provided the ʿaṣabiyya of the Seljuqs and Berberness the ʿaṣabiyya of some North African dynasties. Indeed, the potential threat presented by Kurdish solidarity was not irrational. Following the fall of the Ayyubids in Syria in 1258, there were e,forts on the part of certain elements amongst the Kurdish community of Egypt to a,fect the restoration of a Kurdish dynasty.V^ Thus it seems that the Mamluks hoped to mobilize the Kurdish ʿaṣabiyya and direct it against their Mongol rivals. By implying that the Kurds should reciprocate Muslim solidarity, they kept them away from the core institutions of the state and assigned them the singular task of re-conquering their lost lands in the margins.

Conclusions

These Mamluk policies of ‘ethnic engineering’ should be contextualized within the broader geo-political realities of the fourteenth century. On the one hand, they sought to protect the core institutions of the state from the threat of a Kurdish-Ayyubid restoration. On the other, they aimed to project Mamluk inWXuence eastwards towards the lands held by the Mongols, through supporting Kurdish groups within the borderlands. Thus, the Mamluk strategy

C` al- ʿUmarī, al-Taʿrīf, 147. Ca Martinez-Gros, Ibn Khaldoun et les sept vies de l’Islam; Ibn Khaldūn: Kitāb al-‘ibar, 1: 263. _b James, “Arab Ethnonyms (ʿAjam, ʿArab, badū and Turk . . .),” 708–11. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; cU( surely entered the wide range of tactics used against the Ilkhanids and well described by Reuven Amitai (espionage, propaganda, economic warfare, diplomatic demarches, manipulation of proxies etc.).V* However, it had another implication for the Kurds. While the evolution of the category of Kurdishness and a Kurdish zone pre-dates Mamluk–Ilkhanid rivalries, the circumstances of geographical in-betweenness and border culture were ‘o,-.cialized’ during this period. These processes continued during the subsequent confrontations between the Mamluks and the Timurids, and in many ways laid foundations for the autonomy granted to the Kurdish principalities by the Ottomans and Safavids in the sixteenth century.V+ This in turn institutionalized and perpetuated the category of the Kurds as a distinct people. Hence, the fourteenth century should be regarded as a critical period in the long-term development of the Kurdish identity.

Bibliography

Abū Shāma. Tarājim rijāl al-qarnayn al-sādis wa al-sābiʿ. Damascus: Dār al-Kutub al-Mālikōyah, 1947. Aigle, Denise. Le Fârs sous la domination mongole politique et 12scalité (,---.–,-/. S.). Studia Iranica Cahier 31, Paris: Association pour l’avancement des études iraniennes, 2005. Amitai, Reuven. “Turko-Mongolian Nomads and the iqṭāʿ System in the Islamic Middle East (1000–1400 #q).” In Nomads in the Sedentary World, edited by A. Wink and M. Khazanov, 152–171. London: Curzon Press, 2001. ———. “The Resolution of the Mongol-Mamluk War.” In Mongols, Turks and Others, Eurasian Nomads and the Sedentary World, edited by R. Amitai and M. Biran, 359– 390. Leiden: Brill, 2005. Amitai-Preiss, Reuven. Mongols and Mamluks: The Mamluk-Īlkhānid War, 1260–1281. Cambridge: Cambridge University Press, 1995. Ateş, Sabri. “Towards a History of the Ottoman-Iranian Borderland and the Borderland peoples 1843–1881.” PhD diss., New York University, 2006. al-Balādhurī, Aḥmad b. Yaḥyā. Kitāb futūḥ al-buldān. Edited by M. G. de Goeje. Leiden: Brill, 1866. Barth, Fredrik, ed. Ethnic Groups and Boundaries: The Social Organization of Culture Di0ference. Oslo: Universitets Forlaget, 1969.

_0 Amitai, “The Resolution of the Mongol-Mamluk War,” 359–60. _2 Ateş, “Towards a History of the Ottoman-Iranian Borderland.” cU) 789:;

Barthelemy, Dominique. Chevaliers et miracles. La violence et le sacré dans la société féodale. Paris: Armand Colin, 2004. Baybars al-Manṣūrī. Zubdat al-12kra fī ta⁠ʾrīkh al-hijra. Edited and translated by D. S. Richards. Beirut: Orient Institut, 1998. Bruinessen, Martin van. Agha, Shaykh and State: The social and political structures of Kurdistan. New Jersey: Zed Books Ltd, 1999. Cahen, Claude. “Ḥimāya.” In Encyclopaedia of Islam, Second Edition SSS:394. ———. “Khafāra,” In Encyclopaedia of Islam, Second Edition St:911. Clastres, Pierre. La société contre l’Etat. Paris: Les Editions de Minuit, 1974. Colliot-Thélène, Catherine. Le désenchantement de l’Etat, de Hegel à Max Weber. Paris: Les Editions de minuit, 1992. Eddé, Anne-Marie. La Principauté ayyoubide d’Alep (579/1189–658/1260). Stuttgart: Franz Steiner Verlag, Freidburger Islamstudien, 1999. Elissée,f, Nikita. Nûr ad-dîn, un grand prince musulman de Syrie au temps des croisades. 3 vols. Damascus: Institut français de Damas, 1966. Fried, Morton H. On the Evolution of Social Strati12cation and the State. Indianapolis, Su: Bobbs-Merrill, 1960. Haarmann, Ulrich. “Ideology and History, Identity and Alterity: The Arab Image of the Turk from the ʿAbbasids to Modern Egypt.” International Journal of Middle East Studies 20 (1988): 175–96. al-Ḥamawī, Yāqūt. Muʿjam al-buldān. 5 vols. Beirut: 1955–1957. Heidemann, Stefan. Das Aleppiner Kalifat (89 1261), Vom Ende des Kalifates in Bagdad über Aleppo zu den Restaurationen in Kairo. Edited by U. Haarmann. Leiden: Brill, 1994. ———. Die renaissance der Städte in Nordsyrien und Nordmesopotamien, Städtishe Entwicklung und Wirtschaftliche, Bedingungen in ar-Raqqa und Ḥarrān von der Zeit der Beduinischen, Vohrerrschaft bis zu der Seldschuken. Edited by Wadad Kadi and Rotraud Wielandt. Leiden: Brill, 2001. Ibn ʿAbd al-Ẓāhir (Muḥī al-Dīn). al-Rawḍ al-zāhir fī sīrat al-Malik al-Ẓāhir. Edited by ʿAbd al-ʿAzīz al-Khuwayṭir. Riyadh: Muʼassasat Fuʼād, 1976. ———. Tashrīf al-ayyām wa ’l-ʿuṣur fī sīrat al-Malik al-Manṣūr. Edited by Murād Kāmil. Cairo: Wizārat al-Thaqāfa wa al-Irshād al-Qaumō, 1961. Ibn al-ʿAmīd, al-Makīn. al-Majmūʿ al-mubārak. Translated by A.-M. Eddé and F. Micheau, Chronique des Ayyoubides (602–658/1205–6–1259–60). Paris: Académie des Inscriptions et Belles-lettres, 1994. Ibn al-Athīr. al-Kāmil f;̄ al-tār;̄kh. 3 vols. Beirut: Dār al-kutub al-ʿIlmōyah, 1998. Ibn al-Dawādārī, Abū Bakr. Kanz al-durar wa jāmiʿ al-ghurar. Edited by S. ʿAshūr. Cairo: 1972. Ibn Jubayr. Riḥlat Ibn Jubayr. Beirut: Dār Ṣādir, 1964. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; cUc

Ibn Ḥawqal. Kitāb ṣūrat al-arḍ. Edited by J. H. Kramers. Leiden: Brill, 1939. Ibn Khaldūn. Kitāb al-ʿibar (wa dīwān al-mubtadā’ wa ’l-khabar fī ayyām al-ʿarab wa ’l-ʿajam wa man ʿasharahum min dhāwī ’l-sultān al-akbār). 7 vols. Beirut: Maktabat al-Madrasa, 1959–1961. Ibn al-ʿImād. Shadharāt al-dhahab fī akhbār man dhahab. Edited by M. al-Arna’ūṭ, 10 vols. Damascus: Dār al-Ṭabbāʻ, 1991. Ibn al-Munshī, al-Ḥasan ʿAbdallah al-Ḥiṣnī. Nuzhat al-naẓīr fī rāḥat al-khaṭīr. cod. mixt. 355, HoWyibliotek, Vienna. Ibn Nāẓir al-Jaysh. Kitāb tathqīf al-taʿrīf bil-muṣṭalaḥ al-sharīf. Edited by Rudolf Vesely. Cairo: SR#z, 1987. Ibn Shaddād (Bahā al-Dīn). al-Nawādir al-sulṭāniyya wa l-maḥāsin al-yūsu12yya. Edited by G. al-Shayyāl. Cairo: al-Muʼassasa al-Miṣriyya al-ʻA{mma, 1964. Ibn Shaddād, ʿIzz al-Dīn. Ta⁠ʾrīkh al-Malik al-Ẓāhir (al-Rawḍa al-zāhira fī sīrat al-ẓāhira). Edited by A. Hutait. Wiesbaden: Bibliotheca Islamica, 1983 [also, ms, Edirne, 12679]. al-Iṣfahānī, ʿImād al-Dīn. Kharīdat al-qaṣr. Baghdad: Majmaʿ al-ʿIlmō al-ʿIrāqō, 1955. James, Boris. Saladin et les Kurdes: perception d’un groupe au temps des croisades. Paris: L’Harmattan, 2006. ———. “Le territoire tribal des Kurdes et l’aire iraqienne (|}–|SSS} siècles): esquisse des recompositions spatiales.” In l’Irak en perspective, Revue des Mondes Musulmans et de la Méditerranée 117–118 (2007): 101–26. ———. “Une ethnographie succincte de ‘l’entre-deux kurde’ au Moyen Âge.” Etudes Rurales 186 (2011): 21–43. ———. “Arab Ethnonyms (ʿAjam, ʿArab, badū and Turk . . . ): The Kurdish Case as a Paradigm for Thinking about Di,ferences in the Middle Ages.” Iranian Studies 47/5 (2014): 683–713. ———. Les Kurdes dans l’Orient mamelouk et mongol de 1250 à 1340: entre marginalisa- tion et autonomie. PhD dissertation at Université Paris-Ouest Nanterre-La Défense, 2014. Juwaynī, ʿAṭā Malik. Ta⁠ʾrīkh-i Jahāngushā. Edited by Mīrzā Muḥammad Qazwīnī. Leiden: Brill, 1912–37. Kemp, Percy. Territoire d’Islam; Le monde vu de Mossoul au ,/---. siècle. Paris: Sindbad, 1982. Khazanov, A. M. “Nomads and Oases in Central Asia.” In Transition to Modernity, Essays on Power, Wealth and Belief, edited by John Hall and I. C. Jarvie, 69–91. Cambridge: Cambridge University Press, 1992. Kolbas, Judith. The Mongols in Iran. Chingiz Khan to Uljaytu 1220–1309. New York: Routledge, 2006. Lyons, Malcolm Cameron and David Edward Pritchett Jackson. Saladin: the Politics of the Holy War. Cambridge: Cambridge University Press, 1982 and 1997. cUd 789:; al-Maqrīzī (al-Khiṭaṭ). Kitāb al-mawāʿiẓ wa’l-iʿtibār fī ’l-khiṭaṭ wa ’l-āthār. Edited by Sayyed Ayman Fuad. London: Mu’assasat al-Furqān, 2002. Martinez-Gros, Gabriel. Ibn Khaldoun et les sept vies de l’Islam. Arles: Actes Sud- Sindbad, 2006. Martiniello, Marco. L’ethnicité dans les sciences sociales contemporaines. Paris, Presse Universitaire de France, 1995. Masson Smith, Jr., John. “Mongol Nomadism and Middle Eastern Geography: Qîshlâqs and Tümens.” In The Mongol Empire and its Legacy, edited by Reuven Amitai-Preiss and David O. Morgan, 39–56. Leiden: Brill, 1999. al-Masʿūdī. Murūj al-dhahab wa maʿādin al-jawhar. Edited by Ch. Pellat. Beirut: al-Jāmiʿah al-Lubnānōyah, 1966. Mauss, Marcel. Essai sur le don. Forme et raison de l’échange dans les sociétés archaïques (1925). Paris: Quadrige/Presses universitaires de France, 2007. [English translation: The Gift. Translated by Mary Douglas London: Routledge, 2002]. Melville, Charles. “The Itineraries of Sultan Öljeitü, 1304–16.” Iran 28 (1990): 55–70. Michel le Syrian, La Chronique syriaque. Edited and translated by J. B. Chabot. Paris: Ernest Leroux, 1899–1914/1962. Miller, Konrad. Mappae Arabicae, Arabische Welt- und Länderkarten des 9–13. Jahrhunderts Herausgegeben. Frankfurt: Johann Wolfgang Goethe Universität, 1994. Minorsky, Vladimir. Studies in Caucasian History. London: Taylor’s Foreign Press, 1953. ———. “Kurds and Kurdistan.” In Encyclopaedia of Islam, Second Edition t:449–64. Mustawfī Qazwīnī, Ḥamdallah b. Abī Bakr. The Geographical Part of the Nuzhat al-Qulūb. Edited and translated by G. Le Strange. Leiden: Brill, 1915, 1919. Naṭanzī, Muʿīn al-Dīn Iskandar. Muntakhab al-tawārīkh-i muʿīnī. Edited by Jean Aubin. Tehran: Kitābfurūshō-’i Khayyām, 1336/1958. Patton, Douglas. “Badr al-Din Lu’lu’ and the Establishment of a Mamluk Government in Mosul.” Studia Islamica 74 (1991): 79–103. Paul, Jurgen. “Perspectives nomades, état et structures militaires.” Annales. Histoire, Sciences Sociales 5–6 (Sept. –Dec. 2004): 1069–93. al-Qalqashandī, Abū’l-ʿAbbās Aḥmad. Ṣubḥ al-aʿshā fī taʿrīf dīwān al-inshā’. Cairo: [s.n.] 1912–1913. al-Qazwīnī, Zakariyā’ b. Muḥammad. Kitāb āthār al-bilād wa akhbar al-ʿibād. Edited by Ferdinand Wüstenfeld. Göttingen: Druck und Verlag, 1848. Schacht, Joseph. Introduction au droit musulman. Translated by P. Kempf and A.-M. Turki. Paris: Maisonneuve & Larose, 1983. Szombathy, Zoltan. “Genealogy in Medieval Muslim Societies.” Studia Islamica 95 (2002): 5–35. al-ʿUmarī, Shihāb al-Dīn. Masālik al-abṣār fī mamālīk al-amṣār. Edited by F. Sezgin. 27 vols. Frankfurt: Maʿhad Ta’rōkh al-ʿUlūm al-ʿArabōyah, 1988. 989DEF 8GH 9IGJID K:LMKN:L8D KIDMOMP; cUf al-ʿUmarī. Aḥmad ibn Yaḥyá Ibn Faḍl Allāh. al-Taʿrīf bil-muṣṭalaḥ al-sharīf. Edited by Samīr al-Durūbī. Mu’ta: University of Mu’ta, 1992. Vanly, Ismet Cherif. “Le déplacement du pays Kurde vers l’ouest du |} au |t} siècle, recherche historique et géographique.” Rivista degli studi orientali 50/3–4 (1976): 353–63. Waṣṣāf, ʿAbd Allāh ibn Faz̤l Allāh. Taḥrīr-i Tārīkh-i Waṣṣāf. Edited by ʿAbd al-Muḥammad Āyatī. Tehran: Mu’assasa-i Muṭālaʿāt va Taḥqōqāt-i Farhangō, 1372/1994.

Epilogue

∵

"#$%&'( )* The End of the Ilkhanate and After: Observations on the Collapse of the Mongol World Empire

Charles Melville

The crux of the process was the interplay between the goals and policies of the rulers, on the one hand, and the political orientations and activities of the major social groups, on the other. These con+,icts became especially serious when the rulers emphasized very “expensive” goals which exhausted the available economic and manpower resources, or when the di-ferent social groups developed strong, autonomous political orientations. (.. /. '0.'/.&$1&, The Decline of Empires, 138)

Compared with the question of origin, the problem of the breakup of the Mongol empire seems almost simple . . . Arguably the most important [explanation] is succession practices. As we have seen . . . the Mongol world empire was repeatedly shaken by con+,icts among members of the royal family, including sons and brothers of the khan, for the succession. ($. /. 3#$41(5/, in Spuler, The Mongol Period, xxiii)

Hsiao’s point that the number of Mongols excelling in Chinese cultural pursuits was on the increase in the mid- and late Yuan, and that had the dynasty lasted longer, many more such Mongol literati might indeed have emerged is well taken. Yet, it could also be argued that the brevity of the dynasty was in itself a result of the Mongols’ tendency to avoid immer- sion in Chinese cultural and political mores and . . . of their tendency to govern by their own cultural compass. ('/10"5&&-3'.&, “Aspects of Khitan Liao and Mongolian Yüan Imperial Rule,” 214)

Introduction

The Ilkhan Sultan Öljeitü died on 30 Ramadan 716/16 December 1316; it was not until early–mid 717/1317, several months later, that his only surviving son and designated heir, Abū Saʿīd, was enthroned in Tabriz. It is true that at the time of his father’s death, Abū Saʿīd was in Khurasan, and that he was only 12 years old.? Nevertheless, the delay suggests quite strongly that the senior noyans— notably the amirs Sevïnch and Chūpān—were in no great hurry to maintain the unbroken continuity of Ilkhanid dynastic rule, and that the anxiety normally attendant on the absence of a ruler—leaving the kingdom rudderless and prey to the ambitions of men of violence—had evaporated. We might rephrase this to say, the anxiety ‘normally’ felt by the bureaucratic classes and their historian mouthpieces; the men of violence were evidently less alarmed at the absence of a restraining authority. As is well known, the lack of consensus surrounding Abū Saʿīd’s accession found a resolution only after the so- called ‘revolt of the amirs’ in 719/1319, when Abū Saʿīd won his spurs but at the same time was obliged to accept the tutelage of Amir Chūpān.A It is true that substantial interregna were not uncommon in the Mongol Empire, the most notorious interval following the death of the Great Khan Güyük in April 1248 until the accession of Möngke in the summer of 1251, delays not purely the consequence of the enormous distances involved in BCrst bringing news and then getting all interested parties together. But in these cases, the regency of an imperial wife BClled the gap in authority, BCrst under Töregene Khatun after the death of Ögetei and later Oghul Ghaimish.D Furthermore, this practice had not been followed in Iran after the establishment of the Ilkhanate by Hülegü in 1258. The intervals between reigns were generally reasonably brief and consistent with the time taken to assemble the Mongol ruling elite for the enthronement of the new khan (see Table 1).E On the death of the Great Khan Möngke, there was a contest between two crowned khans, his brothers Qubilai and Ariq Böke. But in Iran, even if the succession were contested to the point of civil war, it was a contest between a crowned khan and a rival claimant, not between opposing pretenders. The same was the case in the BCrst instance on the death of Abū Saʿīd, although on

F There is no consensus on the date of Abū Saʿīd’s coronation in the sources; see Jackson, “Abu Sa‘id,” 374; the latest date, 23 Rabīʿ 00, 717*/5 July 1317, is given by the contemporary Banākatī, Tārīkh, 478. H Melville, “The Revolt of the Amirs.” I See the recent PhD thesis of Bruno De Nicola, “Unveiling the Khātūns,” 68–75. J For most of these dates, see Melville, “The Chinese Uighur Animal Calendar,” 89. KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S *))

&$X4' ) The chronology of the succession to the Ilkhanate

Ruler On date Next ruler Crowned in On date Interval died in was in (days)

Hülegu FY February FHWV Abaqa Chaghan FY June FHWV FHZ Nr. Maragha Mazandaran Na’ur Abaqa IF March FH[H Ahmad Jaghatu FW April FH[H FW Hamadan Kurdistan Ahmad FZ August FH[J Arghun Yuz Aghach FF August FH[J F Yuz Aghach Yuz Aghach Arghun Y March FHYF Gaikhatu Akhlat HH July FHYF FIV Baghcha, Anatolia Arran Gaikhatu HJ March FHYJ Baidu Tabriz H May FHYV IY Pilasuvar Not clear Baidu J October FHYV Ghazan Tabriz W November FHYV II Tabriz Ujan Ghazan F\ May FIZJ Öljeitü Ujan FY July FIZJ WI Qazvin Khurasan Nr. Öljeitü FW December FIFW Abū Saʿīd Tabriz V July FIF\ F\Z Sultaniyya Khurasan Abū Saʿīd IZ November FIIV Arpa Ke’ün Qarabagh V December FIIV V Qarabagh Qarabagh

the failure of Arpa Ke’ün and the rapid turnover of alternatives, eventually a quriltai was held to attempt to resolve the issue in a more traditional way.T These developments have been rehearsed at length elsewhere, perhaps in most detail in my earlier study on the last years of the Ilkhanate, in which I attempted to provide a narrative of the sequence of revolts, battles and con+,icting claims that marked the loss of the ruling dynasty’s authority.U

V Aubin, “Le quriltai de Sultân-Maydân.” W Melville, The Fall of Amir Chupan. The origins of the current paper can be traced to the period I was working on that study, and to a book proposal for the series on ‘Historical Endings’ (Hodder Arnold) that was never accomplished. An earlier version of this paper was presented at the conference “Empire and After: the Middle East and Central Asia,” held in St Andrews, Scotland, on 21 October 2005, at which David Morgan also presented a paper, later published as “The Decline and Fall of the Mongol Empire.” My paper was lightly revised *)] 6789:887

That study was not in itself intended as an exploration of the ‘decline and fall’ of the Ilkhanate, so much as to chart the course of events that brought the Ilkhanate to an end;^ nevertheless, its title suggested my belief in the decline of the centralized power of the Ilkhanate during this period, whereas others, most notably David Morgan, have believed that the Ilkhanate deBCed “Gibbon’s Law” and fell without having previously declined._ While I accept with Morgan that Abū Saʿīd’s reign was not conspicuously or quantitatively ‘worse,’ or much more fraught with internal dissentions than those of earlier khans, and that, in the words of Peter Jackson, after Chūpān’s fall in 1327, “Abū Saʿīd’s last years have an air of greater stability than the era of nominal rule,”` I do not think it necessarily follows that Ilkhanid rule would not have collapsed when it did, “or for long after, had the line of Hülegü continued uninterruptedly after the death of Abū Saʿīd.”?a In one sense, of course, this is obvious—if the line of rulers had continued, Ilkhanid rule would have continued. But my point, on the other hand, and the purpose of this preamble, is to rea-BCrm my belief that the authority of the ruling house had declined and that, as in Hamlet’s Denmark, “something was amiss.” Despite the view expressed by some, that terminal decline set in during the reign of Abū Saʿīd,?? I am content to agree that in material and economic terms (insofar as we can assess the latter), late Ilkhanid Iran was relatively prosperous and even saw an improvement on earlier times, as remarked by some of the Persian chroniclers writing in the turbulent aftermath of the death of Abū Saʿīd. Auliyāʼ Allāh Acmuld̄, for instance, writes “due to the death of the noble soul of Abū Saʿīd, that situation became reversed, and security was exchanged for fear, justice for oppression, ease for toil, cultivation for destruction.”?A The general, or undeBCned, notion of decline is indeed not very helpful; in the words of Bennet Bronson, “How do we know that the symptoms we observe are those of decline? Because the state in question eventually falls. And how do we know that these symptoms of decline are causative agents? Either because we think

for the 35"f'. conference in Barcelona in July 2010. I am grateful to Bruno De Nicola for his persistence in ensuring it has eventually found its way into print. \ Ibid., 3. [ Morgan, Medieval Persia, 78; idem, “Review,” 413; The Mongols, 2nd ed., 50, and “The Decline and Fall of the Mongol Empire,” 7. Y Jackson, “Abu Sa‘id,” 376. FZ Morgan, “Mongols in Iran,” 134–35; “Decline and Fall,” 6–7. FF Marek and Knizkova, The Jenghiz Khan Miniatures, 16, start the decline with the accession of Öljeitü (i.e. after the ‘high point’ of Ghazan); Wilber, The Architecture of Islamic Iran, 26: both are works of art history. FH Ā mulī, Rūyān, 178–79, writing c. 1360. KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S *)* we see them getting worse as the end approaches or because we have deBCned them that way.”?D We have no reasonable way to measure what is a reversible process (it had already been reversed once, by Ghazan Khan; later rulers like Shāh ‘Abbās also restored the fortunes of the Safavid state evidently on the verge of collapse), and we cannot say that the Ilkhanate ‘fell’ to any competitor (unlike the case of the Safavids), so much as came to an end due to its internal disfunction- ing. That end is, nevertheless, of great interest as explaining the nature of the Mongol state as it had evolved in Iran,?E as is what came next. As Rudi Matthee’s recently published masterpiece on the second Safavid century reminds us, regardless of whether ‘decline’ is still fashionable among modern historians, some explanation of failure is necessary.?T Indeed, his discussion of the wider analysis of the Safavid state seems to hold good for much of the earlier periods of Iranian history. While the succession to Abū Saʿīd was clearly the immediate di-BCculty, this was not a new problem; and it is not impertinent to observe that the absence (or presence) of direct heirs did not cause (or prevent) the downfall of Mongol rule in China or the Mongol khanates in the Dasht-i Qipchaq and Central Asia.

Iran and the Mongol World Empire

As is clear from what David Morgan has said about the Mongol Empire as a whole,?U it should be observed from the outset that in concentrating on the Mongols in Iran, it is debatable whether it is appropriate to think in terms of the ‘end of empire’ for the Ilkhanate. While its size alone might qualify the Ilkhanate to be called an empire—it certainly covered much the same geographical area as that ruled by the Safavids—in other respects it was barely a coherent state. At best, it was merely a part of a larger whole, albeit an important part. At worst, it was essentially an independent polity of dubious legitimacy in Chinggisid terms, and itself evidence of the fact that the Mongol Empire had already fragmented and dissolved in the succession crises of the

FI Bronson, “The Role of Barbarians in the Fall of States,” 197–99. FJ I refer here to a long-running discourse on the nature of Mongol rule around the question, ‘who ran the empire?’, for which see, for example, Morgan, “Mongol or Persian”; Fragner, “Ilkhanid Rule,” and Melville, “The keshig in Iran,” FV Matthee, Persia in Crisis, xxiv–xxx. FW Morgan, “Decline and Fall,” 7–10. *)h 6789:887

1250s and 1260s.?^ In particular, with reference to the empire as a whole, we should also note the obvious, that the Ilkhanid regime did not come to an end because of the collapse of the Yuan dynasty in China—in other words, the periphery did not fall with the centre, but actually preceded it by some 20–30 years (depending on when one dates the ‘end’). This of course was because China was not the centre of the empire by this time. It would, however, be interesting to know how the events in Iran in the 1330s–50s were received and recorded in the Yuan capital, and whether the news of the breakdown of central authority in the Ilkhanate had any repercussions in China. I am not in a position to answer this question myself. Regardless of how we view the status of the Ilkhanid polity, comparisons with other parts of the empire are certainly helpful and illuminating, and even if we are to concern ourselves only with why Ilkhanid rule ended in Iran, we should surely look beyond the immediate and apparent reasons, to the underlying structural or fundamental problems that weakened the cohesion of the regime and made it unable to maintain itself at a moment of crisis. We will, therefore, consider examples from the other parts of the empire as appropriate and as evidence permits; the case of China is clearly the most instructive, whereas the Mongol khanates in southern Russia and Central Asia were so dif- ferent as to be interesting largely for the contrasts they o-fer. There are rather few scholarly attempts to draw together detailed information on the last stages of the Mongol Empire in its di-ferent regions and to formulate an analysis of the issues concerned; my remarks are only of a preliminary nature, drawing on a few secondary studies.?_ We should bear some other general considerations in mind as well. First, the fact that descendants of Chinggis Khan reigned over Iran for around a century may seem rather a poor outcome, given the overwhelming force with which they established their rule and the lack of serious military opposition. Looked at in another way, however, it might seem a remarkable achievement. We may recall that the ‘Great’ Seljuqs only maintained their unitary rule in Iran for less than half that time, from their entry into Baghdad in 1055 to the death of Malik Shāh in 1092. I do not necessarily, therefore, equate the collapse of the

F\ As BCrst explored by Jackson, “The Dissolution.” F[ General accounts of the phenomenon of collapse include Roux, Histoire de l’empire mongol, 489–507 (mainly narrative), and Kwanten, Imperial Nomads, 225–58 (a fairly substantial overview). Vernadsky, The Mongols and Russia, 130–37, raises a series of interesting questions on the “inner contradictions in the Mongol Empire,” which are relevant to the wider topics discussed here, although it is not my aim to engage with all these points. KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S *)i

Ilkhanate with its failure. As noted above, the last generally recognized Ilkhans all seem to have been more or less competent and their reigns more or less stable—though it is not at all clear to me now what criteria we can or should use to measure this. I will touch shortly on how contemporaries judged the question, but it is worth remembering here that the bias of the historiographical record certainly presents a one-sided view of the merits of the Muslim khans from Ghazan to Abū Saʿīd compared with their pagan predecessors. We have no version of how the Mongols themselves viewed the situation, except by reading between the lines of the Persian chroniclers. Although leadership qualities alone do not seem to be a major problem, therefore, it is likely that Eisenstadt’s observation (reproduced at the head of this paper) about di-ferent social groups developing “strong autonomous political orientations” provides a useful formulation of the situation.

Some Contemporary Views

Aḥmad A+,ākī, the author of a hagiography of the SuBC saint, Jalāl al-Dīn Rūmī and his successors, explains the end of the Mongol Empire in Persia with the following anecdote:?` Rūmī’s descendant, Chelebi ʿĀrif, went to the court (ordu, camp) in Tabriz during the reign of the last Ilkhan, Abū Saʿīd. There, together with an unnamed Mongol prince, he had an audience with the chief minister, Ghiyāth al-Dīn Muḥammad, and explained to him a few simple needs of the dervishes in Anatolia. He was rebu-fed and “not treated with the attention he deserved” and his requests were not attended to. In a storming rage, Chelebi ʿĀrif and the prince returned to Anatolia, the one regaining Konya and the other embarking at Sinop. This proved the downfall of the Mongols’ kingdom and their fortunes were overturned: “the great men of those regions fell on one another and not one of them remained. Having become enemies, they overturned each other and their unity became fragmented.” They were still at it when A+,ākī was writing, c. 1353 (hanūz dar ān kār-and),Aa and no-one had continued to prosper. Chelebi ʿĀrif himself died in 1338.

FY A +,ākī, Manāqib al-ʿĀrifīn, 2:980–81, referring to the vizier as Shams al-Dīn. French translation by Clément Huart, Les saints des derviches tourneurs, 2:417–18; English translation by John O’Kane, The Feats of the Knowers of God, 686–87. HZ A +,ākī, Manāqib al-ʿĀrifīn, 981. *)k 6789:887

Losing the protection of a saint of the age may not seem to modern readers an adequate reason for the disintegration of a mighty military empire, even though in mediaeval Iran, as in Europe, rulers and saints were the ‘heroes of the age’ and except for the rare occurrence of a saintly king (such as Saint Louis), there is an implicit antithesis between them.A? In its time, therefore, such a belief was not unusual. In the Chinese view of things, rulers enjoyed the ‘Mandate of Heaven’ and prospered under its favour. The forfeiture of this mandate, usually signaled by a series of natural catastrophes—the +,ooding of the great rivers, famines, plagues or civil insurrection, spelled (in retrospect at least) the inevitable demise of the ruling dynasty and its replacement by another.AA We encounter similar beliefs elsewhere. They imply that both the rulers and the ruled believed in the God-given quality of kingship, which could be bestowed and withdrawn, often seemingly at random, but particularly as a consequence of some o-fense against wisdom, justice, or the natural order of things. But realistically, the Mongols were not brought down by the rage of a still rather obscure mystic from the western extremities of their realm. Indeed, there were other saints, closer to home, who had enjoyed the attentions of the Mongol court and the patronage of the chief minister, Ghiyāth al-Dīn Muḥammad. Shaykh Ṣafī al-Dīn (d. 1334) was a frequent visitor to the Mongol court in Tabriz and also received the Ilkhan Abū Saʿīd and other leading men of state in his convent at Ardabil, conveniently situated between the winter pastures of Mughan and Arran and the capital cities of Tabriz and Sultaniyya. The Mongol court was a mobile one, and the rulers lost no opportunity to visit such a charismatic holy BCgure on their way. Whatever the SuBC shaykhs thought of these barbaric warlords, their association held mutual beneBCts: in material wealth and increased prestige for the shaykh, and in a vicarious spiritual sanction for the sultan. So Ghiyāth al-Dīn might have slighted Chelebi ʿĀrif of Konya, but he had not slighted Ṣafī al-Dīn of Ardabil,AD and we can hardly accuse the Mongol court of being unaware of the value of cultivating the approval of local holy men. Perhaps Ghiyāth al-Dīn picked the wrong one: the shaykhs demanded absolute devotion and were vicious and merciless in their

HF See the stimulating collection of articles around this theme, united in Le Go-f, Héros du Moyen Âge. HH Chan, Legitimation in Imperial China, 42–43, 47; also Yates, “Cosmos and Central Authority,” 353. For a detailed account of the “nine sloughs” in Yüan–Ming history, see Brook, The Troubled Empire, 50–78. In Chinese historiography, dynastic history of one regime was compiled from its records by its successor. HI Jackson and Melville, “Gīāt al-Dīn,” 599. KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S *)m treatment of rivals.AE Such a mistake was also the cause of the downfall of the Seljuq regime in Anatolia, again according to A+,ākī.AT Another contemporary view appears at BCrst sight to be far more pragmatic, but its underlying argument is conceived in the same terms of reference. This is the cynical treatment of the subject by the scurrilous satirist, ʿUbayd-i Zākānd̄ (d. c. 1371):

When Hulagu Khan conquered Baghdad he ordered all the inhabitants who had escaped the sword to assemble in front of him . . . He separated the judges, shaykhs, suBCs, hajjis, preachers, noblemen, sayyids, beggars, religious mendicants, wrestlers, poets and story-tellers from the rest and . . . by ordering them to be drowned in the Tigris, puriBCed the earth of their vile existence. Consequently, sovereignty was BCrmly established in his family for about 90 years, and their prosperity increased daily. But since poor Abū Saʿīd was obsessed with the idea of justice and distinguished himself with this quality, before long the days of his monar- chy were numbered, and Hulagu’s house and endeavours disappeared through the aspirations of Abū Saʿīd.AU

This is a most perceptive characterization of the evolution of the Ilkhanate. It provides a reverse side of the coin struck by A+,ākī. Regardless of whether the Mongols fell because they paid insu-BCcient attention to the claims of the dervishes, or because, on the contrary, they became softened by too much concern for justice and for the religious classes, whom earlier Hülegü had dumped in the river, the question is how comfortable the Mongol elites felt, trying to operate within the cultural framework of the conquered populations. The success or otherwise of a ruler or dynasty was measured in the Persian sources with reference to Islamic principles, which has tended to polarize the tensions in society in one set of terms of reference, rather than others such as the dichotomy of Turk vs. Tajik, or nomad vs. urban (or agriculturalist), or cen- tralization vs. dispersed and consensual rule. The intense focus on the person

HJ Aigle, “Charismes et rôle social des saints dans l’hagiographie persane médiéval (no–npo siècles),” 20; she does not include in her survey the miracles of Shaykh Ṣafī al-Dīn, a whole category of which fall under the headings of “miracles of pure severity” and “miracles of kindness and severity,” see Ibn Bazzāz, Ṣafwat al-ṣafā, 356–431. HV A +,āki, Manāqib al-ʿĀrifīn, 0, 146–47; tr. Huart, 0, 113–14; tr. O’Kane, 102–3. A+,ākī himself died in 1360. HW ‘Ubaid-i Zākānī, Akhlāq al-ashrāf, 243; translated by Hasan Javadi, The Ethics of the Aristocrats, 40. *)q 6789:887 or behaviour of the ruler, and his fortune (dawlat), however, largely transcends these characteristics of the society he ruled and which he had to keep in balance. In the Persian context, the profession of Islam went together with the more ancient respect for the unquestionable majesty of the ruler, but hovering over these qualities, which could be a-fected by personal choice, remained the random interventions of fate, which operated according to its own inscrutable rhythms. It is not surprising, therefore, to BCnd little e-fort at real analysis by the Persian chroniclers, however accurately they were able to observe and report the last years of Ilkhanid rule. The overthrow of rulers and changes of dynasty are generally greeted with phrases reminiscent of Firdawsī’s Shāhnāma (‘Book of Kings’), expressing the sentiment that when Heaven’s wheel had determined the end, the eye of wisdom and discernment was stitched up, although general accusations about the failure to heed advice, neglect of the a-fairs of state, and absorption in private pleasures are also recurring topoi.

The End of the Ilkhanate

The death of Abū Saʿīd in 1335 without an heir is normally given as the main reason for the collapse of the dynasty, and there is no great need to challenge this obvious point of view. We may draw attention to two points, however. First, why didn’t he have an heir? It can be seen from Table 2 that the reproduc- tive power of the Ilkhans’ ruling family deteriorated as time went by, despite marriages to several wives and the enjoyment of many concubines.

&$X4' ] The Ilkhans, their wives and their o(fspring

Ilkhan Reign Age at Wives Concubines Sons Daughters (years) Death (survived) (survived)

Hülegü Y JY V Many FJ FF Abaqa FW JY Y W H \ Tegüder I I\ W Many I W Arghun \ II FZ Many J J Geikhatu J VZ W Many I J Baidu F I Ghazan Y II \ F (Z) F Öljeitü FH IW FH W (F) I (H) Abū Saʿīd F[ IF W Z F KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S *)r

John Masson Smith has written about this problem,A^ particularly the role of alcohol in damaging the fertility of the Mongols—a neat but inadequate explanation, as it fails to explain (Turko-)Mongol drinking elsewhere and in other times having no such e-fects. Nevertheless, something seems to have been wrong with the genetic composition of the ruling family. Secondly, however, just because Abū Saʿīd had no direct (male) heir, it does not mean that no Chinggisid, Toluid and even Hülegüid successors were available. As discussed elsewhere, the years immediately after the death of Abū Saʿīd were characterized by the promotion of numerous would-be Ilkhans, respectively descendants of Ariq Böke, Baidu and Anbarji.A_ The last recognized Chinggisid, Taghay (Togha) Temür, descended from Jöchi-Qasar, brother of Chinggis Khan, was murdered by the Sarbadarids in 1353 at his camp in Sultan-Duvin.A` These BCgureheads were sponsored for the most part by rival Mongol military chiefs (see also below). In the past, and in principle, there was no formal restriction on the succession, and a large number of crises and con+,icts between cousins bears witness to the fact that the succession was a free-for-all. A di-ference here was that the pretenders were not mounting bids in their own right, mobilizing their own supporters, but on the contrary were evidently plucked from obscurity and used simply to further the ambitions of the noyans. One could thus suggest that the failure to agree on one candidate shows paradoxically a diminution of respect for the whole notion of Chinggisid legitimacy—that it was now less important to rally round a qualiBCed candidate in the interests of maintaining the dynasty, than to pursue maximum personal advantage and power. Jean Aubin, with his usual insight, goes so far as to describe Abū Saʿīd as the BCrst of the puppet khans, during his minority.Da There are interesting parallels here with the demise of other Islamic dynasties. The case of the Ayyubids is not dissimilar: various Ayyubid puppets, and even the wife of al-Malik al-Ṣālih Ayyūb (Shajar al-Durr) being maintained by rival factions of Mamluks, until the Mamluks themselves took over (a development assisted, ironically, by the Mongol invasion of Syria in 1260).D? One thinks here of the brief career of Sātī Beg, sister of Abū Saʿīd, who was BCrst betrothed to Amir Chūpān, then considered for elevation to the throne on the death of

H\ Smith, “Dietary Decadence,” 50–51. His BCgures di-fer very slightly from my own. H[ Thus Arpa, Mūsā Khan and Muḥammad Khan, cf. Melville, The Fall of Amir Chupan, 44, 46, 51. The repeated bloodlettings that marked the outset of the reign of Ghazan and later episodes did, however, clearly reduce the pool of possible contenders. HY Smith, The History of the Sarbadār Dynasty, 135–39. IZ Aubin, Émirs mongols et vizirs persans, 85. IF Holt, The Age of the Crusades, 82–89; also Irwin, The Middle East in the Middle Ages, 26–29. *]; 6789:887

Abū Saʿīd, but was instead married to Arpa Ke’ün in an e-fort to boost his legitimacy. She was installed on the throne for nine months in 1338–39 in her own right by the Chupanids before being married to their BCnal surrogate, a certain Ilyās said to be descended from prince Süge, who was enthroned as Sulaymān Khan.DA In the case of the ‘Abbasids, however, the domination of the (mainly Turkish) army commanders, also Mamluks for the most part, after the late ninth century never progressed to the stage that they felt able to dispense with an ‘Abbasid caliph altogether. As a result, the caliphate lasted another 350 years, despite being largely powerless for much of this time, despite establishing a strong regional presence in the twelfth and early thirteenth century: such was the prestige of the o-BCce and the established legitimacy of the ‘Abbasid family.DD We observe that the same can evidently not be said of the Chinggisids. One has the sense that the Persian historians and bureaucrats (whose views the historians re+,ected) probably had a greater respect for the imperial regime, as guarantors of order and legality, than did the Mongol noyans, despite their superBCcial lip service to the dynastic right of the Ilkhans. It was not until the rise of Temür, and the later Turkoman regimes, that Chinggisid and Ilkhanid precedents were once more invoked.DE The weakness of the last e-fective Ilkhan must have accelerated this distanc- ing of the noyans from the dynasty. Aged only 12 on his accession, he was dominated BCrst by Amir Chūpān and his son Dimashq Khwāja, then after their fall, by Chūpān’s daughter, Baghdād Khatun (dimashq-i digar gasht baghdād-i ū),DT and the vizier, Ghiyāth al-Dīn, whose ability should have been a su-BCcient guarantee of the proper functioning of government. Rudi Matthee has drawn attention to the high quality and relatively long periods of o-BCce of the senior ministers in the second Safavid century—often labelled as one of steady ‘decline’—which ensured an e-fective administration even when the Shah was unBCt.DU The main problem arose when the Shah failed to fulBCl his essential role as mediator between the various factions and interest groups at court, with a consequent lessening of authority and coercive strength.D^ The same surely

IH Shab ānkāra’ī, Majmaʿ al-ansāb, 293; Āharī, Tavārīkh, 226–27. See Album, “Studies in Ilkhanid History 0,” 77–83 and “Studies in Ilkhanid History 00,” 48–56, for the extent of territories notionally under her aegis. II Kennedy, “The Decline and Fall of the First Muslim Empire,” o-fers an analysis of the early ‘Abbasid caliphate. IJ See Melville, “Between Tabriz and Herat,” 32–33, and earlier, Manz, “Mongol History Rewritten,” esp. 143–46, and her “Family and Ruler.” IV ḤāBCz-i Abrū, Zail, ms. Or. 2885, f. 412v, cited in Melville, The Fall of Amir Chupan, 29. IW Matthee, Persia in Crisis, chapters 2–3, expanding on earlier studies. I\ Ibid., esp. 246–49. KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S *]) applies to the Ilkhanid period and, as I have shown, the remainder of Abū Saʿīd’s reign after the fall of Chūpān was marked by a string of challenges to his authority. While there is perhaps nothing so unusual in that, and while he was able to overcome them, one has the sense that the factional political ambitions of the noyans were already well advanced, especially when their interests were challenged, as seems to have been the case with Ghiyāth al-Dīn’s e-forts at reform.D_ The details are barely known, but this appears to have involved an attempt on the part of the central administration to recover control of the lands and their revenues alienated from the treasury due to their assignment as appanages to the amirs. To this end, he set a BCxed rate for taxes (amvāl) for the province and for expenses for the upkeep of troops and required the balance to be sent to the central treasury. Ghiyāth al-Dīn’s measures, and their failure, anticipated similar e-forts in the next century, when both Qāḍd̄ ʻIsác Sāvajd̄ in the Aq Qoyunlu realms and Majd al-Dīn Muḥammad Khwāfd̄ in Timurid Khurasan undertook such programmes of administrative reform.D` The timing of these two initiatives (1489 and 1487–90 respectively), and the close links between the two courts during this period,Ea suggest a common inspiration. If such issues of BCscal and administrative control were still not resolved by the late BCfteenth century and threatened the cohesion of the kingdom, it is hardly surprising if they were an even greater problem for the Ilkhanid bureaucrats of the fourteenth century. These developments reveal clearly the transitional phase through which the Ilkhanate was passing by the 1330s, how the regime was neither one thing nor the other, and how old loyalties were being undermined but had not yet been replaced by new ones. For the functionaries and bureaucrats, loyalty to the dynasty was facilitated by the Mongols’ conversion to Islam and the patronage of art, letters and architecture that began to revive in the BCrst decades of the fourteenth century.E? The Mongols also satisBCed a deeper desire for a truly imperial regime based on Iranian soil (Īrān zamīn), the BCrst since the collapse of the Sasanians and one that not only attracted the spark of divine charisma (farr) but could be

I[ Mustawf ī, Nuzhat, 147, trans. Le Strange, 146–47; and Jackson and Melville, “Gīāt al-Dīn,” 598. IY Minorsky, “The Aq-qoyunlu and Land Reforms,” 451–58; Woods, Clan, Confederation, Empire, 144–45, 151–52 (Qāḍd̄ ʻIsá);c Subtelny, Timurids in Transition, 89–99 (Majd al-Dīn). JZ Lingwood, Politics, Poetry, and Su*+sm, esp. Chapter 4. JF For an attractive survey of this artistic upsurge, see Komaro-f and Carboni, The Legacy of Genghis Khan, and several studies in the follow-up volume, Komaro-f, Beyond the Legacy of Genghis Khan, 167–368. *]] 6789:887 represented as a new heroic age in terms reminiscent of, and to a certain extent modelled on, Firdawsī’s Shāhnāma.EA The various verse chronicles composed at this time, and especially the Ẓafarnāma of Ḥamd Allāh Mustawfī, bear witness to this development.ED For many Mongol chiefs, however, this transformation was not so complete nor so welcome. On the one hand, respect for their past and their traditions was being eroded by the adoption of new ways—as exempliBCed by Ghazan’s public mausoleum built in the Sham quarter,EE the abandonment of the quriltai as the forum for electing a new ruler,ET perhaps even, too, the abandonment of e-forts to defeat the Mamluks with the peace of 1323,EU and attempts at reform that were aimed at the privileged position of the inner circles of the ruler’s household and the BCnancial gains of the conquerors from the spoils of conquest.E^ On the other hand, erosion of Chinggisid discipline and adherence to the yasa was not balanced by a compensating loyalty to the regime that no longer demanded loyalty in these traditional terms. If the Chinggisid dispen- sation was going to break down, then the noyans would revert to their own ingrained behaviour, the politics of the steppes, rather than to some new and still alien framework of Perso-Islamic expectations. The lack of an heir, or of a consensus about which successor might be acceptable, is thus not on its own a su-BCcient explanation for the end of Ilkhanid rule. In China 30 years later, when, faced with internal military challenges and a total breakdown of authority, the last Great Khan threw in the towel and many took the long way home back to the steppes, the mismatch between Mongolian and indigenous (Chinese) norms was as marked as in Persia. As noted by Elizabeth Endicott-West, once back there, “the nomadic cultural identity of the Mongols was little a-fected by one hundred or so years living within China.”E_ In both cases, the ‘last’ established rulers, Abū Saʿīd (1317–35) and Toghōn Temür (1333–68), both the same young age on coming to the throne, had the longest reigns, which should have been periods of stability and consolidation. In both

JH Melville, “The Royal Image,” 347–51. JI Melville, “Between Firdausī and Rashīd al-Dīn,” and “The Mongol and Timurid Periods,” 193–97. JJ See Haneda, “The Pastoral City and the Mausoleum City,” 144–55. JV Ḥasan-i Buzurg’s complaint against ʿAld̄-Pādshāh, in Shabānakāra’ī, Majmaʿ al-ansāb, 303; cf. Melville, The Fall of Amir Chupan, 66. JW Amitai, “The Resolution of the Mongol-Mamluk War,” esp. 381–84. J\ Rashīd al-Dīn’s reforms under Ghazan were aimed at abuses that had developed since the start of the period, not least among the ruler’s household, see Melville, “The keshig in Iran,” 150–51. J[ Endicott-West, “Aspects of Khitan Liao and Mongolian Yüan Imperial Rule,” 215. KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S *]* cases, however, the court found itself increasingly to be steering a ship on a course set by others and lost sight of the star that had brought them there in the BCrst place.

After the Ilkhanate

As Susan Alcock has written, “empires possess a potent afterlife.”E` This is as true of the Ilkhanate as of many other imperial regimes, and importantly, as true also of the situation in China, despite what one might imagine to have been the opportunity provided by the restoration of Chinese rule and the erad- ication of the last of a series of ‘barbarian’ dynasties.Ta However, it is not my intention here to explore this aspect of ‘empire and after.’ Rather, it might be helpful to look back on the Ilkhanate from the perspective of what followed, of the elements into which the Ilkhanate immediately disaggregated and which, therefore, were necessarily the constituent elements that had temporarily, at least, been glued together by the regime. As already noted, after the death of Abū Saʿīd, various Chinggisid princes were put forward as claimants to the throne. In themselves, they are of little interest (in representing any sort of political agenda, at least), though it is sig- niBCcant that they did not come forward on their own initiative: there was thus no strong leader from within the Ilkhanid family itself. It is their backers who merit attention. There is no need to repeat what has already been gone over in detail elsewhere,T? except to concentrate on the main points. The BCrst con- tender, Arpa Ke’ün, was a descendant of Ariq Böke, son of Tolui, an obscure but seemingly competent commander, a Mongol of the ‘old school,’ who was let down largely by his association with the vizier, Ghiyāth al-Dīn, who inserted him on the throne. This serves to emphasize the incompatibility of the tendency towards a Perso-Islamic framework of government with a more traditional Mongol orientation; also, perhaps, the aversion to a relatively strong BCgure emerging to thwart the aspirations of the noyans. His rival, ʿAld̄-Pādshāh, was Abū Saʿīd’s uncle, but did not claim the throne for himself, advancing instead a certain Mūsā, said to be descended from Baidu. ʿAld̄-Pādshāh, with his royal connections, could also be said to represent

JY Alcock, “The Afterlife of Empires,” 370. VZ Dreyer, Early Ming China, 1–2, instead draws attention to the “intense sense of continuity” with the Yuan. VF Spuler, The Mongol Period, 40–42; Roemer, “The Jalayirids, Muza-farids and Sarbadārs” and Melville, The Fall of Amir Chupan, 43–74. *]h 6789:887 a legitimist (dynastic) strand, but a divisive one. He had previously displayed a disregard for his young nephew, Abū Saʿīd, and was evidently undone by his domineering behaviour, which eroded his support. He also represents, more clearly than most, a tribal identiBCcation, as leader of the Oirats in the northwest Zagros region and Mesopotamia. Both these challenges were ephemeral and tell us relatively little.TA More signiBCcant are the longer-term polities that emerged. Naturally, each had its regional character, in that they formed in di-ferent geographical provinces of the Ilkhanate, but this should not be seen as a manifestation of frustrated regional ambitions surfacing once the restraining hand of central control was removed. I do not consider the size and variety of the Ilkhans’ territories to be a major factor undermining their rule, though it is worth noting that the south of Iran, untouched by the Mongol invasions, was never really under direct central government control (it is perhaps interesting to observe that the Yuan’s control over southern China, even after the BCnal defeat of the Song, was also far less direct and their physical presence much less than in the north). It is su-BCcient to say a few words about each of these regimes or ‘party kingdoms’ into which the Ilklhanate fragmented. The BCrst, the Jalayirids, under their leader Shaykh Ḥasan-i Buzurg, defeated ʿAld̄-Pādshāh but were in turn repeatedly defeated by their leader’s rival and namesake, Shaykh Ḥasan-i Kuchik the Chupanid, before these two opponents established themselves in their own territories. Both had close connections with the Ilkhanid royal family, but they represent opposite and competing tendencies in their attitude to rule. The Chupanids, descended from the former senior amir, Chūpān, established in north-west Iran, Georgia and eastern Anatolia, maintained a violent, aggressive and unprincipled control over the traditional Mongol heart- lands, as ruthless and BCckle in their dealings among themselves as with their opponents.TD That such untamed forces could emerge after 80 years of Chinggisid rule shows how patchy was the accommodation of the Mongols to Iranian conditions. They held sway until 1357, during which time they cynically maintained a series of puppet Chinggisids (at least until c. 1345),TE including for a time, as noted earlier, Abū Saʿīd’s sister, Sātī Beg.

VH Melville, The Fall of Amir Chupan, 44–53. VI Melville and Zaryab, “Chobanids,” 500–502. VJ The last of the Chinggisids promoted by the Chupanids, Sulaymān Khan, was rejected by them, but acknowledged by Ḥasan-i Buzurg Jalayir brie+,y in 1346. The Chupanids’ BCnal puppet, sarcastically nicknamed Anūshirvān, was not a Chinggisid; coins were struck in his name at least until 1355, see Album, “Power and Legitimacy,” 158. KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S *]i

The Jalayirids, named after a prominent Mongol tribe, who BCnally beneBCted from the Chupanids’ disappearance, maintained a more moderate regime that lasted into the BCfteenth century, centred on Baghdad and later also Tabriz. Ḥasan-i Buzurg used only the title he had held under Abū Saʿīd, Ulus Beg.TT The Jalayirids inherited a commitment to stable rule respectful of Islamic principles, while not forgetting their steppe origins, and became notable patrons of artists and writers (including the poet, ʿUbayd-i Zākānd̄). They thus continued along the path tentatively opened up by Ghazan and his successors, towards greater integration. Sultan Shaykh-ʿUways (1356–74) commissioned a verse epic about Ghazan Khan, the Ghāzān-nāma, completed in 763/1362 by Khwāja Nūr al-Dīn b. Shams al-Dīn Muḥammad Azhdarī, a somewhat fanciful work that mythologizes the Ilkhan as an archetypal philosopher-king.TU While displaying contrasting aspects of Mongol rule, these were genuinely ‘Mongol’ successor states, and both advanced the claims of Chinggisid puppet Khans (Ḥasan-i Buzurg dropped the pretence in the 1340s). The Injü’ids, on the other hand, were an indigenous Sayyid family said to be descended from the SuBC scholar ʿAbd-Allāh Anṣārī (d. 1089).T^ Their founder, Maḥmūd-Shāh, governed Fars and raised revenues from crown lands (injü) as an agent of the Ilkhans. They too, were patrons of the arts (and of the poets, ḤāBCẓ and ʿUbayd-i Zākānd̄), though in their tortuous political manoeuvering and aggression they were closer to the Chupanids (with whom they were closely involved) than to the Jalayirids (with whom they occasionally sought refuge), before Abū Isḥāq, son of Maḥmūd-Shāh, was eclipsed by another force in southern Iran, the Muza-farids, in 1357. These were descended from a family of Arab origin and named after their ancestor Sharaf al-Dīn Muẓa-far. His son Mubāriz al-Dīn served in the Mongol provincial governorship around Yazd, where he was recognized as governor by Abū Saʿīd in 1319. Both the Injü’ids and Muza-farids could be thought in some ways to be championing the continuing independence of southern Iran from Mongol rule; they also reveal, however, the extent to which political life had been corrupted by the example of decades of Mongol power.T_ The Muza-farids (like the Injü’ids) disappeared in a blaze of interne- cine BCghting that made their defeat by Temür in 1393 a foregone conclusion. Abū Isḥāq was the BCrst of the ‘successor’ rulers to strike coins in his own name

VV Roemer, “The Jalayirids, Muza-farids and Sarbadārs,” 5. VW Melville, “Gāzān-nāmā,” and “History and Myth,” esp. 134–35 and 142–43. V\ Roemer, “The Jalayirids, Muza-farids and Sarbadārs,” 11–13; Limbert, “Inju Dynasty,” 143. V[ Roemer, “The Jalayirids, Muza-farids and Sarbadārs,” 11–16; Jackson, “Muẓa-farids,” 821; the most comprehensive treatment of events in the south of Iran in this period is by Denise Aigle, Le Fārs sous la domination mongole, 173–99. *]k 6789:887

(from 1344); Mubāriz al-Dīn BCrst recognized Taghay Temür and the ‘Chupanid’ Sulaymān Khan, then Abū Isḥāq Īnjü, before minting anonymous coins and BCnally issues in his own name (from at least 1356). Most interestingly, these last issues recognized the shadow ‘Abbasid caliph al-Muʿtaz̤id in Cairo, suggesting that older concepts of political legitimacy still held some signiBCcance, and certainly allowing Mubāriz al-Dīn to break free of any associations with the now defunct Ilkhanate.T` Quite distinct was the last of the powers to establish themselves on the ruins of the Ilkhanate, namely the Sarbadars of western Khurasan. A group of local landowners, rising up against repressive Mongol tax collecting, took advantage of the breakdown of government after Abū Saʿīd’s death to establish themselves in the region between Nishapur and Sabzavar and to resist successfully the local Mongol chiefs—murdering the last Chinggisid Khan, Taghay Temür, in 1353, as already mentioned. What makes the outlaw (‘gallows-bird’) Sarbadar regime so particularly interesting is its association with the disciples of the local dervish Shaykh Ḥasan Jūrī, an unholy alliance that went through several crises, but nevertheless reveals the growing in+,uence of SuBC shaykhs in political life and their role as leaders of the people in the face of military regimes that proved themselves far from the shepherds caring for their +,ocks that Islamic political theory required.Ua With the Sarbadars, we are not so far from the point at which we set o-f, with Chelebi ʿĀrif’s complaints about the indi-ference of the Mongol court to the needs of the dervishes. These complaints may be irrelevant to the question of the end of the Ilkhanate, but they serve to show some of the forces that the Mongols needed to master to remain in control, and the varied aspirations they needed to satisfy, to retain the loyalty of their military backbone. The di-ferent regimes that sprung up on their demise, while often professing a Chinggisid loyalty, and originating either on the fringes of the ruling family or among their provincial administrations, show that the process of accommodation was still far from complete 40 years after Ghazan Khan took the Mongols into the fold of Islam. In the short term, at least, this seems to have had the e-fect of undermining the authority of the Ilkhanid house among its military commanders, who were not disposed to perpetuate it at a moment of crisis, but sought instead to pursue their own more instinctive and self-interested ambitions. Although, naturally, the details di-fer—and the timing, in that the regional uprisings in China (the period of ‘rival contenders’) occurred while the unitary

VY Album, “Power and Legitimacy,” esp. 167–69. WZ See also recently, Mahendrarajah, “The Sarbadars of Sabzavar.” KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S *]m

Yuan state was still in existence and not already fragmented—it is interesting to note, BCrst, that some parallels may be found with the Sarbadar uprising in the revolt of Zhang Shichen, a salt transport worker who in 1353 killed a corrupt tax o-BCcial, captured Gaoyou and set up a new ‘dynasty’ there, remaining a powerful factor in the competition for power until surrendering to the future Ming emperor in 1367. Elsewhere, the growth of sectarian and messianic groups could also be seen as a response to the breakdown of society and the disorder of the world. SpeciBCcally, the Red Turbans, a Buddhist sect, absorb- ing the ‘White Lotus’ sect, generated a popular messianic vision of the appearance of the Maitreya (‘future’ Buddha) to coincide with the end of Mongol rule. Their founder claimed descent from the Northern Song and after his death in 1351, Liu Futong established his heir as the ‘Prince of Radiance’ (or ‘King of Light’) and formally restored the Song dynasty. Their main success was the capture of Shangdu (Xanadu) in 1359. Another ‘legitimist’ group under Chen Youliang founded a revived ‘Han’ dynasty in the central Yangzi region, while further west, a ‘Xia’ dynasty was founded circa 1357. Whereas in Iran no-one emerged strong enough, until the rise of Temür, to defeat and absorb the di-ferent contenders for power, whether motivated by the desire to establish a dynastically legitimate regime or animated by popular messianic sentiment, in China a leader arose in the person of Zhu Yuanzhang (Yuanchang) who was able to eliminate his rivals (who were competing among themselves as much as against the Mongols), and then the Yuan regime. The support for the latter among the Mongol military elite and regional warlords was as tenuous and self-serving in China as in Iran.U? Finally (unlike in China), we should note that the end of the Ilkhanate did not mean the end of Mongol rule in Iran, merely the end of the Chinggisids. Within half a century, Mongol rule was reasserted by Temür, but it took much longer for the tensions created by the clash of Turko-Mongol and Perso-Islamic concepts of government to be somewhat reconciled.UA

Discussion

In this brief survey of the events surrounding the collapse of Ilkhanid rule in Iran, I have identiBCed a number of issues. It is understandable that desire for

WF For useful surveys of this period, see Mote, Imperial China, 517–41; and Lorge, War, Politics and Society, 98–108. WH See e.g. the BCne study by Maria Subtelny, Timurids in Transition. For the somewhat similar process taking place in China, see Dreyer, Early Ming China. *]q 6789:887 an heir and anxiety on this count was (and remains) common to monarchical regimes throughout Europe and Asia. The question here is not to deny that Abū Saʿīd’s failure to produce an heir is critical in plunging the Ilkhanate into a succession crisis that no-one was strong enough to resolve; rather it is to attempt to understand why the Mongol regime failed to handle the problem when it arose. I consider the answer to lie in the very rudimentary acculturation of the Mongols to the norms of Persian political and social life. It is true that even when there was a mature heir, there were still succession crises in the Ilkhanate, from the death of Abaqa Khan onwards, as indeed there were throughout the Mongol Empire, and nowhere more so than in the Chaghatai Khanate and the Golden Horde, where steppe traditions most naturally persisted; this is not necessarily a weakness, in that it might be assumed to produce the stron- gest candidate. Nevertheless, it betrays an attitude to power that put personal before public interest and promoted extreme factionalism, a characteristic of Mongol government that was already fostered by the multiplication and duplication of o-BCces and therefore spread outwards from the military elite into the bureaucracy.UD The factionalism in the Yuan court is a marked feature of the reign of Toghōn Temür (Shun-Ti, 1333–68), which saw a remarkable oscillation of competing ideologies over short intervals and led to the paralysis of the government.UE To a large extent, these competing ideologies, as in Iran, revolved around the extent to which Mongols adhered to the path of Sinicization opened up by Qubilai Khan (d. 1294) at more or less the same time that Ghazan Khan (d. 1304) initiated the Islamization of the Ilkhanate. One symptom of this was the question of the suppression, reintroduction and manipulation of the Confucian civil service exams that was a running issue under the later Yuan; another was the fact that the court still returned to the steppe capital of Shangdu (Xanadu, where Toghōn Temür was crowned) every summer, spending months away from the principal capital at Dadu (Beijing).UT It will be recalled that even the palace at Dadu contained extensive parks where Mongol-style tents were erected, to create a familiar environment for the court.UU

WI For an example of this in one province of the Ilkhanate, see Melville, “Anatolia under the Mongols,” esp. 98–99. WJ For a recent comprehensive account of the period, see Dardess, “Shun-Ti and the End of Yüan rule in China,” esp. 584–86. WV Ibid., 561–65. WW Rossabi, “The Reign of Khubilai Khan,” 454–57; see also Vernadsky, The Mongols and Russia, 135–36. KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S *]r

Vladimir Barthold observed that the “policy of reconciling two incompatible things—nomadic life and intellectual culture—was the weakest spot in Chingiz Khan’s system and the principal cause of its fall,”U^ and this remains an accurate assessment of the situation in the sedentary lands of China and Iran. Although Jan Rypka’s view, that Ghazan’s reforms and the conversion to Islam of the last rulers and the Mongol regiments could not help to avert the power of the dynasty “from its imminent dissolution . . . [and] . . . The Il-Khans produced not a single poet or prose writer, and their courts displayed not the slightest interest in poetry,”U_ is incorrect so far as the court of Abū Saʿīd was concerned—he was said to write poetry himself—the general tenor of his remark is echoed in the verdict of Elizabeth Endicott-West (quoted at the head of this chapter), concerning the lack of acculturation among the Mongols in China. Jean Aubin also joins the dismal refrain, stating that even though one sometimes speaks, for convenience, of “Mongol miniatures” or “Mongol architecture,” credit for the construction of prestigious buildings, protection of poets and artists, and support for an intense spiritual life should be given to those indigenous counsellors of the Ilkhans and the noyans who guided their pretensions and inspired their tastes, rather than to the Ilkhans and noyans themselves.U` Essentially, then, the Mongols remained aliens in the sedentary lands they conquered and kept themselves physically apart while moving across their territories, with most of the noyans remaining in their encampments and pastures, which they doubtless viewed as conquests to be rightfully enjoyed rather than lands to administer for the beneBCt of the state. The role of religion in this process is exaggerated by the wishful thinking of the historians. Although some beneBCts in terms of acceptability and legitimacy certainly accrued to the last three Ilkhans, this was more for the beneBCt of the sensitivities of the subject population (as was the restoration of the civil service exams in China from 1315). Among the Mongols themselves, Islamization proved divisive. This is clear from the famous reaction in the reign of Öljeitü and the return to traditional Mongol ways under Arpa Ke’ün.^a In Central Asia too, the initial conversion of Tarmashirin in 1325 or 1329 appears to have undermined his regime and been instrumental in the subsequent division of the Chaghatai khanate into western and eastern halves; in the latter, Moghulistan,

W\ Barthold, Turkestan, 461. W[ Rypka, “Poets and Prose Writers,” 555, but see the reference to Abū Saʿīd writing poetry, and Shabānkāra’ī, Majmaʿ al-ansāb, 286–87 for his skill in calligraphy and an example of his verses. WY Aubin, “Le patronage culturel,” 107. \Z Pfei-fer, “Conversion Versions,” 39; and Melville, The Fall of Amir Chupan, 61–62. **; 6789:887

Tughluq Temür enhanced his kingship by converting to Islam (c. 1360), but the process was not complete until the seventeenth century.^? In the ulus of Jochi (the Golden Horde), the conversion of Özbek Khan (r. 1313–41) seems ultimately to have had a unifying e-fect, although it did nothing to prevent the total collapse of consensual rule after the death of Birdi Beg in 1359.^A All the Mongol states, regardless of the extent of their conversion to Islam (or adherence, for example, to Tibetan Buddhism), were essentially secular in character. Peter Lorge writes that “the Yuan collapsed, not because it was Mongol ruled but because it was a Chinese dynasty.”^D This seemingly paradoxical conclusion implies that the Mongols had given up their own methods of governing, failing to maintain their military power and the adherence of the regional Mongol commanders to a common cause of conquest and dominion. In Iran, it is perhaps a moot point whether the Mongols had become a Persian dynasty—essentially, ʿUbayd-i Zākānd̄’s ironic view—or rather, had failed to become so. On the whole, I tend to the latter view: the Ilkhanate remained a Mongol regime, but lost its way and its sense of identity in the turbulence of acculturation. The bureaucrats were unable to save the dynasty in the face of the disillusion and disa-fection of the noyans. Either way, with the dismissal of the last e-fective Yuan chancellor, Toqto, in 1355, a turning point one can perhaps not too fancifully equate with the fall of Amir Chūpān in the Ilkhanate, a strong and loyal agent of central control was lost.

Bibliography

A+,ākī, Shams al-Dīn. Manāqib al-ʿārifīn. 2 vols. Edited by Tahsin Yazıcı. Ankara: Türk Tarih Kurumu Basımevi, 1959; Translated by Clément Huart, Les saints des derviches tourneurs, 2 vols. Paris: Michel Allard, 1978; Translated by John O’Kane, The Feats of the Knowers of God. Leiden: Brill, 2002. Aharī, Quṭbī Najm. Tavārīkh-i Shaikh ʿUways. Edited by I. Afshar. Tabriz: Sotudeh, 1389/2010. Aigle, Denise. “Charismes et rôle social des saints dans l’hagiographie persane médiéval (no–npo siècles).” Bulletin d’études orientales 47 (1995): 15–36.

\F Biran, “The Chaghadaids and Islam,” esp. 749–51; idem, “The Ögedeid and Chaghadaid Realms,” 58–60; and Jackson, “Chaghatayid dynasty,” 346. \H DeWeese, Islamization and Native Religion, for Özkek’s conversion; István Vásáry, “The Jochid Realm,” 78–81. \I Lorge, War, Politics and Society, 95. KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S **)

———. Le Fārs sous la domination mongole. Politique et *+scalité (-.../–-.0/ S.). Studia Iranica Cahier 31. Paris: Association pour l’avancement des études iraniennes, 2005. Album, Stephen. “Power and Legitimacy. The Coinage of Mubāriz al-Dīn Muḥammad ibn al-Muẓa-far at Yazd and Kirmān.” Le Monde iranien et l’Islam 2 (1974): 151–71. ———. “Studies in Ilkhanid History and Numismatics 0. A Late Ilkhanid Hoard (743/1342).” Studia Iranica 13:i (1984): 49–116. ———. “Studies in Ilkhanid History and Numismatics 00. A Late Ilkhanid Hoard (741/1340) as Evidence for the History of Diyar Bakr.” Studia Iranica 14:i (1985): 43–76. Alcock, Susan E. “The Afterlife of Empires.” In Empires. Perspectives from Archaeology and History, edited by Alcock et al., 369–73. Cambridge: University Press, 2001. Amitai, Reuven. “The Resolution of the Mongol-Mamluk war.” In Mongols, Turks, and Others. Eurasian Nomads and the Sedentary World, edited by R. Amitai and Michal Biran, 359–90. Leiden and Boston: Brill, 2005. Āmulī, Awliyā’-Allāh. Tārīkh-i Rūyān. Edited by M. Sotoudeh. Tehran: Intishārāt-i Bunyād-i Farhang-i Īrān, 1338/1969. Aubin, Jean. “Le patronage culturel en Iran sous les Ilkhans. Une grande famille de Yazd.” Le Monde iranien et l’Islam 3 (1975): 107–18. ———. “Le quriltai de Sultân-Maydân (1336).” Journal Asiatque 279/i–ii (1991): 175–97. ———. Émirs mongols et vizirs persans dans les remous de l’acculturation. Studia Iranica Cahier 15. Paris: Association pour l’avancement des études iraniennes, 1995. Banākatī, Tārīkh. Edited by J. Shi‘ar. Tehran: Intishārāt-i Anjuman-i Ā sār-i Millī, 1348/1969. Barthold, W. Turkestan down to the Mongol Invasion. 4th edition, Cambridge: E. J. W. Gibb Memorial Trust, 1977. Biran, Michal. “The Chaghadaids and Islam: the Conversion of Tarmashirin Khan (1331–34).” Journal of the American Oriental Society 122/4 (2002): 742–52. ———. “The Mongols in Central Asia from Chinggis Khan’s Invasion to the Rise of Temür: the Ögedeid and Chaghadaid Realms.” In The Cambridge History of Inner Asia. The Chinggisid Age, edited by Nicolo di Cosmo, Allen J. Frank and Peter B. Golden, 46–66. Cambridge: University Press, 2009. Bronson, Bennet. “The Role of Barbarians in the Fall of States.” In The Collapse of Ancient States and Civilizations, edited by N. Yo-fee and G. L. Cowgill, 196–218. Tucson: University of Arizona Press, 1988. Brook, Timothy. The Troubled Empire. China in the Yuan and Ming Dynasties. Cambridge, f$ & London: Harvard University Press, 2010. Chan, Hok-lam. Legitimation in Imperial China. Discussions under the Jurchen-Chin Dynasty (1115–1234). Seattle and London: University of Washington Press, 1984. Dardess, John. “Shun-Ti and the End of Yüan Rule in China.” In The Cambridge History of China, volume 0., Alien Regimes and Border States, edited by H. Franke and Denis Twitchett, 561–86. Cambridge: Cambridge University Press, 1994. **] 6789:887

De Nicola, Bruno. “Unveiling the Khātūns: Some Aspects of the Role of Women in the Mongol Empire.” PhD diss., University of Cambridge, 2011. DeWeese, Devin. Islamization and Native Religion in the Golden Horde. Baba Tükles and Conversion to Islam in Historical and Epic Tradition. Pennsylvania: Penn State University Press, 1994. Dreyer, Edward L. Early Ming China. A Political History 1355–1435. Stanford: Stanford University Press, 1982. Eisenstadt, S. N., ed. The Decline of Empires. Engelwood Cli-fs, /|: Prentice Hall, 1967. Endicott-West, E. “Aspects of Khitan Liao and Mongolian Yüan Imperial Rule: a Comparative Perspective.” In Rulers from the Steppe. State Formation on the Eurasian Periphery, edited by Gary Seaman and Daniel Marks, 199–222. Los Angeles: Ethnographics Press, 1991. Fragner, Bert. “Ilkhanid Rule and its Contribution to Iranian Political Culture.” In Beyond the Legacy of Genghis Khan, edited by L. Komaro-f, 68–80. Leiden: Brill, 2006. ḤāBCẓ-i Abrū. Ẕail-i Jāmiʿ al-tawārīkh. British Library, ms. Or. 2885. Haneda, Masashi. “The Pastoral City and the Mausoleum City: Nomadic Rule and City Construction in the Eastern Islamic World.” In Islamic Urbanism in Human History. Political Power and Social Networks, edited by Sato Tsugitaka, 142–70. London and New York: Keegan Paul, 1997. Holt, P. M. The Age of the Crusades. The Near East from the Eleventh Century to 1517. London and New York: Longman, 1986. Ibn Bazzāz. Ṣafwat al-ṣafā. Edited by Ghulamriza Tabataba’i Majd. Tabriz: privately published by the editor, 1373/1994. Irwin, Robert. The Middle East in the Middle Ages. The Early Mamluk Sultanate 1250– 1382. London and Sydney: Croom Helm, 1986. Jackson, Peter. “The Dissolution of the Mongol Empire.” Central Asiatic Journal 22/iii–iv (1978): 186–244. ———. “Abu Sa‘id.” Encyclop2dia Iranica 0 (1985): 374–77. ———. “Chaghatayid dynasty.” Encyclop2dia Iranica p/4 (1991): 343–47. ———. “Muẓa-farids.” In Encyclopaedia of Islam, Second Edition p00: 820–22. ———. and C. Melville. “Gīāṯ al-Dīn.” Encyclop2dia Iranica n:6: 598–99. Kennedy, Hugh. “The Decline and Fall of the First Muslim Empire.” Der Islam 81/i (2004): 3–30. Komarao-f, Linda, ed. Beyond the Legacy of Genghis Khan. Leiden and Boston: Brill, 2006. ——— and Stefano Carboni, eds. The Legacy of Genghis Khan. Courtly Art and Culture in Western Asia, 1256–1353. New York and London: Yale University Press, 2002. Kwanten, Luc. Imperial Nomads. A History of Central Asia, 500–1500. Leicester: University Press, 1979. Le Go-f, Jacques. Héros du Moyen Âge, le Saint et le Roi. Paris: Gallimard, 2004. KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S ***

Limbert, John. “Inju dynasty.” Encyclop2dia Iranica n000:2 (2004): 143–47. Lingwood, Chad G. Politics, Poetry, and Su*+sm in Medieval Iran. New Perspectives on Jāmī’s Salāmān va Absāl. Leiden: Brill, 2013. Lorge, Peter. War, Politics and Society in Early Modern China, 900–1795. London and New York: Routledge, 2005. Mahendrarajah, Shivan. “The Sarbadars of Sabzavar: Re-examining their ‘Shi‘a’ Roots and Alleged Goal to ‘Destroy Khurasanian Sunnism’.” Journal of Shi‘a Islamic Studies 5/iv (2012): 379–402. Manz, Beatrice Forbes. “Mongol History Rewritten and Relived.” In Denise Aigle, ed. “Figures mythiques des mondes musulmans” (Special issue), Revue des mondes musulmans et de la Méditerrannée 89–90 (2000): 129–49. ———. “Family and Ruler in Timurid Historiography.” In Studies on Central Asian History in Honor of Yuri Bregel, edited by Devin DeWeese, 57–78. Bloomington: Research Institute for Inner Asian Studies, 2001. Marek, J. and H. Knizkova.The Jenghiz Khan Miniatures from the Court of Akbar the Great. London: Spring Books, 1963. Matthee, Rudi. Persia in Crisis. Safavid Decline and the Fall of Isfahan. London: I. B. Tauris, 2012. Melville, Charles. “The Chinese Uighur Animal Calendar in Persian Historiography of the Mongol Period.” Iran 32 (1994): 83–98. ———. “Abū Sa‘īd and the Revolt of the Amirs in 1319.” In L’Iran face à la domination mongole, edited by D. Aigle, 89–120. Tehran and Paris: 0~(0, 1997. ———. The fall of Amir Chupan and the Decline of the Ilkhanate, 1327–1337: a Decade of Discord in Mongol Iran. Bloomington: Research Institute for Inner Asian Studies, 1999. ———. “Gāzān-nāmā.” Encyclop2dia Iranica n:iv (2000): 383. ———. “History and Myth: the Persianisation of Ghazan Khan.” In Irano-Turkic Cultural Contacts in the 11th–17th Centuries, edited by Éva M. Jeremiás, 133–60. Piliscsaba: Avicenna Institute of Middle Eastern Studies, [2002] 2003. ———. “The keshig in Iran: the Survival of the Royal Mongol Household.” In Beyond the Legacy of Genghis Khan, edited by L. Komaro-f, 135–64. Leiden: Brill, 2006. ———. “Between Firdausī and Rashīd al-Dīn: Persian Verse Chronicles of the Mongol Period.” Studia Islamica 104/105 (2007): 45–65. ———. “Between Tabriz and Herat: Persian Historical Writing in the 15th Century.” In Iran und iranisch geprägte Kulturen. Studien zum 65. Geburtstag von Bert G. Fragner, edited by Markus Ritter, Ralph Kauz and Birgit Ho-fmann, 28–38. Wiesbaden: Dr. Ludwig Reichert Verlag, 2008. ———. “Anatolia under the Mongols.” In The Cambridge History of Turkey, volume 1, Byzantium to Turkey, 1071–1453, edited by Kate Fleet, 51–101. Cambridge: Cambridge University Press, 2009. **h 6789:887

———. “The Mongol and Timurid Periods, 1250–1500.” In Persian Historiography, A History of Persian literature, vol. n, edited by C. Melville, 155–208. London: I. B. Tauris, 2012. ———. “The Royal Image in Mongol Iran.” In Every Inch a king. Comparative Studies on Kings and Kingship in the Ancient and Medieval Worlds, edited by Lynette Mitchell and Charles Melville, 343–69. Leiden: Brill, 2012. ———, and A. Zaryab. “Chobanids.” Encyclop2dia Iranica p:5 (1991): 496–502. Minorsky, V. “The Aq-Qoyunlu and Land Reforms.” 34564 17 (1955): 449–62. Morgan, D. O. Medieval Persia 1040–1797. London: Longman, 1988. ———. “Mongol or Persian: the Government of Īlkhānid Iran.” Harvard Middle Eastern and Islamic Review 3/i–ii (1996): 62–76. ———. “Review of Charles Melville, The Fall of Amir Chupan.” 34564 63 (2000): 430–32. ———. “The Mongols in Iran: a Reappraisal.” Iran 42 (2004): 131–36. ———. The Mongols. 2nd edition, Oxford: Blackwell, 2007. ———. “The Decline and Fall of the Mongol Empire.” 7864 ser. 3, 19/4 (2009): 1–11. Mote, F. W. Imperial China 900–1800. Cambridge, f$ and London: Harvard University Press, 1999. Mustawfī Qazwīnī, Ḥamd Allāh. Nuzhat al-qulūb. Edited by G. Le Strange. London: Luzac, 1915; Translated by G. Le Strange. London: Luzac, 1919. Pfei-fer, Judith. “Conversion Versions: Sultan Öljeytü’s Conversion to Shi‘ism (709/1309) in Muslim Narrative Sources.” Journal of the Mongolian Society 22 (1999): 35–67. Roemer, H. R. “The Jalayirids, Muza-farids and Sarbadārs.” In Cambridge History of Iran, vol. 6. The Timurid and Safavid Periods, edited by P. Jackson and L. Lockhart, 1–41. Cambridge: Cambridge University Press, 1986. Rossabi, Morris. “The Reign of Khubilai khan.” In The Cambridge History of China, volume 0., Alien Regimes and Border States, edited by H. Franke and Denis Twitchett, 414–89. Cambridge: Cambridge University Press, 1994. Roux, Jean-Paul. Histoire de l’empire mongol. Paris: Fayard, 1993. Rypka, Jan. “Poets and Prose Writers of the Late Saljuq and Mongol Periods.” In The Cambridge History of Iran, vol. 5. The Saljuq and Mongol Periods, edited by J. A. Boyle, 550–625. Cambridge: University Press, 1968. Shabānkāra’ī, Muḥammad b. ʿAlī. Majmaʿ al-ansāb, edited by Md̄r Hāshim Muḥaddis.̱ Tehran: Amīr Kabīr, 1363/1984. Smith, J. M. Jr. The History of the Sarbadār Dynasty 1336–1381 69 and its Sources. The Hague and Paris: Mouton, 1970. ———. “Dietary Decadence and Dynastic Decline in the Mongol Empire.” Journal of Asian History 34/1 (2000): 35–52. Spuler, Bertold. The Mongol Period. History of the Muslim World. Introduction by Arthur N. Waldron. Princeton: Markus Wiener, 1994. KL7 7MN OP KL7 :8QLRMRK7 RMN RPK7S **i

Subtelny, Maria E. Timurids in Transition. Turko-Persian Politics and Acculturation in Medieval Iran. Leiden: Brill, 2007. ʿUbayd-i Zākānī. Akhlāq al-ashrāf. In Kulliyāt-i ʿ Ubayd-i Zākānī, edited by M. Ja‘far Mahjub, 231–55. New York: Bibliotheca Persica, 1999; English Translation by Hasan Javadi. The Ethics of the Aristocrats and other Satirical Works. Piedmont, "$: Jahan Books Co., 1985. Vásáry, István. “The Jochid Realm: the Western Steppe and Eastern Europe.” In The Cambridge History of Inner Asia. The Chinggisid Age, edited by Nicolo di Cosmo, Allen J. Frank and Peter B. Golden, 67–85. Cambridge: Cambridge University Press, 2009. Vernadsky, George. The Mongols and Russia. New Haven: Yale University Press, 1953. Waldron, A. N. Introduction to Bertold Spuler, The Mongol Period. History of the Muslim World, vii–xxxiii. Princeton: Marcus Weiner, 1994. Wilber, D. N. The Architecture of Islamic Iran. The Il Khānid Period. New York: Greenwood Press, 1955. Woods, John E. The Aqquyunlu Clan. Confederation, Empire. Revised ed. Salt Lake City: University of Utah Press, 1999. Yates, Robin D. S. “Cosmos and Central Authority in the Early Chinese Empire.” In Empires. Perspectives from Archaeology and History, edited by Susan E. Alcock et al., 351–68. Cambridge: Cambridge University Press, 2001.

Index

Abaqa, Ilkhan 43, 57 n. 15, 62, 66, 67, 70, 73, Akhlat 201, 311 91–94, 120, 144, 145, 145 n. 46, 208, 230, Aksaray 28–29 311, 318, 328 Akşehir 28 ‘Abbasids 320 ʿAlāma Ḥillī 69 Abbasid Caliphate 1, 32 ʿAlī Beg, Turkmen lord 30 Abbasid culture 133 Aljaktu 29 see also al-Mustaʿṣim bi-Allāh 134 Allsen, Thomas T. 19, 23, 79, 80, 84, 87 ‘Abd al-Ḥayy, painter 172 Alp Kutlu Tuna Bilge Masʿūd b. Ḥasan 210 ʿAbd al-Raḥmān (Fiscal administrator of the ʿAmid 24 Mongols) 87 Amir Abū Bakr 262, 265, 289 Abū al-Faḍl Bayhaqī 168, 199 Amir Bakhtiyār 247, 259, 262, 265, 270 Abū Bakr ibn Saʿd, Salghurid Ruler 205 Amitai, Reuven 4, 267, 293, 301 Abū Bakr Muḥammad ibn Zakariyyā al-Rāzi Āmulī, Awliyā’-Allāh 312 197 Anatolia 1, 4–7, 26–31, 56, 59–62, 66, 75, 88, Abū Ḥanīfa Dīnawarī 197 145, 195, 200–204, 208–211, 280, 282, 288, Abū Isḥāq, Injü’id ruler 145, 325, 326 289, 311, 315, 317 Abū Saʿīd, Ilkhan 39, 94, 98, 118, 119, 147, 148, Anbarji 318 172, 245, 287, 289, 310, 311, 312, 313, Andarznāma 199 316–326, 328–329 Angara River 88 administration 27, 33, 55–59, 66, 75, 89, Ankara 25 93–94, 98, 120, 144, 166, 200, 202, 206, Antioch 29, 31, 32 210, 211, 220, 222, 225, 230–231, 238, 241, Anūshīrvān ibn Khālid 200 250, 255, 261, 266, 273, 298, 320, 321, 326 Aor Khan 29 Afghanistan 16 n. 8, 22, 91 Aq-Qoyunlu 45 A*+ākī, Aḥmad 315, 317 ʿAqr (Akre) 289, 294, 296 n. 59 Aghbagha 25 Āqsarāʼī, Maḥmūd b. Muḥammad al-Karīm Aḥmad Mūsā, painter 172 208 Aḥmad of Khojend 85 Arab conquest 2, 195, 196 n. 4 Aḥmad Tegüder 92, 110, 112, 113, 118, 287, 293, Arabian Peninsula 96 294, 318 Arabic script 197–199 Aigle, Denise 56, 58 Arabs 27, 148 n. 67, 196, 197 Airarat 218 Ardabil 70, 316 ‘Ajā’ib al-Hind 177 archery 46 Akbar, Mughal emperor 177, 182 Arevelts‘i, Vardan 226 ʿAlāʾ al-Dīn Kayqubād -, Sultan of Rūm 24, Armenia 6, 7, 22, 24–26, 29–30, 40, 216–232, 209 280 Alamūt 95, 206, 286 Armenia, Greater 216, 218, 219, 229, 231 Alcock, Susan 323 Armenia, Zak‘arid 224, 229 alcohol 319 Armini Khatun 74 ‘Alā’ al-Dawla Simnānī 263, 268 Arghun Aqa 113, 223–225, 230 ‘ālim 272 Arghun, Ilkhan 63, 70, 74, 112, 117, 120, ʿAlī-Pādshāh 323, 324 122–124, 166, 221, 268, 311, 318 Aleppo 25, 31–33, 137, 247, 287 Arighan 91 alginchi (pl. alginchin) 17, 18 Ariq Böke 310, 319, 323 Alp Arslan, Sultan 201 Aristotle 169 112 -3456

Arpa Ke’ün 311, 320, 323, 329 Bayhaqī, Abū al-Faẓl 168, 199 Arran 22, 217, 218, 226, 311, 316 Bayn al-Jabalayn 290 Arsanjānī, Niẓām al-Dīn Aḥmad 203 Beijing 328 Arslan Khan 17 Beirut 34 Assassins 32 Berke Khan 84 Atrpatakan, see Azerbaijan Besütei 28 Aubin, J. 64, 319, 329 Bethlehem 30 Austria 23 Bībī Maryam 96 Āveh 60 Bichikchi 21 al-Awāmir al-ʻalāʼīya fī al-umūr al-ʻalāʼīya 208 Bitikchi 57, 225 Awṣāf al-ashrāf 207 Biqāʿ Valley 33–34 ‘Ayn Jalut 15, 19 Bīrūnī 200 Ayyubids 69, 282, 285 n. 28, 293 n. 51, 300, 319 Bohemund 29–30 Azerbaijan 1, 20, 22, 32, 41, 45, 94, 134, 144, Book of Changes 121 201, 218, 226, 280, 288 Borderlands 34, 300 Azhdari, Nūr al-Dīn 325 Börte 14 n. 20, 81, 84 n. 28 Bosworth, C. E. 38 Badr al-Dīn Lu’lu’ 142 n. 31, 146, 286 Boyle J. A. 26 Badakhshan, southern Tajikistan 91 Browne, Edward G. 175 Baghdad 6, 22, 23, 31, 32, 34, 90, 91, 95, 98, Buddhism 106–107, 119–121, 241, 330 110–111, 115 n. 61, 116–118, 133–136, Buell, Paul 15–16, 20–21 138–144, 146–148, 150, 161, 172, 204, 227, Bugha (Ghara Bugha) 221 246, 261, 286, 287, 288, 314, 317, 320, 325 Bukhara 140, 183, 205, 247 Baghdad Khatun 98, 112, 320 al-Bukhārī, Muḥammad 197 bahādur (pl. bahāduriyya) 42–43 Būlūghān Khatun Buzurg 97 baghatur/ba’atur (pl. baghatut/ba’atut) 42 Bureaucracy 55, 258, 328 Bahman-nāma 175, 176 Bureaucrats 120, 202, 205, 209, 321, 330 Bahrām Gūr 209 Buzurg ibn Shahriyār 177 Bahrām Mīrzā, Safavid prince 172 Byzantines 27, 30 Bahrāmshāh 199 Byzantium 61, 148 n. 67 Baidar 22, 33–34 Baidu, Ilkhan 70, 97, 113, 311, 319, 323, Cahen, Claude 47, 200 Baiju 26 n. 8, 20, 23–24, 27–29, 31–32, 34, Cairo 146, 278, 293, 298, 326 136–137, 225 Cappadocia 29 Bakhshi 120, 121, 123 Carpathian Mountains 23 al-Balādhurī 280 Carpini, John of Plano 42 Balkh 209 Caucasus 60, 216, 222, 225, 231 Banākatī 244 Chächäyigän 82 Bānū Noyan, Baiju 136, 138–138 Chaghatai Khan 328 Bar Hebraeus 29, 114 Chaghatai Khanate 92, 328–329 Baraq 43, 230 Chamchamal 288 Barcelona 3 Charandāb 73 Barthold, Vladimir 329 Chelebi, ‘Ārif 315, 316, 326 Basqaq 21, 216, 220–222, 226, 231 Chen Youliang 327 Batu 20, 21, 226 Ch’i-ch’ing Hsiao 18, 309 Bayan Noyan 237, 241, 261 Chïn-Temür 222–223 Baybars al-Manṣūrī 46, 293 n. 51, 294, 298 China 1, 2, 7, 15, 18, 19, 20, 23, 42, 64, 79, 83, Bayezid -, Ottoman Emperor 210 86, 87, 89, 94 n. 88, 120, 123, 125, 149, 156, -3456 118

216, 222, 228, 232, 237, 244, 247, 252, Doquz Khatun 29 259, 261, 264, 266, 269, 313, 314, 322, 323, Dongping 18 324, 326, 327, 329 Dūst Muḥammad, calligrapher 172 Chinggis Khan 1, 13, 16, 17, 18 n. 18, 22, 39, 80, Durand-Guédy, David 3, 47 81, 82, 85, 86, 89, 99, 109, 114, 116, 140, 159, 160, 161, 164, 206, 216, 217, 314, 319 East Asia 39, 107, 237, 243 Chormaqan 20, 22, 23, 24, 218, 220, 222, 225 economy 5, 28, 44, 57, 223, 228, 280–283, 290 Christianity/Christians 24, 30, 31, 106, 122, Women’s economic activity 79–80, 84, 123, 141, 142, 160, 207, 211, 238 85, 88–89, 93–94, 98 Chūpān, Amir 114, 310, 312, 319–321. 324, 330 Edgü-Temür 223 Chupanids 320, 324–326 Egypt, see also Mamluk Sultanate 7, 33, 34, Central Asia 7, 69, 79, 85, 88, 112, 120, 196, 38, 40, 44, 47, 148, 149, 277, 282, 286, 199, 200, 206–207, 209–210, 217, 222, 288, 295, 297–298, 300 266, 287, 313–314, 329 Eisenstadt, S. N. 309, 315 Cilicia 30–32, 287 Elchi 225, 228 civil government 15, 21, 22 Eljigedei 32 commercial agents, see merchants Endicott-West, E. 322, 329 Concubines 91, 92, 94, 239, 259, 318 Episkopos, Step‘annos 226, 230 Confucianism 107, 119–120 Erzincan 25 Constantinople 27, 30 Erzurum 24–25, 220, 229 continuity 3–7, 38, 45, 48, 55, 58, 75, 90, 310 Ettinghausen, Richard 157 conversion 64, 70, 97, 204, 207, 321, 329, 331 Euthymius 29–30 Crusades 20 Eurasia 23, 84, 89, 90, 99, 149 Europe 19, 23, 26, 83, 88, 99, 149, 244, 316, 328 Dadu, see Beijing execution 106–116, 118–120, 125 al-Dhahabī, Muḥammad ibn Aḥmad 26 Dahuading mosque, see Songjiang mosque factionalism 328 248, 255 al-Fārābī, Abū Naṣr 161 Daiyuan 18 Fars 96, 145, 201, 205, 278, 280, 289, 325 Dalaman (Tālamānī) 30 Faryābī, Ẓahīr al-Dīn 203 Dali 23, 265 fertility 319 Damascus 34, 135, 287 Firdawsi, poet 318, 322 Da-ming 18 Fīrūzshāh-nāma 184 Dārābnāma 177 Fortress, fortresses 19–20, 24, 32–33, 139, 184, Darughachi 6, 21, 22, 25, 31, 32, 216–217, 245, 286, 292 220–222, 224, 230–232, 261 Franks 25 Dasht-i Qipchaq 313 Dāstān-i Qirān-i Ḥabashī 183 Gag 25, 219 Daylamī, Mu‘īn al-Dīn Sulaymān 208 Galatia 29 Diatessaron 207 Gandjeï, Tourkhan 203 death penalty 5, 106, 125 Gandzakets‘i, Kirakos 220, 230 Delhi Sultanate 32, 33, 149 Gaoyou 327 Denizli 28, 30 Gardīzī 199 Diez, Freiherr von 174 Garshāsp-nāma 176 Dimashq Khwāja 320 Gaza 33 Diplomacy 31, 205, 210 Georgia 22, 24, 200, 217–219, 223–227, dīwān-i aʿlā 55 230–232, 278, 324 dīwān-i ishrāf 57 Ghazālī 200 Diyarbakır 27, 60 Ghaznavids 168, 198–201, 211 1:; -3456

Geikhatu, Ilkhan 92, 94, 113, 123–124, 318 Horses 25, 28, 39, 43, 45–47, 59, 91, 98, 223, Ghazan Khan 5, 64, 83, 90, 92, 94–97, 100, 256, 291 111, 119, 167, 286, 313, 325–326, 328 Hotan 116 Ghāzān-nāma 325 ḥudūd 116 Ghazna 200 Hui hui 237, 239, 257–258 al-Ghaznawī, Abū al-Ḥasan Hujwirī 200–201 Hülegü 1, 4, 23, 28–30, 32–34, 39, 42–44, 46, ghilmān 47 55, 64, 70, 80, 90–92, 95, 111–114, 124, Ghiyāth al-Dīn Muḥammad, vizier 315–316 135–144, 226–230, 241, 252, 286–288, Ghiyāth al-Dīn Kaykhusraw -, Sultan of Rūm 310–312, 317–319 203 Humaydiyya 293 Ghiyāth al-Dīn Kaykhusraw --, Sultan of Rūm Hungary 21, 23 24, 26, 203 Ghur 32–33 Ibaqa Beki 82 Ghurids 201 Ibn ʿAbd al-Ẓāhir 297, 298 Ghuzz (Oğuz) 284 Ibn al-ʿAmīd 286 gifts 92, 95, 97, 99, 137–138, 160 Ibn al-Athīr 280 Glkhavor 221 Ibn al-Munshī 288, 292 Golden Horde 92, 95, 166, 218, 328, 330 Ibn Baṭṭūṭa 47, 84, 147, 241, 252 Greeks 25–26, 29–30 Ibn Bazzāz 113 Grigor of Akner, also Grigor Aknerts‘i (Grigor Ibn Bībī 202–203, 208–209 of Akanc‘) 24, 227 Ibn Faḍlallāh al-‘Umarī 43–44, 133, 135, 281, Grousset, René 158 283–284, 290–291, 293, 297–299 Güchülüg 16–17 Ibn al-Fuwaṭī 116, 118, 146 Gulistān 205 Ibn al-Muqa@faʿ 196, 199 Gushāyish va Rahāyish 199 Ibn Ḥawqal 280 Güyük Khan 112, 224 Ibn Jubayr 282 Ibn Khaldūn 284, 300 Ḥā=>ẓ, poet 325 Ibn Sīnā 200 Haithon 26 ibn ʻUmar al-Rādūyānī, Muḥammad 199 Hakkārī mountains 297 ibn Yaghān Ṭughril Bak 247, 264 Ḥalabī, Muḥammad 264, 268 identity 17, 39, 109, 160, 262, 277, 301, 322, Hamadan 201, 287–289, 296, 311 330 Hamayon, Roberte 108 Ilgin 28 Ḥamza-nāma 182, 184 Ikeres 16 Hanaway, William L. 180 ‘Illī, Ẓahīr al-Dīn 203 Hangzhou 7, 237–239, 241, 244–245, Ilkhanate 1, 4–6, 38–45, 48, 55, 59, 64, 70, 90, 249–250, 252–253, 255, 257–262, 266, 92–100, 106–107, 110, 112–117, 119–121, 125, 271–273 135, 145, 149–150, 211, 218, 220, 225, 231, Ḥasan Jūrī, Shaykh 325 248, 252, 309–315, 317–318, 321, 323–324, Ḥasan-i Buzurg, Shaykh 324–325 326–328, 330 Ḥasan-i Kuchik, Shaykh 324 Ilkhanid Iran 80, 90, 99, 146, 312 Hasanwayhids 296 Ilyās Beg 30 Heidemann, Stefan 287 ‘Imād al-Dīn Yaḥyā b. Aḥmad Kāshī 145 Herat 43, 63, 66, 113–114, 173 ʿImadiyya (Amedi) 289 Ḥilla 141 Imām Mawlānā Burhān al-Dīn 264, 268 Hindūshāh, Muḥammad Qāsim 206 Imāmī of Herat 63 Ḥiṣn Kayfā (Hasankeyf) 286, 288, 292 India 6, 22, 32, 83, 96, 120, 123, 201, 205–206, Hö’elün 81–82 209, 211 Homs 34, 45 Inner Asia 42–43, 84 -3456 1:C

Innocent -D, Pope 30 Jarquchi 21 Injü 94, 114, 145, 179–181, 184, 219, 325–326 Jebe 22 Inju’ids 325 Jin dynasty 42, 237 Ipşiroğlu, Mazhar S. 174–175, 178 Jinan 18 iqṭā‘ (pl. iqṭā‘āt) 30, 41, 44, 48, 292 Jöchi-Qasar 319 Iran 1–7, 15, 22, 38–39, 41–45, 48, 55, 58–59, John Vatatzes ---, Emperor of Nicaea 26–27 79–80, 82–83, 88, 90–100, 134, 146, 149, Judaism 106 156, 160–161, 167–169, 181, 183, 195–198, Jujing 241, 244, 247–248, 252–254, 256–263, 200–201, 204–208, 210–211, 216, 222, 265, 269 226–227, 230, 241, 244–247, 252, 259, Jūlmarkiyya 284, 293–294 273, 277, 280, 282, 286–288, 310, 312–314, Jumghur 94 316, 321, 324–325, 327–330 Jurchen, see Jin dynasty Ancient Iran 196, 247 Jurjān 68, 228 Irān zamīn 321 Jüshkab 94 Iranian plateau 202, 206 Juwaynī, ʿAlāʾ al-Dīn ʿAṭā Malik 5–6, 26, Iraq 6, 27, 32, 94, 117–118, 144, 160, 223, 247, 85–86, 116–117, 144–147, 155–169, 263–264, 267–269, 287 206–207, 209, 225, 228 Irbil 135, 139, 144, 286, 288–291, 293–294 Juvaynī, Bahāʾ al-Dīn 61, 144 Irtysh River 17, 217 Juwaynī, Hārūn Sharaf al-Dīn 61, 72, 144, 208 Isfahan 118, 207–208, 247, 263–265, 269–270 Juwaynī, Shams al-Dīn 5, 55–75, 115, 122–123, al-Iṣfahānī 265, 269–270 139, 144–147, 208 Iskandar Sultan, Timurid 170 Jūzjānī, Minhāj Sirāj 114, 201 Islam 67, 70–71, 86, 97, 106–107, 115, 195, 197, 204, 207, 246, 249, 263, 318, 321, 326, Kalīla wa Dimna 6, 173, 175, 185–186, 188, 199, 329–330 209 Islamization 7, 40, 45, 48, 125, 328–329 Karin, see Erzurum Isma’ilis 23, 32, 34, 67, 69, 229 Kart dynasty 113 Istanbul 61, 173–174, 210 Kashan 203 Italians 83 Kashf al-maḥjūb 201 Iwannis ʿIzz al-Dīn Muḥammad b. Muẓa@far Kashmir 32, 120–121 207 Kayseri 25, 219 ʿIzz al-Dīn Kaykāvūs -, Sultan of Rūm Khitai 86, 160, 252 202–203 Kereit/Kereits 80, 82 ʿIzz al-Dīn Kaykāvūs --, Sultan of Rūm keshig 42–43, 265 27–30, 209 Ket-Buqa 23, 33–34 ʿIzz al-Irbilī, al-Ḥasan b. Aḥmad b. Ẓafar 135 Khānaqāh 60, 65, 65, 67–71, 72 ʿIzz al-Dīn Qilij Arslān b. Masʿūd 200, Khānbāliqī, Khwāja ibn Arslān 247, 266, 271 202–203 Khūnās 30 Khatuns 79, 82, 84–85, 87–88, 90, 92–95, Jackson, Peter 156–157, 312 98–99, 100, 148 Jamāl al-Dīn Abū Isḥaq Īnjū 145, 326 Khurasan 6, 22, 66, 79, 141, 146, 158, 162, 198, al-Jamāl al-Mashriqī 146, 148 200–201, 203, 206, 209, 211, 216, 223, Jāmiʿ al-tawārīkh 44, 59, 82, 97, 181, 207 225–226, 247, 270, 310–311, 321, 326 Javānmardī 187 al-Khurāsānī 270 al-Jazīra 145 Khwāfī, Majd al-Dīn Muḥammad 321 Jalāl al-Dīn Khwarazmshah 22 Khwāja Aḥrār 69 Jalayir 16 Khwāja Jalāl 248, 271 Jalayirids 145, 172–174, 176, 179, 188, 230, Khwājū-yi Kirmānī 172–173 324–325 Khwāndamīr 265 1:F -3456

Khwarazm 22, 261 Majd al-Dīn Isḥāq 203 Khwarazmshah 85, 167, 265 al-Malik al-Ashraf Mūsā 34 King David D-- Ulu 223 al-Malik al-Nāṣir b. Qalāwūn 286, 288 Kingdom of Jerusalem 33–34 al-Malik al-Ṣāliḥ Ayyūb 292, 329 kingship 139, 187, 316, 330 Malik Shāh, Seljuq Sultan 199, 314 Kingshü 94 Malik Shāh, Quṭb al-Dīn 202 Kitāb al-adwār (The book of cycles) 134, al-Mamlaka al-Akrādiyya 297, 299 144–145 Mamluk Sultanate 31, 33–34, 38, 40, 147 Kitāb al-Kifāyah fī al-ṭibb 200 Mamluks 4, 7, 38–40, 44–48, 319–320, Kitāb-i Samak-i ʿAyyār 179–181, 184 322 Kitbūghā 286 Manchuria 42 Köketei 20 Mangqut 16 Kolbas, Judith 288 Manuel Comnenus -, Byzantine emperor 26 Konya 25, 27, 29, 203, 209, 315–316 Manzikert 47, 201 Korea 23, 231 al-Maqrīzī, Aḥmad ibn ʿAlī 288 Körgüz 118–119, 223 Mar Sargis 237 Koryŏ dynasty 32 Maragha 287–288, 311 Köse Dagh 25, 27, 31 Marj al-Ṣu@far 46 Kubadabad 209 Marco Polo 95, 243–244 Kucha 19, 246 Mardin 146, 288 Küchlüg 116–117, 160 Martin, H. Desmond 19, 38 Kurdish tribes 7, 278, 281, 283–286, 288, Martinez, Arsenio 38, 41, 58 290–291, 293–294, 300 Marv 162–163 Kurdistan 283, 289, 296, 311 Masālik al-abṣār fī mamālik al-amṣār 50, Kurds 27, 228, 277–278, 280–286, 288, 134, 281 290–301 Mashhad 68 Kutayla 146 Masʿūd, Sultan of Ghazna 199 Matthee, R. 313, 320 Labnasagut 123 Mawarannahr 22 Lagzistān 60 Mawṣil, see Mosul Lahore 33, 201 May, Timothy 7, 38 Lambton, A. K. S. 3, 47, 55–58, 63, 74, 79 Mayyafariqin 94, 288 Lane, George 7, 55–56, 61 Mazandaran 22, 223, 311 Latin States 33 Māzanjāniyya 293–294 legitimacy 1, 7, 64, 70, 71, 143, 313, 319–320, Mediterranean 2, 44, 61, 287 326, 329 Mehmed Beg al-Ūji 30 Levant 2, 43, 96 Melitene (Malatya) 29 light cavalry 39, 41, 43, 45–46 Melville, Charles 7, 42, 48, 60, 70, 155, 202, lingua franca 196 288 Lippard, Bruce G. 26 merchants 14, 83–90, 95–97, 99, 141, 142, 185, Liu Futong 327 247, 249, 257, 259, 262–264, 267–270, Loyalty 57, 60–62, 65–66, 71, 74–75, 178, 256, 273, 282, 287, 291 321, 322, 326 Mesopotamia 204, 222, 278, 280, 282, 287, Lorge, P. 330 324 Mīānīj 61 madrasa 60, 68, 134 Michael Palaeologus, Byzantine Emperor 29 Mahagela, Lamaist Buddhist God of War 241 Middle East 2–7, 13, 19, 28, 32, 38–39, 41, 47, Maḥmūd-Shāh 325 91, 100, 155–156, 216, 218, 277, 291, 296, Maḥmūd Yalavach 87 300, 333 -3456 1:1

Middle Persian 199 Ögetei 13, 16, 30, 86–87, 122, 158, 160, Ming dynasty 238, 250, 256 164–166, 218, 220, 222–225, 227, 310 Minqad 15, 17 Ögeteids 89 Minqan 15 Oghul Ghaimish 88, 310 Miskawayh 200 Oirats 324 Möngke, Great Khan 23, 29, 32–34, 64, Oljäi Khatun 94 89, 90, 91, 95, 97, 99, 100, 165–166, Öljei 70, 74, 311 225–226, 310 Öljeitü, Ilkhan 147, 287–289, 310–311, 329 Mongolia 16, 17, 20, 80, 84, 86, 91, 99–100, Onggirat 16 159, 164, 223 Orbelian, Step‘annos 221 Morgan, David O. 38, 41, 47, 155, 312–313 ordo, pl. ordos 81, 82, 84, 88, 91–95, Morton, Alexander 269–271 97–100 Mosul 60, 146, 286–288, 296 ortaqs, see merchants Mottahedeh, Roy 64 Ostikan 221, 230 Mubāriz al-Dīn, Muza@farid 325–326, 331 Ostrowski, Donald 17 Mughan 28, 34, 126, 218, 316 Otchigin 16 Muʿīn al-Dīn Parvāna 66, 202, 208 Özbek Khan 330 Muqali 16, 20 Musāmarat al-akhbār wa musāyarat Pahlavi script 197 al-akhyār 208 paiza 223, 225 Mustanṣiriyya College 134–135 Palestine 23, 91 Mustawfī, Ḥamd Allāh 60, 69, 114, 207–208, Paphlagonia 28 244, 250, 282, 289, 296, 322 Peacock, Andrew 47 mushrif-i mamālik 57 Persia, see Iran al-Mustaʻṣim, Caliph 111, 134, 142, 145 Persian Gulf 96 al-Mutawakkil, Caliph 110 Persian, language 6, 164, 195–198, 201–203, Muza@farids 325 207, 209–210 Persianization 205–206, 208, 210 Naiman 16–17, 116 Petis de la Croix 145 Nafthat al-maṣdūr 200 Petrushevsky, I. P. 79 Najd 247, 263–264, 267–268 Phoenix Mosque 7, 237–239, 241, 244, 253, Nakhchivān 60 257, 259, 262–263, 267 Nakhchivānī, Muḥammad b. Hindūshāh Pingyang 18 59 Poland 23 naqīb al-ashrāf 148 Pulad (Bolad) Chingsang 244 Nāṣir al-Dīn Fīrūz Shāh 201 Pythagoras 140 Nāṣir Khusraw 199 al-Nāṣir Muḥammad b. Qalawun 45 Qābūsnāma 199 Nawrūz b. ʿAṭā Malik Juwaynī 72 qāḍī 68, 70 Nayan 111 Qāḍī Niẓām al-Dīn Iṣfahānī 207 Nestorian church 237–238 Qala Diza 287 Nestorians 207 Qalan 59, 226–227, 230 Nicea (Nicaea) 26–27, 29–32 Qalāwūn, al-Manṣūr 287, 297 Nishapur 162–163, 326 al-Qalqashandī 297 Nīshapūrī, Ẓāhir al-Dīn 201 Qanā’ī, poet 209 Niẓām al-Dīn al-Ṭayyārī 146 Qaraqorum 42, 86, 166, 266 Niẓām al-Mulk 199 Qara Khitai 17, 22, 120, 216 1:: -3456

Qara-Qoyunlu 45 Saḥib Shams al-Dīn Iṣfahanī 208 Qarluq 17, 19 shaḥna 221 Qing dynasty 229 Sahriyya 291 qiṣāṣ 116 Salghurid dynasty 96, 205 Qipchaq territories 21, 23, 313 Salmas 94 Quanzhou, see Zaytun Samanid dynasty 197–198 qubchur 59, 93 n. 78, 226–228 230 Samarqandī, Dawlatshāh 65 Qubilai Khan 23, 111, 166, 241, 250, 255, 266, al-Samarqandī, Muḥammad b. ʿAlī 210 310, 328 San Yuan Lou 240, 272 Qulān Khatun 82 Sarbadars 326 quriltai 14, 166, 218, 225, 311, 322 Sati Beg 319, 324 Quṭb al-Dīn Shīrāzī 144, 145 Sāvajī, Qāḍī ʻĪsá 321 Qutlughkhanid dynasty 96 Sāveh 60 Qutui Khatun 91, 94 Sayyid ‘Ajall, Shams al-Dīn 262, 265 Secret History of the Mongols 16, 18, 80, 82, Rāḥat al-ṣudūr wa āyāt al-surūr 203, 214 109–110, 116, 168, 217 Raḥba 47, 147 sedentarisation 41 Raphael, Kate 39 Seljuqids, see Seljuqs Rashīd al-Dīn 41–42, 44, 58, 61, 68, 70, 72, 81, Seljuqs 1–4, 13, 198, 201, 211 88–92, 94–85, 97–98, 109, 112, 115, 117, Seljuqs of Rūm 23–25, 27–29, 31, 41, 47, 58, 120–121, 123, 135, 143, 149, 156, 165, 167, 75, 114, 202–204, 209–210, 286, 289, 181, 206–207, 210, 227, 244, 250, 333 300, 314 Rachewiltz, Igor de 16, 168 Sevinch 30, 310 Rāwandī, Muḥammad ibn ʿAlī 203 Shāh ‘Abbās 313 Rawanduz 287, 291 Shāhnāma 6, 167, 173, 175–177, 185–188, 195, Red Turbans, Buddhist sect 327 198, 209, 318, 322 Reid, Robert 38 Shāh Tahmāsp 172 Riccoldo di Monte Croce 123–124 Shahrazūr 282–283, 289, 296, 299 al-Risāla al-Shara-.yya -.’l-nisab al-taʾlī-.yya Shajar al-Durr 319 144 shaman 88, 107, 120, 123–124 Romanus Diogenes 201 Shamanism 107–108, 125 Rubruck, William of 26, 90 Shams al-Dīn Aḥmad al-Suhrawardī 146 Rukn al-Dīn Qilij Arslān, Sultan of Rūm Shangdu (Xanadu) 327–328 27, 29–30 shāṭir 181–183 Rukn al-Dīn Sulaymān Qilij Arslān Shaykh Kamāl al-Dīn Badī‘ al-Zamān Abū (r. 1196–1204) 202, 203 Faḍl Hubaysh al-Ghaznawī (d. 1183) 200 Rūmī, Jalāl al-Dīn 209, 315 Shaykh-‘Uways, Sultan 325 Rus’ 20–21, 23, 232 shiʿism 69 Russia 96, 226, 314 Shi‘ites 141 Rypka, J. 155, 329 Shiraz 173, 176, 179, 205, 246 Shirvan 218 Saʿdī 68, 73, 205 Sikandar Lodi 206 Safarnāma 199, 201 Silk Road 88, 287 Safavids 45, 69, 72, 301, 313 Sindbadnāma 210 Sa@farid dynasty 197–198, Simnānī, family 247 Ṣafī al-Dīn Ardabīlī 69, 316 Sinicization 328 Ṣafī al-Dīn Urmawī 6, 133–134, 144–145 Sinjar 286, 288–289 Ṣaḥib Dīwān 73, 122 Sinop 315 -3456 1:J

Sivas 25, 29, 60–61, 202 Tarmashirin 329 Siwnikʿ 218–219 Tatian 207 Sıȳ āsatnāma 200 Tāzīgūī, Shams al-Dīn 63 Smith, John M., Jr. 38, 41, 280, 319 Tegüder, see Aḥmad Tegüder Smpad 26 Temüge 16 Song Dynasty 249, 327 Temüjin 80 Songjiang 248, 255 Temür Khan (Emperor Chengzong) 122 Sönitei 20 Temür, Tamerlane 320, 325, 327 Sorghaghtani Beki 88–89 Tengri 107 Split (Spalato) 19 al-Ti*+īsi 200 Spuler, Bertold 38, 67–68, 71 Timurids 64, 172, 301 Sübedei 22 Titley, Nora M. 175 su=>sm 188, 201 Toghon Temür, Yuan emperor 322, 328 Süge 110–111, 114, 320 Tokat 27, 60 Suhrawardī, Shahāb al-Dīn Yahya 146, 200 Tolui 1, 22, 64, 80, 88, 244, 248, 258, 319, 323 Sulaymān Khan 320, 326 Tolun Khatun 94 Sultaniyya 147, 166, 288, 311, 316 Toqtemür, Emperor 270 Sultan Aḥmad, Jalayirid 172 Toqto, Yuan chancellor 330 Sultan-Duvin 319 Toquchar 16–17, 20 Sultanate of Rūm 13, 23, 31 Töregene Khatun 87, 89, 112, 310 sunni 64, 67–69, 268 transformation 2–5, 7, 40, 79, 90, 97, 99–100, sunnism 69–71, 264 125, 155, 172, 176, 277, 322 Świętosławski, Witold 38 Transcaucasia 23–24 Syria 19, 23, 32–34, 38, 40–41, 44, 46–47, 91, Transoxania 39, 205 94, 207, 227, 247, 263–264, 267–268, Trebizond 26, 31–32, 229 277–278, 282–283, 286–287, 289–290, Tsunami strategy 4, 13–14, 22–23, 31–32 297, 300, 319 Tughluq Temür 330 Syriac language 40, 207 Tümälün Khatun 82 tuman 219, 227, 231, 244 Ṭabaqāt-i Nāṣirī 201 Turkomans 45, 289 al-Ṭabarī, Muḥammad b. Jarīr 197 Turkey 201 Tabriz 66, 70, 72, 95, 144, 160, 181, 205, 225, Turkmen 26–27, 30 248, 288–289, 310–311, 315–316, 325 Turki=>cation 45, 48 Taghar 226, 228–230 Turks 19, 27, 64, 110, 195, 202, 205, 208, 225 Taghay (Togha) Temür 319, 326 Turnbull, S. R. 38 Taimaz 22 Tus 69–70, 209 Tamgha 59–60, 117, 229–230 Ṭūsī, Aḥmad b. Muḥammad 209 Tamma 4, 13, 16–21, 23, 32–34 Ṭūsī, Naṣīr al-Dīn 69, 91, 144–145, 206–207 Tammachi 17–19, 21, 23–24, 28, 32, 34 Tangut 81 ‘Ubayd-i Zākānī 317, 325, 330 Tao Zongyi 240, 257 Uighur 17, 19, 120, 206, 222–223, 225, 238, Tārīkh al-wuzarā 200 252, 256, 258 Tārīkh-i Firishta 206 Ujan 288, 311 Tārīkh-i guzīda 207 ʿulamā 67–69, 124 Tārīkh-i Jahāngushā 6, 207 ulugh bitikchi 57, 225 Tārīkh-i Masʿūdī 199 al-ʿUmarī, Shihāb al-Dīn 133, 135 Tārīkh-i Sīstān 199 urbanization 166 Tārīkh-i Waṣṣāf 207 Uriyangqai 23 1:N -3456

Urmia, lake 280–282, 287 yam system 86, 89 Urmiya 134 Yāqūt al-Mustaʿṣimī 145 Uru’ut 16 Yāqūt al-Ḥamawī 282 yarghuchi 220 Van, lake 281, 294, 296–297 yarligh 220, 225 Vengeance 112, 120, 123–124, 161–162 Yassur 31 vernacular language 197, 205 Yazd 63, 228, 325 vilayet 218–220, 231 Yazdī, Majd al-Mulk 59–60, 63 vizier, vizierate 5, 55–59, 61–63, 65, 67–68, Yelü Chucai 87, 89, 122 70–72, 74–75, 90, 94, 97, 115, 117, 123, 135, Yidu 18 148, 180, 200, 203, 207–209, 320, 323 Yuan dynasty 79, 107, 121, 228, 249, 252, 314 Volga-Ural Region 20 Yuan Shi 18

Wallachia 23 Zagros Mountains 287, 296, 324 Wādī al-Khaznadār 45 Zarzāriyya 284 Waqf (pl. awqāf) 63, 68, 71–73 Zayn al-akhbār 199 Waṣṣāf, ʿAbd Allāh b. Faḍl Allāh 43, 45, 113, Zayn al-Milla ʿAbd al-Rashīd 199 139–140, 145, 207–208, 244–245 Zaytun 252, 256, 259–262, 266 Wazi, pl. Wazis 240, 272 Zeki Velidi Togan 61 West Asia, see Middle East Zengids 286 Whaldron, A. N. 309 Zhang Shichen 327 White Lotus, Buddhist sect 327 Zhenting 18 Zhu Yuanzhang, Ming emperor 327 Xi-Xia Kingdom 81 Zoroastrian communities 196